Mapping Hypertext Analysis, Linkage, and Display of Knowledge for the Next Generation of On-Line Text and Graphics Robert E. Horn Reviews and Comments About Mapping Hypertext Mapping Hypertext by Robert E. Horn is a tour de force in several respects. First, it is an amazing example of "graphic language," the use of complex graphics as an inherent part of the communication. In fact, this book is as close as a paper-based document can be to hyper-graphics. It simulates the rich graphical environment of current-day and future object-oriented computer workstations. Second, Horn's book is a desktop publishing miracle. Created entirely on the Macintosh computer with MacDraw II, it is a remarkable display of textual and graphic information, including hundreds of icons, complex drawings, tables, and structured text. The author deserves some kind of award for design and development of the document itself, independent of its content. Finally, as a contribution to the literature, Mapping Hypertext is a unique and seminal work, covering the history and conceptual underpinnings of hypertext, suggesting applications and design principles capable of stimulating hypertext and hypermedia design for years to come, and providing hooks to the Information Mapping® method, a systematic approach that promises to take much of the guess-work out of hypertext development. Mapping Hypertext was a finalist for the 1991 Outstanding instructional Communication Award, and for good reason. What fewer people may know is that Horn won the NSPI Outstanding Research Award in 1976 for the work which led to his trademarked Information Mapping method. - Carl Binder, Performance and Instruction, October 1991 "This book will change the way people think about their current information and the hypertext revolution." — Ken Blanchard, co-author of the best selling The One Minute Manager "Bob Horn suggests an antidote for the problem of disorientation that often comes with navigating through hypertext..." —Patricia Seybold, founder of Patricia Seybold's Office Computing Group and sponsor of the Seybold Office Computing Conferences. Quotes are from Paradigm Shift: Patricia Seybold's Guide to the Information Revolution. Boy, do I wish we'd had this book when we were designing the CD-ROM Electronic Whole Earth Catalog. With so much textual (and graphic) information now available in electronic formats, how can we develop, organize, display and interlink any collection of such information in a useful manner? This book is the most thorough survey of solutions thus far. And it is organized in a highly visual hypertext-like format which effectively illustrates many of the principles being discussed. An absolutely first-rate work. —Keith Jordan, Whole Earth Review, Summer 1991 With both words and illustrations, Bob Horn has come to the rescue. In Mapping Hypertext, Horn has provided a methodology for dealing with the crucial task of organizing hypertext in ways that will make complex bodies of knowledge readily available to a user. -Robert F. Mager, Mager Associates, Inc., author of the Criterion-Referenced Instruction course and many books on training I am convinced that the future of man's knowledge production and utilization will be deeply emplanted in the structure, conventions and methods associated with the descendants of today's hypertext. Bob Horn has produced a notable step toward that end. --Doug Engelbart, Bootstrap Project, Stanford University ; first person to implement hypertext on a computer system Mapping Hypertext is a thoughtful and provocative overview of both hypertext and Information Mapping, full of useful advice and interesting bits of history. It is a must read for anyone concerned about how computers can become effective tools for human communication. --Paul Saffo, The Institute for the Future; columnist, Personal Computing Mapping Hypertext The Analysis, Organization, and Display of Knowledge for the Next Generation of On-Line Text and Graphics Mapping Hypertext The Analysis, Organization, and Display of Knowledge for the Next Generation of On-Line Text and Graphics by Robert E. Horn A Publication of The Lexington Institute For Andrea and Jenny © 1989. by Robert E. Horn All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Creation of derivative works, such as information abstracts, unless agreed to in writing by the copyright owner, is forbidden. Copying of authorized derivative works by any means, including electronic recordings, is prohibited. ® Information Mapping, InfoMap and IMAP are registered trademarks of Information Mapping, Inc., 303 Wyman Street, Waltham, MA. 02154 (617) 890-7003. Apple, HyperCard, and Macintosh are registered trademarks of Apple Computer, Inc. GUIDE and GUIDEReader are trademarks of Owl International, Inc. IBM is a registered trademark of IBM. Xanadu is a registered trademark of Autodesk Inc. To Order This Book Call (617) 890-7003 or Write The Lexington Institute 80 Marrett Road Lexington, MA 02173 Library of Congress Catalog Card Number 90-060088 ISBN 0-9625565-0-5 1989 R. E. Horn • ' f > Introduction This Book: An Overview of Two Relatively New Frameworks for Thinking For a number of years I have been working with two relatively new frameworks for thinking about and presenting information, which show a great deal of promise for improving human communication. They are: Hypertext a form of organizing text in computers that permits the linking of any place in text (or other media) to any other place and the rapid retrieval of information by following trails of these associative links. Information Mapping's Method of Structured Writing a methodology for analyzing, organizing, writing, sequencing, and formatting information to improve communication. It provides a way of describing the structure of subject matters that is very useful throughout the communication process. Since no good description of these two methodologies exists in a single book, I have written chapter-length overviews of them. I also present extensive examples of their applications in business and academia. Summary of the Argument In addition to these overviews, I also make a case for the future importance of the coming together of these two approaches. The argument can be summarized as follows. Hypertext will help knowledge workers to better organize the information they manage and to find it when they need to. Hypertext may even become the basis for a large new public medium of information exchange. There are problems associated with the design, development, use, and implementation of hypertext knowledge bases. Some of these problems co-arise out of the very nature of hypertext and thus may not have completely satisfactory solutions. Other problems will be solved by the creativity and hard work of people in the field. Many of the major issues are resolved by the Information Mapping method. Information Mapping's method is a mature, extensively tested, and widely used (in industry and government) methodology for analyzing, organizing and writing documents. It represents a new approach to thinking about the fundamentals of rhetoric (the study of the principles and rules for written and spoken composition). The method produces measurable improvements in human performance through better communication. Modem argumentation analysis has been on the scene since 1956. When put into the context of the software systems described in this book, it can be regarded as an application of what I call structured hypertext. My claim in this book is that argumentation analysis will contribute to representing and solving some of the problems of analysis of disputes and that it can be used as a major linkage device to other domains of discourse we describe in this book — experimental (the domain of empirical science) and the domain of relatively stable subject matter (that which we place in our textbooks, procedure, policy, training, and documentation manuals). Introduction, continued I then propose that the ideas in Information Mapping's method and argumentation analysis that have proven so useful in other areas could be applied to certain parts of the communication of scientific information used in hypertext software systems. This point is illustrated with a sketch for the redesign of basic scientific reports and abstracts. I then provide prototypes of these documents so the reader can consider their usefulness. Another implicit claim of this book is that Information Mapping's method and hypertext can be used in books like the one you have in your hands. So I have simulated insofar as possible on paper what this book might be like if it were actually being presented on a computer screen of a hypertext system of the near future. Also implicit is the claim that the liberal use of graphics contributes to the effectiveness of communication of our ideas. In fact, in the last chapter I venture a forecast that such visual language will become a major communication methodology increasingly integrated with the words we use every day. (If a picture is worth a thousand words, then the approx. 600 illustrations in this book provide 600,000 extra words, which, if you figure about 300 words per page, is about 2,000 extra pages, which makes the book well worth its price.) I might mention that this book also demonstrates what can be accomplished with inexpensive modern computer graphics. All of it was done on the Macintosh computer with MacDraw II software. Serve Different Readers This book will be of use to many different readers: the general reader who wants to take a look at important developments in communication — present and future managers who are planning to convert some text or graphic data bases to the computer for on-line access the knowledge worker who is planning to design, buy or implement a hypertext system, use the Information Mapping method or argumentation analysis scholars, scientists, teachers, students, writers or consultants who are planning to use hypertext, structured writing, or argumentation. Advice for Reading This Book One of the advantages of the Information Mapping method of structured writing is that it is written to be browsed. Readers are able to look at the headings (and the graphics) and get a broad picture of the content without reading every word. So, I urge you to skip around in this book as much as possible and don't feel you have to read every sentence in order to use this book properly. There are summaries at the beginning of each chapter that give you a quick high level idea of what is in the chapter and review how the chapter fits into the overall book. I have claimed that the combination of hypertext and Information Mapping is a very important way of enabling readers with different backgrounds and different interests to get what they need from the same text. This book provides an opportunity for you to test that hypothesis. Introduction, continued This book is not □ a report of comparison shopping on current hypertext software systems, for they change far too often for books to keep up (so watch your computer magazines for these consumer reports). □ a way of teaching you how to do either hypertext or Information Mapping's method (you can browse both well here, but if you want some minimum amount of competency or fluency, you will need to take a course or two and then practice). □ an analysis of the latest comments on the issues in these burgeoning fields (for this you need to go to the conferences and read your magazines). In this book you will find the more fundamental long term issues. □ an investigation of the specific computer interface issues that arise as a result of hypertext implementations on specific systems (for this, you need to become acquainted with the specific constraints of each hardware and software system). Robert E. Horn Lexington, Massachusetts November 1989 Digitized by the Internet Archive in 2019 https://archive.org/details/mappinghypertextOOOOhorn Contents Part 1. Hypertext - Hypermedia, New Opportunities □ Introduction to Hypertext and Hypermedia, 1 0 Current Issues with Hypertext, 37 Part 2. The Method of Information Mapping 3 Introduction to Information Mapping's Method of Structured Writing, 75 u Navigating Structured Hypertrails, 125 U Resolving Some Hypertext Problems, 149 Part 3. Some Applications of Structured Hypertext u Relatively Stable Discourse: Documentation and Training, 16' l\ Disputed Discourse: Argumentation Analysis, 185 Experimental Discourse: Part 4. So What? What Next? Scientific Information, 209 □ Mapping Future InfoSpace: Summary and Trends, 233 Appendix 0 Some Historical Notes, 251 Note: On these facing pages are two different views of the book -- detailed table of contents on this page and corresponding visual summary on the right. PART 1. Hypertext - Hypermedia, New Opportunities - Detailed Table of Contents Chapter 1, Introduction to Hypertext and Hypermedia The Basics of Hypertext Overview of This Chapter 2 A Brief History of Hypertext 2 Sources of Hypertext 4 What is Hypertext? 6 Essence of Hypertext: Links, Nodes and Buttons 8 Features of Hypertext 10 Some Early Applications of Hypertext 12 Public and Private Hypertext Systems 14 Navigation Through Information Space Metaphor 16 Hypermedia What is Hypermedia? 1 8 Hypermedia Application: New Product Marketing 20 Case Study: Hypermedia for Shakespeare 22 Current Concepts of Hypertext Next — More Detailed Dimensions of Hypertext 24 What Are Some Types of Links? 26 Dimensions of Hypertext Systems 28 Paper Metaphors for Hypertext Linkages 30 Computer Metaphors for Hypertext 32 Ways of Using Hypertext 34 Chapter 2. Current Issues with Hypertext Overview of This Chapter 38 System Design Issues What Shall Be the Size and Contents of Nodes? 40 What Shall the Links Connect? 42 Characteristics of Links 44 Where to Put How Many Buttons of Which Kind? 46 User Issues Inadequate (and Missing) Reading Cues 48 Special Provisions for Common Reader Behavior 50 Branching Difficulties of Serialist Readers 52 Poor Metacognition Skills May Limit Training Uses 54 Lost in Hyperspace 56 Overchoice & Cognitive Overload 58 Chaos for Titles for Documents and Their Parts 60 Development Issues Labor-Intensive Creation 62 Labor-Intensive Database Maintenance 64 Additional Skills Needed for Hypertext Authoring 66 Relating to Other On-Line Documents and Training 68 Managing the Creation of Different Versions 70 Rewrite Text or Convert to On-Line Text "As Is"? 72 Summary and Plan of the Book PART 1. Hypertext ■ Hypermedia, New Opportunities Chapter 1. Introduction to Hypertext and Hypermedia Hypertext lill l Ills mmmrn m III :lli!S Hypertext is an important development in the storage and retrieval of text and graphics on computers... . It enables us to link... nodes ...different places in text called "nodes"... ...with other nodes.. ...through the use of buttons... butto: ..any place in a text... ...to any other place... ...any time... si m ssss ...and get there fast... Chapter 2. Current Issues with Hypertext Issues in Hypertext Among the major issues of hypertext implementation and use are: What shall the nodes contain and the links represent? Lost in Hyperspace Overchoice and Cognitive Overload Some Issues Serialist & Holist Readers Labor-Intensive Maintenance Multiple Skills and New Rhetoric Needed ...these issues and others are addressed by the Information Mapping method described in the next few chapters... Note: On these facing pages are two different views of the book — detailed table of contents on this page and corresponding visual summary on the right. PART 2. The Method of Information Mapping - Detailed Table of contents Chapter 3. Introduction to information Mapping's Method of Structured Writing Overview of This Chapter 76 Some Problems Addressed with the Method 78 What is the Information Mapping Approach? 80 The Problem of Human Short Term Memory 82 The Method What are Information Blocks? 84 How Four Principles Constrain Block Construction 86 Examples of Information Blocks 88 Block Replaces Traditional Definition of Paragraph 90 Developing Guidelines and Standards for Blocks 92 Hierarchy of Chunking and Labeling Principle 94 Examples of Maps Displayed on Paper 96 Change Traditional Document Structure 98 Managing Completeness by Key Block and Topic 100 Improves Efficiency Throughout Analysis Process 102 Discourse Domains 104 Brief Discourse Analysis (Stable Subjects) 106 Information Type and Block Type Analysis 108 What are the Information Types? 110 Key Blocks Provide "Completeness Templates" 1 12 Results Greater Ability to Specify Rule Domains 114 Top Down and Bottom Up Analysis 116 Reference-Based Training 1 1 8 The Mapping Metaphor: Subject Matter Structure 120 Meeting the Criteria for Better Communication 122 Chapter 4. Navigating Structured Hypertrails Overview of This Chapter 126 Prerequisite Hypertrails 128 Classification Hypertrails 130 Chronological Hypertrails 132 Geographic Hypertrails 1 34 Project Hypertrails 136 Structure Hypertrails 138 Decision Hypertrails 140 Definition Hypertrails 142 Example Hypertrails 144 Hypertrail Webs into Linearized Sequences 146 Chapter 5. Resolving Some Hypertext Problems Overview of This Chapter 150 At the Nodes, Blocks and Maps Structure Hypertext 152 Clustering Documents From Different Domains 154 Addressing Lost in Hyperspace and Overload 156 Addressing the Major Reading Cues Problem 158 Addressing Creation and Maintenance Issues 160 Addressing Group Analysis and Writing Issues 162 Some Navigational Options 164 Summary and Plan of the Book Chapter 3. Introduction to Information Mapping’s Method of Structured Writing Information Mapping's method is a mature technology for analyzing, organizing, writing, sequencing, and formatting information. It introduces a novel way of modularizing information and provides a modem alternative to traditional concepts of approaching the writing of many types of documents in business, technology and academia. We show how four principles generate information blocks and maps. Then we show how the contribution of other principles and research enables us to build a systematic process for subject matter and task analysis. Finally we describe tools for the organization and presentation of learning and reference materials to users. The ability to analyze the structure of subject matters gives the method its particular usefulness. Chapter 4. Navigating Structured Hypertralls The larger structure of the subject matter can be shown to users by sequences or weblike structures of links called hypertrails. These hypertrails can also be used to specify various linear paths through the networks of hypertext. The nodes of hypertrails are information blocks and maps. Chapter 5. Some rnmmm Using Information Mapping's structured writing method and hypertrails provides systematic guidance in the analysis and specification of nodes and links and in mapping the larger picture. Thus it enables us to resolve some of the major issues with hypertext raised in Chapter 2. . Note: On these facing pages are two different views of the book -- detailed table of contents on this page and corresponding visual summary on the right. PART 3. Some Applications of Structured Hypertext - Detailed Table of contents Chapter 6. Relatively Stable Discourse: Documentation and Training Overview of This Chapter 168 Operations and Technical Manuals 170 Personnel Manuals and Policy Manuals 172 Product Knowledge Case Study Introduction to Product Knowledge Case Study 174 Product Knowledge Case Study: Main Menu 176 Product Knowledge Case Study: Specifications 178 Product Knowledge Case Study: Control Panel 180 Access by Task-Driven Procedures 182 Chapter 7. Disputed Discourse: Argumentation Analysis Overview of This Chapter 1 86 Brief History of Argumentation Analysis 186 Claims 188 Grounds (Data) 189 Warrants 190 Backing 191 Rebuttal 192 Qualifiers 193 Applications of Argumentation Analysis to Well and Poorly Structured Problems Argumentation Analysis for Four Principles 194 Argumentation Analysis for Three More Principles 196 Useful in Representing Ill-Structured Problems 198 Case Study of a Poorly-Structured Problem 200 Case Study Brings Together Opposing Viewpoints 202 Comparing Ill-Structured and "Tame" Problems 204 Conclusions: Argumentation and Hypertext 206 Chapter a. Experimental Discourse: Scientific Information Overview of This Chapter 210 The Science Information System 2 1 2 Application: Scientific Reports and Articles 214 Case Study: Scientific Abstracts Applying Information Block Analysis to Abstracts 216 Miller: Short Term Memory Limits and Chunking 218 Simon's Tests Show Chunking Size to be 5 to 7 220 Hartley and Trueman: Headings Aid Retrieval 222 Shaffer: Information Mapping's Methodology 224 Reid and Wright: Superiority of Visual Structuring 226 Other Potential Benefits of Hypertext in Science Information Facilitate Identifying Problems at Science Frontiers 228 Linked Comments Will Highlight Deficiencies 230 Summary and Plan of the Book Chapter 6. Relatively Stable Discourse: Documentation and Training We present examples of the use of the Information Mapping method to analyze critical and complex documents in business and industrial situations, such as personnel and policy manuals, operations and training manuals, and product knowledge databases. Chapter 7. Disputed Discourse: Argumentation Analysis 7 Argumentation analysis is a method of representing Grounds [Typically Data) Claims \ Rebuttals (Counter - Claims in a hypertext network \ \ sentences that are the components of a reasoning process about a dispute. mm Si Si Relations Warrants 1 Backing QualifierS[ Among the examples presented in this chapter is an argumentation analysis of seven fundamental principles of Information Mapping methodology. We show how argumentation analysis provides "trlmkage between relatively stable subjects and discbtit^e about scientific experiments. Chapter 8. Experimental Discourse: Scientific Information We apply the principles of Information Mapping's method to a provisional redesign of scientific abstracts to improve their communication possibilities in a hypertext environment. Abstract Abstract Abstract Abstract Among the examples provided of abstracts are some of the research papers and theoretical articles that support the principles and practices of Information Mapping's analysis and writing methodology. Note: On these facing pages are two different views of the book — detailed table of contents on this page and corresponding visual summary on the right. PART 4. So What? What Next? - Detailed Table of Contents Chapter 9. Mapping Future InfoSpace: Summary and Trends Summary of the Argument 234 Trend: Integrate Communication and Computing 236 Navigating Through Whole Subject Matters 238 Navigating Along Hypertrails 240 Looking from Multiple Points of View 242 Virtual Reality — A New Tool 244 Travelling in Large Visual Landscapes 246 Heading into Future Information Landscapes... 248 Appendix A. Some Historical Notes Bush: Inventor of the Concept of Hypertext 252 Engelbart's Augment: First Operational Hypertext 254 Engelbart: Edison of the Personal Computer 256 Nelson: Name-Giver of the Word "Hypertext” " 258 Van Dam and Brown: First University Instruction 260 Zog Group at Camegie-Mellon: Menu Interfaces 262 Negroponte and Bolt: Spatial Dataland 264 Brown and Guide: Hypertext for PC and Macintosh 266 Sculley: Vision of the Knowledge Navigator 268 Atkinson: First Commercial Hypertext "Hit" 270 Notes 273 References 277 Acknowledgments 281 Index 283 Chapter 9. Mapping Future InfoSpace: Summary and Trends 'UNI Appendix A Some Historical Notes 9 The Future Finally we combine all of these approaches to take a look at mapping and navigation of hypertext and hypermedia in the future.. From Vannevar Bush in 1945 who first suggested the possibility of hypertext, to Ted Nelson who named it hypertext and hypermedia, through Doug Engelbart who built the first system, through the other visionaries and builders who have given us the possibilities of hypertext. A Little Bit of History Where did hypertext come from? Who were the people who came up with the key ideas? Who were the people who first implemented the technology? When did all this take place? A Index References Notes Chapter 1. Introduction to Hypertext and Hypermedia The Basics of Hypertext Overview of This Chapter 2 A Brief History of Hypertext 2 Sources of Hypertext 4 What is Hypertext? 6 Essence of Hypertext: Links, Nodes and Buttons 8 Features of Hypertext 10 Some Early Applications of Hypertext 12 Public and Private Hypertext Systems 14 Navigation Through Information Space Metaphor 16 Hypermedia What is Hypermedia? 18 Hypermedia Application: New Product Marketing 20 Case Study: Hypermedia for Shakespeare 22 Current Concepts of Hypertext Next -- More Detailed Dimensions of Hypertext 24 What Are Some Types of Links? 26 Dimensions of Hypertext Systems 28 Paper Metaphors for Hypertext Linkages 30 Computer Metaphors for Hypertext 32 Ways of Using Hypertext 34 Chapter 1 Introduction to Hypertext and Hypermedia Essence of hypertext software Paper and computer metaphors for hypertext Definition of hypertext In this Chapter ...an overview of hypertext and hypermedia Definition of hypermedia Dimensions and characteristics of hypertext systems Paper metaphors for hypertext Computer metaphors for hypertext Navigation metaphors | - — — On « Jiflnl f 1 Overview of This Chapter Introduction Hypertext is a novel way of constructing computer-supported, non-linear writing. It can also be thought of as a new way of developing text databases. It provides many new capabilities for linking all sorts of electronic media. In this chapter we provide a detailed description of this important new computer tool. Treatment: Textbook Chapter on Hypertext and Hypermedia Since the concepts of hypertext and hypermedia have only emerged in the last 20 years and since they are still not well known to most people, our approach in this chapter is to present a relatively detailed introduction to this field. Approach: Simulate the Look and Feel of Hypertext in Print We all assimilate information in different ways and, hence, we often need different paths through a document in order to get the most out of it. Hypertext gets some of its power from the almost instantaneous ability to jump from one place in the text to another. In order to provide you with the best idea possible on the printed page of what hypertext looks and feels like, we have simulated computer screens and linkages in this book. Commentary: Major Argument of this Chapter This chapter will give you a broad picture of what hypertext is all about. As this book is written (in early 1989) there are only a few hypertext systems in operation. We are in a stage of technology development just before the rush. Hypertext will emerge as a major way of organizing text and other media despite the obstacles that I spell out in Chapter 2. This obviously is not a startling hypothesis for those who have been following recent developments in the field. But for others, to contemplate having the flexibility and the choices of hypertext routinely available on their computers can be an exhilarating thought. (REH) A Brief History of Hypertext 1935 1940 1945 1950 1955 1960 Chapter 1. Introduction to Hypertext and Hypermedia Author's Commentary: Personal Judgments for example icons ...simulate what in hypertext are called "buttons," places in the text on a computer screen you would be able to click your mouse, depress a key, or otherwise indicate that you want to go to the place indicated by the button. I faced a number of unique problems in writing this book. First, the goal of giving an introduction to hypertext, argumentation analysis, and Information Mapping implied a goal of "just the facts." But to give perspective on the issues, I needed to write more personal or more judgmental remarks like this. So I decided to simulate how I might "overlay" my comments on an otherwise somewhat more neutral approach to the subject. In this way, you'll be able to see how such commentary might work in a hypertext system where you and other people can add notes to the original text. We use a distinctive shade of border and sign the notes. In some cases, you will see that I have included quotes from other writers as if they were comments on my text. In these cases the author will be so noted and references can be found in the back of the book. My comments will be signed: (REH). * Forecast (1967) Multiple Ways to Read Hypertext Ted Nelson, who coined the word hypertext, has suggested: "To be useful, the hypertext medium requires some ... variety in the ways and sequences that the same material can be connected together or explained. Indeed, these texts may be made big and. diverse enough for study by specialist and beginner alike; with many entrances, tracks, and specially oriented meanders. Thus the user’s previous background and level of knowledge could be taken beneficially into account by the author-editors." (Nelson, 1967) First Commercial Hypertext Systems Dataland and other hypermedia experiments at MIT Other early experimental systems Brown University and Owl's Guide and Apple's HyperCard Nicholas Negroponte Carnegie Mellon The word "hypertext" coined and the vision expanded Theodor Nelson Sources of Hypertext Introduction: Human Thought is Multi-Dimensional; But Conventional Text is Linear Human thought has many aspects, many dimensions. Our memories are associative; we connect many different things in quite unpredictable and idiosyncratic ways. And often our thinking appears as a kind of free association or even free juxtaposition. Other memory and thought is quite structured and hierarchical. We reason from goals to means. From causes to consequences. We plan rationally. We build large organizations on highly structured principles. The question this book addresses is: How do we best represent all this information and all these connections to ourselves and to others for communication, learning, and problem solving? People inevitably bring forth multiple viewpoints. @ VO 00 vO 73 tn si o 3 Commentary: Memory is Associative and It Fades j "Our ineptitude in getting at the record is largely caused by the artificiality of systems of indexing.... The human mind does not work that way. It operates by association. With one item in its grasp, it snaps instantly to the next that is suggested by the association of thoughts, in accordance with some intricate web of trails carried by the cells of the brain. It has other characteristics, of course; trails that are not frequently followed are prone to fade, items are not fully permanent, memory is transitory. Yet the speed of action, the intricacy of trails, the detail of mental pictures, is awe-inspiring beyond all else in nature..." Vannevar Bush, 1945 Chapter 7. Introduction to Hypertext and Hypermedia ...our familiar school subjects are many- sided conglomerates... Subject matters have many layers, dimensions, and approaches, some contrary, some consistent. ...SSSf: iiiffiS- . . y/'f" ' ' . ¥:*:*>: mm /mm v ...but we've tried to put everything we know into linear text... Textbooks, journal articles, encyclopedias- all have taken on similar linear text format, although the thoughts they convey [ are not necessarily linear. . :• • • ■? S&ffi ^-1 Putting computer-based links into text is what hypertext addresses. 5 What is Hypertext ? — — - button symbol In this book, buttons are simulated with the A and these little "see page" icons. In computer software, you would be able to click your mouse or otherwise indicate that you want to go to the place indicated by the button. y Douglas C. Engelbart V> i — / v/ / hi per tekst Definition A hy • per* text 1. the ability to link any place in text stored in a computer with any other place in the same or different texts, that permits rapid access through buttons . A and other tools across non-linear pathways. First suggested by Vannevar Bush A ip 1945, terrp Coined by Theodor Nelson Ajn 1965. First j,jOT|piemented between 1962 and 19*75 by DpugHas Engelbart ./A 2. a subset of the bFpader . concept ” hypermedia." A. First Implementor of a hypertext system For Details Chapter 1. Introduction to Hypertext and Hypermedia "Hypertext is both an author's tool and a reader's medium. With hypertext software, authors or groups of authors will be able to link information together, annotate existing texts, and create footnotes that allow readers to see either bibliographic data or the body of the referenced text. Readers will be able to browse through linked, cross-referenced, annotated, footnoted texts in an orderly, but nonsequential manner: an automated encyclopedia of sorts ..." (Brown University, 1985) Essence of Hypertext: Links, Nodes and Buttons Definition Links connect nodes in the hypertext software by computer-supported relationships that permit rapid, easy movement across the network of nodes. Examples of Some \ WnjKJ ' „ Kinds of Links There are a great variety of links in hypertext systems. Here are some types of links: • the internal document organization (e.g., connect two pieces of text in same document) • the external organization (e.g., connection of one document to other documents) • annotation via pop-up windows • table of contents to document • index to document • local table of contents to a part of a document. Three Important Features of Hypertext Software The essence of hypertext software is 1. a network of nodes A„, which may be text and/or graphics . 2. software methodology that facilitates building of and access to nodes via links. A 3. interface tools that facilitate the creation of arbitrary linkageyln the text with buttons A"- and (frequently) the easy manipulation of chunks of text--'and media through windows. Chapter 1 . Introduction to Hypertext and Hypermedia Buttons Nodes Definition Buttons are specific locations in the hypertext or on other media that permit the user to jump along a link to another node, usually with the click of a mouse or the pressing of a key. In one sense, buttons are the user-visible manifestations | of links. Example W77| button Definition 1. Nodes are the part of the hypertext network where the text or other media are located. 2. For some software implementations, a node contains one idea or one sentence; for other implementations the node may be a whole document as long as a book or chapter. At present, the node is not a well defined concept except in certain very structured contexts. One node may include composite nodes where, for example, a node is a subnode. Nodes may have different display metaphors, such as cards, pages, windows. [mm I I I I | I I T I 1 1 I 1 . . . . ** . . Features of Hypertext Introduction Links, nodes and buttons are essential characteristics of hypertext. But there are other characteristics that give hypertext its distinctive flavor, and distinguish it from just another text database. Three Features of Hypertext ill .. : : v: tim! sipssi Focus on Rapid Browsing mmmm mim Focus on Non-Linear Discourse Description One of the values that hypertext theorists have always espoused is the ability of the user to conduct rapid browsing and navigation. The user must be able to get around in the hypertext quickly and easily. This means that the user interface must enable the user to navigate across the links in the system, usually with a couple of clicks of a mouse or a couple of key strokes. \\ I / ^ - CLICK - Example It is assumed that users will do something else when they get to the place they've jumped, i.e., to use the information for... Comment @ — Little is made of these other functions in the hypertext S literature. Rather, in that literature, rapid browsing and navigation are often valued for their own sake, m x o 3 10 Description Some hypertext software authors value highly the ability to have information partially or non-linearly structured. Here these distinctive definitional qualities enter the structure of the contents as well as the link-node structure of the software. They celebrate the non-linear structure of the text and non-symmetrical organization. Example of This Thinking Not only is it possible that a story could have several endings, but it is claimed or implied that it is somehow better if it does have several endings. Authors Who Have Tried Non-Sequential Writing These three authors have written novels with different paths for reading. Nabakov Pale Fire Comment The claim is made that, because human thought has the characteristic of being able to associate anything with anything else, somehow it is implied that all or most text should have "the freedom to associate." Chapter 1 . Introduction to Hypertext and Hypermedia Focus on Two Making Options ea BJ Read-Only Hypertext Description Some hypertext systems, particularly those used for policies, procedures, training and documentation in organizations, permit users to read and follow links without being able to write or make links. Example author supplied link only Comment Personalized Hypertext Description Many hypertext systems can be personalized. For example, the user can modify the hypertext database by making links and adding text, graphics, and numbers to the documents and windows. Thus the document becomes increasingly the personal document of the user, rather than one solely produced by the orginal author. Different levels of security can be supplied so that access can be permitted or denied to different users. The Original Document Personalized Document (with User-Created Links ) You can't change the company's personnel manual unless you are the Vice President for Human Resources or somebody designated by the V.P. Similarly, while feedback may be welcome, you may want it to be private and addressed to the V. P. rather than posted on the pages of the manual. I Comment Little has been done to help readers figure out how they might want to organize such links, which could rapidly represent a large personal information management problem. Some Early Applications of Hypertext Introduction To what has hypertext been applied so far? What kinds of experimental and operational situations have been tried out? On these pages we note a number of the early explorations of hypertext. Bibliographic references in the back of this book will lead you to more detail. A number of projects have been launched to put all of the documentation of given systems on electronic storage. There are several large projects to provide information systems for all of the maintenance manuals, costing, and troubleshooting of all of the models of a given make. Maintenance and Repair of Business and Government Applications @ s 00 VO TO m x o 3 On-Line Procedure and Operations Manuals 'Zj For an example 12 Chapter- I . Introduction to Hypertext and Hypermedia On-Line help messages are now routine. Connecting them with hypertext links to more detail will improve documentation and training. Olllllllllll Introduction to Computers and Computer-Based Training On-Line Software Documentation S* . . *S Science, Technology and Academic Applications Large Databases of Text or Graphics Help Scholars and Writers Organize Notes Encyclopedias M Si A couple of encyclopedia-size works have been organized with hypertext links and stored on compact disk storage. A current project at Harvard is linking- all of the text on Ancient Greece. ; Course Material for Introductory Courses One of the early research programs focused on the metaphor of organizing the electronic equivalent of note cards. 13 Public and Private Hypertext Systems Introduction Hypertext concepts may be applied at different scales or sizes of systems as well as in public or private access situations. Two Major Usage Contexts for Hypertext Systems □ cb [7] Private (or in-house) Systems Definition Private (or in-house) hypertext systems are systems in which the access to usage is limited to a designated group of users (usually within a single organization). Example One Company Reference System on Products and Services Example Two Company Reference System on Research & Development Sales and marketing could provide hypertext facilities that give sales people and customer service people access to hypertext databases on all the products and services of a :j Example Three Company Reference System on Personnel Policies Personnel handbooks and other human resource information ® can be built that enable all branches in a particular § organization to access the hypertext that provides the information supervisors and administrators need to make ^0 « • decisions and take action. A large research and development laboratory could provide a hypertext facility for the documents pertaining to various projects being worked on at the laboratory. Chapter 1. Introduction to Hypertext and Hypermedia imm % mmm s • . Definition Public access hypertext systems provide facilities for scholars, specialists, practitioners, and students in a particular field to share a common hypertext knowledge base system. Public Access Hypertext Systems Example A scientific field or sub-field could establish a hypertext knowledge base to provide up-to-date scientific information in hypertext form. This would provide significant improvements to the current science information systems. I 1 | Research Knowledge Base in BioGenetic Computing liSIlSi! Commentary: Linkage The researcher using the public and.-th,e""’ private R & D knowledge bases A' could not make linkages across these two systems directly. Presumably, some kind of software system could be made to link the researcher's personal links across the two systems. (REH) : ' : ! Chapter 8 contains a more extended treatment of scientific information systems 15 Navigation Through Information Space Metaphor Introduction Ted Nelson's gigantic vision of all the world's literature in one massive hypertext system has inspired science fiction writers to imagine what it might be like to maneuver around in all the world's electronic literature. The result is a metaphor based on the science fiction concept of "navigation" through hyperspace, that is, traversing through time warps that enable space travelers to cover vast amounts of time-space in rapid maneuvers. Metaphor Getting to the information you want resembles steering from the bridge of a spaceship. Various subject matters appear at different distances, e.g., (1) galaxies at a great distance, (2) huge abstract constructions at middle distance, and (3) vast integrated displays close up. Chapter 1 . Introduction to Hypertext and Hypermedia Commentary: Navigation as Metaphor I sometimes wonder if we have chosen the best metaphor for moving around in a hypertext database. A lot of people can't read maps and get lost taking shortcuts through big city neighborhoods. But we are very much creatures that move through 3-dimensional space and those metaphors are likely to feel as comfortable as any. Hyperspace is yet to be developed as a metaphor. (REH) 17 1989 R. E. Horn What is Hypermedia? Definition hy •per#me*di*a me dea 1. an extension of the idea of hypertext that incorporates other components such as video, illustrations, diagrams, voice and animation, and computer graphics. Typically an author creates computer-supported links between text, graphs, diagrams, photographs, video, music, film and other media. The author may be the user or learner. Linkages ...any place in any of these ,,:l media can be linked to any other place... This book will address more fully the issues of hypertext than it does hypermedia, although we will present brief introductory examples of hypermedia in the next few pages. Interactive Media Examples Commercial Example (real estate) Interactive electronic media have a continuing convergence with hypertext. Academic i-; Example *- (Shakespeare) !•!•!•!•!•!•!•! v!v!!vTv?-!"v!-!v,!v!-!vlv!v,l-,v'l-'-1 >x<<*x::*::::x:&vx^x^ ?x:x¥#^^ XX-X-XXrXX-XX-X-XvXvXvX-XvX-XvX Hypermedia Chapter 1 . Introduction to Hypertext and Hypermedia 19 Hypermedia Application: New Product Marketing Introduction Hypermedia is finding a fair number of early application areas in the marketing of new products. Convention exhibitions and point-of-sale marketing are important areas where hypermedia concepts have been applied. Below we present a schematic of a new product release concept with hypermedia. Potential clients use hypermedia software on computers to make choices about what they want to see on the laser disk and to get detailed text from the computer's memory. VO OO VO 73 tfl X o 3 Clients can see still photos of different parts of the building and its spaces. They can also see animations and movies of movement through the building to visualize their space. All can be stored on the same disk and controlled by hypermedia links from the computer. 20 Chapter 1. Introduction to Hypertext and Hypermedia Clients view video film, animation, and slides on video disk. In this example, commercial real estate can be explored for leasing before or while it is being built. | Commentary and music can | accompany both the computer and laser disk displays. Any segment is available in any order on the disk and in the computer. Case Study: Hypermedia for Shakespeare Introduction Hypermedia is also beginning to make a mark in education. On these pages we present a brief schematic overview of what hypermedia course materials might be like. The subject is Shakespeare's plays and their historical and critical context. Begin by watching the play scene by scene on videodisc... 22 Chapter 1 . Introduction to Hypertext and Hypermedia Jump to the sc^ne in the text and fojhjiw along... Skip ahead or back on screen or in the text... ) Jump to related scenes in other plays... / Read commentary linked to specific places in the play... index Jump to interpretations of Shakespeare in modern dress... original text / Jujrpp to other places in other plays... commentary Jump to the history of Elizabethan theater... 23 1989 R. E. Horn Next -- More Detailed Dimensions of Hypertext Chapter 2. Some Issues with Hypertext In chapter 2, we look at four different kinds of issues having to do with hypertext: □ System Design Issues • What size shall the nodes be and what shall they contain and why? • What shall the links connect? • Characteristics of links • Where to put how many buttons of which kind? Development Issues • Labor-intensive creation • Labor-intensive database maintenance • Additional skills needed for hypertext authoring • Relating hypertext to other on-line documentation and training 3 User Issues Normal reading cues may be inadequate in hypertext and these are often missing Poor metacognition skills may hinder usefulness of hypertext in business training Serialist readers may have considerable difficulty when forced to branch Lost in hyperspace Overchoice and cognitive overload Chaos in titles for documents and their parts |~J] Implementation Issues Managing the creation of different versions Rewrite or convert to on-line text "as-is" 24 Chapter 1 . Introduction to Hypertext and Hypermedia This Chapter So Far We have looked at • what hypertext is • what its main features are • some examples of applications • what hypermedia is • and some typical examples. These are the basics. Now you have a choice point. J Author's Commentary: Choose Your Detail Level Go to Chapter 2 and start looking at the major issues concerning this new medium. Choice Point Continue reading this chapter which now will go into a lot more detail about some of the characteristics and dimensions of hypertext systems. Other Options Of course, you always have the option to browse, skim, and dip into whatever you choose. (REH) 1 3 1 pi : The Rest of This Chapter The rest of this chapter focuses on the following topics. |j— ] What are Some Types of Links? 2 Dimensions of Hypertext Systems go an Computer Metaphors for Hypertext |~^~[ Paper Metaphors for Hypertext Linkages |~5~| Ways of Using Hypertext Bro^ing_ Briefing Learning . Grazing Referencing Analysis & Writing 25 © 1989 R. E. Horn What Are Some Types of Links? Introduction We now turn our attention to a more detailed level of examination of hypertext. We revisit the concept of links to observe some of the kinds of links that have evolved in some of the systems developed thus far. m §§ Three Types of Hypertext Links r- LJ m illllil in System-Supplied Links x-x'x'x:x •X;X;X;: X; gillll Definition System-supplied links are those that are automatically supplied by the hypertext software at all times or are created by the software according to predetermined criteria. Examples Here are some examples of the system-supplied links: [71 command and control pathways through the textual knowledge base (e.g. , next buttons) back next o e> I home menu [~2~| automatically created tables of contents |~3~j automatic tracking of what user has seen and showing this track upon command Here's where you've bee OHEZinillinED [~4~| automatically created user profiles and suggested sequences Based on your recent viewing , you may be interested in this -.t. Chapter 1. Introduction to Hypertext and Hypermedia User-Created Links Definition User-supplied linkages are those links created by the users (using system-supplied facilities) to link text for their own purposes. Examples Some examples of user-created linkages include: [T] detours and shortcuts ....skip to . ,1 2 notes, commentary and reminders 1 Note |~3~j analogical linkages 0 4 new text Add this text here |~5~1 links to other knowledge bases Link my manuscript to this part of this knowledge base This is like . Note: These links are described from the viewpoint of the reader. They may be implemented in the software in quite different fashions. 3 Author-Created Links Definition Authors. We refer to authors as those who create text and hypertext for others to read to distinguish them from end-users. Author-created linkages are links which authors insert in text that are "pre-prepared" for the user to traverse. Usually they are links that authors anticipate users will need frequently. Examples Some of the major types of author-created links include: |~1~| links to prerequisite knowledge hierarchical links based on classification of information [ 3 | chronological links 1987 J ub / seepagej> 1 32 27 Dimensions of Hypertext Systems Introduction software. Different systems have been built (and imagined) from small, single user authoring systems for a single task, to large, multi-purpose networks with a large number of users. Three Dimensions of Hypertext Systems ^ Information Sharing Definition Scope of coverage refers to the number of users supported by a given hypertext system. Single user Small work group Large network of users □ u i i n □ UkX n L. i A K 1 L _ _ _ 1 L _ _ _ * m3 ,\ / The concept of hypertext covers a fair amount of territory — any way you can segment text and any way you can link text is of interest to hypertext theorists. So, clearly we need some ways of characterizing the dimensions of hypertext and its Chapter 1 . Introduction to Hypertext and Hypermedia :;>> xx;v;: sc? .iiiiiiiiiiliil Definition Modes are different ways of interacting with hypertext. 1. Using Modes These include browsing, - learning, referencing, etc. a J3LJriQbTn cj r-,0 a a □ Authoring Mode 0 Editing Mode \ IN. [sete pagg ►34 29 3J Applications ^L. 0 General purpose hypertext 0 All the world's literature Theodor Nelson A ..ha§„ suggested that eventually a . hypertext system should be built to connect all of the world's literature, so that you could jump from one place to literally any other place. Definition Scope of software refers to the generality of the hypertext and its applicability to a very limited or larger group of tasks. 0One specific task (e.g., argumentation analysis) Argumentation analysis is one way of arranging the sentences to support a 1'fitionale or analysis of a disputed point. Systems wn<\ph handle argumentation analysis are quite task specific!), Many of the current hypertext implementations handle a variety of text and graphics with fixed or variable sized chunking. Paper Metaphors for Hypertext Linkages Introduction Paper metaphors identify the ideas used on paper that are implemented in different hypertext systems. Usually a specific hypertext software system has a dominant metaphor and may have facilities for other metaphors. Eight Metaphorical Sources of . Hypertext-like Linkages in Paper :vx$iv8::: |~i~| Library Card Catalogs Example 2J Footnotes Explanation Footnotes are, perhaps, one of the original link types in paper text, used to identify sources of information and to amplify or comment on topics in the main text. Example 5 Commentaries Definition Commentaries are extended discussions of one text in another text, usually in parallel fashion. Explanation @ Indexes Indexes are a kind of linkage system because they enable the reader to go to specific places in the text. original text index Chapter 1 . Introduction to Hypertext and Hypermedia Commentary Ever since the invention of writing, people have been trying to get their associative links onto paper. Over the ages, we've used quite a few devices to convey these associations. Note on this page how these associations have been integrated into the link and node structure of hypertext quite easily. (REH) l~3~1 Cross-References Example [4] Sticky Notes Explanation Those little yellow stick-ons can be found all over with new information added to original document. Example [7| Quotes Explanation GO Anthologies Definition Anthologies are books which are made up of chapters and long quotes from other books. They thus provide a kind of metaphor for hypertext of gathering parts of other books together for some purpose. r 31 Computer Metaphors for Hypertext Description Computer metaphors have been used in conceptualizing several specific hypertext systems. Usually, but not always, a given hypertext software system has a dominant metaphor and may have facilities for other metaphors. Eight Computer Metaphors Used Up as Sources for Hypertext Links [5555555& fi~| Linked Note Cards Example — 1 ■ - ■ '/sj Number of Cards Per Screen --Two Types • one card per screen • multiple cards per screen Size of Card-Two Types • fixed size • user scalable sizes 2J Popup Notes Example @ 3 00 VO pa m X o 3 Semantic Nets Definition A semantic net is a knowledge representation method consisting of a network of nodes (which represent concepts) and links (which represent relationships between the nodes). Example is example of . CAT Dino {' . ( . '' / : 1 . ' / has is - a (|i'’ y _ mammal Henry / kittens is example of is - a Tabby chordata 1~6~| Branching Stories Definition Branching stories are a genre of fiction, normally presented on computers, that permit the reader to make key choices along the way and hence influence the outcome, plot, or action of the story. They have influenced the design of some hypertext systems. 32 Chapter 1 . Introduction to Hypertext and Hypermedia Commentary One of the intriguing things about hypertext software is how it brings together a number of ideas that have been around quite a while and integrates them into something that has quite different dynamic properties. (REH) 3[ Linked Screens or Windows PH Stretch Text (Outline) Before pressing button After pressing button [71 Relationai Databases 8 Simulations Definition Relational databases are databases constructed in suchaway that if any 2 files have a common field, then the data base can link these files to other files also sharing common fields, and keep track of these links so that users can find what they are looking for along many different paths. Example ■ . 1 • . . . . 1 1 Definition Discrete state simulations have the capacity to branch in many different directions and lead to many different outcomes. This property appears in some of the hypertext simulations. Example 33 Ways of Using Hypertext Introduction Too often, discussions about hypertext fail to identify the purpose the user has for looking at the text. While user surveys are scarce, we can identify a small number of ways that users typically read text. Different ways of using hypertext will call forth the need for different features and produce different problems, as we shall see in the next chapter. Ll Browsing Mode Description Browsing is skimming, usually quickly, over large amounts of text to find regions of interest. We distinguish it from referencing, which is the search for very specific information or places known to exist. Characteristics: Speed / Highlights Only Implication Browsers need a variety of tools to enable them to see different views of subject matters and documents. They need large scale maps of the subject matter as well as very well defined routes. Modes OD Training Mode Introduction Training covers a fairly wide variety of activities with quite different implications for the design of hypertext. Description Training suggests that there are organizational goals for specific accomplishment levels within a specific time. Implication Hypertext can be useful in training, but the most important variables will continue to be the setting of goals, the use of practice exercises, and the provision of feedback. Comment Jan Walker, one of the readers of an early version of this manuscript, wrote: "Thanks to unthinking use, the word 'browse' is too vague to be useful anymore. Anyone trying to read the litrature needs to be alerted to this. The original English-language implications of the term have been drowned in drek. When most implementors say 'browse' they mean nothing more than 'use.' " Grazing as a Kind of Browsing Some observers have also described a kind of information grazing where the user simply meanders munching on whatever is nearby. Some think of it as a sort of slow-motion browsing. (REH) H] Help Mode Description Help mode is a special kind of referencing mode that provides (usually) brief amounts of information to aid users of computers to accomplish their current tasks. Implication Hypertext can be useful for linking users to glossaries and to other forms including computer-based training. 34 Chapter 1. Introduction to Hypertext and Hypermedia Commentary: Too Much Focus on Browsing? It appears that sometimes hypertext discussions take place as if browsing were the only important way that human beings interact with text. We probably do browse more in this age of information overload. But hypertext must be able to support many ways of learning, not only browsing. (REH) Briefing (Presentation) Mode Description Briefing (or Presentation) Mode is generally not a user directed mode. It is, of course, user chosen in the sense that the user chooses to ask for a briefing. The system presents an overview and summary of the subject, project, organization, sequence of events. Characteristics: Speed / Highlights Only Implication Building good briefings into each hypertext region is an important value in good hypertext production. A whole separate set of guidelines and standards applies to the design of presentation briefings. HI Learning and Analysis Mode Description The learning and analysis mode is that mode in which users are solving problems or making decisions. They are typically involved in the collection, analysis, and rearranging of information. Examples Scholars gathering and rearranging notes for a book or paper; scientists and engineers working on problems. Characteristics Users in this mode often have time and inclination to explore alternatives that may be provided by hypertext. Implication Hypertext may be of considerable use in this mode, especially in the exploration of different viewpoints, alternatives, and data. Referencing Mode Description In the referencing mode, users are focused on trying to retrieve information from the hypertext database in response to a specific question or a specific need. Perhaps they have seen it before. Or they have been told that it exists. Or, knowing the status of the knowledge base, they conjecture that it exists. Characteristics Highly focused, fact-oriented. Users are usually looking for a specific place and stop referencing when that information is found. Implication This mode relies on more traditional methods of information retrieval, such as indexing, tables of contents, or state-of-the-art reviews, than it does on hypertext linkages. Chapter 2. Current Issues with Hypertext Overview of This Chapter 38 System Design Issues What Shall Be the Size and Contents of Nodes? 40 What Shall the Links Connect? 42 Characteristics of Links 44 Where to Put How Many Buttons of Which Kind? 46 User Issues Inadequate (and Missing) Reading Cues 48 Special Provisions for Common Reader Behavior 50 Branching Difficulties of Serialist Readers 52 Poor Metacognition Skills May Limit Training Uses 54 Lost in Hyperspace 56 Overchoice & Cognitive Overload 58 Chaos for Titles for Documents and Their Parts 60 Development Issues Labor-Intensive Creation 62 Labor-Intensive Database Maintenance 64 Additional Skills Needed for Hypertext Authoring 66 Relating to Other On-Line Documents and Training 68 Managing the Creation of Different Versions 70 Rewrite Text or Convert to On-Line Text "As Is"? 72 36 1989 R. E. Horn Chapter 2 Current Issues with Hypertext Overchoice and Cognitive Overload Serialist & Holist Readers Labor-Intensive Creation In this Chapter ...an overview of hypertext issues... Labor-Intensive Maintenance Lost in Hyperspace Design & Authoring Issues j Poor : Metacognition Skills May Limit Training Use Multiple Skills and New Rhetoric Needed m What shall the nodes contain and the links represent? Ml i" 1 4==J v> If 37 Overview of This Chapter Introduction Hypertext certainly provides features and advantages that many of us can use in our work and study. But these advantages do not come without their corresponding problems. It is important to sort the big problems from the smaller ones. Many of the smaller ones will go away as hypertext software designers and authors work on upgrades and new systems. But there are some more fundamental problems, those that have to do with the limitations of human information processing capacity. These must be addressed in a more basic and more structured fashion. In this chapter we survey the major issues that have been raised in connection with various hypertext implementations. These two pages present an overview of the rest of the chapter. Section System Design Issues In this section we point out that the software designer must solve certain very basic issues about text construction: • What size shall the nodes be and what shall they contain and why? • What shall the links connect? • Characteristics of links • Where to put how many buttons of which kind? @ 00 73 m x o 3 Commentary All hypertext authors will have to pay attention to these issues as well. We will have to understand our text better and understand better how our users will use on-line hypertext and hypermedia. It's not only a system designer's issue. Recent history is littered with software systems that have been left in the dust because the designers made the wrong decisions about the fundamentals. (REH) 38 Chapter 2. Current Issues With Hypertext We focus in this section on some of the major problems that have been raised by researchers and users: • Normal reading cues may be inadequate in hypertext and these are often missing • Poor metacognition skills may hinder usefulness of hypertext in business training • Serialist readers may have considerable difficulty when forced to branch • Readers may become lost in hyperspace • Users may experience overchoice and cognitive overload • Chaos in titles for documents and their parts rnmmmmmmmmmmm Commentary 1 This sections says in another way as well as "know thyself." (REH) "buyer beware! m In this section we look at what it takes to do the intellectual work that makes hypermedia and hypertext possible and the knotty problems of organizations or groups of people trying to use hypertext systems: • Labor-intensive creation • Labor-intensive database maintenance • Additional skills needed for hypertext authoring • Relating hypertext with other on-line documentation and training • Managing the creation of different versions • Rewrite or convert to on-line text "as is" The real world uses accounting and you have to pay attention to your costs as well as the benefits you think you will derive. Another real world issue: Once you start making revisions on multiple drafts of a document you get a spaghetti-like tangle of links. "A plate of spaghetti looks the same from every angle," someone has said in the context of hypertext. (REHj ■ h •5 What Shall Be the Size and Contents of Nodes? Introduction Many of the problems that we will discuss in this chapter are related to more general questions about nodes, links, and buttons. In this opening discussion we will examine nodes. Major Questions The major questions about nodes include: • What shall the nodes contain? • What principles shall we use to determine contents of nodes? • On what basis should size decisions be made? • Is there any systematic way of determining "natural" divisions of a subject matter that will help us? Definition: Granularity Granularity refers to the amount of the information contained in a node relative to the large size of the information. Loosely, it is the "amount of information in a node." Seven Different "Sizes” of Nodes (as they appear to a user) c55£^55& 1 One sentence 2 Text of arbitrary 3 Index card 4 size (e.g., an article) size of the screen Description Description Description Description Some experimental hypertext systems enable authors and users to use single sentences as nodes. Example In Chapter 7, we present a description of argumentation analysis which is based on sentence node size. Some hypertext systems have been built that enable authors and users to insert nodes at any point in text. The typical size of the node follows current practice in document composition, i.e., chapters, articles, paragraphs, etc. Some hypertext systems have used the fixed index card as the size of the node. Some hypertext systems have used the entire screen as the definition of the size of the node. Commentary In many cases the fixed card size should be avoided. Far better is some form of scrolling that permits chunk sizes that can more fully cover sufficiently large thought patterns. (REH) VO 00 pa tfl X o 3 40 Chapter 2. Current Issues With Hypertext Commentary While I call these system design issues, these are issues that software purchasers have to pay attention to as well, because they need to get the maximum flexibility and usefulness from their systems. (REH) Commentary Point of View Each piece of hypertext software has to incorporate some way of indicating nodes - independent of other aspects such as the interface. Since the focus of this book is on how the hypertext appears to the user, we present a classification of these views below. Description Some hypertext systems permit the screen to scroll a sizable amount, sometimes up to chapter size, yet allow the insertion of buttons for links at any place in the text. Description Some hypertext systems have flexibly sized chunks that correspond to the individual author's view of the "size" of the subject matter. This permits virtually any size (or shape) of display. 5 Scroll of any length 6 Variable size 7 Variably sized, precisely and flexibly chunked Description It is possible to devise a methodology to segment subject matter into variably sized chunks that also provide the user and the author with a systematic way of chunking the subject matter. It would then be useful to have a hypertext system that "understood" this chunking methodology and would provide such facilities for both author and user. Commentary: Our Bias Precise yet variably sized blocks is the bias of this book, as the reader will see in Chapter 3 when we explain Information Mapping's method of structured writing. (REH) 41 What Shall the Links Connect? Introduction Hypertext software creates links between the nodes rapidly and conveniently. But if the author and user can link anything to anything, what is to stop them from "over-linking" or "mis-linking" to the ultimate confusion of everyone? The question about "which links?" is a significant one. Major Questions The major questions about links are: • Which kinds of links to implement? • How many links should one use? • How can we implement different hyperlinkage networks of the same node? • How shall the links be represented? (This question is an interface issue which we take up under the Four Kinds of Links p (from the user’s standpoint) [fTfi lTQ Description Organizational links implement hierarchical tree linkages within the hypertext network. Terms are parent, child, sibling. Organizational links include tables of contents and other such hierarchical structuring. Example Description Keyword links are links created by the system and permit the users to find the location of specific words in the text. They aid in search for strings of information. Example index Chapter 2. Current Issues With Hypertext i— ill 3 Referential links Description Referential links connect points or regions of text and are non-hierarchical. In the terminology of some in the field, referential links go from link source to link destination. Link source is also called a link point or a link icon. Destination is also called a link region. Example 4 Cluster links ; : ;< ■■ inp-:; ■ v ; - Description Cluster links are links that enable a user to organize a group of shorter pieces of information as they proceed through the project. They may resemble file boxes, groups of notes and the like. Example 43 Characteristics of Links Introduction In hypertext systems, the designer of the system must consider a number of issues, such as, whether you can get back to where you jumped off from, the kinds of links permitted, the value of particular links, and the size of the node to which the links are made. Two Fundamental Issues With Links r-- □ mm mm ■ llsllllil^lllilllillilllll Directionality of Links •• ’ :• • : • y: • . : . '• • : • ••:•••• '*■ Two Types of Link Directionality : [~A~| One-way Directionality B Two-way Directionality Definition A one-way link will take the user in one direction after pushing a button and will leave the user there. Users will not be able to return to the original button from which they came. Definition A two-way link enables users to go to another place and return. Example A button which goes to another place in the text but does not return. Example A button which has a comment on it and then goes back to A major interface design issue is to show which kind of button you have. Category Filter B Voting Filter Definition In this type of filter, the user would pick a category from a list provided and the software would follow links associated with that category. Example Definition In this type of proposed filter, the system keeps track of user choices and asks for an evaluation after the user has used a link to establish over time some kind of priority for linking. VO 00 VO 50 m x o 3 Chronological Classification Structure Geographic Most traveled links click here A 44 Chapter 2. Current Issues With Hypertext Multiplicity of Links Problem Statement Some systems which have been proposed would conceivably have a large number of links that different people have attached to a single point in a given text. It is possible that everybody would want to comment and link. For users, this would produce a bewildering array of choices. Proposed Solution One proposed solution is to have a dynamic filtering system selectively displaying links. Definition A filter is a user controlled software feature that selects and displays some, but not all, of the links at a given node. Four Types of Proposed Filters ^ Expert Filter Menu of Links Definition Authorities on a particular subject would examine a given area of the hypertext and provide a series of links that would give the reader a path to follow. Definition Menu links would provide users with a dynamically constructed display of links available when a particular button is activated. - Available Links: A Experts (3) A Standard Hypertrails A Glossary 45 Where to Put How Many Buttons of Which Kind? Introduction A third major question area is the button. In principle, you can sprinkle them around anywhere in the text and any place on the rest of the screen. But that raises a whole series of questions described on these pages. User Issue: How to Spot the Button How do you tell where an "activated area" (i.e. , a button) is on a screen, particularly in the graphic areas? Putting in the game of "hide and seek" for the buttons is of some limited use in playful hypertexts, but not in those for serious purposes. Ear Buttons Soon One trend is for you to be able to talk to your computer, especially to give it commands. You’ll be able to push buttons by voice soon. Will they be called "earcons”? Icons so why not Earcons? I could be a button... Maybe I'm a button too... IP' . (Zn van in in ■1 III III a hi ■■■ \ i j ■II 111 lllllpll,, J 1 Ill Help Next Phone Deepej Erase Get MorSj Send 1 Recurse Tracks More Back Fast Calenda Map | Find BringBacfllfip | Signal ML. Mil, SO 00 vj o 73 m x o 3 The button clutter problem is the one of too many buttons scattered aimlessly or packed densely on a screen, forcing users to study their choices on each screen with considerable diligence. Already in some systems we see a confusing array of choices that begin to look like this. 46 Chapter 2. Current Issues With Hypertext The User’s Question: Can I Trust This Button? Interface Designer's Question: How Can I Possibly Know What This User (or Any User) Wants? Where does it lead? What happens when I push it? Where will it take me? What will be the nature of the text when I get there? Will it meet my needs? Will I just get lost in a mish-mash of confusing buttons and screens and links and not be able to get back here? Another wild goose chase... another fishing expedition. Maybe I shouldn't even push this button at all... How can I tell what the user wants? • The screen next in prerequisite order? • The screen(s) previous in prerequisite order? • A definition only? • An example? • Both a definition and example? • Something opposite? • A menu of choices? Jo o O O User Issue: Seductive Buttons Writers work hard at what they do. Quite naturally they want to have their words read by somebody. Designers want people to use all of their system features — often whether the user needs them or not. Sometimes designers go beyond what is necessary to seduce the user to follow a particular path. We need some sort of "truth in button advertising" law. 47 Inadequate (and Missing) Reading Cues Introduction Hypertext — especially unstructured hypertext — may run into difficulties by the very nature of its linking and branching facilities. These remove some of the discourse cues that provide readers — particularly initial learners of a subject — with information essential for making learning efficient and effective. Definition: Discourse Cues Discourse cues are different elements of text that give readers orientation information to guide the process of reading. Discourse Cues That Hypertext Destroys or Disrupts Some of the discourse cues that hypertext may destroy or disrupt are: Hierarchical Text Organization When we read, research suggests that we tend to build an hierarchical framework for ideas. Many discourse cues, such as outlines, patterns of subheadings, words such as "initially, next, finally, firstly," etc. provide cohesiveness and may also provide cues as to where in the author's hierarchy, the reader is. These aid reading and may be unused or meaningless if the reader jumps across links that leave them dangling out on the furthest limb of some hierarchical structure. i Explicit Transitions Overviews, Introductions, Summaries Authors are taught to prepare readers for what is coming up in text by providing overviews and introductions that summarize or preview the content and structure of the document. Traveling across links may cause users to miss these transitional remarks or to fail to find them easily. Most authors alert readers to transitions. Readers expect to have transitions made explicit. These transition words such as "however, moreover, then, when, more importantly, etc." often refer back to previous text structure. When they are absent or when they refer back to someplace readers have not visited (because they arrived here by a link), they experience a jarring sense of disorientation or a feeling of being lost. Text contains signals about local organization such as "There are seven kinds of..." Readers arriving from links in the middle of such text structures must do extra work to take advantage of the cues they provide. It would be very helpful to have an "overview button" or some other kind of structure cues. Overview 48 Chapter 2. Current Issues With Hypertext mmmmmMmmmmmmmmmm Commentary mmmmmmmmmmmmmmmm Any author reading this will begin to draw conclusions about how to write and, perhaps most importantly, where not to put links. Contrast and Similarity Cues Introducing new concepts frequently requires the distinguishing of the current idea from other ideas. Conventional text provides cues to the reader that this is about to take place (or is taking place). Hypertext may leave the reader puzzled as to what these cues refer to if the reader has just arrived via some link. One Idea Another Idea How are they similar? How are they different? Pronouns as Cohesiveness Cues Metaphors The extended metaphor running through pages of text can be a useful text organizer. It is of less use if readers arrive in the middle of the text from some other link. 8 Content Schemas One of the ways that conventional text provides cohesiveness is through the use of pronouns which refer back to material that the reader is assumed to have read because linear reading is assumed. Authors of hypertext may not be able to make that assumption. There are different conventional schemas for organizing text (e.g., narratives). Discourse cues such as "once upon a time..." or "she looked deeply into his eyes" tell us that a fairy story or a love story follows. These may be less meaningful in hypertext that is heavily laced with links. 49 Special Provisions for Common Reader Behavior Introduction How do novice readers approach the task of learning from text? Are there important patterns in the behavior of average readers that may affect how hypertext is written? Novice Readers Some research suggests that a significant number of readers who are novices in a subject matter exhibit the following behavior: Novice readers may stop reading too soon. Many who are new to a subject matter may think they understand a subject before they have read all they should. They jump to premature conclusions. Novice readers are often misled by superficial features of the subject matter. Readers new to a field may be unable to tell the important from the less important, either undervaluing or overvaluing particular passages. Nothing important here... It all looks important to me... Novice readers rarely seek non-contiguous information. Novices to a subject matter rarely are observed to go to another page to look at the diagram when the text says "See diagram number ..." @ VO 00 7) tn X o 3 50 Chapter 2. Current Issues With Hypertext Research on Reading Gives Insights on How Average Readers Behave Recent research on reading suggests these generalizations about how many readers go about the task of reading: When tested later, readers tend to remember the information in the higher levels of the hierarchies better than lower level information. Readers depend on repetition of key words. Readers tend to use the words that are repeated in successive sentences to build their meanings. | I I - Then, catalysts.... So, catalysts 51 Branching Difficulties of Serialist Readers Introduction Research by Pask and Scott has shown some interesting results on how different kinds of people learn. They find two basic learning strategies which they designate serialist and holist. They find that serialists and holists are quite different in their characteristics and in their approach to text. Pask and Scott found that about 50% of readers fall into each of the categories of holist and serialist. Serialist Learners Definition Serialist learners proceed through learning tasks starting from the beginning and taking each task in turn. They will fulfill all prerequisites necessary at one level in order before starting the next level. How a Serialist Reads Serialists almost always start a book on page one of chapter one and siart reading at the upper left hand comer and proceed sentence by sentence. They often will not go to the next sentence until they have fully understood the sentence they are currently reading. 1....2....3....4....5....6....7.... Holist Learners Definition Holist learners like to understand the big picture before getting to the details. So they may try hard exercises before they are prepared to solve them. They skip around in learning, trying out what interests them, making their own survey and connections or they may read the author provided overviews and then decide where to dip in. vC oc vO m x o 3 How a Holist Reads Holists almost always open a book in the middle and jump around, scanning, forming impressions. ;5.. :6.. j -A. '■ 2 . -3. 52 Chapter 2. Current Issues With Hypertext Path of the Serialist Reader Implications for Hypertext Serialists will have some difficulty with hypertext. They will resent the forced choices. Traveling to new links they will be thrown into confusion and disorientation much more easily than holists. They may not use links nearly as much and as a result may not be able to take full advantage of the system. They will resent the introduction of words and concepts they don't understand. They will not accept large "conceptual maps” of the subject matter because they contain terms they don't understand. They will benefit from the definition links more than holists because they will be able to satisfy their need for certainty about the meaning of unfamiliar terms. M.... ' ' 53 Poor Metacognition Skills May Limit Training Uses Definition: Metacognition Metacognition can be defined as persons' "knowledge about their own cognitive processes and their ability to control these processes by organizing, monitoring, and modifying them as a function of learning outcomes." (Weinstein and Mayer, 1986) In other words, metacognition refers in part to those skills that have been called "study skills" and "learning how to learn." Applying this definition to learning tasks "metacognition refers to an individual’s ability to accurately determine the goal of a given task, apply appropriate strategies to reach the goal, monitor progress towards the goal, and adjust strategies as necessary." (Clark, 1988) Determine Learning Objectives Ability to set realistic learning objectives. Implications It is clear from looking at this model that every learner brings to the learning task a more or less complex set of metacognition behaviors. It is certainly possible to observe a wide range of variability among learners. Learners with poor metacognitive skills are unlikely to be able to make the choices involved in self-instruction that are required by hypertext and are likely to be among the chief sufferers of problems of cognitive overload and "lost in hyperspace." It is possible that hypertext environments specifically structured in particular ways may help learners with poor metacognition skills improve these skills. This, in particular the ability of hypertext to rapidly reward such things as exploration and curiosity and the ability to put together topics of different kinds, may improve metacognition skills. On the other hand, some kinds of hypertexts and interfaces could really interfere with organization and consolidation. Manage Time in Learning Ability to realistically determine how long various tasks will take to learn. Understand Sequence Ability to sequence learning goals effectively. Determine Prerequisites Ability to determine prerequisites required to learn a specific set of task of knowledge items. @ >3 00 sO 7> m x o 3 54 Use Learning Resources Ability to determine when to use the instructor or other experts effectively as a resource. Self-Monitoring Skills Ability to monitor progress towards achieving goals with reasonable accuracy. Chapter 2. Current Issues With Hypertext Kinds of Skills Generally Included in the Concept of Metacognition General Learning Skills Organizational Techniques Ability to determine when, where, and how to put something into long term memory. Ability to read, to take notes, practice, review, take self tests, check feedback, in order to make remembering efficient and effective. Ability to scan rapidly to make decisions about whether to commit something to long term memory. Ability to determine whether outside evaluation is necessary to determine level of accomplishment. Ability to devise and use mnemonics and visualization to memorize. Ability to paraphrase, summarize, and abstract information. T Metacognitive Tactics Ability to organize learning tasks into manageable chunks. Ability to choose which of these tactics to use in specific situations. |S Ability to reorganize new information in integrating it with what is already known. Ability to draw appropriate diagrams to show relationships among concepts, processes and other ideas. Ability to adequately use categorization techniques to break down information and learning tasks into manageable subunits. Memory Consolidation Techniques Ability to insure knowledge is integrated into current understanding of a field of study or inquiry. Ability to add own questions and comments of the material to integrate it with current knowledge. Ability to create more visual or more personal examples of purely abstract or verbal information in order to retain it in long term memory. Ability to use metaphors and analogies to link what the reader knows to the new material as well as to move beyond the new material. Motivational Awareness z Z z 55 Ability to monitor motivation in learning so as to conduct learning under the best circumstances possible. Lost in Hyperspace Introduction In some of the early hypertext systems users frequently reported that they had the problems of knowing • Where am I? • Where have I been? • Where am I going? • What are my options? These together are also known as the "context" or "lost in hyperspace" problem. Hypertext authors and designers often responded with "solutions" that merely compounded the problems. see Notes "Solutions" that haven't solved the fundamental problem Show them all of the connections If the reader is lost, some hypertext systems have been built that show all of the connections to a particular piece of text (card, window, article). 56 Chapter 2. Current Issues With Hypertext "Solutions" that haven't solved the fundamental problem Welcome to the home card. From here you can get anywhere in your hypertext. "Solutions" that haven't solved the fundamental problem When readers are lost, some hypertext systems have facilities that show them the path they have followed or enable them to flip through the screens they've just seen. This works sometimes, but often avoids the fundamental problem of what the structure of the subject matter is. We will now show you where you've been. Go back to where theyVe started If readers are lost, some hypertext systems take them back to where they've started. That works sometimes but wastes a lot of the user’s time. 2 57 Overchoice & Cognitive Overload Introduction In the experimental hypertext systems of the late 1980's, readers frequently report being overwhelmed by too much information and being constantly bombarded by overchoice. They were presented with more links than they needed for their purposes and found themselves wasting time with a great deal of extraneous information. 1 i Overwhelmed by Windows In some windowing systems the reader can pile up many windows until the screen looks like a messy desktop. That tends to look like this.... Chapter 2. Current Issues With Hypertext Overwhelmed by Nodes and Links Some card-based or screen-based systems can only display one small bit of information at a time. Users have great difficulty determining just where they are in the overall context of the topic they are browsing or learning. Many users in many jobs and learning situations can't handle and shouldn't have to handle a lot of linkage choices. Overwhelmed by Endless Scrolls In some large on-line text systems, the dominant metaphor is the scroll, with few markers to tell you where you are or where you are going. It is easy to get lost and it takes considerable mental effort to keep going under these circumstances. r a (=U v- B Chaos in Titles for Documents and TheirParts Introduction One of the major causes of wasted time in scanning documents is the failure of authors to consider user needs when they provide titles for documents and provide subtitles. People like to create cute, idiosyncratic, short, often cryptic names for files. This is the opposite of what is needed for rapid, easy browsing. How will we be able to prevent the building of millions of electronic Towers of Babel? To illustrate the problem, consider two simple tables of contents for a training manual below. Compare this table of contents... with this table of contents. Table of Contents for the Best Little Tool You’ll Ever Use 1 . Starting to Work 2. Getting It Going 3. Making It Work in the Long Run 4. Help 5. Finding Mistakes 6. When Nothing Goes Right 7. Anybody Can Run It 8. Happiness is a Tool Like This r Table of Contents for the Training Manual for the Model 501 KX Network Integrator 1 . Setting Up the Equipment 2. On-site Testing and Initial Operation 3. Putting the 501 KX into the Network 4. Troubleshooting the 501KX 5. What to Do When Local Troubleshooting Does Not Work 6. Preventive Maintenance for the 501KX 7. Operating the Network Integrator 8. Specifications for the 501 KX From which table of contents would it be easier to locate what you are looking for? @ tn Commentary: Multiple Representational Ecologies Chris Dede has warned that hypertext must face and solve the problem of different people using quite different clusters of terms for roughly the same phenomena. He calls this the dilemma of "multiple representational ecologies." I call it naming chaos. Hmmm. (REH) I Commentary: Label Standards |For the reasons spelled out on these pages, guidelines land standards for labeling have been a part of [information Mapping's method from its inception. [They are relevant as much to paper as hypertext. (REH) 60 Chapter 2. Current Issues With Hypertext The major question is how will we get people to feel as obsessive about putting fully informative titles on their documents as they are about finding spelling errors? Compare the Titles in These Two Files File 1 Plan for the Human Resource Handbook Handbook First Draft Employee Task Force Comments Specific Rebuttals to the Task Force Comments Legal Department Comments File 2 Plan First Cut Reactions Betty's Remarks ^Comments on Comments Which is ore informative? The Seven Commandments for Titling Thy Documents and Their Parts 1 . Thou shall name thy documents neither too general nor too specific, neither too long nor too short. 2. Thou shalt use the language we all use unless thy document is only for a technical or special audience. 3. Thou shalt itemize all thy possible readers and consider whether each shall be able to understand thy titles and thine other subtitles and labels. . ■' 4. Thou shalt not surrender to the adolescent urge to ■ label with cuteness or silliness, for the rest of us might suffer at not getting thy joke. 5. Thou shalt not put the rest of us through the hell of vague, mislabeled documents lest thou be thrown into that hell thyself for an eternity. 6. Thou shalt honor thy peers, thy boss and thy subordinates and all the sentient beings that may have to browse thy writing by providing them with well considered titles. 7. Thou shalt use the same words in the table of contents, the titles, on the pages, and in references within the text. 61 Labor-Intensive Creation Introduction Text is created by people sitting at workstations and writing. The intellectual labors of creating hypertext are no different. But in some ways, hypertext adds to the time required for creation. 1 Additional Text Preparation 1 for Nodes Description For hypertext that is prepared in advance for learners and referencers, all of the major textual information has to be developed. This is a labor-intensive project even without any hypertext requirements. Just to write a competent training manual, good report or good technical documentation requires training, methodology, management, and a lot of effort. The requirements for such projects usually limit the effort to "need to know" information. Every jump, comment, or branch in a hypertext database must go somewhere and presumably text must be there if someone jumps or browses to that place. Adding possible hypertext requirements of analyzing additional "nice to know" or "nice to browse" information can add a considerable cost to the information gathering, organization and writing phases of projects. Implication For most projects in business and industry, it is unlikely that much additional writing will be done. Efficiency and "need to know" criteria will prevail. Additional Work to Create Linkages Description Hypertext software has made the creation of buttons and links rather automated, although we would say that it is not "highly automated." At present, many links must be decided upon and created by hand. Implication Adding links has many implications both for the user and for the maintainer of the hypertext network. Primarily, they add cost. The strategy of cost benefits decision-making is yet to 62 Chapter 2. Current Issues With Hypertext itional Quality rol Requirements fiiisttifj llllilllii Description Links and nodes must be correct and sensible, especially if large numbers of people in your organization are going to use them. If the nodes don't give users the information they want and if links don't get them there efficiently, users will stop using the hypertext system. Implication Quality control will become more important than ever in the construction of text. Greater Need for Standards When Groups Author a Document Description When several people in a work group or task force work on preparing a document, editorial tasks increase. The group has to develop agreements that function as standards for style, format, organization, level of detail, and many other criteria. This is rarely easy, not only because of the requirements of a high level of cooperation and planning skills, but also because the intellectual tasks of conceptualizing a large network of writing is difficult. Somebody must understand the whole structure before writing, and the whole team must have a common language and a common understanding of the structure of the project. Implication Hypertext increases the demands upon these skills and will require carefully planned systems of standards that are managed in a disciplined way to provide useful publications. Labor-Intensive Database Maintenance Introduction Not only does hypertext make larger demands in the text development phase, but it also increases many of the time, cost and quality requirements for the maintenance of the database after it is in operation. Bigger Job of Routine Updating Description Information goes out of date. New facts are discovered. Concepts are revised. Classifications are expanded to encompass new specimens. Whole views of subjects are turned on end by new technology and new ideas. All of this takes routine updating of the hypertext knowledge base. Companies specifically select projects that need constant updating for hypertext because updating is so difficult to do with paper. But putting the manual in a computer will not eliminate the cost of updating. As of now, there are only a few software aids to help in this process. Implication Such routine updating of hypertext files will have to be carried on. It will be a labor-intensive job. The more branches and links that are connected to a node, the more labor it will require to keep the knowledge base current. Additional Editing for Quality of Updates and Comments Description Most print-oriented information sharing networks have institutionalized such jobs as newsletter editor that help the members of the network maintain some semblance of order and quality to the information they receive. Those networks that do not have this, exhibit the worst qualities of bulletin boards on many college campuses and corporate hallways where layer upon layer of old lecture notices compete with important announcements. Implication For personal networks, mild chaos will prevail. For institutional networks, analogues from paper-based systems will be there. We will have editors, but their jobs will be more difficult because of the difficulties of overchoice. A few researchers have suggested that the job of network hygiene or network maintenance be automated or that some kind of "automated filters" be developed to help people navigate a crowded network of dubious quality. Chapter 2. Current Issues With Hypertext Additional Pruning and on Remarks 3 Filtering Links Hypertext Description With the possibility of more links and nodes come the problems of rampant growth. One can imagine metaphors of the growth of weeds in untended lawns or of the jungle reclaiming the carefully cleared villages. V . Every link will demand an answer to the question: "Shall we leave it in or take it out?" We will need to understand the value of each link (i.e., make an assessment of its costs and benefits). Implication Remark on all comments M|/// r Comment 5 on Remark 3 on Comment 24 Additional (and Difficult) Indexing Issues Description Indexing has taken large strides with the development of automatic indexing software. We are now able to get around quite usefully — but by no means easily without professional help — in large bibliographic databases such as Lexis, Nexis, and Dialog. And the indexing of large textual documents with full text searches is a routine if not really widespread facility. The indexer's difficulty is the creative communicator's delight. You can describe the same event or thing with many different words. You can even create new terms for the same old things. But how do you find your way back to terms you were working with last year and have forgotten? The cost-benefits of links will become a minor new branch of economics, with the focus on the cost of intellectual labor in creation, maintenance, and use of links. Many have pointed out that time is becoming one of our most valuable resources. This implies that economists will have to examine the opportunity costs of actually following many branches. The cost of traveling in hyperspace is partially the cost of not being somewhere interesting. Hyperspace traveling is not free. In terms of the high cost of professional labor, we may see demands for superhighways. Toll booths will certainly follow. Implication We will have to build automatic indexers that can incorporate new documents as they are created and integrate these into larger search strategies that are somehow constrained so that the search space doesn't become too large. This will not be an easy task if hypertext databases grow as rapidly as it now appears. Additional Skills Needed for Hypertext Authoring Introduction "Authoring" in hypertext is different from ordinary writing. Developing hypertext and hypermedia will require considerable additional skills. These pages suggest some of those skills. 1 Knowledge Base 2 1 Management Skills Ad Hypertext Rhetoric & Analysis Skills Description Knowledge base management skills include a knowledge of the theories included in this book about the organization of knowledge and a practical working knowledge of how hypertext systems work. Description Hypertext rhetoric and analysis skills include a working fluency in developing structured analysis and writing. It also involves understanding how to prepare special parts of hypertext knowledge bases that help readers connect widely separated and perhaps loosely connected meanings. List of Skills and Knowledge Some of the skills needed, in varying degrees, are: List of Skills and Knowledge Some of the skills needed are: • Planning databases • User analysis • Instructional design (in some applications) • Hardware and software knowledge • Network knowledge • Knowledge of database construction. ® VO oo vO r° tn X o 3 • Awareness of the limits that hypertext places on writing done in large chunks • Understanding how to integrate effective visuals with text • Understanding of most of the skills taught in the Information Mapping® seminars A . • Knowledge of how to create and use hypertrail structures A . see Notes 66 Chapter 2. Current Issues With Hypertext Interface Design & Human Factors Skills Graphics Skills Description Hypertext implementations will provide great flexibility to the user to create and modify human interfaces. Already we have seen problems arising from the lack of knowledge about human factors in interface design. List of Skills and Knowledge Some of the skills needed are: • Ability to use results of human factors research that influence effective communication to design effective screens • Ability to apply the same research to the design of links and larger subject matter organizations. Description Systems will have provisions for highly graphic communication. Users will begin to expect more than just text. The writer or team that produces hypermedia will need to have considerable graphic skills. List of Skills and Knowledge Some of the skills needed are: • Knowledge of aesthetic guidelines for use of color • Recommendations for effective use of typefaces • Ability to integrate graphics with text • Ability to create simple visuals and specify more complex ones. 67 Relating to Other On-Line Documents and Training Introduction Who will use the knowledge base and for what purposes? That is the basic question organizations planning the conversion of large amounts of paper-based text to on-line storage have to ask. Not all software meets every need. How does hypertext relate to these different types of software products? On these pages we present a description of the four major kinds of on-line text and describe some of the problems and possibilities that hypertext will provide for each. Definition Definition Computer-based training software is software that is designed to give learners specific performance situations and exercises in which they can practice the new skills and knowledge presented in the course using the computer. The computer also provides feedback on the outcome of their answers to the exercises. Full text search of computer-stored documents is software that examines every single word of a document and prepares an index either prior to the search or as part of the search. The result of using such systems is that the computer provides a list of all the places where specific words or groups of words are found in the text. The user may then examine one or more of the specific places in the text where the word appears. Advantages The major advantage of this kind of search is its completeness. Users who must find every appearance of a term, find this facility essential. Another major advantage is that you can transfer existing paper documentation to on-line search without extensive rework, important in some applications (such as legal cases) where you cannot justify extensive rewrite (or in paper evidence in legal cases where no rewrite is needed). Disadvantages Full text search also calls for relatively sophisticated search strategy skills. The disadvantage of full text search is that users may receive, very often, a larger list of places to look than they may want. Moreover, users often don't know all of the words to search for and can never feel assured that they've found everything. Relation to Hypertext Full text search facilities would be important in hypertext environments but are no substitute for the linking of hypertext. It is the thesis of this book that structured hypertext will be the most important form of hypertext and that, for particular purposes, full text search will be important. Indexes of most documents will continue to be essential research devices. Advantages An advantage to computer-based training systems is that they provide carefully designed lessons that insure that if the learners do the exercises they will be able to use and apply the new skills and knowledge rather than simply "know about” them. The computer also tracks the success of the student, provides feedback, and flexibly inserts branching for more practice or for review. Disadvantages Many current computer-based learning systems do not provide for rapid browsing for better prepared students. Relation to Hypertext Computer-based training software will benefit from, hypertext links, particulady'for glossaries of specific terms that the learner may npCknow or may have forgotten. In the future, hypertrails a, such as those described in this book, will be inserted in the computer-based training courses for more flexible learner-controlled exploration of the knowledge base. Also, on-line hypertext systems will give CBT learners a reference manual that can be used with the system they're learning about, which is one of the major problems with current CBT implementations. 68 Chapter 2. Current Issues With Hypertext Definition On-line help messages are components of an increasing number of software packages which provide users with instantaneous access to information that will help them in doing whatever task the software has been written for. Advantage Search Large Databases With Keywords or "Controlled Vocabulary" Definition On-line structured databases, such as those containing large numbers of bibliographic entries and abstracts, provide users with the ability to find up-to-date information on a variety of sources. They are called structured not only because each item is quite similar in form, but also because the index terms used to index the database are controlled by a thesaurus which generally aids search strategies. On-line help messages are increasingly regarded as the primary kind of documentation supplied with computer software because they are much more frequently referred to than is the paper documentation. Disadvantages On-line help messages are by their nature quite short and in most systems do not link users to more extended learning situations. Relation to Hypertext Hypertext linkages and buttons will get users to follow a trail to on-line help to computer-based instruction courses. Advantages The advantages of the structured databases are that the reader has excellent navigational control over what to expect of the data and also has good navigational control over the vocabulary search. Disadvantages To find items in such massive databases requires professional skills of an on-line search specialist, although the ordinary lay person or scientist can learn these search skills. Relation to Hypertext Commentary One of the reasons that much paper and on-line documentation is not used is because they are incomplete and badly written, a topic we take up in the next chapter. (REH) These databases show the advantages of structure in writing and in search methodologies. They give greater precision and efficiency in the search. V ^ 69 Managing the Creation of Different Versions Introduction Seldom does a writer write only one draft. In fact, teachers and writers often say that good writing is produced by editing and revision, not by writing. Frequently in organizations, writing is a group project. Some new software permits groups of authors or authors and their editors to comment upon current drafts and keep track of previous versions. Some authorities — including Ted Nelson — have classified "versioning" as one of the problems hypertext should solve. Definition: Versions Versions of documents are drafts of all or parts of documents that are produced during different stages of the analysis and writing project. Issue: Linking Alternative Sections of Drafts One of the important problems in hypertext is to show the links that are all connected to a draft and whether or not to carry all of these links ahead to the next draft. This set of issues is not an easy one and is illustrated on this page by a schematic that shows comments and corrections on a single draft, alternative versions, and other suggestions which are then carried over to another draft. It shows the complexity of the problem and the difficulty of solution. Issue: When is a Version a Version? Suppose just one little change is made on one page of a draft. Does this create a new version? If so, who names it? What is it named? How are the other people — if any-- who are working on the draft to know that there is a new version? How are they to know where the change was made? And by whom? @ S 00 'vO 70 m Version 1 ■ ’ ' : v - \ Chapter 2. Current Issues With Hypertext Issue: Database Design Once you've answered the basic questions about what will be permitted and what is a version, the next problem is how to represent this usefully in a database, so that the information can be tracked automatically. Issue: Interface Design Assuming that you can keep all of the versions in the database straight, how do you display to users all of the differences in the versions and the actual changes as well as the suggested changes? / Commentary In general, keeping track of versions is very difficult even with a good system. You don’t usually want to keep a lot of old drafts around, because digging through them is excessively time consuming. There are exceptions of course. CAD documentation is one such example of needing it because of the multiple versions of drawings and their critiques. (REH) Rewrite Text or Convert to On-Line Text "As Is"? Introduction Now that most people have a computer on their desks, in many companies the idea springs to many managers' minds that you just put whatever text you have into a database — paper into electronic storage with no pain. A number of major failures have resulted from this oversimplified idea. These failures have resulted in on-line text systems that go unused because they are so bad. Three Options in Converting the Text Keep Text "As Is" (No rewriting or revision) It is cheap just to scan in the text with an optical character reader. BUT OThe "keep as is" choice raises most of the user issues raised in this chapter -- lost in hypertext, cognitive overload, lack of normal reading cues, etc. Superficially Chunk and Label Adding more labels, leaving the text "as is" but divided into more parts is less expensive than rewriting. BUT QThis choice may reduce some of the user issues but it is my claim in this book that the major costs come on the user side rather than on the costs of putting text on-line in the first place. Reanalyze and Rewrite as Necessary ■ This option will cost more on the "make" side than leaving as is, but it will save money on the "use" side. BUT This choice will reduce costs on the "use" side, and you may end up with a more useful paper document as well. Commentary: The Major Cost Tradeoff @ Co 00 VO 73 m X o 3 Costs of Creation The major tradeoff is between the cost of creation and the cost to users. Usually the cost of creation is • a more visible budget of the salaries of a small task force who are assembled to make the conversion • a one time cost. The elapsed time of the project is also a major factor. Costs of Use Usually the costs of use are the • costs of a large number of high priced professional users whose time is their most valuable resource being frustrated by the system • costs of users not using a system because it is not easy enough to use • costs of errors generated from poor analysis, organization and writing of the text database These costs are often hidden in overhead— although some studies have revealed their magnitude. (REH) 72 Chapter 2. Current Issues With Hypertext Commentary: Computer Created Links Will Not Solve Many Problems H Some people believe that we will be able to write computer software that will H automatically search text and make hypertext linkages between appropriate places. H While this is possible to some degree, especially with Information Mapping's H method, it is difficult in relatively unstructured text. (REH) Four Major Options in Choosing the Retrieval Method □ Use Full Text Search This software searches the entire text and retrieves every mention of a search word. There are a variety of commercially available packages that will do this kind of search. BUT In very dense technical, administrative or business files, the software would present far too many "hits" on words that are frequently used. (E.g., if the search of the text in this book were on "hypertext," there would be hits on almost every page. Not very useful.) Too many synonyms is also a problem. Even with Boolean search capability, the searches would require a relatively high level of search skill and considerable time in large databases. Create Keyword Index The text is linked to specially selected keywords decided upon by a human indexer which provides a more controlled search and a likelihood of more precise retrieval. BUT a Develop Hierarchical Structured Index(es) These are one or more tables of contents or similar hierarchical lists of subjects in the text. Usually not much of a problem to create and very helpful. BUT OThe cost of creation is higher than for automatic full text searches. Q Hierarchical indexes, even very good ones are not adequate for searches of very large text databases. They are best for searches of a relatively small number of pages, for example, the size of a single volume manual. Develop "Applications Overviews" These are job aids, checklists, procedures that might link to many different parts of a text. They are very helpful in task-oriented organizational situations. BUT □ They require a good structuring of the original database or they create a messy tangle for the user. Successful implementation depends on the subject judgments of a few writers. Commentary: Choice Depends on User Requirements The choice of which of these four (including the choice to use all four) depends on the analysis of user needs and the objectives of the system. All, but the full text search option, are labor intensive on the creation side, but less costly to the organization on the use side. (REH) 73 Chapter 3. Introduction to Information Mapping's Method of Structured Writing Overview of This Chapter 76 Some Problems Addressed with the Method 78 What is the Information Mapping Approach? The Problem of Human Short Term Memory The Method What are Information Blocks? 84 How Four Principles Constrain Block Construction 86 Examples of Information Blocks 88 Block Replaces Traditional Definition of Paragraph 90 Developing Guidelines and Standards for Blocks 92 Hierarchy of Chunking and Labeling Principle 94 Examples of Maps Displayed on Paper 96 Change Traditional Document Structure 98 Managing Completeness by Key Block and Topic 100 Improves Efficiency Throughout Analysis Process 102 Discourse Domains 104 Brief Discourse Analysis (Stable Subjects) 106 Information Type and Block Type Analysis 108 What are the Information Types? 110 Key Blocks Provide "Completeness Templates" 1 12 Results Greater Ability to Specify Rule Domains 1 14 Top Down and Bottom Up Analysis 1 16 Reference-Based Training 118 The Mapping Metaphor: Subject Matter Structure 120 Meeting the Criteria for Better Communication 122 80 82 © 1989 R. E. Horn Chapter 3 Introduction to Information Mapping's Method of Structured Writing mmm 40 million pages Definitions Definition of Information Blocks Results In this Chapter ...an Overview The Process of Analysis Definition of Information Maps Information Mapping's Method The Mapping Metaphor What are Information Types? Discourse Domains ■ - ■ ■p 1 ■ j ■ m m m ■ Tf*' ll,«, ■ ■r m ■■ ■i / JB mm ■ ■ m UJQH ^ X 6S6I @ Overview of This Chapter Introduction In this chapter, we provide a description of Information Mapping's approach to the analysis, organization, and presentation of information in different kinds of documents. The reader can expect a "what is" rather than a "how to" level of detail. Information Mapping’s method is a synthesis of tools and techniques conceptual tools to... determine exactly what is to be communicated to whom by a precise analysis of the users of information and the content conceptual tools to... identify and describe ail of the parts and relations essential to a particular document conceptual tools to... organize and manage the information through various revisions and reorganization phases It does this with a novel and precise way of chunking information and... It also has a novel way of categorizing information content types... It does this with a set of tools to keep track in large documents of what is known and what is missing... Brief History of Information Mapping 1965 1970 1975 Period of Invention Period of Evaluation Initial Period of Applications Regular Teaching — Testing and Teaching Method and Vali< The Core of Information Mapping's Method Was Developed During This Period Information map concept The information block concept which replaces the that provides the basis for organization and for hierar- writing. |see pag^ . 84 The information types which provide a powerful analytic tool for establishing content . (see pag' |see ] . 94 Guidelines and rules for Approaches to sequencing methods. [seepage Complete set of hints, ques¬ tions and content-topic and block type relationships applied recursively to each new Initial specification presentation formats for paper-based procedures, policies, and instructional documents. |see ; . 96 precise cnunking, an and writina N. aivsis chunk of content. .. .112 110 [see pag^ . 86 |see pagg Chapter 3. Introduction to Information Mapping's Method of Structured Writing Commentary: Goal of Information Mapping 's Method My goal in developing the method was to enable people to produce better communication documents in business, industry, government, technology and science. (REH) conceptual tools to... prepare presentation sequences of the information for different purposes (such as learning and reference) It contains a set of sequencing strategies for different types of documents. conceptual tools to... develop effective formats for final presentation Commentary What gives Information Mapping's approach its strength and wide applicability is not any one of these but the careful synthesis of the whole group of tools. (REH) ...with a group of formats each designed for different documents and different media (e.g., paper and computer screen). 1980 1985 ; of the lation Research Mill : : : y; . Syj fl Period of Use of the Method and Wide Application in Training U.S. Business Period of Integration of the Method into On-Line Text and Hypertext Examples of Applications ] 1 - 1 - 1 - procedures policies training materials documentation _ N 1 see page*" > 170 Examples of Applications hypertext on-line documentation electronic mail messages see pag^ive 77 1989 R. E. Horn Some Problems Addressed with the Method ■* Commentary: Ask People What Is Wrong With Writing 1 As I have travelled around the country giving speeches and workshops, I have often asked people to tell me, "What are the main problems with the documents that come across your desk?" This is what they tell me. (REH) Where are the key ideas? They're buried somewhere around here. You can't tell what's important. Reader Problems I'm not sure what I'm supposed to do with this information. These examples are too complex. And sometimes they aren't even there. 1 can’t tell what the organization of this document is. Commentary: To Solve These Problems When you compare the problems of readers and writers, you can see that there is a relationship. Many of the same things that writers complain about are exactly issues that readers have with writing that comes across their desks. As we shall see in this chapter, these problems are the ones that Information Mapping's method has addressed. (REH) Often, all I need is one or two sentences. I wish this were set up so I could scan it. I don't have time to read it all. The words the author uses are unnecessarily technical. There is just too much to read! The report is just too long. The sentences are too long. The whole thing is too long Who can read it? They send down these instructions that are so vague we can't figure out what they mean. Parts of the same thing I need are scattered all over the document Chapter 3. Introduction to Information Mapping's Method of Structured Writing Commentary: Ask People About Their Writing Difficulties The other side of the coin, of course, is, "what are the difficulties that writers face?" I ask people that question too. Here is what they say. (REH) 1 always have trouble getting started. Sometimes that blank piece of paper can keep me from writing all morning Writer Problems When it gets longer than a page, I don't know how to get it organized It takes a long time to put it all together. I have trouble being concise. I'm afraid someone will misconstrue what I write so I write each point into the ground. Commentary: Connection With Hypertext I never know how much detail to use. People tell me I overwrite Since the mid-seventies, Information Mapping's method has been used on an increasingly wider scale. It appears that it will resolve a number of the hypertext issues that we discussed in the previous chapter. But before we can describe the solutions, we have to look at the basic structure of the method itself. (REH) Getting it to flow. I can't seem to tie it all together the way my English teachers kept telling me I had to. I can't always figure out a good sequence to put things in. I have trouble knowing when to stop. I spend an inordinate amount of time getting my thoughts down .on paper. I can't figure out how to write for diverse audiences. 79 1989 R. E. Horn What is the Information Mapping Approach? Definition in • for • ma* tion map* ping (in 'far ma'shan map* ping) n. I. a U. S. trademark A designating certain products and services, specifically training courses in business communication, documentation, and training. 2. the name of the leading U. S. documentation engineering and business communications company, Information Mapping, Inc. A founded by Robert E. Horn A . 3* in particular, an informal . name given to the use of the information map A and the information block A to replace the paragraph as the basic unit oif meaning in functional written communication . 4. an informal name given to Information . Mapping’s method ”, “A... # synthesis of tools and techniques for the analysis of complex subject masters aftdjobs, and the group of standards and techniques for the management of large amounts of rapidly changing information, and to the procedures for planning, organising, sequencing and presenting communications. 5. a metaphor used to compare the way geographical maps represent terrain with how information maps represent the key /features of the information landscape . A. . 6. (adj.) an informal name given to any one of thousands of .. . applications .A 6f the methodology to different jobs and communication documents . (as in an "information mapped" document) (t. L: s. informdtion; r. ME enformacion , t. OF) i(. . . Examples Science Information Product &/ Services / | see page^> 174 Knowledge Base Individual Pagesrseepage>% and Screens Procedures, Policies, Documentation Training Explanation of The Methodology ...at each stage precision, completeness, clarity, and consistency... Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Pre-writing Analysis H Information Gathering Analysis * Organization Analysis Sequencing Analysis * Presentation Format 80 Chapter 3. Introduction to Information Mapping's Method of Structured Writing . . Information Mapping ® Seminars and Courses Information Mapping ® Training and Documentation Services Info-Map ® Publications see Notes Information Mapping, Inc. the company f Robert E. Horn originator of the method; founder of the company 1965 . '**»* . . see pagej> 94 . Definition Examples Information Maps collection of blocks about a specific topic . . see pag 84 Definition Examples Information Blocks precisely defined chunking of information 81 The Problem of Human Short Term Memory Introduction Problem |Let me see... \ 1492 times 1776... I need to C remember the partial answers. How am I going to do that without a pencil? 1492 x1776 8952 10444 10444 AA 22 _ 2539792 Part of the rationale for structured writing starts with the limitations in human short term memory, limitations which every human being has. We introduce the problem with an example (first used by Herbert Simon). Example One While in the shower, multiply 1776 by 1492. Next remember the answer long enough to check it. f This situation highlights the limitations of human short term memory. It will probably take you several minutes to do the multiplication task, if you succeed at all. Example Two You look up a phone number. A little interruption and you've forgotten the number... "Where did you put the coffee, dear?" @ VO 00 VO 70 m x o 3 Why do we have these memory problems? Answer Human short term memory capacity has severe limitations. Two Estimates of the Size of Short Term Memory Every thought process that requires what we call "attention" has to be held in short term memory and human beings can hold only a small number of "chunks" of information in short term memory. George Miller, A an outstanding communications psychologist suggested in 1956 that the number of chunks you can hold in short t^rm memory is 7 (plus or minus 2). \ Research by Herbert Simon, the Nobel ppze winning economist and information scientist, suggests the number is smaller — around 4 to possibly 9 chunks. Whatever the size, The Chunk Definition A chunk is any familiar pattern. Chunk size itself depends on your prior learning. Examples For one person for one subject matter, a chunk may be one sentence, while in another subject it may be several sentences. 82 For younger children, a chunk might be reading a single word; for still younger ones, simply recognizing a single letter would constitute a chunk. Chapter 3. Introduction to Information Mapping's Method of Structured Writing Long Term Memory Overcoming transfer time limits In order to do the "1776 x 1492 in the shower" problem, you would have had to transfer some of the information to long-term memory (i.e., some of the intermediate products of the multiplication). How does transfer to long term memory take place? What does our brain do with the contents of our short term memory in order to turn it into long term memory? We don't know exactly, but somehow we have to gather these 5 to 9 chunks of information, add some sort of identifier and link them in with other previous experience. All this takes place on an unconscious or partially conscious level in every human being every day. Miller is saying that all human beings everywhere chunk information in order to transfer it to long term memory and this process has severe limits in its capacity. How long does transfer to long term memory take? To store 5-9 chunks of information in long term memory takes, according to Herbert Simon, A^5to 10 seconds per Principle: Organize thought so as to stay within memory limits "We must," Herbert Simon says, "organize our thought processes so they do not require us to hold more information than 4 to 7 chunks in short term memory simultaneously." How do we overcome limits? Simon points out that we do it • by external aids to memory (computers and calculators, paper and pencil) • by aids to long term memory, such as books, reports, manuals • by discovering and using strategies for accomplishing thinking tasks. Definition: Recoding Miller suggests that the process of grouping many chunks of information into larger chunks by conscious and uncon¬ scious processes could be called re-coding. Example How do we recode our experience? Miller suggests, "Pro¬ bably the simplest is to group the input events, apply a new name to the group and then remember the new name rather than the individual input events." Importance And he says, " . . .1 am convinced that this process is a very general and important one for psychology ..." He says, "The point is that re-coding is an extremely powerful weapon for increasing the amount of information that we can deal with. In one form or another, we use re-coding constantly in our daily behavior." Three Types of Recoding a. verbal code (rephrasing in our own words) b. seeing something and then making a verbal description of it. Miller says, "I suspect that imagery is a form of re-coding, too, but images seem much harder to get at operationally and to study experimentally than the more symbolic kinds of re-coding." c. presupplied codes (such as the labeling system of Information Mapping A 1 ^ ("" | see pag^> 92 83 What are Information Blocks? Introduction We have made it a principle that we must help the reader by "pre-chunking" the information into blocks. This not only permits readers an easier way of taking information in, but also helps writers in their analysis of the information. Definition in • for • ma • tion block (in far ma shan blok) n. 1. any one of 200 precisely defined kinds of information that together make up the basic subject matter of a manual, book, or course, used in the analytic method developed by Robert E. Horn and taught by Information Mapping, Inc. 2. the basic subdivision of a subject matter, replacing the paragraph as the fundamental unit of analysis and presentation in functional and task oriented text. 3. composed of one or more sentences and/or graphical structures, but not more than (usually) seven sentences, identified clearly by a label; blocks are constructed according to four principles, the chunking, relevance, consistency, and labeling principles; blocks are normally a part of a larger structure of organization (called an information map A ). - Icon for Blocks In this book we sometimes use this icon to indicate the information block. So, in this book each of the labeled chunks of information is an information block. For Other Examples of Information Blocks Examples For example, a definition block is an example of a block. Information Blocks: Different Types The large differences between these blocks introduce the concept that blocks are of quite different types, depending on the kind of document they are a part of. And each block type A may have quite different rules and guidelines for analysis and writing. s he 00 VO 70 W X o 84 Chapter 3. Introduction to Information Mapping's Method of Structured Writing : : ' : Four Principles for Constructing Blocks □ Chunking Principle Group all information into small, manageable units, called blocks and maps. Relevance Principle Include in one chunk only information that relates to one main point based on that information's purpose or function for the reader. Consistency Principle For similar subject mat¬ ters, use similar words, labels, format, organi¬ zations, and sequences. Labeling Principle Label every chunk and group of chunks according to specific criteria. 1 Commentary: Intuitive Chunking vs. Precision Modularity^ One of the reasons that Information Mapping's method is so useful and powerful in managing large amounts of information is the precision modularity provided by information blocks. Precision modularity is a term we use to distinguish information blocks from the practice of "intuitive chunking." Intuitive chunking is forced upon people by screens of limited size. Intuitive chunking is simply dividing information into small chunks without knowing exactly why you are doing that. Precision modularity means following specific piinciples and guidelines throughout the analysis. The four principles act as constraints A to the construction of blocks, and the information typology A we develop in forthcoming pages provides "engineering-like" guidelines for the construction of information blocks. 85 How Four Principles Constrain Block Construction Introduction We have introduced the information block as the smallest unit of analysis in Information Mapping's method. We construct blocks by applying four basic principles to any piece of information in a subject matter. This constrains how any block is made and what it may contain. The process of using the four principles transforms mere intuitive chunking into precision modularity. The diagram on these pages depicts the four basic principles and how they perform this constraining function. When combined with a strong system of types A of information, these block construction principles provide the foundation for the entire methodology. Icon We sometimes use this icon to represent the four principles Chunking 0 o c <0 ill ara B n Consistency | Labeling Principle Label every chunk and group of chunks according to specific criteria. The labeling principle requires that a clear label, the word "definition" in this case, and the term being defined would be set apart from the sentences that define the term. The contents of a definition block would clearly explain the term(s) being defined. 00 VO 73 m X o 3 86 Constraint Chapter 3. Introduction to Information Mapping's Method of Structured Writing Chunking Principle . Group all information into small, manageable units, called blocks and maps. The chunking principle suggests that all of our text will be divided into relatively small units of information, about the amount that humans can handle with their short term memory limitations. Constraint wfmim Definition: Cash Value Cash value is the amount of money a life insurance policy is worth at a specified time. Constraint _ ■ " llll Relevance Principle * Include in one chunk only information that relates to : . one main point based on that information's purpose or function for the reader. vzMmMSmik ligllillllpp - ■ ■ The relevance principle is interpreted in this instance to say "don’t put anything other than definitional sentences into the block.” If you have information that is nice to know, or contains examples or commentary, the relevance and consistency principles demand that you put it some place else and label it appropriately, but do not put it in the definition block. Consistency Principle isMmw • . ^ m _ put in a different block For similar subject mat¬ ters, use similar words, labels, format, organi¬ zations, and sequences. sistency principle says "Do this ng for every other definition in your manual, book, hypertext." In other it would be interpreted in this ample as "Treat all definitions alike, i.e„ hem in blocks by themselves labeled _ (e.g., this sentence, "Cash value is only one of the kinds of value an insurance policy has," would go in some other block, not in the definition block because the sentence is not definitional.) 87 Examples of Information Blocks Size and Content An information block may be one or more sentences long. It may also be a list, a fairly complicated table, or other kinds of graphic structure. On these pages, we present examples of each kind. Information Block: Smallest Meaningful Chunk What do all of these blocks have in common? All three represent the smallest meaningful chunk for most readers. So, to repeat, a block may have one to seven sentences (very occasionally more). It may also contain a table or simple graphics so long as it meets the criterion of meaningfulness, relevance, and consistency. Three Kinds of Information Block Contents [Tl Example of a One-sentence Information Block Definition: Cash Value Cash value is the amount of money a life insurance policy is worth at a specified time. Example of a Block that Contains a Table @ S 00 vO *3 tn X o 3 IF the book is . . . THEN send the patron . . . AND send . . . available the book an invoice to the Billing Unit. not available • never owned • lost Form 25 — checked out with no waiting list Form 66 a copy of Form 66 to Circulation Desk. checked out with a waiting list Form 66 and Waiting List Notice a copy of Form 66 to Circulation Desk. Chapter 3. Introduction to Information Mapping's Method of Structured Writing 3j Example of an Information Block Using Graphics Sequence of Events in the Receiving Department Carrier Shipment Arrival Check in Staging . mu ijjjUb-ia — l Receiving Department O Sign in Packing Slips 1 WwScP and just about everything else. Ope of his lesser known, but most influential, books was on rhetoric — the rules of written and spoken argument, / ^ 1 see page Modified by Philosophers & Literary Critics Of course, over the years some of these rules were modified and refined. New rules were added to the study of rhetoric. The ideas of the originators of many of these ideas were formalized by philosophers of the Middle Ages where rhetoric was one of the Big Three core subjects - called the Trivium -- at universities. Literary essayists of the seventeenth and eighteenth centuries in Western Europe also contributed to the subject Habit Keeps Us Using It Habit keeps human beings using what they have done in the past even if it barely works. If it gets us by, we don’t bother to learn a new skill. You Wouldn’t Use Aristotle’s Science in your R & D Program, So.,, Passed Along To Us By Habit As wise as he was for his time, you wouldn’t use Aristotle's natural history for your biochemical research planning. Too much science has developed. Most of these rules were then automatically taught by your teacher and mine. We started learning them in the first grade and we’ve kept getting the same Aristotelian rules right on through college. But, we are still using Aristotle to do our hypertext and our communication. And we are hampering ourselves, just as if we were using him as a consultant for our research and development. 91 Developing Guidelines and Standards for Blocks Introduction As we have noted, Information Mapping's method does not have vague, fuzzy rules for writing paragraphs. Information must be part of a precision-developed information block for it to fit into a larger document and to be managed in an efficient manner. So, for each domain (and often for a specific document type) it has been important to specify the most frequent types of blocks. After the blocks have been identified, we then need to develop the guidelines that will make the consistency principle operational for both the contents and the labeling of all blocks of a similar type. On these pages, we present an overview of how we approach this task of developing guidelines that enables us to ensure that basic communication functions are accomplished on a consistent basis. Note that these are samples of the guidelines and apply only to the labeling. Space in this book does not permit reproducing all of the specifics of block development. Two Basic Kinds of Block Labels □ dn Eb Analytic Labels Definition Display Labels Definition Analytic labels are those block labels that are used in the content analysis phase of document or training development to manage completeness and (often) share the bulk of information. Where to Find Examples Lists of analytic labels (but not the criteria for sorting information to identify them) are provided in this book for • documents in the relatively stable subject matter domain scientific papers, reports, abstracts and presentations. Display labels are those block labels that are used, depending upon document type, to make it easier for the user to understand what the block contains than if it had analytic labels. Sometimes, and in some cases even frequently, the analytic labels are used as display labels. But often, words must be added to analytic labels to make contents readily accessible. Example A definition block would have the analytic label "Definition" as a part of it. If a display label were called for, the writer would use "Definition" (followed by the term being defined) as in "Definition: Cash Value." Three Kinds Display Labels rh K rh Subject Matter Independent Labels OD Subject Matter Labels 0 Combination Labels @ VO 00 VO PO m x o 3 Definition A subject matter independent label is a label that identifies the purpose or function of the information in the block for the reader. A subject matter independent label is independent of the content of the block. It could appear on many different documents regardless of the subject. Definition A subject matter label is a label that describes the content of the information in the block. Definition A combination label is a label made up of two parts: • a subject matter independent label that indicates the purpose or function of the information, and • a subject matter label that identifies the content. 92 Chapter 3. Introduction to Information Mapping's Method of Structured Writing Implication of the Principles of Relevance and Consistency Insofar as possible, block labels should be consistent and relevant not only for a single block but across all blocks of a simitar type. And there may be guidelines for labeling that apply to all blocks. Usefulness of Block Labels For the reader, block labels perform three functions: • to act as "advance organizers" about the contents of the block for learners • to act as "access tools" for persons who are scanning or browsing • (in some very limited discourse domains) to act as a tool to attract the reader's attention. For the writer, block labels facilitate and constrain analysis and organization of the subject matter. c ‘5 k_ +* (A C o o Constraint Definition Definition Definition Definition: Cash Value Cash value is the amount of money a life insurance poli¬ cy is worth at a specified time. Constraint c <5 k. « c o o Some Guidelines That Apply to All Block Labels Clear: Use labels that clearly describe the function or the content of the Block. Brief: Make labels brief. Shorter labels are better than longer ones as long as they don't introduce ambiguity. In general, use no more than 3-5 words. The label should not be so brief as to be meaningless. Consistent: Use the same vocabulary in the label that you use in the Block. Familiar: Use vocabulary that is generally familiar to anyone in the audience. Avoid technical jargon unless you are certain all readers will know the jargon. Appropriate: Make sure that the label reflects the significance of what you want to say. "Comment” may work well as a label for optional information. However, if the Block contains crucial information, ''Important’’ or "Caution" would be a better choice. Guidelines for Managing the Size of the Message Growing Out of the Principle of Chunking Insofar as possible, blocks should be no more than 7 plus or minus 2 sentences. Insofar as possible, information maps should be no more than 7 plus or minus 2 blocks. Insofar as possible, parts of chapters or short chapters should be no more than 7 plus or minus 2 maps. Here we follow the suggestions of the short term memory research. Independent: Make each label stand alone and act as an advance organizer for a single Block. Avoid labels that act as transitional devices. Other guidelines have been developed for specific types of blocks and for the three types of labels described on the facing page. Other Conventional Quality Control Factors Used in Blocks Grammar Sentence length & complexity (i.e., readability index) Syntax Spelling Word usage Style (formal, informal) 93 Hierarchy of Chunking and Labeling Principle Introduction When we begin to deal with several blocks about a similar topic, we apply another principle, the hierarchy principle. And, of course, we continue to apply religiously the other four principles of chunking, labeling, relevance and consistency A Principle: Hierarchy of Chunking and Labeling Organize small, relevant units of information into a hierarchy and provide the larger group(s) with labels. Rationale As the number of groups of subject matter chunks grows beyond the seven plus or minus two limits of short term memory, readers have difficulty comprehending or remembering the information. Information Map Introduction These clusters of seven plus or minus two blocks of information are as different from other forms of organization of writing as the blocks are different from paragraphs. Definition: Information Map An information map is a collection of two or more, and usually no more than nine, information blocks about a limited topic. Approximate Page Size In general, we can think of an information map as approximately one to two pages in length, but some maps of certain well specified types run several pages in length. For Definitions and _ _ ^ Examples of the Foui\ see~page Principles , . . • chunking • labeling • relevance • consistency yr Icon for W Information Maps ^ In this book we sometimes use this icon to indicate the information map © vO 00 VO 73 m X o 94 Chapter 3. Introduction to Information Mapping's Method of Structured Writing Apply The Hierarchy of Chunking and Labeling Principle to Larger Aggregates Introduction Size is always important. We must keep applying the hierarchy principle and the other four principles for grouping larger and larger aggregates. Guidelines We have said that if you get more than 5 to 9 sentences in a block, it is time to chunk them into two blocks (or, in certain circumstances, into two or more sub-blocks). Group all Blocks into Maps If we get more than 5 to 9 blocks in a map, make two, give them an overview and call it a "section." Chunk every 5 to 9 sections into chapters and every 5 to 9 chapters into "parts" and so on until a document is complete. Sequencing Blocks When Presenting Maps The definition of a map only requires clustering of similar blocks together. It does not specify their sequence. This permits us to display them in different order for, say, initial learning by novices in a subject matter, and from the order used for reference. And it encourages the temporary formation of map-like clusters during the early stages of information gathering and subject matter analysis. There is no demand for strict sequencing at these early stages. W« can always think of two distinct levels of blocks (or collections of blocks, maps, parts, etc.): Two Kinds of Block Clustering £ Analytic Clustering Description Display Description The most elementary analytic task is to cluster blocks of information around the concept of the map. We assert that "this block belongs to this map." Map Title Block 1 Block 4 Block 2 Block 3 Block 5 When actually presented to a user, blocks need some kind of order and sequence. This is a separate set of decisions from those used to cluster blocks into maps. .for reference Map Tit e Block 1 Block 4 Block 2 Block 5 Block 3 ...for initial learning... Map Tit e Block 1 Block 2 Block 3 Block 4 Block 5 95 1989 R. E. Horn Examples of Maps Displayed on Paper Introduction Information Mapping's method works equally well on paper as on computer screens. We have presented many examples of the method in computer screens in this book. The reader may be wondering about some of the ways that examples of this method look when displayed on the printed page. We present two typical pages. from a’ procedure manual How to Prepare Data for an Audit One of the most important procedures in an audit is preparing the data. ntroduction Procedure Careful preparation ensures that the data is correct and that each step of preparation has been carried out. Follow the steps below to prepare for the audit. Step 1 Action For data items selected for the audit, obtain the following: • source document samples, and run data from the computer room. Verify the source documents sample by comparing the samples to the original list. Record on a worksheet sufficient descriptive information to provide accurate identification for future audits. I t Minimum Information Required Examples Attributes of the sample j • Sales Territory • Effective data Description of each data item • Account Name • Account Number • Type of Business Compare data samples to related documents and record on the worksheet any source of error or difference. Compare . . . To... sample data 1 i 1 • programming instructions in effect when source document began • company requests • statistical guidelines. i j source data run data printouts. Prepare a summary sheet that • lists each difference or error found, and • analyzes each data item to compute accuracy ratios for the audit sample data items. 96 J ill Chapter 3. Introduction to Information Mapping's Method of Structured Writing While this book emphasizes the display of structured hypertext, most of the work today with the methodology — millions of pages — has been done on paper and paper will probably continue to outrank electronic display for some time. Rules for Defining Audit Accuracy Ratios Terms and Definitions Rule 1: state data items Rule 2: note any changes Rule 3: use consistent definitions Introduction In order to compare the statistics of one audit to the next, the accuracy ratios must be defined and presented consistently. Each audit must contain the following ratios. Ratio Data Field Verifying Ratio Transaction Accuracy Ratio Account Accuracy Ratio Definition The number of times a given data field is correct, divided by the total number of transactions reviewed The number of times all data fields reviewed on a transaction are correct, divided by the total number of transactions reviewed The number of times all data fields audited are correct on all transactions relating to a single account, divided by the total number of accounts reviewed In your audit report, you must clearly state in the accuracy ratios which data items are included in the Data Field Verifying Ratio. When the company requests changes or additions to the Data Field Verifying Ratio list, you must note these changes in your Audit Report. Rationale • Any changes will affect the results of the transaction and account accuracy ratios. The more data fields reviewed per transaction, the lower the accuracy ratios. • Comparison of audits would be impossible unless the changes were listed. When you are comparing transaction accuracy ratios over a specific time period, you must use the same transaction definition for all audits going on within the time frame. Example: If the Cash Receivable transactions included international receivables as part of the transaction during the first audit, then you must include international receivables in the next audit. from a procedure manual 97 Change Traditional Document Structure Introduction We have made the case that hypertext requires a re-examination of writing. And in this chapter we have introduced Information Mapping's method as a research-based thoroughly tested and proven way of analysis and organization and presentation of writing. It may have occurred to the reader that we are introducing nothing less than a complete overhaul of the traditional rules of rhetoric. We are doing just that. Definition: Rhetorics Rhetorics are rules for writing particular documents for particular purposes. Comment We use the classical definition for "rhetoric." The reader should note that this meaning is quite different from the more common meaning of the term, i.e., "tricky special pleading." Summary of Rules for Organizing Newspaper Articles As usually taught, the rules of rhetoric for newspapers are as follows: 1 . Grab the reader in the headline by a. summarizing the important point in the headline b. an unusual human interest angle. 2. Use any kind of cute, attractive, possibly in¬ formative, subheads to attract the reader. Occasionally, subheads can be an informa¬ tive summary of what is coming up. The rule for insertion of subheads is break up the gray space. 3. Organize the entire story by decreasing im¬ portance. That means that the most impor¬ tant elements of the story are first and the less important elements and details of the story are last. (The reason for this is so that a newspaper editor can cut it at any place to put more ads in the newspaper.) 4. Summarize everything in the lead paragraph @ with answers to the questions who, what, where, when, why, and how. Kinds of Rhetoric Usually, rhetoric has been thought of as a single unitary subject. However, it is clear that we use different sets of rules for different kinds of writing situations. Think of how different the rules for writing for newspapers and popular magazines are from those for technical writing. We can't go into a complete analysis of all of the different kinds of rhetorics in this book. That isn't our purpose. But we do want to contrast the approaches taken in hypertext and in Information Mapping to the traditional approaches. We have done this with regard to the paragraph A and below we spell out some of the major rules for organization of prose in newspapers and magazines so the readeKcan contrast them with what we are saying about Information Mapping in this chapter. Obviously, we do not usually use\the rhetorics for newspapers and magazines for books, training manuals, and business letters. The point of the contrast is that there are very clear rules for rhetoric in each document domain and we must be very clear about these guidelines. \ S* X The New Tim Economy Contir Drug War Accelerates Experts Puzzled by New Changes in U.S.S.R. 98 Chapter 3. Introduction to Information Mapping's Method of Structured Writing Business Needs are Different Business has special needs. There is no need to have everybody read every word of a technical or administrative report, so we don't want to use journalistic gimmicks to attract readers. Business documents are often much longer than newspaper stories, and require much more sophisticated methods of organization. They, in fact, require the kind of systematic approach we have used in Information Mapping's method. Our Conclusion: Informed Access Is Better Than Tantalizing Hints In magazines, as we pointed out below, the whole idea is to hint at fascinating material coming up later and hide it in the back pages. In business, we assert the absolutely opposite principle. The reader should be able to locate quickly any key information and make decisions appropriately. Moreover, we would recommend that the piece be organized, not by a continual series of hints, but by very clear chunks clearly designated by subheads. In newspapers and some magazines, editors do provide informative headlines and readers are able to scan and skip to what they want to read. es Times lues to Boom Deficits Pile Up But People Keep Buying Terrorists Bomb Restaurant Summary of Rules for Organizing Magazine Articles Magazine rhetoric is much like newspaper rhetoric with a few exceptions. 1. Grab them in the title. 2. Grab them again in the first paragraph with a human interest angle. Do not give away all of the information in the first paragraph. 3. Write to keep the reader by a. not revealing everything right away b. hinting at fascinating stuff coming up. 4. Use cute, not very informative subheads to break up the gray. Purpose is to Keep Reader Turning Pages to See Advertising The purpose of magazine rhetoric is to suck readers into the article and keep them reading so they will "like the magazine" and continue to buy it. A secondary purpose is to make sure that the reader flips through the magazine, continuing to read the story to the end, increasing the probability that the reader will look at some of the advertising on the back pages. 99 Managing Completeness by Key Block and Topic Introduction An important factor in managing information is the analysis of subject matter into precise blocks. This enables us to develop a comprehensive analytic structure in which the subject matter is divided according to specific criteria and principles into small labeled chunks by which complex information can be managed in all stages of the communication process. vO m P° ra x o 3 Matrix Topic topic A topic B topic C topic D topic E topic F topic G topic H topic 1 topic J topic K topic L block type 1 block type 2 block type 3 block type 4 block type 5 block type 6 block type 7 block type 8 block type 9 block type 1 0 block type 1 1 block type 12 == = ' = = mu mu = ' = EEEE — EEEE — mu — = _ = — in hi hi = = s= = = = — — = ± === EEEE — EEEE 1 ...topic E has block types 4, 7, 9, and 1 1 associated with it. Block Types Axis of Matrix Description For training manuals, procedures, policies, and documentation, there are 40 blocks into which approx. 80 per cent of the content of most subject matters can be analyzed with considerable reliability — if the analyst fol¬ lows the classification criteria of the method. Only 12 are shown in this schematic of the system. Over 200 common block types for different document types have been identi¬ fied and described. 100 ■ Chapter 3. Introduction to Information Mapping's Method of Structured Writing How to Determine What the Topics Should Be Topics are identified by specific criteria for specific document type. » How to Understand This Diagram — n. if block type 1 is "definition" then topics A, D, F, H, I, and L have definitions. ■Hi i 111111! Note: This matrix is a conceptual rather than a physical tool. If actually put on a wall the matrix for a 100 page book would measure 4 feet by 50 feet. When does a blank space on the matrix indicate missing information? A Description Specific key block templates for specific in¬ formation types have been developed which permit the analyst to know when a blank space in the Block Type-Content Topic matrix indicates missing information that the analysis has not produced as yet. Example A template of key information blocks for concepts in the "relatively stable subject matter" discourse domain would include these blocks: • Definition • Example • (optional) Non-Example. Guidelines, Rules, and Standards Guidelines indicate which terms are to be regarded as concepts for a particular audience for a particular type of document. Additional guidelines indicate how many examples and non-examples would need to be provided in the analysis. Other guidelines indicate the kind of examples and other standards. 101 / Advantage f The advantage to systematic analysis of a subject matter ’■ is obvious: it is much bet¬ ter to know precisely when you are missing information from your analysis. Improves Efficiency Throughout Analysis Process Introduction The fundamental change in the basic unit of rhetorical analysis (i.e., the information block) produces many beneficial effects throughout the document development process. We use the term precision modularity to distinguish the information block from "intuitive" chunking where the criteria for chunking are The stages of the process of developing a document Stage 1 Pre-writing Analysis a. audience and job analysis b. initial document analysis and specification c. planning for scope and staffing of project How precision modularity affects this stage ... Enables analyst to obtain information from client to specify more precisely the need, purpose, audience, use, etc. Enables analyst to build job-task matrix initially and to modify throughout the doc¬ ument development process. not clearly formulated or where chunk size may be forced upon the writer by the fixed size of a computer screen. The precision modularity of blocks facilitates technical and administrative communication, improving its clarity and efficiency. Stage 2 Preliminary Information Gathering Analysis a. gather and sort information into preliminary block analysis b. use information type analysis to identify missing information How precision modularity affects this stage ... Enables analyst to identify where information goes in the Content Topic-Block Type matrix. User- Topic- Task Matrix User 1 User 2 < * 2 Task B u J* C/2 a H [I ask A | [Task B | Task C Tnpir 1 Topic 2 Tonic Tonic 4 Subject Matter Expert ii QO sp 5jb pn i Enables analyst to establish need for and to define effi¬ ciently project specific standards. Project Specific Standards Content Topic - Key Block Matrix n — □ HH cj HI — 1 LJ HH m 1 HH i H ttFM Hi Hi H HH m H HI HI 1 H iH Hi n Hi HI r~ si HH Hi IH Hi L-l n Hi iH a H hh m m HI ~i Hi IH Hi LJ ^H Hi H H Hi lj ■1 Hi i H Hi o Hi ra H HI HH HH tim HH IH r- i HI Hi H HI PH PH ■H ■1 LJ HI a Hi iH Hi H HH rn HI LJ ITj Hi Hi ■cfc Hi n HI nn H H m IH □a EH iH Hi Hi Hi HI l l wm IH ■■ Hi n H IH Hi IH IH IH hh IH LJLJ ■i HH IH LJ Hi Hi n LJ Enables analyst to develop specific and relevant questions for the subject matter expert before and during the information gathering process, making that process much more efficient and complete. ??? Enables analyst to identify which information is missing. Chapter 3. Introduction to Information Mapping's Method of Structured Writing Stage 3 Organization Analysis a. determine what information is needed for each task and for each job b. do quality control check for completeness on entire information gathering process How precision modularity affects this stage ... Enables analyst to do a completeness check on all information. User- Topic - Task Matrix User 1 User 2 < * C/D a Task B u C/D CT3 h < C/D C3 [Task B | Task C Topic 1 Tonic 2 Tome 3 T ODic 4 •Topic 5 Enables analyst to filter information blocks through User Topic-Task Matrix to determine precise informa¬ tion needs of each job task. Enables analyst to identify and manage all information in a complex project, and to track changes. i | i 1 Stage 4 Sequencing Analysis a. based on different types of documents and user types, determine sequence requirements b. complete sequencing analysis 1 | 2. How precision modularity affects this stage ... Enables analyst A. to select sequencing templates | B. to consistently and effectively sequence information blocks within maps C. to sequence information maps into larger units, sections, chapters and documents. 103 Stage 5 Presentation Analysis a. use research-based principles to determine presentation format for paper-based delivery b. develop computer screens for computer-based delivery c. develop audio visual delivery L How precision modularity affects this stage ... Enables analyst to determine delivery of information through appropriate media... Standup Training IF Audience / Goals AND Cost / Efficiency THEN delivery media Policies, Procedures Reports, Proposals Documentation Reference Based Training Quick Reference = m Computer-Based Training Programs Video Training On - Line Documentation Hypertext Hypermedia Discourse Domains Introduction How does a report of a scientific experiment differ from a sales presentation or a policy manual? In many ways: who the authors are, how they have come to know the subject matter, what they assume about their audience, what level of detail is used, what content is communicated. These are some of the ways. And how are all reports of scientific experiments alike? How are all sales presentations alike? The analysis of these similarities and differences is what we call domain analysis in Information Mapping's method. It involves examining the relationships between author and reader of different kinds of documents and the "stances" and points of view that can be seen as a result. Definition: Domains of Discourse A domain of discourse is the specification of information blocks of a particular class of documents, all of which share the same type of author-reader assumptions and the same stance or point of view towards subject matter. Examples A training manual differs quite radically from a sales presentation. Both differ significantly from scientific documents. The chart on these pages spells out some of these distinctions which provide the basis for identifying different discourse domains. Domain of Discourse Topic and Stance of the Discourse Examples Typical Documents Relatively stable subject matter We think we know \ enough to teach it. 1 \ T • Well established subject matters (e.g. algebra) • How equipment functions • Training and orientation • Documentation • Procedures • Policy manuals Experimental knowledge We have conducted scientific , experiments to confirm and we would like to present our results and how we 4*‘ came by them to you. ~ ® • New results in physical, biological, and social sciences • Reports of scientific experiments • Abstracts of reports Part of subject matter under debate or consideration We know enough about it to chart disagreements. • Rhetoric of debate • Propaganda analysis • Analysis of systems • Argumentation analysis • Systems analysis reports • Analysis of arguments • Legal briefs 104 Chapter 3. Introduction to Information Mapping's Method of Structured Writing Note: The three discourse domains in the table below are not the only possible domains of discourse, nor is this a complete and exhaustive analysis of the domain. Importance of Identification of Domains An analysis of domains usually finds that there are similar clusters of information block types across many documents of the same domain. This enables us to clarify the similarities of each of the information block types in the new domain and thus to write and to teach the writing of these documents much more precisely and easily. (REH) Reader’s Understanding of the Subject Matter For detailed examples of discourse domains look in this book at... Does not know (most of) the Running through the chapter is a partial subject matter. analysis of this domain. For more examples Knows a lot about the subject matter but not about this particular experiment. An extended example of domain analysis on scientific abstracts May know a fair amount about the subiect matter but perhaps not the „ . , . ... , of argumentation analysis subtleties. May be looking to J change mind about something (e.g., to buy a new product, a new plan or concept). 105 Brief Discourse Analysis Introduction We introduced the concept of discourse analysis as an examination of the components of documents all of which share the same type of author-reader characteristics and the same stance or point of view toward the subject matter of the document. Here we examine briefly one such domain, that of relatively stable subject matter. It is the domain that we have been using as an example frequently throughout this chapter. Definition: Relatively Stable Subject Matters A relatively stable subject matter is one which • has been arbitrarily decided upon by an organization (e.g., administrative policies or procedures), or • results from a design process (e.g., technical information about manufactured products), or • has been agreed upon by a particular scholarly or scientific field and which is taught as the established information in that field. Comment This is the domain of discourse that has been most deeply studied and most widely applied by using Information Mapping's approach. It covers the broad range of high volume "information transfer" documents in administrative, technical and scientific subject matters that are regarded to be known well enough to teach. Block Types for Document Types Most documents in this domain are served by approximately 40 well understood information block types. @ (Stable Subjects) Typical Document Types 1 1 1 1 r 1 i 1 II :J|> m i ■ i i 1 i H i 1 i r | L r i i i m | 1 1 I L 1; E « 1 Procedures Manuals User Guides Job Aids Policy Manuals Operations Manuals Desktop Procedures Equipment Manuals Troubleshooting Manuals Screen Design Standards and Manuals Tutorials Instructor’s Guides Course Administration Guides Training Manuals Textbooks System Manuals Installation Guides Systems Documentation Systems Standards & Functional Specifications Computer Language Manuals Applications Manuals Product Specifications and Descriptions Chapter 3. Introduction to Information Mapping's Method of Structured Writing Different Document Types Have Different Proportions of Block Types If you got together a hundred procedure manuals or a hundred policy manuals, you would see a great similarity of the pattern of the kinds of information block types appearing in them. The Data Commentary What is the value of this kind of data? We can compare content types across subject matters. We can have some idea of what kind of writing tasks to look forward to and plan for. (REH) The two documents described in the charts below have different purposes and hence have quite different clusters of information blocks. One is a procedure manual and the other a mathematics textbook. Key Mathematics Textbook Procedure Manual a» a H Introduction Definition Example Notation Formula x i 5 Diagram Procedure Table Decision Table Stage Table Description Use Comment Other w 0 10 15 20 25 30 35 40 Number of Instances in Document 180 107 Information Type and Block Type Analysis Application: Relatively Stable Subject Matters Another principle that Information Mapping's methodology uses with consistency is to chunk at various levels. We have already introduced the information map as the next level of hierarchy larger than a block. The diagram below shows the general approach to chunking information at different levels. The sentences in these kinds of documents (all relatively stable subjects)... This is the domain of ^relatively stable subject matter. ...can be divided into two types of information for purposes of early analysis phase... Research has found that broad classes of documents have quite different types of information blocks, so it is important to analyze them separately into clusters that have similar blocks. @ v D 00 VO 73 cn X o 3 Basic Information Definition Information which can be regarded as being "new" to most of the users in a given user group description. Supplementary Information Definition Information which is added to the basic in¬ formation to help ini¬ tial learners, refer¬ ences, or browsers to find, understand or leam the basic infor¬ mation. 108 kWhy make this distinction? Because during analysis of a subject matter, ana¬ lysts should concern themselves only with the basic information, identifying it, refining it, and making sure they have all of it. Other information gets in the way of that process. Chapter 3. Introduction to Information Mapping's Method of Structured Writing ...and the sentences (and diagrams and illustrations) of basic information can be sorted into seven information types which can in turn be sorted into block types... Information Types Structure Concept Procedure Process Classification Principle Fact Information Added for Initial Learning Definition ! Types • Overview • Compare and Contrast • Course Objectives • Prerequisites to Course, Chapter, Map, Block • Learning Advice • Review and Summary • Test Questions, Practice Exercises & Performance Situations Initial learning in¬ formation aids learners in their first pass through the subject matter when their goal is to be able to apply or recall. Information Added for Reference Definition Information added for reference aids users to review or look up forgotten material or to find information. Types • Tables of Contents • Indexes • Lists of Notation • Lists of Formulae • Special Purpose Tables • Menus Most Frequently Used Block Types (in domain of relatively stable subject matter) Analogy Block Diagram Checklist Classification List Classification Table Classification Tree Comment Cycle Chart Decision Table Definition Description Diagram Example Expanded Procedure Table Fact Flow Chart Flow Diagram Formula Input-Procedure-Output Non-example Notation Objectives Outlines Parts-Function Table Parts Table Prerequisites to Course Principle Procedure Table Purpose Rule Specified Action Table Stage Table Synonym Theorem When to Use WHIF Chart Who Does What Worksheet 109 Further Types and Distinctions There are obviously finer dis¬ tinctions, criteria, and standards for each of these information blocks that must be mastered to obtain the advantages of the methodology. procedure What are the Information Types? Definition in • for# maotion types (in *far ma ^shan tips) n. 1. the seven basic classifications into which sentences and/or diagrams of basic blocks of information in a subject matter associated with training or educational textbooks, procedure or policy books, manuals, and other similar forms of documentation may be sorted, namely, concepts, procedures, processes, classifications, facts, structures, and principles; each type has certain key information block types associated with it. 2. a classification system into which the information blocks in Information Mapping's method may be sorted so as to ensure completeness and efficiency in the analysis of subject matter. 3. the seven types are sometimes used, in their pure form, to refer to information maps by the same name, e.g., procedure map, process map, etc. Use The information types theory is used to help the analyst identify specific information that is needed for each topic. These information type templates specify the key infor- mation blocks that are needed to ensure completeness and accuracy of the analysis. The templates also ensure a reader - rather than a writer-based document. The information type analysis also guides the specification of feedback and practice questions for training materials. process <& structure concept WB ' - X -1 *— z J classification H — i m n Ti Chapter 3. Introduction to Information Mapping's Method of Structured Writing Definition a set of sequential steps that one person or entity performs to obtain a specified outcome. This includes the decisions that need to be made and the action that must be carried out as a result of those decisions. Example ...first, place the trim piece on the jig, second, take your power tool in your right hand... When to Use... When the reader needs to know "how to do it" a series of events or phases which take place over time and usually have an identifiable purpose or result. ...when the transmission shifts from neutral to first, the following events occur... When the reader needs to know "what happens" a physical object or something that can be divided into parts or has boundaries. ...the spark plug is composed of the following main components... When the reader needs to know • what something looks like • what its parts are a group or class of objects, conditions, events, ideas, responses or relations that • all have one or more attributes in common • are different from one another in some other respect, and • are all designated by a common name. ...acceleration can be defined as... When the reader needs to understand a term, idea or abstraction a statement of data without supporting information that is asserted with certainty. ...the wheel base of this car is 5 feet 3 inches... When the reader needs state¬ ments of data without supporting information the division of specimens or things into categories using one or more sorting factors. ...we can divide this repair manual into the following parts... When the reader wants to organize and qualify a large group into kinds or types based on some aspect of the group a statement that 1 . tells what should or should not be done such as • rules • policies or guidelines • warnings or cautions ...the principle of road safety can be stated... When the reader needs to know about what should or should not be done 2. seems to be true in light of the evidence such as • generalization • theorems 3. is unprovable but implied by other statements, such as • assumptions • axioms • postulates. Ill Key Blocks Provide "Completeness Templates" Introduction To ensure a more complete capture of the subject matter, we have identified certain key chunks of information that go with each of the information types. Definition Key blocks are information blocks which help the writer capture and analyze critical content and present it in the best way for the reader. One or more key blocks are associated with each of the seven information types. Example For those content topics identified as concepts, the key blocks are: • Definition Example Non-Example vO 00 vo JO m X o 3 Getting All of the Subject Matter: Completeness by Successive Approximal 112 Chapter 3. Introduction to Information Mapping's Method of Structured Writing Myth of Completeness All of the subject matter is in the source documents, or, if not, then the technical expert will tell you what you need to know. Our clients tell us that we have all of the information long before that is so. A survey done in Information Mapping's consulting division showed that we never got more than 45 to 50% of the subject matter from the source documents. The technical experts have a different point of view than the user. They look at the subject matter differently. Therefore, they never tell you all of what the user needs. In fact, our surveys show they will only tell you about half of what you need to put in your written communication — if unassisted by analysts using the Information Mapping method. And the half that is in the technical material and the half the expert tells you do not add up to 100%. Often they add up to the same 50%. So the corollary to the myth of completeness is that "Half of what they tell you is irrelevant." SliliBS!!!!! ion ■HfeC ' j tf#l . If! . . . . ■■pig which are then used to... Pllll ; - hh ill - WwA ■ which are used to User— Topic- Task Matrix Topic 1 Topic 2 Topic 3 Topic 4 Topic 5 User I User 2 manage complexity and size and determine completeness of information by comparing with level of detail requirements that come from the user analysis which are used to This procedure, when successively applied to the results of the analysis at each stage, enables us to capture approximately 95% of the subject matter. Typically, we get 80% or better in the first pass, and another 10 to 15% in the second. Then, technical review of the final draft document by technical experts produces the rest. More traditional, less rigorous approaches yield considerably less completeness. 113 Greater Ability to Specify Rule Domains Introduction If you are a writer or an editor, you want to know just when to apply a particular writing guideline. Traditional teaching of paragraphs is often too general to provide precise domains in which to apply a particular guideline. Establishing precise principles for constructing information blocks together with the specific block types for different types of documents has resulted in an increased ability to identify just where and when a guideline applies. How this is different from most conventional approaches to writing and analysis Note how different this approach is from the conventional approaches to writing paragraphs, which have rather vaguely specified guidelines and rules (e.g., have unity and coherence, or have topic sentences). Diagrammatically, we can express conventional writing analysis as a kind of cloudy set of constraints that literally permit almost anything... Blocks do not necessarily have to have "topic sentences." Blocks can be one sentence long if that is what does the job. They can contain tables if that does the job. The ability to specify the subcomponents of the document (e.g., maps, blocks, specific types of information) gives us the ability to avoid sweeping generalizations and make each standard, guideline, or rule count. @ 00 VO m Documents Chapters & Sections Maps Blocks m 1 1 114 Chapter 3. Introduction to Information Mapping's Method of Structured Writing Examples of Quality Control Domains Here are some examples of the domains which we have found to be the most useful in specifying standards, guidelines, and rules for analysis and writing: AU documents of a specific type (e.g. , all policy books, all procedure manuals, ail reference-based instruction manuals) All chapters All sections of chapters All parts of sections All information of a particular information type ... 10 All map titles All maps All maps of a particular kind IF the book is . THEN send the patron . . . AND send . available the book an invoice to the Billing Unit. not available Form • never owned • lost 25 checked out with Form a copy of Form 66 no waiting list 66 to Circulation Desk. checked out with Form 66 and a copy of Form 66 to a waiting list Waiting List Notice Circulation Desk. Whether to use? When to use? How many to use? What kind to use? What contents? What type of label? What linkages? What limits in content? What criteria? What standards? etc. All blocks All blocks with lists in them AH blocks with tables in them All definition blocks All block contents All block labels Definition Cash value is the amount of money a life insurance policy is worth at a specified time. All sentences of a particular type All sentences of a particular type in a particular type of block AH sentences 115 Top Down and Bottom Up Analysis Introduction Most writing projects in the domain of relatively stable subject matter have as their source of subject matter either a subject Jw0 w Qf Ana|yzjng Subject Matter r-S_ matter expert or a pile of documents. The analyst/writer can CI1 L_J proceed in two distinct ways to begin to develop an analysis of the subject matter. Definition Top down analysis of a subject matter proceeds from the top of a hierarchy of information breaking the subject matter into parts hierarchically. Example j To prepare a new procedure book on accounting, an analyst | might look at the project from the point of view of what has to be accomplished by the people following the procedure and then continue to subdivide a task into more and more detail. X o Chapter 3. Introduction to Information Mapping's Method of Structured Writing mmmmrnmsm 2 Bottom Up Analysis Definition Bottom up analysis of a subject matter proceeds by doing an information block analysis of pieces of the subject matter as they become available either from the subject matter expert or from documents. Example One An analyst may interview a subject matter expert (SME) who is the only source of information (e.g., the design engineer on a product or the manager in charge of a new service). The SME may ramble from one topic to another in the interview. The analyst can nevertheless "track" the subject matter acquisition by immediately putting the information into provisional information block assignments which assume that other maps and blocks of infor¬ mation will be obtained later. Thus the experienced Information Mapping analyst always has a dynamically changing framework to attach the subject matter chunks they obtain one by one. I | The analyst can approach a stack of source docu¬ ments with a "fishing net" of block type and content topics, leaving for later the task of initially organizing and identifying what is missing. This enables the analyst to move rapidly and quickly through a subject matter without great repetition andwith great efficiency. Example Two Content Topic - Key Block Matrix 117 Reference-Based Training Definition — Reference-Based Training Reference-based training is a way of arranging training that relies on reference manuals, on-line documentation, and on-line help messages, rather than on instructors or other media, to provide information needed to perform skills. The method recognizes that many tasks do not need to be or cannot be efficiently learned ahead of the time they are needed for use. Rather, users need good reference manuals and the confidence in their skills of finding the information when they need it. A key element of the training design is that the practice exercises during training require the user to look up information in the user guides, reference manuals, and on-line help so that the user gets accustomed to using them and finds them helpful, thereby making it more likely that they will use them on the job after the training. Reference-based training may be self-instructional and used at the work-site or may be partially used in the classroom with instructors Factors That Can Trigger Consideration of Reference-Based Training information that is not used frequently lack of trained staff of instructors geographically diverse group of trainees large number of trainees to be trained in small amount of time training costs must be kept to a minimum replacement training important computer-based system (can use on-line and in paper-based reference) Typical Components of Reference-Based Training Systems On-line Help for Systems provides quick information and training for shorter, low-prerequisite skills Reference Manuals provide all of the information needed to operate the system in readily accessible form Chapter 7 Training Exercises provide a sequence of prac¬ tice situations that ensure that the trainee will come away with skills Exercises Reference Manual Simulation Exercises Tutorials Tutorials provide careful initial sequence of training familiarization with the job and with the reference materials Simulations provide practice situations that are similar or identical to those done on the job Chapter 3. Introduction to Information Mapping's Method of Structured Writing Why is Reference-Based Training emerging as an important alternative? • Software is proliferating — nobody can learn every function of the products they use. • Excellent reference-based training can cut user reliance on hot lines, personal handholding, need for trained instructors. • Contemporary documentation engineering makes it possible to create excellent reference manuals efficiently and consistently. Why is Information Mapping’s method important for Reference- Based Training? Reference-based training only works if the reference manuals are complete, accurate, easy to use, easy to access, task-based, and easy to maintain. The Information Mapping's methodology is the only complete documentation engineering methodology that will deliver this. Benefits of Reference-Based Training To the Trainee • Greater confidence in the system. ..sooner • Less frustration in using the new system • Shorter formal training time (if any) • Greater confidence in user's ability to retrieve and learn To the Training Development Group • Cost savings in preparation of training materials • Cost savings in training delivery • Can use supervisors (or other non-instructors) as trainers For the Organization • Better use of sophisticated software/hardware • Cost savings for the hotline • Increased productivity • Less supervisory time spent correcting errors and answering questions • Increased customer satisfaction • Fewer errors in system use • Lower technical support requirement • Improved morale Important Conditions for Success... V ■ Reference manual must be task-oriented and written from user point nf : I _ I Reference manual must be complete, easy to access, easy to learn from PT ■ 1 Exercises must be close in level of simulation to on-fbe-job tasks On-line help messages must be extraordina¬ rily dear, provide ac¬ cess to more detail and more context Tutorials must take beginner through only the main path, not the detailed ex¬ ceptions 119 S) 1989 R. E. Horn The Mapping Metaphor: Subject Matter Structure Introduction One of the important things about Information Mapping's analysis method is that it follows the underlying structure of the subject matter, much like a geographical map follows the exact contours of the terrain. Geographical Maps Show Point-to-Point Correspondence On geographical maps, the most important places have the most prominent graphic features such as the heaviest type. Major superhighways and airports are easy to spot. The smaller roads are indicated by smaller lines and the smaller towns are in smaller type sizes. We can follow the larger structure of the territory without getting confused by the detail. On a geographical map, the shape of the territory is shown by a similar shape on the map. If the city streets or the airport runways are laid out in a rectangular fashion, then that regularity is represented on the map in a point for point correspondence. Chapter 3. Introduction to Information Mapping's Method of Structured Writing Information Maps How do information maps and their components, information blocks, follow the terrain? What is the terrain? The contours of a subject matter are individual concepts, structures, organizations, processes, procedures, etc. To have a correct "map" of such a territory, you have to have (metaphorically) the right viewpoint and the right way of representing the subject matter. And you must follow the contours of the subject matter. If it is a very regular one, then the result of an analysis is straightforward. If the contours are uneven, then the information maps must follow this structure. Information Maps show point-to-point linkages to... he underlying organization of a subject matter Meeting the Criteria for Better Communication Commentary: The individual user and the organization have different criteria for a method of writing. On these pages, we present some of the most important criteria that have come out of our research on Information Mapping's method. For discussion , - h-sj Notes Criteria of the Individual Relevance to Purpose of Communication iCTT Does the document contain the right information for what I need to know or do? Completeness frit Does the document contain all of the information I need? Accuracy Is the information in the document correct? Comprehensibility j rif Is the document written so that I can compre- L^hJI hend the information easily and quickly? Learnability M Is the document written so that I can learn what I |y| need optimally when I am completely new to the subject ? Accessibility and Easier Scannability -M Is the document organized and formatted so that [A/I I can easily and quickly find what I am looking for without having to read everything? Improved Decision Making | Is the document structured so that I can make s2ShJ| better decisions? Particularly, are the recom¬ mendations, supporting data, and evaluations immediately available? Improved Analysis £ Does the document help me analyze my pro- |V| blems and help me to spot easily what is not there? Fewer Errors Does this document help me to make fewer per- \VJ formance errors? Improved Electronic Filing and Retrievability Does the document help me retrieve what I'm '-mJ looking for more quickly and easily? Improved Creativity by Reducing Routine Work » Does the document help me think more cre- iy/1 atively by reducing the amount of brain work I have to do to figure out what is there and how to structure it? Chapter 3. Introduction to Information Mapping's Method of Structured Writing The Organization's Criteria for a Communications Method Increased Revenues and Profits Does it contribute to our sales and profits, and lower our overhead? Lower Product Support Costs Does it help us lower what it costs us to provide good maintenance and excellent customer support? Optimally Effective and Efficient Communication and Training Does the method help us be more effective in the human side of business -- communication — and does it do so efficiently? Lower Training Costs Since our products and services are changing so much more rapidly, does the method help us keep training and retraining costs in line? Maintainability Are the documents developed such that, if information needs to be added, deleted, or changed, this can be done efficiently and effectively? Teachability for Group Use Can the method be taught quickly to a level that different members of a group can work together to compose a document? Comprehensiveness Is the methodology appropriate for a large number of different kinds of documents? 123 Chapter 4. Navigating Structured Hypertrails Overview of This Chapter 126 Prerequisite Hypertrails 128 Classification Hypertrails 1 30 Chronological Hypertrails 132 Geographic Hypertrails 134 Project Hypertrails 136 Structure Hypertrails 138 Decision Hypertrails 140 Definition Hypertrails 142 Example Hypertrails 144 Hypertrail Webs into Linearized Sequences 146 © 1989 R. E. Horn Chapter 4 Navigating Structured Hypertrails lillilill S-: ^l|| Geographic Hypertrails Prerequisite Hypertrails Structural Hypertrails Classification Hypertrails In this Chapter ...we ask: "What are some kinds of generally useful linkage structures in hypertext? ( _ ) is a Definition (_)witha Hypertrails ( _ ) iSx-x-ixSxSxS: SiSiSxSiSiixSi illlllllillli 111:!"" SSSSs, siswi, iliilil iliili rnmrn . . . is iPii: Decision Hypertrails Project Hypertrails Example Hypertrails polygon A A 125 Overview of This Chapter Definition: Hypertrails A hypertrail is a set of links between chunks of information, such as units, chapters, articles, books, or courses (and in the context of Information Mapping's method, blocks and maps), that organize and sequence information about a particular function or characteristic of subject matters. Use Hypertrails enable users to take different, yet structured, paths through hypertext knowledge bases. Usually, there is a dominant trail for use of particular kinds of users such as initial learners or referencers. Multiple hypertrails have the advantage of providing both structure and the free-browsing modes of knowledge base use. Making Hypertrail Networks Linear: Definition Linearizing hypertrails is a process of selecting from the networks of links a series that makes sense as the primary hypertrail for the user who selects a command to be guided through the detail of the hypertrail. Prerequisite Hypertrails Classification Hypertrails Chronological Hypertrails Geographic Hypertrails ■ | see page> 134 Chapter 4. Navigating Structured Hypertrails Commentary Other hypertrails have been identified. But these will serve to provide a good overview of the functioning of hypertrails when used with information blocks and maps. (REH) Prerequisite Hypertrails Introduction Prerequisites are one of the basic ways of organizing information for users of hypertext learning systems. Definition A prerequisite hypertrail provides a set of linkages between information maps, information blocks (or other larger chunks of information, such as units, chapters, articles, books, or courses). These connections specify which maps learners must understand (or which tasks they must be able to do) in order to understand more advanced topics or accomplish more advanced skills. Example Block ^ XI Block Block Block Block Use The prerequisite hypertrail is typically used to sequence chunks of text in a linear way so that the learner may always be sure of having encountered significant information needed to understand the present information. Thus, if the learner has informed the system that he or she is an initial learner and knows little or nothing about the subject matter, then the text in the system should be linearized using the prerequisite hypertrail linkages. The prerequisite hypertrail is of most use to the user who: 1. 2. 3. 4. is interested in the prerequisite structure of a subject matter, OR has never studied the subject before, and is a serialist learner A OR has never studied a similar subject before, OR is interested in th&'5^ost efficient (i.e., straightest) path through the material. VO 00 VO ? O m X o 3 Two l evels of Prerequisite Analysis Course Level Prerequisites Example Here is an example of a prerequisite hypertrail at the level of courses in math and physics: 128 Chapter 4. Navigating Structured Hypertrails m/* h mm mm Concept Level Prerequisites WmMmmsmi WmMm WffimmmMm , , Example In analyzing the beginning concepts in the topic vector mathematics, we find these prerequisite relationships: This ASPECT has these concepts as PREREQUISITES Displacement • Distance • Direction Vector • Direction from a point • Magnitude Magnitude • Number • Unit of measurement Fairness to Learners Principle Synonym The Principle of Guaranteed Access to Prerequisites Introduction "If ya’da told me that first, I woulda understood what you were talkin’ about." Almost everybody has an intuitive notion that you have to learn some things before you can learn others. Too often in textbooks and training manuals, readers are stranded at mid-text by the introduction of concepts they are expected to be able to understand without ever having previously encountered the concepts, the words or experience. Principle We take as a basic principle for sequencing initial learning materials the principle, called "Fairness to Learners Principle." It says: Learners should have previously encountered the prerequisites to all concepts they are presented with in learning materials, or should have immediate access to that prerequisite information. Implementation Access to all major prerequisites is provided by 1. developing a prerequisite hypertrails that can be activated by learners if they want that kind of sequence through the text, and 2, providing hypertext linkages from the currently displayed text to prerequisite information such as where terms are introduced. 129 Classification Hypertrails Introduction Classification is one of the basic ways of organizing information. It uses the basic principle of grouping similar things together into classes and distinguishing these classes from other classes by differences. The ability to follow and display large classifications is an important characteristic of efficient hypertext systems. Definition A classification hypertrail is a set of linkages in a hypertext database that enables a user to find any linkages higher or lower on a classification tree for a particular subject display a classification structure of a given hypertext region. Example of a Classification Hypertrail In this example we show maps about the topic of "scientific models" Models Mental Models Symbolic Models Physical Models @ vO 00 VO 73 PI X o 3 Verbal Models Mathematical Models 130 Classification Hypertrails Shown as Links Between Maps Enacted Models Analog Models Mimetic Models Mental Models Chapter 4. Navigating Structured Hypertrails Classification Tree of Models for Biological and Social Sciences All Models Note: We show only a portion of this tree structure here. Physical Models Symbolic Models Enacted Models Other Mimetic Models Analog Models Verbal Models Mathematical Models Components Types Constants Variables Parameters Mathematical Simulations Equations Exogenous Endogenous Type of Predictability Deterministic Probabilistic Micro Level of Aggregation Macro 131 Chronological Hypertrails Introduction Chronological hypertrails resemble the familiar time lines. But they go beyond these graphical tools in organizing large amounts of data. Definition Chronological hypertrails are linkages of nodes that organize information with time. Example of a Chronological Hypertrail Development of Hypertext Ideas about 1948 First use of computer-stored text 1277 Hugo de St. Caro Initial and 500 monks use Concordance for of the Bible indexes for books. 1965 Nelson suggests term "Hypertext" 1962-75 Engelbart invents or first implements • hypertext software • word processing • outliner software • mouse • windows Vannevar Bush suggests the Memex machine see page> 254 1499 Routine use of page numbers for books Aldine Press Aldus Manutius (1450-1515) Venice ® SO 00 SO & pi Throughout the Middle Ages Widespread use of commentaries written on the margins of books Note: We show only a portion of the chronological hypertrail here. Chapter 4. Navigating Structured Hypertrails Three Kinds of Chronological Hypertrails ^ g ^ Sequence of Events Hypertrail Description A sequence of events hypertrail follows some time measurement, such as Example • Time line of events in a Presidential campaign • Sequence of events in a complicated industrial manufacturing process Natural Development Hypertrail Description A natural development hypertrail follows the sequence of development of a particular process or system. Example • Evolution • Development of an organization Storyline Hypertrail Description A storyline hypertrail tells a sequence of occurrences in the life of a particular person or group of persons. Examples r • scenarios • stories • docudramas Narrative has been classified by literary critics in many ways. Here is just one such classification: Three Kinds of Stories ^ ^ 1. Adventure 2- Discovery or 3. Struggle with Exploration Adversity Two Kinds of Adversity Stories 0 Struggle with Internal Forces (psychological) B Struggle with External Forces Comment In hypertext, branching can produce stories with different trails. 133 Geographic Hypertrails Example of a Geographic Hypertrail Globe to local region to Introduction One of the major ways that we organize information is spatially. We draw maps to help us get around in space. We use word descriptions of how to get from one place to another and to describe the contents of geographical space. Geographic hypertrails link these information blocks. Definition Geographic hypertrails link together descriptions and maps of geographical information. Contrast with Structure Hypertrails Note that one of the other hypertrails, structure hypertrails, is closely related to geographic hypertrails. The major difference is that in geographic hypertrails we are linking spatial relationships between different structures. In structure hypertrails we are linking the subparts to the larger structure. Here is an example of a geographic hypertrail that zooms in on the White House starting from a look at the earth from space. 134 Chapter 4 Navigating Structured Hypertrails the White House to ...the Oval Office nr nt >C-t I It iji UJL 135 Project Hypertrails Introduction One of the ways we frequently organize our work is by projects. We think of projects as work organized around a specific goal that will take a period of time longer than a simple task. Hypertrails should follow natural work linkages, so a project hypertrail becomes a necessity. Definition Two Kinds of Display Metaphors for Project Hypertrails Project hypertrails are specific kinds of chronological hypertrails that link planned and past events all focused on a personal or group project. Example of a Document Event Network Hypertrail In this example we show project hypertrails for a survey research project : 0> 00 >£> 73 Prt SC o 3 136 Chapter 4. Navigating Structured Hypertrails Example of a Document Event Network Hypertrail In this example we show project hypertrails for a survey research project : Report □ / Presentation for Sponsor Develop Objectives Get Sponsor's Agreement Develop Formal Hypotheses Define Universe Develop Sampling Plan Develop Presentation for Sponsor Make Presentation to Sponsor Develop Pool of Items Develop Selection of Items Plan Develop Form for Survey Try Out Form for Survey Revise Form Develop Administrative Procedures Select Interviewers Schedule Interviews Conduct Field Interviews Conduct Follow-Ups Code and Input Data Run Statistical Programs Interpret Results Write Report Make Presentation to Sponsor 137 Structure Hypertrails Definition: Structures Structures are physical objects. They have boundaries and occupy physical space. They have parts. A useful hypertrail links the parts to the large structures. Definition: Structure Hypertrails Structure hypertrails link specific substructures described in information blocks to the larger structure. A user can begin searching a structure hypertrail from any part of the structure or substructure in the hypertrail. Boundaries: Alternative Way of Defining Structure Hypertrails A variation on the standard structural hypertrail is a structural hypertrail that links a structure by its physical boundaries, i.e., by the name of the boundary lines or by the boundary itself (where the two structures or substructures meet). Example of a Structure Hypertrail Here is a structure hypertrail of a wind tunnel arranged names of the subassemblies: by Wind r \mnel Shed Wind Driver Control Equipment Here is a structure hypertrail of a wind tunnel by picture of the equipment subassemblies: Support Blade Assembly Motor Blades Rotor Assembly Rotor Shaft Rotor Shield Bearing Assembly Buttons at Each Node ® A button can be put at each SO node on these trees so that 00 so more detail comes up on the ?° screen when a position is w clicked. ^ X o see page)> 8 CD c CCO 138 Chapter 4. Navigating Structured Hypertrails Decision Hypertrails Introduction Organizations are shaped around processes for making decisions. They must have orderly ways to prepare for, make, implement, and document decisions. In the framework of hypertext systems using Information Mapping, decision hypertrails help track all information about a given decision. Definition: Decision Hypertrails Decision hypertrails link all of the information (blocks, maps, or documents) about a particular decision that a person (or organization) has made (or is in the process of making). Layering and Structuring Required To make a reader's job manageable, decision hypertrails need to be layered and structured. That is, they require the specification information blocks and maps to provide the components of the trail. Types of Decisions There are many kinds of decisions and many "sizes" of decisions. The approach to assembling and displaying information for each type of decision must follow an analysis of these kinds of decisions. We do not go into these details in this book. Example of a Decision Hypertrail Following is an example of a decision hypertrail linking documents in the purchasing process of a large organization: 140 Chapter 4. Navigating Structured Hypertrails Example of a Decision Hypertrail A policy decision hypertrail could display links between relevant documents at different "layers" of information about that decision. Background Summary Our Objectives in Area Political Economic Defense Scientific Option 3 Option 2 Option 1 Premise — General Approach Planned Programs Planned Actions Contingent Actions Pros Cons Geography Industrial Military Technological History — — — — — — — i —1.. i ■ . _ i — =*■ ===== — !!=!■ , ... — — Demography Trends - - — — — — ===== - — 141 Definition Hypertrails Introduction Most words are ambiguous. They have more than one meaning. For the 500 most-used words in the English language, the Oxford dictionary lists 14,070 meanings. Thus, each has an average of almost 30 meanings. So, dictionaries have proven their worth over the centuries. Using definition hypertrails within the context of an Information Mapping hypertext knowledge base will prove to be a significant advance for dictionary users. Definition: Definition Hypertrails A definition hypertrail provides links between the different meanings of a single term in a hypertext document, or between related terms in one or more documents. Context designators for each of the meanings are provided, as well as examples of the use of such words in context. Comment The process of creating definition hypertrails is quite easy since all definitions in Information Mapping are segregated into separate information blocks and are distinctively labeled. The dictionary hypertrail approach makes a particularly useful form of dictionary since the other links (such as example blocks) are readily available. Obviously, it may be useful to load a large conventional dictionary into the knowledge base as well, and link it to the growing definition hypertrail. Example of a Definition Hypertrail In Chapter One of this book, for example, you can find these definitions that comprise a definition hypertrail: • hypertext • links (hypertext) • nodes (hypertext) • buttons (hypertext) • system-supplied links • user-created links • author-created links • semantic nets • branching stories • relational databases @ • simulations commentaries • kO 00 kO • anthologies pa pn X o 3 • hypermedia 142 Chapter 4. Navigating Structured Hypertrails Example of a Definition Hypertrail The blocks below illustrate taking definitions from several related information maps to form a definition hypertrail: Models Definition Models are abstract ways of representing the experience of our inner and outer worlds enabling us to manipulate and sometimes predict from the representations rather than from the systems they represent. Mental Models Definition Mental models are representations consisting of ideas and concepts in the human mind. Physical Models Definition Physical models are models constructed from tangible material. Mimetic Models Definition A mimetic physical model resembles in physical characteristics a system it represents, but does not necessarily behave like the system. Analog Physical Model Definition Analog physical models are made of concrete, tangible materials that are built to act like a real system even though they may not resemble one. Simulations (Using Models) Definition A simulation consists of the construction of a history of the changes in state of a previously defined model. Enacted Models Definition Enacted models are processes in which people perform the roles and actions of systems by making decisions and taking actions through time with a particular outcome. Simulation Game Models Definition Simulations are simplified models of reality so that reality can be better understood. Games are competitive interactions among participants to achieve pre-specified goals. Simulation games are simulations in which participants act out roles and make decisions as if they were making them in the actual or real world situation. Cooperation among participants and teams as well as within groups may also be properties of such simulation games. 143 Example Hypertrails Introduction Some text is organized so that there are one or more running examples throughout. Different topics are introduced and defined, then examples are presented. Some examples are "extended," i.e., they exemplify many of the topics made. Definition: Example Hypertrails An Example Hypertrail is a linking of • all of the different appearances of a single extended example that appears in a single document • all of the specific appearances where the same example is used in different places in different documents. This permits the user to request of the system: "Show me all of the places where this example is used." Example One In a book on dream theory and dream interpretation, specific dreams are the examples. Each dream has a name (e.g., Grandma on the Ceiling, My Boat Sinks, etc.). The full dreams are described the first time they are introduced. Parts of the dreams may be requoted in specific sections (e.g., in sections on "symbols," or on "different ways of interpreting dreams"). The dreams may also be referred to by name in other discussions. Clearly a useful hypertrail would be to link all of the appearances of a given dream on a single trail and to be able to link all of these individual example hypertrails into a large master example hypertrail. Hypertrails like this were used by Walter Bonime in his book, The Clinical Use of Dreams. In the book there is a "dream index" so that you can go look at all of the places where the dream occurs. Example Two In many business procedure manuals, we use the same example in a procedure that runs several pages. These individual examples linked together in an example hypertrail form a case study of the use of this procedure. @ oo VO 50 m x o The arrows identify the appearance of the same example hypertrail in a sequence of different Information Maps. 144 Chapter 4. Navigating Structured Hypertrails /■ g/t " < ¥,S'S %**/> y c. tf'p/VM'-X />, ««**&■•) <■>'-■•✓ yrx^ s <*/ vm v'v/s- « "w/w* ., , Example of an Example Hypertrail Here, we show how the description of a single simulation game about political process at the local level can be used as an example on several different information maps. All of these maps together provide the procedure for making an educational simulation game. Note that only the map title and an example block are presented in this example, not the whole map. 1. Define the Problem Area to Be Simulated Example The educational simulation selected is a local political process. 2. Define the Objective and Scope of the Simulation Example Objectives: To understand how the mayor and city council get elected in our city. Scope: Our major interest is in examining the electoral process at the local city level and to see the other influences of other local organizations on the selection of city officials. 3. Define the People and Organizations Involved Example Organizations: Political parties, government departments, the local newspapers, television, radio stations, community action organizations, etc. Roles: Incumbent mayor, other mayoral candidates, president of city council, powerful local police chief, state political party chairman, heads of local political clubs. 4. Define the Motives and Purposes of the Players Example In the local political elections game, the first goal of the incumbent mayor is to get reelected. The motive of the state political chairman is to see that the candidates loyal to the party and who have a good chance of winning are nominated. 5. Define the Resources Available to the Players example not used for this map 6. Determine the Transactions to be Simulated and the Decision Rules to Be Followed Example "1° the local elections game, the first playing period might cover a pre-nomination period of a few months, where potential candidates for two parties lobby for the nomination. In the second playing period, players gather at conventions to nominate candidates (or perhaps primary elections are held). In the third playing period, players organize and hold a campaign, which is followed by the fourth and shorter period in which the election is held and the winner determined." 7. Formulate the Evaluation Method example not used for this map 8. Develop Simulation Games Prototype example not used for this map 9. Try Out and Modify the Prototype example not used for this map 145 Hypertrail Webs into Linearized Sequences Introduction Because human beings live in a linear, time-sequenced world, we must always have some "next" event in our lives. In hypertext that means whatever the button is called, it is always in some sense a "next" button. If the structure of the subject matter is a two-dimensional or multi-dimensional network or web, we must nevertheless follow some next link in the net. We must go to some next node. This raises the question of how we shall linearize the nodes of hypertrails, because there is always some limit to the size of the web that can be displayed on a single screen. Definition: Linearized Sequences for Hypertrails A linearized sequence for a hypertrail puts each of the elements of a hypertrail into a sequence in such a way that the user can be shown some next information. As an automatic facility of a hypertext system, it must also provide organizing elements of a document for the user, such as a table of contents, index and other "maps" of the structure of the subject. Example of a Linearized Sequence Here we show a structure hypertrail of a series of 146 Chapter 4. Navigating Structured Hypertrails 4. 147 Chapter 5. Resolving Some Hypertext Problems Overview of This Chapter 150 At the Nodes, Blocks and Maps Structure Hypertext 152 Clustering Documents From Different Domains 154 Addressing Lost in Hyperspace and Overload 156 Addressing the Major Reading Cues Problem 158 Addressing Creation and Maintenance Issues 160 Addressing Group Analysis and Writing Issues 162 Some Navigational Options 164 148 1989 R. E. Horn Chapter 5 Resolving Some Hypertext Problems Group Writing Issues Clustering Maps into Documents Reading Cues Issues In this Chapter ...an overview of How Information Mapping's Method Resolves Some of the Hypertext Issues Information Maps as Nodes Lost in Hyperspace Issue Linking Information Blocks Cognitive Overload Issue Interface Issues x:x:x:x-:'x Overview of This Chapter How Information Mapping's Method Addresses the Major Hypertext Design Issues The three major system design issue areas we introduced in Chapter 2 can now be looked at from the perspective of Information Mapping. We will look at nodes, links, and buttons and ask how Information Mapping's method would deal with each. The Three Major Categories of Design Issues Outlined in Chapter Two The fundamental questions about nodes we suggested Chapter 2 were: Nodes What shall the nodes contain? What principle shall we use to determine contents of nodes? These were summarized as "granularity issues On what basis should size decisions be made? What size specifically should various chunks be? Is there any systematic way to determine "natural divisions of a subject matter that will help us? Links The fundamental questions about links we suggested in Chapter 2 were: • Which kinds of links to implement? • How many links should one use? • How can we implement different hyperlinkage networks of the same node? • How shall the links be represented? | see page These were summarized as fundamental organization of documer to provide context and meaning questions. 00 m X o 3 The fundamental questions about buttons we suggested j'seepage in Chapter 2 were: v*'" ♦ What kinds of buttons should be used? . Buttons These were summarized as human interface issues. Where do good interface design principles suggest that buttons should and should not be put? How do we prevent users from being overwhelmed by the number of buttons? How to distinguish different kinds of buttons? What should be the role of graphic icons and words for a particular kind of button? 46 150 Chapter 5. Resolving Some Hypertext Problems How Nodes Links Buttons Information Mapping's Method Addresses these Issues: . . see page Information Mapping 's method suggests that information blocks A""are to be defined as the fundamental nodes in the Hypertext network. The information block is the best way of defining the node because of its properties of providing meaningful precision chunking of relationships between sentences. The rhetorical guidelines and standards for constructing blocks and larger units called maps provide the detailed, well-tested approach to analysis and writing hypertext. The block as the smallest node level is probably sufficiently fine-grained for user commentary even if the comment is about a specific single word in the block. Information maps A, .form a second layer of fundamental nodes. These maps are a clusters of'fclpcks about a related topic that (in general) should be displayed together . blocks with which they are associated. see page Links are connections between map na Blocks can be linked together to form larger nodes called >iinfpjmatjx>n.-fnaps"and''th'ese nodes are linked together in hierarchical order to form chapters and larger documents such as reports, textbooks, manuals, etc. Other links connect maps to chapters, chapters to larger documents. 104 . . see page_ Discourse domains A .provide' the framewol^m— ^-Jifying types of blocks needed for different messages and documents in business, science, and technology. We show a full example in Chapter 8. In addition, structured sets of hypertrail linkages A such as prerequisite and classificatory linkages provide important connections that give ameaningful point of view to the subject matter. These Information Mapping and hypertrail linkages are regarded as fundamental. Other links such as comments, critiques and rebuttals are provided under the rubric of argumentation structures. — - \ Within the structured writing context, specific types of buttons A will provide an orderly, familiar, useful, and general way of navigating different discourse domains. At the Nodes, Blocks and Maps Structure Hypertext Introduction Information Mapping's method provides a powerful and well-tested method for precision modularity that is suitable for defining the size and content of the information at the nodes of hypermedia. On these two pages we show how the blocks and maps appear in hypertext. Examples of this abound in this book— in fact the whole book is an illustration of this concept. Definition: Structured Hypertext Structured hypertext is text that is written according to the methods and criteria of Information Mapping's method. The term is intended to distinguish the text from various forms of partially structured or relatively unstructured free association writing. Example Here is a schematic of the map on the opposite page Related Maps Regular Data Values, and. Introduction ■ . 11 in Definition Non-Regular Data Values 152 Chapter 5. Resolving Some Hypertext Problems Comparing 17.1.0 Regular Data Values, and . . . Non-Regular Data Values Introduction Some data have patterns. They progress by fixed increments. Some data do not show any pattern of intervals between the values. Definition Data are called regular when the values of a data vector progress from some initial value with some fixed interval to another value, and then optionally from that to still other values by even increments. Data are called "non-regular” data when they have no systematic pattern of intervals between them. Example One Time data show frequent regularities. Samples of blood collected from a laboratory animal every hour on the hour might be called SAMPLEHRS and might look this way: Most measurement data do not exhibit systematic regularities that are fixed intervals between values, so they are usually non-regular data. Here is an example: SAMPLEHRS = 6, 7, 8, 9, 10 LABMEAS - .01, .09, .04, .3 Use This Input Statement Input with Computed Clause Statement Standard Input Statement Comment This statement permits you to input regular data in a very compact form and is much quicker to type than a normal input statement. This statement should be used for normal data entry. Related Pages Input with Computed Input Statements, 22 Standard Input Statement, 21 Variables, 19 153 Clustering Documents From Different Domains Introduction With information blocks and maps at the nodes structured hypertext has a form that is considerably more useful than many other possibilities. On this page we show how different forms of blocks, maps, and documents link together to form documents from different domains. On this page, we show schematics of documents from three such domains. Chapter 5. Resolving Some Hypertext Problems Document from the Domain of Subject Matter Under Debate and Consideration Shown here is a schematic of an argumentation analysis. Title of Argument 0^^—— . —— »— — ■» TT" " ,..11 1 1 claim counter-claim grounds warrant Title of Abstract n purpose - — u - —p method nr ^ background results n f __ n 1 ICbUllb new techniaue [ II =Z ■ ll n | treatment cjroups| [P conclusions P controls limits of study data subjects implications author inflations title of publication | y Addressing Lost In Hyperspace and Overload Introduction We noted in Chapter 2 A-Jhat two of the major user issues in the current use of hypertext were-prpblems of knowing • Where am I? • Where have I been? • Where am I going? • What are my options? These questions have been generally known as the "lost in hyperspace" problems. In addition, readers have been overwhelmed by too much information and constantly bombarded by overchoice. Information Mapping's approach significantly reduces these two problems. On these pages we will describe how Information Mapping's approach reduces these two problems to manageable proportions. Problems of Being Lost Problems of being lost are both local (that is they have to do with the page or screen one is currently looking at) and more general (having to do with what chapter or what document I am looking at). Overchoice Some users have been overwhelmed by the inability to track tiny chunks of information. Chapter 5. Resolving Some Hypertext Problems How Information Mapping Prevents Many of the Overload and Lost in Hyperspace Problems The standards and guidelines of Information Mapping's structured hypertext method introduced such items as • uniform careful chunking • uniform careful labeling of each component • hierarchical structuring and titling of larger structures • explicit hypertrails of different kinds • regularly provided overviews, introductions and summaries • consistency and relevance in all titling and labeling • table of contents that are created from the labeling and structuring • important limits as to the kinds of linkages permitted (although readers may — in some software implementations — insert any kinds of linkages that are not exhibited to new users unless the new user asks for them) • similarity of structures across subject matters for different kinds of discourse domains. The method, while explicitly requiring chunks of information, also provides at all times an explicit context for linking them and a presentation method for assuring that the reader understands the context of the chunks. 157 Addressing the Major Reading Cues Problem Introduction {"seepage >ag^>48 We have already noted that the very nature of hypertext, its links and buttons and the ability to jump from one place to another, may provide many readers, especially poor readers, with more difficulty than they have with ordinary text. Hence, their ability to learn from hypertext may be diminished. Because we want to be able to use the advantages of hypertext without losing the coherence and discourse cues of normal text, we have suggested Information Mapping’s approach. On these pages we show specifically how Information Mapping's method provides solutions to many of the problems that hypertext raises by destroying or disrupting normal discourse cues. Discourse Cues That Hypertext How Information Mapping’s Method Deals Destroys or Disrupts With These Difficulties Hierarchical Text Organization Some research suggests that readers build hierarchical frameworks in their minds as they read. Discourse cues, such as outlines, patterns for subheadings, and tip-off words such as "initially, next, finally," etc. which provide clues to the structure of the text, are disrupted by readers following hypertext linkages. The Information Mapping's method introduces systematic ways of outlining, precise guidelines for providing headings and subheadings, and in general provides significantly more chunking and hierarchical structure to the presentation of text. Moreover, since Information Mapping's method is similar across subject matters, the reader is familiar with what to expect when moving to new content. Explicit Transitions Readers traversing hypertext networks run into transitions which link back to structures which may be difficult to find. since we have thoroughl; discussed this, we shall not go into the details here.. Transitions in Information Mapping's method are placed explicitly in highly visible places such as in introduction blocks and overviews. These transition locations are governed by specific guidelines as to how frequently they must appear. XL Overview XL roduction <\! VVf1/ !/ 3 > Sequence Signals Normal relatively unstructured text may contain signals about organization, such as "there are four types of . . .," but readers traversing hypertext webs may find themselves in the middle of a text which says "fourthly . . ." @ VO oo vO 73 tfl X o n fourthly. Information Mapping's method highlights every sequencing signal by explicitly making visible the structure of the document through a carefully designed framework of labels for all major portions of the subject and by making visible sequencing signals such as "types of, kinds of," etc. The reader can see in this book many of the ways Information Mapping's method accomplishes this. Example Four Kinds of Classifications a cfi" b b & 158 Chapter 5. Resolving Some Hypertext Problems Discourse Cues That Hypertext Destroys or Disrupts How Information Mapping's Method Deals With These Difficulties Contrast and Similarity Cues Discussions of similarities and differences may be scattered over large areas in con¬ ventional text. Comparisons may cover several pages and the reader may jump into the middle of such a comparison. in contrast- — similarly, . S Compare and contrast is done with a tabular arrangement. Either the table is put within a block, or if the compare and contrast is extensive, a complete map A will be used to do the comparison. \ Pronouns as Cohesiveness Cues for an example Conventional text uses pronouns (they, he, she, we) to refer the reader back to the material the reader is assumed to have read because linear reading is assumed. Readers in hypertext may arrive in such a sequence and not know to what the pronouns are referring. The structured nature of Information Mapping's method and the requirements of its guidelines for constructing blocks to be as self-contained as possible reduces the reference of pronouns outside of, usually, the block which is currently on display and never outside the information map. Therefore Information Mapping’s method avoids this problem. a He.... ....she... ...they... Metaphors Conventional text sometimes has an extended metaphor running through many pages of text. It provides a useful organization through text but provides difficulty for readers who arrive via hypertext link into the middle of such a text. Precise analogy blocks permit the use of extended metaphors because they can be linked by an extended example or hypertrail. Content Schemas Conventional text generally has an organization which is not entirely evident to the reader. So a reader arriving in the middle of such conventionally written material does not have sufficient cues to orient to the overall structure of subject matter. Information Mapping's method, because of its always explicit structure, enables the reader to see the major hypertrail which is organizing the content. And with appropriate facilities in the software, the reader may also be able to see other hypertrails that are available. Addressing Creation and Maintenance Issues Introduction We have already noted Athaf„}iypertext generally increases the amount of intellectual labor needed-tQ,, create and update knowledge bases. On these pages we make th'&"case„J;hat, just as Information Mapping's method has reduced the cosf'Of producing text on paper, it will vastly outdistance less structured approaches in the creation of hypertext knowledge bases. Careful Limits to Text Preparation Information Mapping's methodology limits the amount of additional text that needs to be done to only that which is absolutely essential and specifiable by the well tested guidelines of the methodology. This analysis is complete. Specify Necessary Links Information Mapping's methodology specifies the particular places and facilitates the creation of specific links . Rational Boundaries on Linkage Creation While we have indicated that there is considerable cost from proliferating linkages, the specifications developed for Information Mapping's method limits these in commercial situations and will provide hierarchical document structures for hypertrail linkages. This puts required boundaries around the problem of overlinkage. Permits Greater Automation of Link Creation The careful definition of the information blocks and information maps will permit higher automation of linkages than would otherwise be provided in automating less structured text. 'C 00 70 m 35 o 3 WM : ■ ills Please wait a few microseconds while I create some more links 160 Chapter 5. Resolving Some Hypertext Problems Additional Quality Control Requirements Information Mapping's method, because of its specification of quality control guidelines, makes the quality control process of creating text much more manageable. We have noted that the precise definition of components of the text permits establishment of readily determinable standards. As the reader has noted in previous chapters, the method places a very strong value on managing the size of the message using the "seven plus or minus two" rule of thumb t9 limit the size of chunks. The method also recommends using strict hierarchies so that every block in a document has an identifiable place. Rapid Cost-Effective Indexing Information Mapping's method will provide the opportunity to do "quick and dirty" indexing of titles and labels, rather than full text. When these have been written according to the guidelines of the methodology, the indexing provides almost the level of quality of a professional indexer. The additional discipline of labeling and titling blocks and maps, thus, produces significant advantages in the indexing process. Modularity and Rationality Aids Data Base Maintenance Because of the careful specification of types of information that belong to specific discourse domains, and because of the modularity of all material written with Information Mapping's method, the updating and revision task becomes much more manageable, efficient, and effective. It also makes following branches and links that are connected to particular nodes rational. 161 Addressing Group Analysis and Writing Issues Introduction Managers have begun to recognize the importance of the many situations where many people have to contribute ideas and actually write portions of a single document. Proposals, plans put together jointly by several departments, task force reports, and documentation systems written by different design groups all share this requirement. As a byproduct advantage, Information Mapping's method provides a framework at the right level of detail for addressing many of the major problems of groups working together. Major Issue A major difficulty in group writing projects is that members of the group do not share meanings of key concepts. The method offers assistance in this area. How the Information Mapping Method Helps Groups to Work Together Stage 1 Prewriting Stage Enables all in a group to agree on goals, scope, etc. in advance. Provides a framework for discussion of the implications of these decisions for the document. Stage 2 Preliminary Information Gathering Different assignments are easily made for different information gathering and preliminary information block writing. Enables different team members to work on different parts with little overlap. Greater assurance of completeness. Stage 3 Organization Analysis Can more easily identify quality control issues. Can see what is missing, what needs to be done. Chapter 5. Resolving Some Hypertext Problems Commentary: Problem Analysis Groups generally don't have so much difficulty with agreeing on the subject matter or with agreeing at the sentence level (i.e., we can generally agree on grammar, syntax, and spelling). They have problems with style. But even more difficult are the problems of context and point of view in the early phases of group analysis projects. Then, dividing the project into parts and giving assignments with little or no overlap becomes the issue. More problems arise when the organization, sequencing and presentation stages arrive. It is here that the Information Mapping method helps the most as we illustrate below. In fact, the Information Mapping method gives the group a common way of looking at the information to be communicated and a common language to discuss that information. Therefore, the process is smoother, and the end product more consistent. wssm Stage 4 Y''. /! Sequencing Analysis • Sequencing issues are made more visible by the chunking and labeling approach. Stage 5 Presentation Analysis * In many instances, presentation formats have been tried out and extensively tested and different ones have been made available as templates on word processing 163 Some Navigational Options Introduction Depending upon the needs of the situation and the hypertext software available, the user can access information blocks and maps with many different access paths. Some or all of the access methods described on these two pages could be used in a hypertext system based on Information Mapping's method. @ 3 00 70 m x o 3 for example Access Through Hierarchical Table of Contents and Classificatory Hypertrails £ sm i5j5Ej& Access by Simple Hypertext Links or Structured Hypertrails Prerequisite Hypertrail Access by Task-Driven Procedures- Access by Hierarchical Menus Main Menu • Products • Services • Industries • Applications 238 >130 ’■‘"HI r 1 _ . for example Access Through Overviews, Summaries and Information Landscapes Chapter 5. Resolving Some Hypertext Problems Commentary -- Summary of Chapter The ability of Information Mapping's method to address so many of the problems and issues raised by hypertext makes it a key tool in every developer's and user’s tool kit. The multiple methods of access shown on these pages show how the method facilitates access because of its systematic approach. Access by Keyword Index Access by Full Text Search Find: engineering AND documentation Access Through Paths Suggested by Expert Systems j see pagej> 69 if.. .then < — < — < — < Access by Personalized Hypertrails Access by Semantic Networks Make Link: Chapter 6. Relatively Stable Discourse: Documentation and Training Overview of This Chapter 168 Example Pages of Relatively Stable Discourse Operations and Technical Manuals 170 Personnel Manuals and Policy Manuals 172 Product Knowledge Case Study Introduction to Product Knowledge Case Study 174 Product Knowledge Case Study: Main Menu 176 Product Knowledge Case Study: Specifications 178 Product Knowledge Case Study: Control Panel 180 Access by Task-Driven Procedures 1 82 1989 R. E. Horn Chapter 6 Relatively Stable Discourse: Documentation & Training On-Line Product Knowledge On-Line Reference Based Training On-Line Procedures In this Chapter ...a Series of Applications to Relatively Stable Subject Matters... On-Line Personnel Policies On-Line Documentation Computer-Based Marketing |~j Tools Computer-Based Training Corporate Applications [c |n o HHI — 1 =J 4k \\ 'A . 167 Overview of This Chapter Introduction We defined one of the major discourse domains as that of relatively stable subject matter. In business this is where we find procedures, policies, documentation and training materials. When writing about these areas we take the stance that the subject matter is stable — not changeless forever, but not going to change every day. In this chapter our primary aim is to present some case studies and examples of different applications of Information Mapping's method to on-line hypertext information retrieval situations. This will give the reader a more concrete idea of the method and its applications to on-line text. Contrast With Paper To see examples of paper-based display of the Information Mapping method see nil Chapter 6 We present two examples of pages from operations and technical manuals: @ S 00 VO 70 m x o 3 168 Chapter 6. Relatively Stable Discourse Personnel and Policy Manuals We present two examples of pages from personnel and policy manuals: Vacation Policy £ Documentary Credits On-Line Product Knowledge DataBase: Case Study We present a series of three examples of an on-line data base from an hypothetical company: 169 1989 R. E. Horn Operations and Technical Manuals Introduction On the following four pages we present examples of screens from manuals that fall under the classification of relatively stable subject matter. The manuals from which they are extracted have been prepared with Information Mapping's approach. 170 Chapter 6. Relatively Stable Discourse documentation How to Remove and Replace the Connector Terminals Introduction Some KRN connectors contain a micro-deck terminal as shown here. Diagram Inspect the terminal. Replace if necessary and then replace the cable by inserting it into the locking tang. Example: Insert the K8889 tool through the hole on the opposite side and gently pull the cable out. Example: Push the cable forward until it will no longer slide. Example The terminal is held in place in the connector cavity by a locking tang. The attached cable allows you to move and position the terminal. Pull Cable v\ I / - CLICK Required Tools To remove the terminal, use the special tool K8889 Movement Via Links In this illustration, the user can click on illustrations contained in the screen which act as buttons. The user may also click on specific words that can be highlighted by pushing a function key. \\ I / -CLICK from technical 171 1989 II Horn from a personnel manual Movement Via Hypertext Links No vacation days will be accumulated from one anniversary to the next. No payment will be made for vacation days not taken Amount Years Exempt Employees Nonexempt Employees vacation 2 weeks 3 weeks 2 weeks 4 weeks 3 weeks 5 weeks 4 weeks 6 weeks 5 weeks W I / - CLICK Holidays during vacation If company paid holidays occur during a vacation period nonexempt employees are eligible for an additional vacation day, but • exempt employees forfeit the holiday. \\ I I ^ -CLICK — . A h85 Claim Integrated Graphics Principle (Information Mapping) Use diagrams, tables, pictures, etc. as an integral part of the text, not as an afterthought added on when the writing is complete. thus V) E o Q. a supports « thus supports Grounds Chunking Research All human beings naturally chunk information as a part of their short term memory process. Pre-chunking aids learning. (Miller, 1967. Simon, 1979)/ A Grounds Hierarchy Research Readers construct hierarchical representations of the text they read. (Van Dijk and Kintsch, 1983) VO 00 VO m x o 3 Grounds Some graphic information structures are better than prose in conveying the same information. (Reid and Wright, 1973 )*A Survey of graphic research literature (MacDonald-Ross and Smith,/ 1977, Smith & Mosier, 19&S) 196 Chapter 7. Disputed Discourse Claim Accessible Detail Principle (Information Mapping) Present detail to support abstractions in a manner that makes it readily accessible to users. Put examples and diagrams close to the text they illustrate. thus o a supports s Claim If you do not provide detail, people will naturally fill in the detail from their own experiment, quite likely incorrectly. thus supports Grounds People "fill in" detail from their own experience when given abstractions in communication situations. They generally believe that their "filled in examples" are correct without further consideration. (Bandler and Grinder, 1975, provide one summary of this research) Grounds All language is ambiguous. There are 14,070 meanings for the 500 most frequently used words in the English language. (Horn, 1967) 197 Useful in Representing Ill-Structured Problems Introduction Only recently have researchers begun to study what can be called ill-structured problems, problems that defy easy definition and boundaries, and have little consensus as to their nature. Definition: Ill-Structured Problems Ill-structured problems are those about which different people have very different perceptions and values concerning their nature, their causes, their boundaries, and their solutions. They are the problems that bring out two or more points of view from the first mention of them. Definition: Well -Structured Problems Example of the Analysis of Part of an Ill-Structured Problem l On the next few pages we present part of a case study in the ethics of using nuclear weapons. This is a field that is ill-structured. It meets many of the characteristics noted on this page. — see page*> 200 ^1_J Well-structured problems are textbook problems, problems which are most often used in training of scientists and engineers. There is widespread consensus as to their nature. They are logically coherent and consistent. Characteristics Ill-structured problems exhibit many of the characteristics shown on these pages. VO 00 vO 73 m X o 3 Complicated & Ambiguous i, 1 V Ideological constraints . . . . l j f No unique y P rfi . ) "correct" view / b 1 e n the problem Great constraints.- * ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 198 Chapter 7. Disputed Discourse 199 Case Study of a Poorly-Structured Problem Introduction We present on the following pages the basic arguments that were argued in the 1980's over the ethics of the policy of nuclear deterrence to illustrate the application of argumentation analysis to ill-structured problems. Deterrence Ais an Immoral Policy The current debate about nuclear ethics is focused in large part on a re-examination of the policy of deterrence. For some people, threatening to use nuclear weapons is an immoral policy. For others, it is the only moral position. With such contradictory positions, we can identify nuclear ethics as an ill-structured problem and apply the tools of argumentation analysis to it. chart i. Claim Any threat to use nuclear weapons is immoral. No moral country will use such threats. □ Agree D No D Not Sure Who makes claim o thus supports objects to Grounds Threatening to use nuclear weapons even with the purpose of preventing their use -- depends on the possibility that they might actually be used — i.e., that the threat will be carried out. □ Agree D No D Not Sure C^see chart BUT Rebuttal (Counterclaim) Deterrence is moral because it is the best policy we have to prevent the use of nuclear weapons. □ Agree D No D Not Sure Who makes claim? . . . . thus supports . Grounds see p^ge'2 . . . increases Warrant If we do something that the possibility of nuclear war, we are just as knowingly immoral as if we killed people ourselves. : HI Agree D No D Not Sure © 3 00 VO PO PI X o 3 Key ♦ Indicates Rebuttal Indicates Support Backing _ T It is wrong to intend to do what one Rebuttal (Counterclaim) Nuclear deterrence decreases knows is wrong to do. the probability of nuclear war. D Agree D No D Not Sure D Agree D No D Not Sure supports Grounds £ see chart 200 Chapter 7. Disputed Discourse Definition Nuclear Deterrence 1 . a condition of the modem age of nuclear powers such that each superpower realizes that, if they started a nuclear war, the other superpower has sufficient invulnerable weapons to retaliate and potentially destroy their military forces, culture and cities. 2. any policy of a nuclear nation that tends to promote or continue the condition of nuclear deterrence. It is a policy in which both superpowers think: "We will not start a nuclear war because the other side threatens to retaliate and destroy us and we think they could and would do that." 201 Case Study Brings Together Opposing Viewpoints Introduction These two pages are a continuation A of the basic arguments that were argued in the 1980’s over the ethics of the policy of nuclear deterrence to illustrate the application of argumentation analysis to ill-structured problems. / chart 2. Deterrence A prevents war; therefore it is the only morally acceptable policy. Definition . — N. :hart The counterclaim of the immorality of deterrence is that deterrence is moral. The argumentation analysis outlining the I see > 1. main structure of this argument is presented on this page. Rebuttal (Counterclaim) Deterrence has too many risks in the face of the possibilities of inadvertent outbreak of nuclear war. BUT □ Agree □ No D Not Sure Who makes claim? see chart. Warrant Pragmatic Argument for Deterrence Policy Before throwing out what has worked to keep the peace and prevent nuclear war, you must come up with a better moral alternative. Otherwise it is morally better to keep the policy you have. D Agree D No D Not Sure see chart Claim Deterrence is moral because it is the best policy we have to prevent the use of nuclear weapons. D Agree □ No □ Not Sure Who makes claim? ■ BUT thus supports Warrant Deterrence is less dangerous than its alternatives . Deterrence may not be perfect as a policy but it is less dangerous than its alternatives. n Agree D No D Not Sure see chart @ 00 vo 73 tn X o 3 Key Indicates Rebuttal Indicates Support Grou nds Deterrence prevents nuclear war by making both sides afraid of starting a war because they will surely lose more than they could possibly gain and could conceivably completely destroy their own country. D Agree D No D Not Sure [^see chart Grounds The policy of deterrence has worked for 40 years. Since 1945 there has been no nuclear war and no conventional war between superpowers. n Agree D No D Not Sure C^see chart 202 Chapter 7. Disputed Discourse Rebuttal (Counterclaim) Any threat to use nuclear weapons is immoral. No moral country will use such threats. D Agree D No D Not Sure Who makes claim? nS see chart 1 BUT Rebuttal (Counterclaim) Nuclear war risks the future of the human species and risking the future of human species is not worth protecting the values claimed to be protected. D Agree D No D Not Sure Who makes claim? see chart _ . Who Makes This Claim? "Nuclear war is such an emotional subject that many people see the weapons themselves as the common enemy of humanity. Nuclear weapons are intrinsically neither moral nor immoral, though they are more prone to immoral use than most weapons. But they can be used to accomplish moral objectives and can do this in ways that are morally acceptable. The most obvious and important way is to use them or their availability to deter others from using nuclear weapons. The second — of much lower, but still significant priority — is to use them to help limit the damage (human, social, political, economic, and military) that could occur if deterrence fails. Anything that reduces war-related destruction should not be considered altogether immoral." 203 © 1989 R. E. Horn Comparing Ill-Structured and "Tame" Problems Introduction. Ill-structured problems can best be seen if we look at them in comparison with "tame" or "well structured" problems, as in the chart below. Characteristics Tame Problems Ill-Structured Problems Ability to formulate the problem Can be formulated exhaustively and written down definitively. No definitive formulation Ability to devise and conduct definitive tests Can be tested. Mistakes and errors can be identified. No single criterion to determine correctness. Difficult to determine when a solution is a solution or even whether a test is applicable. appropriate test? Correct i Incorrect is this the best boundary? only or best alternatives? Relationship between problem and solution Problems can be formulated separately from solutions. Solving the problem is synonymous with understanding it in the first place. Each formulation of an ill-structured problem contains a definition of the solution. 204 © 1986 R. E. Horn Chapter 7. Disputed Discourse Characteristics Tame Problems Ill-Structured Problems Ability to determine whether problem has been solved Have a clear ending point and a determinable solution. Yes done? ^N„ A clear rule or test can be stated to determine completion. No stopping criteria. ..the problem may be ongoing and continuously changing, so there is no way of determining completion. Problem Solution Tractability Exhaustive list of operations used to solve problem exists. No list of operations exists for solving ill-structured problems. Relationship between explanation and solution Can be stated as a discrepancy between what is and what could or ought to be, and an explanation exists for every gap. Problem — ► Gap -*►- Solution Explanatory Framework Many possible explanations and each one "contains" or "implies" a different solution. More on this table on next page 205 1989 R. E. Horn Comparing Ill-Structured and "Tame" Problems ^ continued Characteristics Uniqueness or reproducibility of problem Repeatability of solutions Level of analysis Tame Problems Ill-Structured Problems Problems can be abstracted from the real world and similar solutions can be found. Each problem and each solution is unique. ^xplanation ^ Attempts to solve can be made You cannot undo what you have tried, repeatedly until one works. so that each solution is unique and changes the nature of the problem. Problem Problem Identifiable, "natural" form with high degree of certainty. ..level of detail for solving the problem can be found. ..and boundaries for the problem are reasonably easy to agree upon. No identifiable causes. ..every "symptom" is a problem and vice versa. ..level of detail and approach are not easy to define. ..little agreement on setting boundaries of the problem. Problem > X boundary C c 3 Level of Detail problem sympto symptoi problei cause? level of detail for ^analysis? J 206 Chapter 7. Disputed Discourse Conclusions: Argumentation and Hypertext Summary Argumentation is a different kind of discourse from relatively stable subject matter. We have seen in this chapter that it is useful to clarify the components of a disagreement by identifying exactly what the claims, grounds, warrants, etc., are. And it is useful to use a more graphic way of displaying these components. Connection With Other Types of Discourse How does argumentation analysis relate to the other major types of discourse we have presented in this book? On this page we show the major connections with other types of discourse discussed in this book. jS m Commentary: Usefulness of Argumentation Analysis It is quite possible that argumentation analysis, as described in this chapter, will provide a method for slowing down disputes and looking very carefully at the merits of different points of view. Obviously, many disputes can be conducted without it. In other disputes we will be able to use argumentation analysis as a kind of "microscope" to look at the argument quite closely for any flaws or weaknesses. For that, it will become a significant tool. (REH) Many of the statements in textbooks and training manuals arrive there after a process of argumentation and experimentation. This is particularly the case with facts, generalizations, and explanations of process. see chapter After a scientific experiment has been planned and carried out, it is written up in articles which we describe in the domain of experimental discourse. j — i Scientific Article: and Papers see chapter These results may provide the grounds for new claims that clarify the argument and may provide clear facts or generalizations that find their way into the documents of relatively stable discourse. Disputed discourse arises when individuals disagree with either the claims, grounds, warrants, or backing of a given argument. They propose a rebuttal (or counter claim) which may become the subject of more systematic observation or a scientific experiment. _ _ Grounds (Typically Dal Rebuttals (Counter - Cl< Claims Relations see chapter .Qualifiers 0 vr Warrants f Backing 207 Chapter 8. Experimental Discourse: Scientific Information Overview of This Chapter 210 The Science Information System 212 Application: Scientific Reports and Articles 214 Case Study: Scientific Abstracts Applying Information Block Analysis to Abstracts 216 Miller: Short Term Memory Limits and Chunking 218 Simon's Tests Show Chunking Size to be 5 to 7 220 Hartley and Trueman: Headings Aid Retrieval 222 Shaffer: Information Mapping's Methodology 224 Reid and Wright: Superiority of Visual Structuring 226 Other Potential Benefits of Hypertext in Science Information Facilitate Identifying Problems at Science Frontiers 228 Linked Comments Will Highlight Deficiencies 230 1989 R. E. Horn Chapter 8 Experimental Discourse: Scientific Information W May Be Able to Point the Way to Advances The Science Information System May Be Able to Identify Problems In this Chapter ...a Group of Scientific Abstracts on Presenting Information Application to Scientific Reports Application to Scientific Abstracts Jr Abstract of Reid and Wright on Graphic Communication Structures Abstract of Simon’s Paper on Chunking Abstract of Miller's Paper on Chunking 209 Overview of This Chapter Introduction Unlike the relatively stable discourse domain (Chapter 6) which has seen literally tens of thousands of applications of Information Mapping's method, our applications to the discourse of experiment is somewhat speculative and theoretical. We propose that the approach of structured hypertext with Information Mapping's method could be used to approach scientific abstracts and other elements of the science information system. We do this by showing prototype abstracts. It is speculative in that we have only tried this on a pilot basis in our own lab with documents on human factors research and a few other small projects. By extension, we suggest that scientific articles may be written with many of the guidelines and principles of Information Mapping's method, and be accessed through associative hypertext networks. The Science Information System Applying Information Mapping Principles to Scientific Reports, Articles, and Presentations To provide a context for this discussion, we present a simplified seven level model of how "new" information flows in science from untested theoretical ideas to what is taught in the classroom. Analysis & Synthesis Level The major difference from current practice in writing scientific reports, articles, and presentations with Information Mapping's approach is a much smaller chunking size (as well as a more useful set of block types). At this point, we present a list of potential block types for scientific reports. 0 Experimentation Level Reporting Level Abstracting Level What experiments did we perform? What procedures? What results? Summary Level VO 00 m x o 3 0 Teaching Level 210 Chapter 8. Experimental Discourse Definition: Discourse Domain A,_of Experimental Knowledge The domain of experimental knowledge discourse descriptions of scientific experiments and the discussion of the results of these experiments in abstracts, state of the art reviews and theoretical papers. Applying Information Mapping Futuristic Possibilities 3 Principles to Abstracts of 4 Suggested by Hypertext Scientific Papers and Information Mapping Writing informative abstracts about scientific abstracts is quite similar to the process for scientific papers. We first present an overview of the kinds of blocks that might be expected in such abstracts. And then we present five examples of such abstracts on psychological research related to the subject of this book. Hypertext and Information Mapping may help us explore the frontiers of science and identify the most productive problems and flag them in more visible ways. We present an outline of one way of looking for productive problems in science and then consider how that might look in a hypertext system. Abstract Miller Abstract Simon Abstract Reid & Wright Abstract Hartley & Trueman Abstract Schaffer Find and Mark Problems and 211 The Science Information System Introduction To provide a context for this chapter, we will review on these pages the general structure of the international scientific information system. It is substantially similar for most of the scientific disciplines and subspecialties. The Levels of Scientific Endeavor Path of "New" Information Analysis and Synthesis Level At this level of work the scientist is asking questions, formulating new ideas for research, and putting together results from other workers to form new theories. Experimentation Level Reporting Level At this level of the science process, the scientist writes up the results of experiments and presents the data and conclusions in papers and reports. Summary Level This is the level at which books and articles summarizing advances in the field are written. ©— At this level of work the scientist carries out the experimental procedures formulated at the previous stage, preparing notes, collecting and processing data. Abstracting Level At the abstracting level of the science information process, professional writers prepare abstracts and summaries of the reporting level documents. Teaching Level 00 50 m E o 3 At this level the textbooks and other teaching materials are developed. 212 Chapter 8. Experimental Discourse Documents that are Used at Each Level of Science Proposals Theoretical Artide — • Identify Problem JP * Suggest Expedment(s) • Request Funding * Reinterpret Data if ♦ Critique Theory * * Analyze Problems - - — 1 J Speculative Article • Identify Problems • Suggest Ways of Approach • Present New Theory • Lab Notes • Documentation of Experiments • Data Conference f Papers and Proceedings 1 • * Journal Articles V * Reports Present Data Offer Conclusions For examples of the Information Mapping hypertext approach applied to scientific reports Abstracts Summarize articles and papers For examples of the Information Mapping hypertext approach applied to scientific abstracts Advances in The Field In a limited area * What do we know? • What hase been done? * What remains unexplained? • What might be done? Science News Announce most significant results Textbooks and Handbooks Integrate what is known in didactic or handbook format This is the domain of relatively stable subject matter. 213 Application: Scientific Reports and Articles Introduction If you walk into the offices of most research scientists, you see large piles of scientific papers. When I ask the scientist about the contents of the stack of reports, I usually get the reply that each pile represents what they have to read before going on to the next experiment or writing the next proposal. These piles are not small. They often contain 50 or 100 papers. How many papers can a scientist read in a week? Not many if they are doing other things, writing, doing experiments, and teaching. There would be a great advantage to a million scientists and engineers if we could improve the scannability of basic scientific reports. If scientists could quickly determine which of the 50 or 100 reports contained information that they needed to read carefully, they could do their jobs better. The principles of Information Mapping's approach are specifically aimed at improving such communication situations. How would that work? What is the situation now? When we look at contemporary primary scientific reports today, we typically see five main divisions: background, method, data, conclusions, and discussion. Compared to the amount of information in the report, this is not enough chunks. There are typically too few labels to provide rapid scanning. A preliminary analysis On these pages, we present our preliminary analysis of reports that contain experimental data. We suggest a group of information blocks that would provide the working scientist with better guidance for rapidly scanning scientific reports. When we applied the chunking principle, we divided the information in each report into much more fine-grained pieces. The labels focus on a content-independent set of categories that working scientists are interested in. Status of this work We should point out that, unlike the results we presented in Chapter 3 for relatively stable subject matter, this information block analysis is preliminary. We have not extensively tested it either deeply in one field or broadly across several fields of science and technology. We believe that many of the categories would hold up well in such an evaluation, but there are likely to be additions to the set of information blocks from such a test. © 00 v© JO m X o Background Information Leading Up to the Research • Argument (leading up to hypothesis development, background) - Related Research - Theoretical Propositions - Experimental Evidence - Implications (for this project) - Bibliographic Citations • Definitions (of novel key terms) • Definitions of Abbreviations & Notation • Examples (of novel key terms) • Prerequisite Technical Terms (not defined but used in this report) • Main Questions • Theoretical Model • Assumptions • Formal Hypothesis - What is New About Hypothesis? - Parameters of Theoretical Model to be Varied - Independent Variable(s) - Dependent Variable(s) • Controls - Conditions of Testing - Subjects - Selection Procedures ill iliSIII xxxx.x.x: :::::x:::x: x x:: •x ;X;XXX; .’ ’ ‘ X ;X * X •• • x . ■ • ; . .. ■_ ; : ::x:xxxxxxx ;x:x:::x:::xx:x ::*x:x::£&xS xx yxxx; x- XXX ;X-':X:X; &:£-i£i-x£:^ x •- • ••• x'x'x-x :XXX: j:g|§§§ ■■:X;V:xv; ::::::::x:x;xvx:: 214 Chapter 8. Experimental Discourse Preliminary Analysis of Blocks Needed for Reporting Experimental Data Procedure for Conducting the Experiment Site Year Experimental Setup Equipment or Apparatus - New Apparatus - Unusual Aspects of Apparatus Experimental Design Procedure (Methods of Data Collection) - Measures Used - New Techniques - Unusual or Significant Aspects of Procedure Subjects - Age - Sex - Race - Income - etc. Outcomes of Experimentation Data Collected (results) Data Reduction(s) - Sample Size Mathematics Used for Data Reduction Computer Program Used Main Conclusions Secondary Conclusions - Unexpected Findings - Intuitive Grasps of the Data or Process 1 Significant Negative Results 1 Implications - Important (to whom?) - Utility (for what?) ■ Limitations and Shortcomings - Limits of Generalization of the Study - Technical Flaws Discovered in Doing the Experiment - Possible Errors - Critical Comments ■ Theoretical Implications • Research Which Needs to be Done as a Result of This Experiment Bibliographic Information • Title • Author(s) • Affiliation and Location of Author - Address (complete, zip) - Phone • Journal • Citation 11111 :X;X:X:X:X:X: • Presentation at Meeting • Contract Number(s) and Acknowledgement of Financial Support • Acknowledgements of Assistance • Document Identification Numbers • Suggested Indexing Terms II mil i*;***?;*:* ;>x;xjxmx; Ssississ si*?;***:*: 215 Applying Information Block Analysis to Abstracts Introduction On the previous page we provided the results of a preliminary examination of scientific reports. Here we examine, in similar fashion, the scientific abstract. We ask, what types of blocks are essential to restructuring the scientific abstract? If standard block types can be developed for different types of scientific abstracts, the abstracter will be able to provide information that can be easily scanned and summarized by investigators who use the abstracts. Our research in this area is preliminary, but suggestive of how we might proceed. Standard Information Blocks for Scientific Abstracts On these pages we present a list of the information blocks that appear to be important for all scientific abstracts presenting experimental data. We would encourage the abstracter to use them unless there are overriding reasons for not doing so. To be regarded as fully following the principles of Information Mapping's method, we would have to develop and test specific standards, guidelines, and rules for construction for each of the twenty-five or thirty blocks that abstracts might contain. Examples On the following pages we present examples of several abstracts developed with this approach. @ 00 VO 70 M X About the experiment • Purpose • Method • Results (Findings) • Conclusions These blocks also appear to be important in describing the experiment in longer abstracts • Implications • Data (i.e., a simplified data table, chart or graph of the data collected) • Limits of Study • Instrumentation Modification • New Techniques or Equipment • Controls • Other Observations • Background (theoretical or other setting for experiment) • Caution • Consequences - Benefits - Safety - Cost - Convenience 216 Chapter 8. Experimental Discourse Information Blocks • Risk • Status • Criteria • Recommendations • Installation Requirements • Environmental Factors • Advantages/Disadvantages • Trade-offs • Feasibility • Promising Lines of Research for Abstracts For Specific Specialties Specific scientific specialties would require blocks for important information items that always or frequently appear. For example, in psychology or education the following would be useful: • Subjects (with additional sub-blocks such as number and characteristics) • Time Span • Task • Treatment of Groups Bibliographic Information • Author - Affiliations • Address • Title of Publication • Citation (include citation in the convention of the field) About the Abstract • Abstracter • Data Miller: Short Term Memory Limits and Chunking Problem What are the limitations on the amount of information that we can receive, process, and remember? Background Psychology and Information Theory Information theory can provide a quantitative approach to questions raised in the psychology of communication. What psychology calls experiments in absolute judgment, information theory calls experiments on the capacity of people to transmit information. In a communication system, the term "amount of information" is used to express what psychology would call the "variance" in either input (stimuli) or output (response). "Amount of transmitted information" is the term used for the relationship between input and output. Information Measurement In experiments on absolute judgment, the capacity to transmit information, the experimental problem is to increase the amount of input information and to measure the amount of transmitted information. Input can be increased in rate or in amount. An increase in the amount of information is an increase in the number of alternative stimuli. The observer is considered a communication channel, and has a "channel capacity" — the upper limit on the extent to which the observer can match response to stimuli. That is, there is a limit to the capacity to accurately transmit received information. As information input is increased, the observer’s transmitted information will at first increase, then level off at some asymptotic value as errors in transmission increase. Bits To test accuracy of transmission, the observer is given a discrimination task; a judgment must be made between alternatives. To describe this judgment quantitatively: The amount of information needed to make a decision between two equally likely alternatives is one bit. For instance, to decide if a man is less than 6 feet tall or more than 6 feet tall, if we know that the chances are 50/50, requires one bit of information. 2 alternatives require 1 bit of information 4 " 2 bits 8 3 16 4 32 5 Each doubling of alternatives requires one additional bit of information. @ 3 00 VO JO m X o 3 Research In experiments of absolute judgment using unidimensional stimuli — pitch or loudness of auditory tones, saltiness, position of a pointer on a line — channel capacities had a mean of about 2.6 bits, or about 6.5 alternative categories. Channel capacity increases when dimensions are added. In experiments for two-dimensional stimuli — dots in a square, saltiness and sweetness, loudness and pitch — capacity was about 3 to 5 bits. For multidimensional stimuli, capacity varies; capacity for colors that vary in size, hue, and brightness was around 4 bits; for 6 different acoustical variables, capacity was about 7 bits. While adding variables increases total capacity, it decreases the capacity for any particular variable. In other words, we can make relatively crude judgments of several things simultaneously. Analysis The number seven recurs in various experiments. The "span of absolute judgment" (limit to accuracy in identifying the magnitude of a unidimensional stimulus variable) is about 7 alternatives. The span of immediate memory is about 7 items in length. The span of attention encompasses about 6 items. While it is easy to postulate that a single process underlies these three spans, that is not the case. There appears to be a difference between the process of absolute judgment and the process of immediate memory. Absolute judgment is limited by amount of information, or bits. Immediate memory is limited by number of items, or chunks. A chunk is a coding unit which groups bits. The span of immediate memory seems to be almost independent of the number of bits per chunk, though chunks themselves are limited to about 7. 218 Chapter 8. Experimental Discourse Importance of Miller's Paper Miller's paper is a classic and changed the way of studying short term memory. New Definition: Chunks Chunks are familiar units -- a word is a chunking of phonemes. In Morse code, dit and dah can be separate chunks for the beginner, but they are only part of larger chunks — letters, words, phrases — for the experienced operator. In communications theory such ordering of information is called recoding. Input is recoded into another code that contains fewer chunks, with more bits per chunk. 7+2 chunks Conclusions The span of absolute judgment and the span of immediate memory impose severe limitations on the amount of information that we are able to receive, process, and remember. By organizing the stimulus input simultaneously into several dimensions and successively into a sequence of chunks, we manage to break (or at least stretch) this information bottleneck. The process of recoding is a very important one in human psychology and deserves much more explicit attention than it has received. Information concepts have already proved valuable in the study of discrimination and of language; they promise a great deal in the study of learning and memory. It has even been proposed that they can be useful in the study of concept formation. There may be something deep and profound behind the "magical number seven," or there may only be a "Pythagorean coincidence." Author George A. Miller Citation "The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information" Psychological Review, Vol. 63, No. 2, March 1956, pp. 81-89 Comment: Type of Abstract This abstract is an example of an informative synoptic abstract of a paper we would classify as a theoretical paper. The abstract was developed using Information Mapping's approach Simon's Tests Show Chunking Size to be 5 to 7 Problem How can the parameters of the human information processing system be determined? Background Shift in Experimental Approach to Memory: The examination of problem-solving processes is part of the attempt to understand complex cognitive behaviors. Short term memory and the transfer of information to long term memory (fixation) is crucial to problem solving. To examine memory, experimental psychology is taking a new view of experiments. Traditionally, an experiment tests a hypothesis by focusing on the relation of a dependent variable to independent variables manipulated over a set of experimental conditions. Such an experiment produces one bit of information for instance, that the ease of learning nonsense syllables is related to their meaningfulness. It does not give the strength of the relation — the parameters that tell whether running up the scale of meaningfulness from 0 to 100 reduces learning times by 5%, 50%, or 100%. Now experiments are shifting from hypothesis testing to parameter estimating. Analysis The Chunk Studies cited by George A. Miller in 1956 pertaining to short term memory were mostly of the parameter estimating type. He postulated the "chunk" as the unit held by short term memory, and found that the capacity of short term memory, measured in chunks, appears constant. The capacity is also apparently independent of the content of the chunks — whether words, digits, colors, poetry, or prose. However, unless chunk size can be measured independently of memory span, the assertion of a fixed chunk span loses all empirical content. Instead, there is a definition: a chunk of any material is what short term memory will hold five of. Determining the Size of the Chunk: The chunk is then not a directly observable quantity. However, if the hypothesis of the unobservable chunk is combined with a hypothesis of an observable quantity, such as learning time, empirical evidence for the chunk can be derived. 1st hypothesis: The span of immediate recall is a constant number of chunks. 2nd hypothesis: Learning time is proportional to the number of chunks to be assembled. For instance, take the short term memory span for a particular test situation, e.g., for nonsense syllables, and compare it with the memory span for another test situation, e.g., simple words. Using Brener's data, the ratio of word span to syllable span is 2.2. This ratio can then be compared to learning-time ratios for the same materials. It is commonly observed that there is a 2.5 to 1 advantage in learning simple words over nonsense syllables, for fixation of information in long term memory. The near agreement of these ratios lends support to the hypothesis that there is a chunk of constant size underlying the process of short term memory and the process of fixation of information in long term memory. Not all experiments show such agreement, however; e.g., the comparison of nonsense syllables with the digits. @ V© OO SO JO m 220 Chapter 8. Experimental Discourse Quick Summary Simon's work essentially confirms Miller's hypothesis. The exact number of chunks is less important than that we know that the size of short term memory is approximately 5. 5 ± 1 chunks 5 to 10 seconds per chunk to fix in long term memory. Comment: Type of Abstract This abstract is an example of an informative synoptic abstract of a paper we would classify as an experimental paper. The abstract was developed using the Information Mapping approach. Implications The chunking hypothesis has implications for the increase in digit memory span with age (possibly due to shortening of encoded strings by use of learned chunks). The chunking hypothesis has been used to explain the ability of chess grand masters to reproduce the pattern of pieces on a chessboard after a brief exposure. This may not be an extraordinary perceptual ability, but a "vocabulary" of chunks similar in size to an educated adult's vocabulary in his native language. The chunking hypothesis can be related to the strategy of using paradigms estimated for simple tasks to predict performance on complex tasks. Conclusions To summarize, the estimates of relative chunk size for nonsense syllables, words, and prose obtained from immediate recall experiments agree very well with the estimates obtained from rote learning (long term memory) experiments. There is serious disagreement, however, between the two estimates of digit chunk size; data for estimating chunk size for colors and geometric figures are apparently not available from the rote learning paradigm. The psychological reality of the chunk has been fairly well demonstrated, and the chunk capacity of short term memory has been shown to be in the range of five to seven. Fixation of information in long term memory has been shown to take about five or ten seconds per chunk. These two basic constants organize, systematize, and explain a wide range of findings about both simple tasks and more complex cognitive performances that have been reported in the psychological literature over the past 50 years or more. Author H. A. Simon . Citation . "How Big is a Chunk?" in Models of Thought, New Haven: Yale University Press, 1979 pp. 50-61 Biographical Note Herbert Simon won the Nobel Prize in economics for his studies in decision making. 221 Hartley and Trueman: Headings Aid Retrieval Problem How does presence or absence of text headings affect recall, search, and retrieval? Definitions Recall — amount of information recalled. Search — time taken to find information in an unfamiliar text. Retrieval — time taken to retrieve information from familiar text. Material Used Text was adapted from the Sunday Observer Magazine, approximately 1,000 words. Topic was television viewing habits in the United Kingdom. Text was a report on a questionnaire, and contained a large number of facts and figures. Text was typed with one and a half spacing on about 3 1/2 pages. There were 2 versions: one with headings and one without. Both versions had 12 paragraphs. Text with headings had either marginal or in-text headings approximately every 2 paragraphs. Headings were either questions or statements. Flesch reading ease score of text: 55, regarded as "fairly difficult" or suitable for 15- to 17-year olds. Theoretical Model No Headings Marginal Headings Embedded Headings — i- - r — — - 7 i J " . . . work . . . conducted from an a-theoretical position." Task To read text and answer questions about it on a short-answer test with 12 items. Task was structured toward either recall, search, or retrieval. Evaluative Criteria Test questions required specific answers: "What percentage of viewers were dissatisfied with BBC 1 programs?" Method of Presenting Task Recall studies: Participants read through material carefully once, and then answered test questions without referring back to the text. Search studies: Participants first did a practice task as a group and then did the main task individually. Task in both cases was to find and circle in the text the answer to each of the test questions. Retrieval studies: A practice task preceded the main task. Both times, participants read the material first, found and circled the answers to each of the test questions in the text. Task was timed. Chapter 8. Experimental Discourse Subjects 1,270 fourth-year British comprehensive school students, 14 and 15 years old, male and female. Experiments took place in classes, usually "organized by the school into different ability groups." No remedial students took part. Participants were allocated randomly, but approximately equally by sex, to each condition of the experiment. Results Headings aided recall, search and retrieval. Headings or not: The average participant who read material with headings "performed better than 66 percent of the participants in the no-headings groups." General or specific effect of headings: Could not be determined for search and retrieval. For recall tasks, effects were general. Position of headings: Whether marginal or embedded had no significant effect. Kind of heading: questions or statements. Could be directly compared in only 1/3 of the experiments. No significant effect. Other Findings Comment In later studies, (Hartley 1989, personal communcation) the results with low-ability participants was not confirmed. (REH) Ability: "... a suggestion that different tasks might have produced different results with the low-ability participants." Data inconclusive. "Headings thus proved effective for aiding recall, search and retrieval. This was the case whether or not headings were embedded in the text or positioned in the margin, and whether or not headings were written in the form of statements or in the form of questions." Authors J. Hartley and M. Trueman, Department of Psychology, University of Keele Citation "The Effects of Headings in Text on Recall, Search and Retrieval," British Journal of Educational Psychology, 1983,53, pp 205-214, 223 Schaffer: Information Mapping's Methodology Problem What is the value of a "structured modular writing technique (Information Mapping)" in the utilization of instruction? Background: Information Mapping "Information Mapping (IM) is a structured modular writing technique." Informational needs of users, content specification, and guidelines for writing and formatting are essential components of the method. Purpose To determine the "potential benefit" from the revision of an instruction by trained professionals utilizing IM. Comment: Size of Task Rarely do studies use reference and training materials of such a large size as the ones used in this study. This, together with the closeness of the experimental situation with on-the-job use of similar materials, makes this an especially significant study. Also relevant here is the concept of reference-based instruction for training and on-the-job reference materials. (REH) . _ _ Subjects Ten subjects: 7 female, 3 male Age: Average 38.6 Job Type: 5 clerical, 4 management Average Length of Employment: 10.2 years Knowledge of Tasks: Subjects were screened to prevent inclusion of individuals familiar with the specific time reporting instruction tested on IM. Materials Used Original Version "140 page Time Reporting Instruction" "The selection was based upon the high quality of the existing document, its technical complexity, and the size of the user population. The current version has few errors in content and an exceptionally clear writing style. Also, the widespread use of the instruction multiplies the importance of the human performance characteristics of the document." Information Mapping Version "185 page revised version" developed by Information Mapping, Inc. Tasks "The tasks were generated by randomly selecting time reporting deviation codes from a pool of codes that are not generally known. The selected codes were then inserted at random into one of two formats. One question format involved the determination of a code's meaning using a multiple choice presentation. The other question format involved the determination of the appropriate code for a given situation. After each item a space was provided for the subject to record the time. In this way, two equivalent sets of tasks were developed, each containing three multiple choice items followed by three code determination items." Other Tasks A pre- and post-semantic differential evaluation form to assess perception of task format and feelings about the material. 224 Chapter 8. Experimental Discourse Method of Presenting Task Group testing: Subjects were asked to evaluate two different versions of the materials. 1st phase: "Equal numbers of the two versions of the instruction were then distributed in a random fashion." 2nd phase: First semantic differential. 3rd phase: "A second trial was then conducted following the above procedure with each subject evaluating the other version of the instruction and performing the second set of tasks." Analysis Time and error data compiled. Semantic differential scaled on 1-7 scale. Results Time: "The instruction version had no significant effect on the time required to complete the tasks." Errors: "Subjects made 54.5% fewer errors in the tasks when using the Information Mapping version of the instruction." Subjectivity rating: The Information Mapping version was reported to be more 'modern,' ’clear,1 'not frustrating,' 'friendly,' and 'good.' "Of the 25 items on the final evaluation instrument, 12 revealed significant differences between the versions of the instruction (P<.05)." "The Information Mapping version was described as . . . • text rambles less • more broken into logical parts • table of contents easier to use • type font less 'too small’ • more 'trustworthy' and 'friendly' • made subject feel more 'satisfied,' 'confident,' and 'in control' • more 'easy to use' • more 'easy to learn from' • more of a 'good quick reference.'" Conclusions "Although the current version is generally considered by management to be in 'good shape' the Information Mapping version was significantly superior. Although the scope of the study is limited the importance of writing quality is clearly demonstrated." Author Eric M. Schaffer Citation "The Potential Benefits of the Information Mapping Technique" NSPI Journal , February 1982, p. 34 - 38. 225 Reid and Wright: Superiority of Visual Structuring Introduction To what degree can we make an abstract visual? And will addition of visual material aid in the rapid comprehension of the abstract.? We present an experiment here for readers to judge. Problem How do alternative ways of presenting technical contingencies? information affect reader ability to determine the outcome of complex Materials Used Examj>le^> The same basic instructional information was presented in four formats: - bureaucratic prose - algorithm (flow chart) - list of short sentences - row and column table. "Fictitious material was invented so that subjects had no option but to read the written information to solve each problem; the problems could not be solved from any previous knowledge." Prose When time is limited, travel by Rocket, unless cost is also limited in which case go by Space Ship. When only cost is limited an Astrobus should be used for journeys of less than 10 orbs, and a Satellite for longer journeys. Cosmocars arq recommended, when there are no constraints on time or cost, unless the distance to be travelleq exceeds 10 orbs. For journeys longer than 10 orbs, when time and cost are not important, journeys should be made by Super Star. Short Sentences Where only time is limited travel by Rocket. Where only cost is limited travel by Satellite if journey more than 10 orbs. Travel by astrobus if journey less than 20 orbs. Where both time and cost are limited , travel by Space Ship Where time and cost are not limited travel by Superstar if journey more than 10 orbs. Travel by Cosmocar if journey less than 10 orbs. Task 36 Problems Easy Difficult Typical Problem Determine the appropriate mode of travel, given information on the traveler's available time, affordable cost, and the journey's distance. Subjects Prose Group Short Sentences Group Algorithm (Flow Chart) Group 68 adults, 17 in each of the ffTffTfinPf fTrffffTf fffTftnrl 4 experimental groups. 32 were male, 36 female. nnnnnnnn fiilfftfi nnnnnnnn ititfitit nnnnnnnn liittmt Data Errors Type of Problem Prose Algorithm Short Sentences Table Straightforward 34.4 18.1 19.1 14.7 Difficult 41.7 26.0 41.7 35.8 Conclusions More Errors with Prose on Straightforward Problems Errors Prose 226 Chapter 8. Experimental Discourse Commentary: Conclusions about graphic communication Does such evidence hold across different kinds of visual communication devices? Yes. (See Smith and Mosier, 1986, for a summary of many of the research findings.) (REH) Algorithm (Flow Chart) Is time limited? i Yes 1 No 1 Is 1 cost limited ? J 1 Is cost 1 limited ? r Yes 1 No r wes i 1 No l Is traveling Is traveling distance more distance more than 10 orbs? r-J - , than 10 orbs? tra\ 'el 1 1 Yes No 1 1 Yes No by travel travel travel travel travel Space by by by by by Ship Rocket SatellitAstrobus Superstar Cosmocar Table If journey less than 1 0 orbs If journey more than 10 orbs Where only time is limited travel by Rocket travel by Rocket Where only cost is limited travel by Astrobus travel by Satellite Where time and cost are not limited travel by Cosmocar travel by Super Star Where both time and cost are limite i travel by Space Ship travel by Space Ship Method of Presenting Task "An independent group design was used, each subject working with only one of the information formats. Subjects were tested individually." For each of the three sections of the experiment, problems were given at two levels of difficulty. Simple problems gave information on time, cost, and distance directly; difficult problems gave the information implicitly. Judgment Criteria Success of subjects was judged according to the accuracy of their problem-solving, since "Clearly the most important datum is error rate. If information cannot be used accurately there is little value in it being used speedily." Table Group mnni 4141111#! Author Reid, F. , and Wright, P. Citation "Written Information: Some Alternatives to Prose for Expressing the Outcomes of Complex Contingencies," Journal of Applied Psychology , 1973, Vol. 57, No. 2, 160-166 This summary covers one of two experiments reported in this paper. Fewer Errors with Algorithms on Difficult Problems mMmmmmm Commentary: Conclusions about Abstracts iiiiiiiiiiimiiiniiiiiiMiiiiiM In this section, we have presented five scientific abstracts. They demonstrate how Information Mapping's method can be used to improve the scannability and ease of use of this kind of document. (REH) Errors Algorithm 227 Facilitate Identifying Problems at Science Frontiers Introduction Science advances by identifying "productive problems" on which experimental or theoretical work can be accomplished. Scientists ask: What are the big problems now? What are the "tractable" problems on which we can work? What are the frontiers that appear a little beyond our capabilities now? An important property of a fully useful scientific hypertext would be the ability to describe the frontiers of science. Problem of Problems Root-Bemstein (1982) describes the following nine types of problems as ones which characterize opportunities for advances in science. Niue Types of Potentially Fruitful Problems (1) problems of definition How shall we classify the phenomena? (2) problems of theory How well does a given theory explain the data? (3) problems of data Do we have sufficient data in sufficient quality to verify the hypothesis? (4) problems of methodology I I I or technique How adequate are the tools and techniques of data collection? (5) problems of criteria How can we integrate two or more theories or collections of data? How can we borrow ideas, data, techniques or tools from other branches of science for this problem? How comparable are two or more definitions, theories, or data collection techniques? How adequate is the "interpretation, meaning or validity" of items 1 through 4 of this list? Are the tools and methods used in an experiment appropriate to the question being asked? @ VO 00 VO 73 m X o Hypertext Application Structured hypertext may help in providing the facilities for individuals and groups to analyze and communicate the frontiers of particular scientific problems through webs of text and graphics such as the logical trees linked to abstracts as shown in this chapter. Example of Annotations of Frontiers What would a survey of the frontiers of science look like? 228 Chapter 8. Experimental Discourse Logical Tree Technique for Visualizing Problem Linkages at Scientific Frontiers Root-Bemstem is interested in how to pick "big problems" to work on, i.e., those which have major implications or which include several problems at once. He suggests that scientists build "logical trees" of problems (rather than solutions) that will better give them a picture of the strategic selection of problems. Example of a Logical Tree (Immune System) First order problem How does the immune system work? Second order problem What constitutes the immune system? (Definitional Problem) /\ What are some immune system reactions that we can observe? (Data, Technique) How do we explain the data? (Theory) Can we explain these reactions in terms of present theory? (Integration) Lower order problems Anatomical Physiological Observations Observations Get the Right Tree Graft Trees Scientific "problems may only be solved when the techniques, data, theories, or concepts exist for solving them, " says Root-Bemstein. "The trick of problem solving, then, becomes the ability to propose a tree of logically connected (i.e., 'nested') problems so constructed that one or more branches or twigs connect with the known. The solution of one or more subproblems may then provide the basis for the solution of the problem next in the 'order.' In a very well-connected 'logical problem tree,' the solution of a single minor problem may create a 'domino effect' or 'chain reaction' leading to the solution of an entire problem area." . «*£> "Logic is not sufficient to the resolution of problems... To resolve one problem requires knowledge that raises another problem that requires knowledge that raises another problem. ..In such instances, infinitely regressing 'problem trees' will only be useful if the 'tree' may be grafted at some point onto another 'tree' that connects to the known. One way of making such grafts is by analogy: 'This problem, about which we know nothing, is like that problem which can be solved; perhaps, therefore, the solution to that problem provides an analogy by which this problem may be resolved.'" 229 Linked Comments Will Highlight Deficiencies Technique Problem Recent studies show data gathering tool is unreliable. Need to repeat experiment. Definition Problem Needs new operationalization of definition. Commentary Classification System of Most Useful Types of Comments It is quite possible that we will need a classification of comments and suggestions along the lines of what we have suggested for frontier scientific problems. Example: Trigg’s Typology of Comment Links Trigg (1983) has suggested a typology of comment type links. (REH) Data Problem Only one study. Small sample. Needs replication. if x-t4 ■jr' Introduction Scientists advance knowledge by noticing the deficiencies, errors, omissions in particular theories, data, technique, interpretations. One of the most powerful capabilities suggested for hypertext facilities will be the ability for analysts to add comments to other workers' presentations. On these pages we present a schematic which shows a portion of a subfield of science with annotations attached to the Information Mapped hypertext. Ability to See Holes in Arguments "Perhaps the most important (yet least vivid) benefit of hypertext will be a new ability to see absences. To survive the coming years, we must evaluate complex ideas correctly, and this requires judging whether an argument is full of holes. But . today we have trouble seeing holes." . Chapter 8. Experimental Discourse Ability to See Absence of Holes JET The philosopher Gregory Bateson was fond of asking, "What is the message of no message? What does it mean when your mother does not write you a letter — especially when you are used to getting a regular letter from her?" Drexler points out that in most discourse it is very difficult to see "the absence of fatal holes, yet this is the key to recognizing a sound argument. Hypertext will help us. Readers will scrutinize important arguments, attaching conspicuous objections where they find holes. These objections will make holes so consistently visible that an absence of good objections will clearly indicate an absence of known holes. It may be hard to appreciate how important this will be: the human mind tends not to recognize the problems caused by our inability to see the absence of holes, to say nothing of the opportunities this inability makes us miss." Ability to See Whether Something Has Already Been Said A*" v nc Drexler points out, "For example, imagine that you have an idea and are trying to decide whether it is sound and worth publishing. If the idea isn't obvious, you might doubt its truth and not publish it. But if it does seem obvious, you might well assume that it has already been published, but that you just can't find out where. Hypertext, by making things much easier to find, will make it easier to see that something has not been published. By making holes in our knowledge more visible, hypertext will encourage hole-filling." Comparison Problem These two data sets were collected by quite different means, hence they are not comparable. W Theory Problem The proposed theory does not cover all of the data. Criteria Problem The criteria used do not permit us to distinguish important phenomena. 231 Summary of Chapter We have attempted to demonstrate here the benefits of Information Mapping's approach in helping structure scientific and technical information so that accessibility and usability is increased. We have also shown how the use of hypertext linkages will provide the new ability to look at problems from different points of view and to chart and manipulate constructive commentary on the problems, thereby creating opportunities for new understanding. Chapter 9. Mapping Future InfoSpace: Summary and Trends Summary of the Argument 234 Trend: Integrate Communication and Computing 236 Navigating Through Whole Subject Matters 238 Navigating Along Hypertrails 240 Looking from Multiple Points of View 242 Virtual Reality — A New Tool 244 Travelling in Large Visual Landscapes 246 Heading into Future Information Landscapes... 248 .© 1989 R. E. Horn Chapter 9 Mapping Future InfoSpace: Summary and Trends mM : m l»g| - I ■■ |ii|l 1 . In this Lizr Chapter we sum up and look at the trends - Summary of the Argument We have made the case in this book that Hypertext will help us get our on-line text organized in a new way, following associative trails that are more like the way human memory operates. nodes Information Mapping's method is mature technology for analyzing, organizing, writing, sequencing and formatting of information. We have shown how it forms an appropriate rhetoric for the writing of hypertext and hypermedia databases. Argumentation analysis, a still evolving methodology of understanding disputes, helps in particular ways when we are looking at important issues. Putting these three emerging technologies together may very well help the science information system with its problems of complexity and information overload. 70 m x o 3 Abstract Abstract Abstract Abstract 234 Chapter 9. Mapping Future InfoSpace buttons. — SH" .,(!■ . y ==!!= — ~7T ' ' ' == = We have also shown that there are a number of problems that emerge out of the technology of hypertext, problems such as... And we have shown how Information Mapping's method resolves some of these problems. What shall the nodes contain and the links represent? Overchoice and Cognitive Overload Labor-Intensive Maintenance 235 Trend: Integrate Communication and Computing Introduction We can see the increasing integration of a number of technologies, hardware, software, and "mindware" that will bring about a much more flexible, convenient, and fertile medium for knowledge workers in the next decade and beyond. Much of the discussion in this book has been about the "mindware" aspects of these developments. But they all interact. What visionaries have "seen" twenty or thirty years ago is coming to pass. @ vO 00 VO 7> ffl X o 3 New Interactive Multi-Media Capabilities of Computers Increasing Electronic Integration of Home, School, and Workplace 236 Chapter 9. Mapping Future InfoSpace increasingly Powerful Workstations and Personal Computers Increasingly "Intelligent" Software Ability of Knowledge Workers to Manage targe Amounts of Information Through Analytic Techniques Increasingly Visual Language Approach to Communication Everything Needed for Widespread Use of Hypertext and Hypermedia Mindware Trends Better Understanding of How Computer Screens and Interaction Can Be Designed to Make Them Easier to Use Software Trends 237 Economics, operations research, programming, games Biology and behavioral sciences Systems, control Information and communication, circuits, automata Subject: Mathematics Hypertrail: Major Branches General History and biography Logic and foundations Set theory Combinatorics, graph theory Order, lattices, ordered algebraic structures General mathematical systems 'Number theory \ Algebraic number theory, field theory and polynomials Commutative rings and algebras Algebraic geometry Linear and multilinear algebra; matrix theory Associative rings and algebras Nonassociative rings and algebras Category theory. homological algebra Group theory and generalizations Topological groups, Lie groups Functions of real variables \ Measure and integration Functions of a complex variabli Potential theory I Several complex variables and I analytical spaces \ Ordinary differential equations Navigating Through Whole Subject Matters Introduction On the next few pages we will present a simulation of what it might be like to use some of the capacities of a future hypertext system with hypertrails. This wall¬ sized conference room is described in a previous chapter. The Classification of Mathematics 238 J Ip m 1 1 III till —Ml HtalMnr Chapter 9. Mapping Future InfoSpace Subject: Probability Hypertrail: Prerequisite Algebraic topology \ Manifolds and cell complexes _ Global analysis, analysis on manifolds / Probability theory and / stochastic process / Statistics I Numerical analysis — - - \ Computer science / \ General applied / \ mathematics Mechanics of particles and systems Mechanics of solids Fluid mechanics, acoustics Optics, electromagnetic theory Subject: Set Theory Hypertrail: Classification Classical thermodynamics, heat transfer Quantum mechanics Statistical physics, structure of matter Relativity Astronomy and astrophysics .Geophysics Geometry Convex sets and geometric inequalities Differential geometry General topology Qaaator. thfinrv Sequences, series, summability Approximations and expansions Fourier analysis Abstract harmonic analysis Integral equations Functional analysis 1989 R. E. Horn Navigating Along Hypertrails Introduction This is a prerequisite hypertrail of part of the subject matter of basic sets and probability theory. For more information on prerequisite hypertrails f~ Venn Diagrams Universal Counting the number of subsets in a Subset Three ways of describing a sample Sample Sample Elements of a set Simple Experiment Outcome of a simple experiment Event Disjoint sets Ways of specifying sets Intersection Operations with two or more sets Union Difference mmmaw}} Complement of a set Every set is a subset of itself The null set is a subset of every set Chapter 9. Mapping Future InfoSpace The probability of the complement of an event Probability assignment based on theoretical assumptions: equally likely outcomes Complement of an event Joint probability Probability assignment based on empirical assumptions relative frequency Two conditional probabilities can be calculated for any two events Empty event Assigning probabilities to elementary events The numerical measure of probability Reduced sample space Independent of more than 2 events The probability of an event Conditional probability Elementary event Independent events Mutually exclusive events Independent events are not mutually exclusive events General addition law Special case of the general addition law: mutually exclusive events Experiments with equally likely outcomes The intersection of 2 or more events is also an event Elemeiv events! combin form ot] events * The union of 2 or more events is also an event 1989 R. E. Horn Kinds of sets which can further be sorted into . . . Subject: Set Theory HypertraH: Classification Relationships among sets which can further be sorted into . . . Identity Laws Complement Laws Associative Laws Looking from Multiple Points of View Introduction This is a classification hypertrail of part of the subject matter of basic sets and probability theory. For more information on classification hypertrails | see page> 130 r 1 1 - 1 . . . . - 1 Universal Null Sub set set s et Unit Equivalent Proper sets sets subsets \ Disjoint sets Complement of a set Chapter 9. Mapping Future InfoSpace SETS Elements of a Set can be sorted into Ways of specifying sets which can be further sorted into .. Listing the members Giving a rule Venn Diagrams Algebra of Sets which can be sorted into . . . Operations with two or more sets which can further be sorted into . . . 1 1 Union Intersection Difference Distn \ Vgan's Commutative Ln ji € i i 251 Bush: Inventor of the Concept of Hypertext Introduction World War II is over. The Director of the U.S. Government's Office of Scientific Research and Development, science advisor to the President, writes an article in the Atlantic Monthly in which he sketches his vision of a tool that will aid individual knowledge workers. "Consider a future device for individual use, which is a sort of mechanized private file and library. It needs a name, and, to coin one at random, memex will do. A memex is a device in which an individual stores all his books, records, and communications, and which is mechanized so that it may be consulted with exceeding speed and flexibility. It is an enlarged intimate supplement to his memory." With these words, Dr. Vannevar Bush describes what is to become the personal computer and hypertext systems of today and tomorrow. All quotes are from Bush’s 1945 article "As We May Think." Scanning as Input In the Bush machine, input was done by photography. The user would place books, photos, handwritten notes, etc., face down on a transparent glass plate, then "the depression of a lever causes it to be photographed onto the next blank space in a section of the memex film..." Display Screens Bush visualized having two display screens so that you could compare data from two documents. Mass Storage Bush was writing before the digital computer was fully invented and produced, so he conceived of micro¬ film as the mass storage medium. Inside the memex is the microphotographic storage device. Bush speculated, "...if the user inserted 5000 pages of material a day, it would take him hundreds of years to fill the repository, so he can be profligate and enter material freely." Purchase Published Documents Bush thought there would be a market for books and articles published on microfilm that could be simply dropped into the memory. "Business correspondence takes the same path," he suggests. I 252 Appendix A The idea is published Vannevar Bush Adding Personal Links Bush also foresaw the idea of user-created links. "A special button transfers him immediately to the first page of the index. Any given book of his library can thus be called up and consulted with far greater facility than if it were taken from a shelf. As he has several projection positions, he can leave one item in position while he calls up another. He can add marginal notes and comments, taking advantage of one possible type of dry photography..." Retrieving Trails of Links Bush had a vivid idea of how the retrieval of links would take place. He wrote, "...associative indexing, the basic idea of which is a provision whereby any item may be caused at will to select immediately and automatically another.. ..When the user is building a trail, he names it, inserts the name in his code book, and taps it out on his keyboard. Before him are two items to be joined, projected onto adjacent viewing positions... The user taps a single key, and the two items are permanently joined... Thereafter, at any time, when one of these items is in view, the other can be instantly recalled merely by tapping a button... Moreover, when numerous items have been thus joined together to form a trail, they can be reviewed in turn, rapidly or slowly, by deflecting a lever...." Keyboard Input "There is a keyboard, and a set of buttons and levers Otherwise it looks like an ordinary desk." Rapid Browsing Part of the attractiveness of the idea of the memex for Bush was rapid access to the scientific and technical literature. "There is, of course, provision for consultation of the record by the usual scheme of indexing. If the user wishes to consult a certain book, he taps its code on the keyboard, and the title page of the book promptly appears before him... On deflecting one. ..lever.. .to the right he runs through the book before him, each paper in turn being projected at a speed which just allows a recognizing glance at each." Engelbart's Augment: First Operational Hypertext Introduction Douglas C. Engelbart, then with Stanford Research Institute, built the first working and usable hypertext system. His Augment hypertext system, currently marketed by McDonnell Douglas, has supported a group of a thousand or more knowledge workers over 20 years. It provides the most sophisticated demonstration of the structured hypertext principles as well as the idea of an on-line community of knowledge workers that has been implemented. Here we present a brief overview of the Augment system and salute Engelbart for his accomplishments. Throw-Away Mai) « Comments * Letters « Memos External Documents (e.g. , printed books, films, videos, research papers, etc. ... in other words, all of the world outside the Augment System.) « Community intelligence analyses • Indexes • Reports • Meeting • Notes Shared Files . . . . . . . Journals i . . Personal Workspace and Files/ for Each User ) Private mail Personal files Personal working space Notes Ideas Personal Workspace 1 J|i f|| I **1 Jt 11 jMf js- Appendix A First Implementation of Hypertext 1 962-75 Douglas C. Engelbart on the Central Mainframe Computer Engelbart: Edison of the Personal Computer Introduction Doug Engelbart is the Edison of the personal computer. He not only invented many of the familiar devices we have on our PC's and workstations, but also was the first builder of a working hypertext system. His research program was built on an extraordinarily broad vision of "augmenting human intelligence." Here we record just some of the major accomplishments of Engelbart and his colleagues at Stanford Research Institute. His Augmentation Research Laboratory began in 1962 and had a working personal computer with the hypertext system and on-line group work environments by the mid-sixties. Among the accomplishments of Doug and the Laboratory are the following: Invention of the Mouse First Major Implementation of Electronic Mail Invention of Multiple Window on Computer Screen First Implementation of Word Processing Invention of On-Line Integrated Help Systems Invention of Outlining Software And Idea Processors 256 Appendix A Invention of Computer Supported Group Conferences Invention of Composite Text- Graphic Files Invention of Many Interface Elements Including a Very Efficient 5-Finger Input Device First Implementer of Hypertext Links and Nodes Invention of Shared Screen Teleconferencing Invention of Remote Procedure Call Protocol for Efficient "Reach Through" Integration of Functions 257 Nelson: Name-Giver of the Word "Hypertext" Introduction Ted Nelson coined the terms "hypertext" and "hypermedia" in 1965 and has acted as an evangelist for the concepts ever since. His definition of hypertext is "computer-supported non-sequential writing." His visionary idea of a "docuverse" containing all of humankind's documents linked has inspired a generation of researchers and educators. "Imagine everything available and tied together. Grand visions come to mind of what things will be like when 'it’s all there and linked.' The thought of that great body of material calls to us, calls to us like the ocean." •Xmjtfo "But the ocean of universal hypertext is not enough: we want free sailing on it.... A world of open hypertext publishing promises extraordinary new freedom for the mind, a new empowerment of humanity." >>#!#<&. The Xanadu Vision Xanadu® is Nelson's plan for a "world-wide network, intended to serve hundreds of millions of users simultaneously from the corpus of the world's stored writings, graphics, and data." "Xanadu is not a large centralized software system but rather an idea for software for running a decentralized network." As Nelson says, "It is a design for a new literature and a system of order to make such a network understandable, usable, and readily expansible to any degree...." Storage System Xanadu is a concept of a storage system that permits documents to be stored only once in a "universal data structure to which all other data may be mapped." Address and Linking System The address and linking system permits "any s^ans of bytes in any document or file, on any server, (to be) linked to any other spans of bytes, in any document or file, oh any server, by a link type which is unique or used elsewhere in the system." Authoring The system would permit 1 . allocation of credit of authorship rind publishing 2. allocation of payment of royalties based on the reader’s use of documents 3. quotability of any document, yit easy tracing to the source of the quotation via hypertext links. Appendix A The Words Hypertext and Hypermedia Theodor Holm Nelson ' Everything is deeply intertwingled. "Imagine making your own notes and connections any way you choose in this great interconnected corpus; so that any time you want to reopen this great hypertext world at any of these private annotations that make it your own, it will be like opening a book to a bookmark." . . "Universal or grand hypertext, then, means a new publishing system -- an accessible great universe of linked documents and graphics (and audio recordings and video and movies). This is an idea many people now share — the idea that we can get to everything, keep track of everything, add to e.v.ety. thing, tie everything together, that we can have it ail." . • MM »•».»' . . . . . . . 1 ”By ’hypertext* I mean non-sequential writing.” mililUfeii Van Dam and Brown: First University Instruction Introduction Since the late 1960's, Andries van Dam and a team at Brown University have created several generations of experimental hypertext and hypermedia systems. Their focus has been on the use of these systems in college instruction. English Poetry One system was used in the early 1970's to teach an English poetry class. Students worked together on the same hypertext document, reading and writing on computer terminals that displayed the hypertext consisting of poetry and commentary. Biology and English Literature Two more classes largely supported in hypertext have been developed. The Brown team has in the last few years built a group of multi-media workstations and taught courses in cell biology and English literature on the system. Prototype sections of other courses have also been implemented. This work has produced important information on how to integrate hypertext documents into normal teaching-learning environments. Poem Instructor's commentary 71 Poem Sludcnl commcnlai'\ F= structor's commentary \ Poem/ Historical Events Satirica Style and Comparisons Biography of Authors and Contemporary Figures @ s 00 PC m x o 3 Time Line for History 1755 1760 260 Appendix A Working Hypertext in Instruction Andries van Dam and Brown University Group Student commentary Student commentary X ^ \ _ L _ \ Student commentary Instructor's commentary Instructor's ' commentary \ Student commentary Instructor's commentary Poem/ Instructor's commentary Student commentary Instructor's commentary 1765 1770 1775 j i i 1785 1790 -i — h 261 Zog Group at Carnegie-Mellon: Menu Interfaces Menu-Driven Interfaces for Hypertext Screen-Sized Frames In 1972, a group at Carnegie-Mellon University that has included Allen Newell, Donald L. McCracken, Robert M. Akscyn, and George G. Robertson began building a series of experimental hypertext systems that were given the collective name Zog. Their work was focused on making a system that would produce rapid response in large networks through a simple menu selection interface. Zog was designed to serve a large community of users. The database in KMS consists of screen-sized frames which may contain "any mixture of text, graphics and image items, each of which may be linked to another frame or used to invoke a program." These frames may be stored in the memories of different workstations on the network. Here we show displays of different frames on the screens of several workstations. Nuclear Aircraft Carrier Application The group was given the opportunity in 1980 to implement its work on the new U. S. Navy nuclear-powered carrier, USS Carl Vinson. They developed a new version that supported the ship's organization and regulations manual and a planning and evaluation application. Knowledge Management System Out of the work on the USS Vinson, a commercial version of the Zog system has been marketed since 1983 under the name Knowledge Management System (KMS). It is implemented on Sun workstations. Current Version of KMS The current version of KMS is particularly well suited to the joint creation of documents on different workstations in a network, such as when many engineers have to work on a O' 00 VO m sc o 3 262 Appendix A Multi-User Menu-Driven Interfaces with Large Database July Fg72 The Zog Group Hierarchical Structure Emphasized KMS emphasizes hierarchical structures and retains fairly conventional implementations of tables of contents, menus, and indexes as key interface devices. Non-hierarchical links are possible. x ■M is: X m. Negroponte and Bolt: Spatial Dataland Managing Information Spatially in Dataland uiMwnwMiim . . . i Si Joy Stick on each armrest for directing travel around the screen. The Architecture Machine Group at MIT in the late 1970's build a number of experimental information environments that expanded the vision of what the possibilities of interacting with the computer can be. They called their information space "Dataland" and it operated in a room where almost everything was manipulatable information. The room, noted Bolt, is the computer terminal. Many of the functions, such as calendars and calculators, that we routinely use on our visual computer interfaces were first demonstrated in Dataland. Strictly speaking, the experiment was not about hypertext but about hypermedia. The ability to switch media and move around in an information environment was the key demonstration. We diagram the room-terminal on these pages. Cursor The "You Are Here" cursor located on the monitor. for "touch travel." They enable users to point to places on "key maps" to navigate in the information space, for example, to control types of functions such a telephone or calculator. Touch sensitive TV monitors Spatial Information Management Principle One major concept used by the dataland experiment is called the "managing things spatially" principle. People "have a place" for information, suggested George Miller. We keep our messy desks because we remember where things are. If we straighten it up, we lose our spatial memory cues. 264 Appendix A The MIT Architecture MachineGroup Managing Information Spatially Nicholas Negroponte and Richard A. Bolt Each Object On Display Can Be Activated Each of the objects displayed can be "zoomed in" on for greater detail. Display Screen Whole wall is display screen. Voice Activated Commands User wears a speech recognition microphone for voice travel. User can say things like "go to the book in upper left," "create a green circle. ..there," "move data A to green circle," and so on for "copy, delete, make smaller.. .larger, call that...,"etc. Loudspeakers Four loudspeakers located in wall provide wrap-around sound. Brown and Guide: Hypertext for PC and Macintosh First Commercial System for Two Major Personal Computer Systems To Owl International, Inc., which was founded in 1985, goes the credit of bringing out the first hypertext system to work on both of the major personal computer platforms, the IBM PC and Apple's Macintosh. Peter Brown of the University of Kent (U.K.) was the inventor of the Guide system. The products are based on further development work at Office Workstations Limited of Edinburgh, Scotland. Owl has continued to increase the flexibility of Guide and to equip it with a family of support products including Guidance, which provides a context-sensitive environment for online reference and tutorials and Guide Reader, a low cost version that permits reading hypertext, but not authoring. Guide also supports multi-media connections and the ability to link not only between documents but also between applications so that, for example, a user can link a text document with a Appendix A Scroll and Outline Architecture Guide relies heavily on a software architecture of scrolls of variable length, an outline structure of the document, and user-controlled expansion of that outline, which are revealed by clicking on portions of the outline. Other link types, such as the ability to link to other places in the text to pop-up notes and to activate other media are also part of the system design. Hypertext for Apple Macintosh and IBM PC DOS Computers 1986 Peter Brown and Owl International, Inc. Before pressing buttoi After pressing button | [button [ - — = -■= This text, normally hidden, is displayed when button is activated. Display takes place by making space available in outline. Sculley: Vision of the Knowledge Navigator Introduction John Sculley is a different kind of visionary. He is CEO of a major computer corporation. Yet many of his speeches have dealt with how we must change ourselves and our information environments in order to compete in the new information age. Sculley inspired and sponsored futuristic work at Apple on the Knowledge Navigator, which describes the possibilities for personal computing in the years beyond 2010. The computer as envisioned by Sculley is driven by voice-activated commands. The computer responds with computer-created speech through the little moving picture of the man in the bow tie. We picture here a sketch of the Knowledge Navigator, which is a book-sized personal computer which has access to large knowledge bases of information. Large, Flat, Full Color Screen "Large, flat display screen.. .full color, high-definition, television-quality images, full pages of text, graphics, computer-generated animation." The original Knowledge Navigator scenario was made into a videotape that simulated the functions of the computer and showed how the computer took its owner through a day that included an exploration of the problems of the destruction of the Amazon rain forests. Impact of the Knowledge Navigator on Education Sculley suggests, "Education will not simply be a prelude to a career, but a lifelong endeavor. Some of the important elements that will promote this new paradigm for lifelong learning are: (1) the development of conceptual skills, and the ability to test reality against multiple points of view; (2) the nourishment of individual creativity and the encouragement of exploration; (3) the encouragement of collaboration, and an emphasis on clear communication." PPIlilPIIBpW 1 . : 1 ■11! I 1 ' ' @ 3 Customize Knowledge "Most important, the Knowledge Navigator will customize knowledge for you. ..to make navigating through information and ideas as interesting and understandable as possible." Plan Network Appendix A Tools Schedule Agent Vision of the Knowledge Navigator John Sculley Atkinson: First Commercial Hypertext "Hit" Introduction HyperCard, developed by Bill Atkinson, is a multi-functional software tool that includes many hypertext properties. Apple Computer made it the first hypertext "hit" by deciding to give it away with the purchase of a computer. Appendix A One Million HyperCard Programs in One Year Bill Atkinson HyperCard HyperCard rapidly became the hit of 1987, far outstripping competing hypertext systems and enabling enthusiasts and commercial applications to hook up to laser disks and CD/ROM's to tap enormous text and graphics files. Card Architecture As the name implies, HyperCard relies on a software architecture and interface that appears to the user as a stack of index cards. However these cards are linked in a great variety of ways that give considerable flexibility in the final development of hypertext and hypermedia on it. HyperCard’s Key Components Atkinson had the genius to put the metaphor of hypertext together with an easy-to-use programming language, a simple word processor, a painting program and an elegant, inviting interface. The ease of use and the combination of functions of HyperCard provided a significant jump for hypertext. HyperCard Focused Attention HyperCard almost singlehandedly brought the idea of hypertext into the minds of well over a million people in one stroke, when Apple Computer's John Sculley decided to give it away with each purchase of a Macintosh computer. While HyperCard is much more than hypertext software, it put hypertext on the map. 271 Notes Chapter 1. Introduction to Hypertext and Hypermedia p. 6. What is Hypertext? "Hypertext is both an . . Brown University (1985). "By 'Hypertext' I mean non-sequential writing." Nelson (1974). p. 16. The Navigation Through Information Space Metaphor. "Ted Nelson's gigantic vision . . ." Nelson (1974). p. 20. Hypermedia Application: New Product Marketing. Lou Casabianca of Hypermedia Magazine first showed me an application like this. p. 22. Case Study: Hypermedia for Shakespeare. This discussion of hypermedia for Shakespeare was inspired by an implementation at Brown University and by a similar one at Stanford University p. 29. Dimensions of Hypertext Systems. "Theodore Nelson has suggested ..." Nelson (1974). Chapter 2. Current Issues With Hypertext p. 47. Where to Put How Many Buttons of Which Kind. The seductive buttons were inspired by a slide from Theodor Nelson's dynamic slide show on hypertext. p. 48-51 Inadequate (and Missing) Reading Cues. Material on these four pages follows the excellent paper by Chamey (1987) and summarizes this paper. p. 52. Branching Difficulties of Serialist Readers, see Pask (1976) p. 56. Lost in Hyperspace. These issues were most compellingly raised by Conklin (1987). p. 58 . Overchoice and Cognitive Overload. These issues are also well put in Conklin (1987). p. 60. Chaos in Titles for Documents and Their Parts. The author thanks Michael J. Steinback for formulating commandment number 7. Chapter 3 Introduction to Information Mapping's StructuredWriting Method p. 76. Overview of This Chapter. Information Mapping, Inc. (for further information on the products and services of the company and licensing of the methodology for software or training, contact Information Mapping, Inc., 303 Wyman Street, Waltham, MA. 02154 or call 617-890-7003) Brief History of Information Mapping. See items listed under Horn in the bibliography. Other Examples of Applications. Application of Information Mapping's Methodology to Philosophy. Several authors in Horn, ed., (1983) use the methodology in essays on metaphysics, cybernetics, and logic. p. 82. The Problem of Human Short Term Memory. For further information on these pages see Miller (1956) and Simon (1979). p. 96-97. Examples of Maps Displayed on Paper. The author acknowledges the permission of Information Mapping Inc. to reproduce these two sample pages of Information Mapping and other example material in this chapter. p. 107. A Brief Discourse Analysis (Stable Subjects). The data in the two examples are from unpublished data of Horn. p. 110. What are the information types? Six of seven of these were first suggested by Horn (1965). See also Horn (1969), (1971, 72, 76) for further information. 273 p. 122. Meeting Criteria for Better Communication. Among the data supporting these claims are eleven doctoral and two masters dissertations on the Information Mapping Method. See references: Burrell (1978), Cheung (1980), Falk (1980), Fields (1982), Hauck (1985), Mcclung (1985), Olivares-Guerrero (1985), Reid (1984), Romiszowski (1977), Skelly (1982), Soyster (1980), Stelnicki (1980), Stuart (1979). Other research conducted on the method has been reported by Webber (1979), Jonassen (1979, 1981), Jonassen and Falk (1980) and Schaffer (1982). Other studies have focused on components used in the method (for example, see Smith and Mosier, 1986). In addition, approx. 20,000 people have been trained in the methodology through courses given by Information Mapping, Inc. Chapter 4 Navigating Structured Hypertrails p. 128. Prerequisite. Mathematics example adapted from Kemmeny (1965) see also Horn et. al. (1969) p. 144. Example Hypertrails. Example One (on dreams), see Bonime (1982) "Example of example hypertrail" is from an article which first appeared in Horn (1976). Chapter 5 Resolving Some Hypertext Problems p. 152. At the Nodes, Blocks and Maps Structure Hypertext. The author acknowledges permission of Information Mapping, Inc. to publish the map on the example of an information map. p. 158. Addressing The Major Reading Cues Problem. The best sources are Horn (1976) currently used as course manuals in Information Mapping's courses. Chapter 6 Relatively Stable Discourse: Documentation and Training p. 170-171. Operations and Technical Manuals. The author acknowledges permission of Information Mapping, Inc. to reproduce these two maps. p. 172-173. Personnel Manuals and Policy Manuals. The author acknowledges permission of Information Mapping, Inc. to reproduce these two maps. Chapter 7 Disputed Discourse: Argumentation Analysis p. 186-187. Overview of This Chapter. The discussion on this page is from Toulmin (1958) as well as the discussion on the next three pages of claims, grounds, warrants, backing, rebuttal, and qualifiers. Extensive use was also made of Toulmin et. al. (1979). p. 200-204. Case Study. The example is from an unpublished study of using argumentation analysis in examining the ethics of using nuclear weapons done at the Lexington Institute. The most extensive use of Toulmin structures in hypertext has been done by Cathy Marshall (1987). p. 204-206. Comparing Ill-Structured and "Tame" Problems. The material on these pages is from an excellent discussion in Mitroff et. al. (1983). Chapter 8 Experimental Discourse: Scientific Information p. 218. Miller: Short Term Memory Limits and Chunking. The chart is from Miller (1956). The author gratefully acknowledges permission of the American Psychological Association to reproduce it. 274 Notes p. 224. Schaffer: Information Mapping's Methodology. The quotes are all from Schaffer (1982). The author gratefully acknowledges permission of the NSPI Journal to reproduce the quotes. p. 226. Reid and Wright: Superiority of Visual Structuring. The four diagrams are from Reid and Wright (1973). The author gratefully acknowledges the permission of the Journal of Applied Psychology to reproduce the four examples in the "material used section." p. 228. Hypertext May Facilitate Identifying Problems. The two quotes are from Root-Bemstein (1982). p. 230. Linked Comments Will Highlight Deficiencies. The references made are to Drexler (1986). Chapter 9 Mapping Future InfoSpace p. 238-239. Navigating Through Whole Subject Matters. Subjects of Mathematics. The categorization of mathematics on this page and the following page is adapted into graphic form from Davis and Hersh (1981) who attributed it to a report in the Mathematical Reviews (1979). p. 241. Navigating Along Hypertrails. Prerequisite Hypertrail . The diagram on this page is from Horn (1969). p. 244. Virtual Reality— A New Tool . The quote is from Walker (1988). p. 246. Travelling in Large Visual Landscapes. The graphic possibilities of large landscapes like this have been suggested to me by Jim Channon and David Sibbet. I have taken their 2-D work and applied it to the 3-D world of virtual reality. Appendix A: Some Historical Notes p. 252. Bush: Inventor of the Concept of Hypertext. All the quotes from this page are from Bush (1945). The author gratefully acknowledges permission of Atlantic Magazine to reproduce these quotes. p. 258. Nelson: Name-Giver of the Word "Hypertext." The quotes on this page are from Nelson (1988) and Nelson (1974) p. 260. van Dam and Brown: First University Instruction. This account is from Yankelovich (1985). p. 262. The Zog Group at Carnegie Mellon. This discussion is based on Newell et. al. (1981). p. 264. Negroponte and Bolt: Spatial Dataland. Details of the material on this page can be found in Bolt (1984). p. 268. Sculley: Vision of the Knowledge Navigator. The quote on this page is from Sculley (1989). Other quotes are from Sculley (1987). 275 References Hypertext A Critical Assessment of Hypertext Systems, (1988) ACM CHI '88 Proceedings Akscyn, R., McCracken, D., Yoder, E. (1987) KMS: A Distributed Hypermedia System for Managing Knowledge in Organizations. Hypertext '87 Papers. November 13-15 1987 Barrett, E., ed. (1988) Text, ConText, And Hypertext. Cambridge, MA: MIT Press. Beeman, W.O., Anderson, K.T., Bader, G., Larkin, J., McClard, A.P., McQuillan, P., Shields, M. (1987) Hypertext and Pluralism: From Lineal to Non-lineal Thinking. Hypertext '87 Papers. November 13-15, 1987 Binder, Carl. (1988) Hypertext: Features and Benefits. Managing End-User Computing Vol. 1, No. 9. April, 1988 Binder, Carl. (1988) Hypertext: Sorting Out the Software Options. Managing End-User Computing Vol. 1, No. 9. April, 1988 Binder, Carl. (1987) The Promise of a Paperless Workplace. Optical Insights Fall 1987 Boston Computer Society, Hypermedia Group, Hypermedia Resource Base, Available as indexed bibliography (paper or on disk) BCS, Hypermedia Group, One Center Plaza, Boston, MA, 02108 Brown, P. J. (1987) Turning Ideas into Products: the Guide System. Hypertext '87 Papers. November 13-15 1987 Bush, V. (1945) As We May Think. Atlantic Monthly 176.1 (July 1945): 101-108 Chamey, D. (1987) Comprehending Non-Linear Text: The Role of Discourse Cues and Reading Strategies. Hypertext '87 Papers. November 13-15 1987 Conklin, J. (1987) Hypertext: An Introduction and Survey I EE Computer Sept, 1987 Drexler, K. E. (1986) Engines of Creation New York: Anchor Press, Doubleday Drexler, K. E. (1987) Hypertext Publishing and the Evolution of Knowledge. Unpublished paper Ehrmann, S.C. (1987) Hypertext as a Medium for The Three Academic Conversations Unpub. Paper Nov. 24, 1987 Engelbart, D. C. (1984) Authorship Provisions in Augment COMPCON '84 Digest, Proceedings of the 1984 COMP CON Conference. Feb. 27-March 1, 1984. 465-472 Engelbart, D. C. (1963) A Conceptual Framework for the Augmentation of Man's Intellect Vistas in Information Handling, Vol. 1 . editors. P. D. Howerton and D. C. Weeks. Washington, D.C., Spartan Books, 1963 1: 1-29 Frisse. M.E. (1987) Searching for Information in a Hypertext Medical Handbook Hypertext '87 Papers November 13-15 1987 Gans, D. (1988) Ted Nelson And The Ultimate Information Machine Microtimes April, '88 Glushko, R.J., Weaver, M. D., Coonan, T.A., Lincoln, J.E. (1988) "Hypertext Engineering": Practical Methods for Creating A Compact Disc Encyclopedia. ACM Conference on Document Processing Systems. Santa Fe. Dec. 5-8, 1988 Goodman, D. (1987) The Complete HyperCard Handbook New York, Bantam Books Halasz, F. G. (1987) Reflections on NoteCards: Seven Issues for the Next Generation of Hypermedia Systems Hypertext '87 Papers. November 13-15, 1987 Hanson, R. (1987) Toward Hypertext Publishing Unpublished paper, August, 1987 Nelson, T.H. (1988) The Call of the Ocean: Hypertext Universal and Open Hyper Age, Vol. 1, No. 2, May-June, 1988 Nelson, T. H. (1974) Dream Machines: New Freedoms through Computer Screens - A Minority Report. Computer Lib: You Can and Must Understand Computers Now Chicago IL., Hugo's Book Service (reprinted by Microsoft Press, 1988) Nelson, T. H. (1981) Literary Machines Swathmore, PA Oren, T. (1987) The Architecture for Hypertexts Hypertext '87 Papers. November 13-15, 1987 Raskin, J. (1987) The Hype in Hypertext: A Critique Hypertext '87 Papers November 13-15, 1987 Remde, J.R., Gomez, L.M., Landauer, T.K. (1987) SuperBook: An Automatic Tool for Information Exploration - Hypertext 1 Hypertext '87 Papers. November 13-15, 1987 Rheingold, H. (1985). Tools For Thought New York: Simon and Schuster. 277 References Schecter, G. (Ed) (1967) Getting It Out of Our System Nelson, T. Information Retrieval: A Critical View Wash. D.C. Thompson, 1967 Schneiderman, B. (1987) User Interface Design for the Hyperties Electronic Encyclopedia Hypertext '87 Papers Nov. 13-15. 1987 Scully, John (1987) Odyssey New York, Harper and Row Slatin, J. M. (1988) Toward a Rhetoric for Hypertext Presentation for Hypermedia '88, Univ. of Houston, Clear Lake, Johnson Space Center, NASA, 1988 Trigg, R. H. (1983) A Netxs’orked-based Approach to Text Handling for the On-line Scientific Community College Park, MD, Univ. of Maryland, 1983. Univ. Microfilms #8429934 Walker, J. H. (1987) Document Examiner: Delivery Interface for Hypertext Documents Hypertext '87 Papers Nov. 13-15, 1987 Waller, R. (1986) What electronic books will have to be better than. Information Design Journal, Vol. 5, No. 1, 1986 Weyer, S. A. (1982) The Design of a Dynamic Book for Information Search International Journal of Man-Machine Studies 17.1 (July 1982): 87-107 Yankelovich, N., L. N. Garrett and K. E. Smith et al. (1988) Issues in Designing a Hypermedia Document System: The Intermedia Case Study Interactive Multimedia Redmond, WA. Microsoft Press The Information Mapping Method Burrell, L. O. (1978) Effectiveness of self-assessment guides as a method of teaching critical care nursing to professional students. Unpub. Ph.D. dissertation. Univ. of Georgia, 1978 Cheung, Raymond Yuk-Ming, (1980) Development of and Information Mapped Text on Basic Unpublished M.A. Thesis, Concordia University. Montreal, Canada Falk, L.M. (1980) Retrieval of information from self study texts, Unpub. Ph.D. dissertation, Temple University Fields, Alan, (1982) An empirical and conceptual investigation of Information Mapping® techniques. Unpub. Ph. D. dissertation, the Open University (U. K.) Hartley, J. and Trueman, M. (1983) The Effects of Headings in Text on Recall, Search and Retrieval British Journal of Educational Psychology, 53, 205-21 Hauck, L. S. Differences in Information Mapping® strategies in left and right brain learners, (1985) Unpub. Ed. D. dissertation, Pennsylvania State Univ. Horn, R. E. (1966) A Terminal Behavior Locator System Programmed Learning (U.K.) Feb. 1966 Horn, R. E. (1969) Come Along With Me Into My Custom-Made It's-Up-To-You Browsing-Leaming-Growing Move-Around Information Environment (If You Want To) reprinted shorter version as "Experiment in Programmed Learning" in Runkel, Harrison, and Runkel (Eds) The Changing College Classroom: Some Innovations in the Teaching-Learning Process, San Francisco, Jossey, Inc., Horn, R. E., Nicol, E., Kleinman, J., and Grace, M.(1969) Information Mapping for Learning and Reference Cambridge, I.R.I. (A.F. Systems Command Report ESD-TR-69-296) Horn, R. E., Nicol, E., Roman, R., et.al.(1971) Information Mapping for Computer-Learning and Reference Cambridge, I.R.I. (A.F. Systems Command Report ESD-TR-7 1-165) Horn, R. E.,(1972) Interview with R.E. Horn in Industrial Training International (U.K.) pp. 232-234 Horn, R. E., (1972) Information Mapping: New Tool for Learning and Research Improving Human Performance : A Research Quarterly Vol. 1, No. 2, June 1972 Horn, R. E.(197_) Information Mapping Japanese Journal of Educational Technology Spring Horn, R. E.( 1 97 1 ) Information Mapping: New Training Technology for the 1970's, a speech presented to the First National Conference on Technology of Higher Education Rio de Janeiro, June 1971 Horn, R. E.(1974) Briefing on Information Mapping Training Magazine March 1974, Vol.II, pp. 27-30 Horn, R. E. (1974) Information Mapping: New Tool to Overcome the Paper Mountain Educational Technology Vol.XIV, No. 5, May 1974, pp. 5-8 Horn, R. E.(1975 Information Mapping, Datamation January 1975 pp. 85-88 Horn, R. E. (1975) Is Information Mapping for Me? Training Magazine 1975 Horn, R. E. (1976) Developing Instructional Materials and Procedures: 1967. Waltham, MA: Information Mapping 2nd edition, 1979; 3rd edition, (new title: Developing Procedures, Policies, and Documentation) Horn, R. E. ( \911)Writing Reports , Waltham, MA, Information Mapping. 2nd edition 1978; 3rd edition retitled Effective Reports, Proposals, and Memos Horn, R. E. (1979) Information Mapping, A presentation to the National Society for Performance and Instruction National Convention, April 1979 278 References Horn, R. E.(1985) Results with Structured Writing Using the Information Mapping Writing Service Standards, an invitational presentation at Designing Usable Text, a conference sponsored by the Open University, Nov. 3-7, 1980 Toward More Usable Text: an Applied Research Perspective, Duffy, T.M. and Waller, R. Academic Press, 1985 Horn, R. E.(1980) Information and Decision Management through Structured Writing, a concept paper for Delta Force, U.S. Army War College Horn, R. E.(1980) Structured Writing Possible Solution to Documentation Problems, a presentation to the IBM Guide UsersConference, Nov. 13, 1980. Miami, Florida Horn, R. E. and John N. Kelly (1981) Structured Writing— An Approach to the Documentation of Computer Software) The Newsletter of the Special Interest Group on Documentation of the Association for Computing Machinery. Horn, R. E.(1982) Structured Writing and Text Design, in Jonassen, D.H. The Technology of Text: Principles for Structuring, Designing, and Displaying Text Englewood Cliffs, N.J. Educational Technology Press, 1982 Horn, R. E.(1985) Recent Perspectives on the Information Mapping Method Waltham, MA. Information Mapping, Inc. Horn, R. E .The Engineering of Documentation— The Information Mapping Approach (1987) Waltham, MA. Information Mapping, Inc. Horn, R. E. (1987) Twelve Myths That Lead to Poor Documentation Waltham, MA. 1987, Information Mapping, Inc. Jonassen, D. H. (1979) Recall and Retrieval from Mapped and Programmed Text, Paper presented at AECT Convention, New Orleans, LA, Feb. 1979. Jonassen, D.H., and Falk, L.M. (1980) Mapping and Programming Textural Materials Programmed Learning and Educational Technology . Feb., 1980 Jonassen, D. H. (1981) Information Mapping: A Description, Rationale and Comparison with Programmed Instruction, Visible Language , XV, 1, 55-66. Jonassen, D. H., ed. (1985) The Technology of Text Englewood Cliffs, N. J.: Educational Technology Publications Mcclung, L. G. (1985)77zc church growth! church planning study guide: a two-phase reading and self-study course unpub. D.Miss.dissertation, Fuller Theological Seminary Olivares-Guerrero, R. A., (1985) Manual for the evaluation of the teacher education programs in Chile Unpub. Ed. D. dissertation, Columbia University Teachers College Reid, U.V. (1984) Instructional systems development: a new approach to education planning in the health systems of the commonwealth Caribbean Unpub. Ed. D. dissertation, Columbia University Teachers College Romiszowski, A.J. (1977) A study of individualized systems for mathematics instruction at the post secondary levels Unpub. Ph.D.dissertation. Loughborough University of Technology (U.K.) Romiszowski, A.J. (1981) Designing Instructional Systems New York: Nichols Publishing Romiszowski, A.J. (1986) Developing Auto-Instructional Materials New York: Nichols Publishing Romiszowski, A.J. (1984) Producing Instructional Systems New York: Nichols Publishing Ross, B. (1987) Reference-Based Training Managing End-User Computing. Information Mapping Dec. 1987 Schaffer, E. M., (1982) "The Potential Benefits of the Information Mapping Technique" NSPI Journal February 1982, p. 34 - 38 Skelly, W. H. (1982)77ze effects of a comparative advance organizer on knowledge acquisition and retention and its interaction with ninth grade male and female students of differing academic abilities in a unit utilizing the Information Mapping® approach to instructional design Unpub. Ed.D. dissertation. Temple Univ. Soyster, Thomas G. (1980)4 comparison of the effects of programmed instruction and the Information Mapping® method of instructional design on learning and retention of students with different mental abilities Unpub. Ed.D. dissertation. TempleUniv. Stelnicki, Michael. (1980) The effects of information-mapped and standard text presentations with fact and concept levels of learning on low general ability adult learner cognition Unpub. Ed. D. dissertation. Northern Illinois Univ. Stuart, Teresa Habito. (1979) The effectiveness of Information Mapping® compared with the conventional paragraph in communicating technical information Unpublished M.A. Thesis. University of the Philippines at Los Banos Webber, N. Some results of Using the Information Mapping Writing Service Standards at Pacific Telephone Co, Paper presented at NSPI Conference, Washington, D.C. 1979 279 References Argumentation Analysis Marshall, C. C. (1987) Exploring Representation Problems Using Hypertext. Hypertext '87 Papers. November 13-15, 1987 Mitroff, 1. 1., Mason, R.O., Barabba, V.P. (1983) The 1980 Census: Policymaking Amid Turbulence. Lexington, MA Lexington Books Perelman, C., Olbrechts-Tyteca, L. (1958) La nouvelle rhetorique: traite de V argumentation. Bruxelles, 1' Universitete Bruxelles Perelman, C., Olbrechts-Tyteca, L. (1958) The New Rhetoric. A Treatise on Argumentation. South Bend IN., University of Notre Dame Press Smolensky, P, B. Bell and B. Fox, et al. (1987) Constraint-based Hypertext for Augmentation. Hypertext '87 Papers Nov. 13-15. 1987 Toulman, S., Rieke R., Janik, A. (1979) An Introduction to Reasoning. New York: Macmillan Publishing Co. Inc. Toulman, S. (1980) The Uses of Argument. London: Cambridge University Press van Eemeren, F.H., Grootendorst, R., Kruiger, T. (1987) Handbook of Argumentation Theory. Dordrecht, Holland Foris Publications Other Works Cited Ambron, S., Hooper, K. eds. (1988) Interactive Multimedia. Redmond, WA. Microsoft Press Bandler,R. and Grinder, J., The Structure of Magic: A book About Language and Therapy, (1975) Palo Alto: Science and Behavior Books Bates, M. J. (1980) Idea Tactics. IEEE Transactions on Professional Communication , Vol. PC-23, No. 2, June, 1980 Bates, M. J. (1979) Information Search Tactics. Journal of the American Society for Information Science, July, 1979 Bolt, R.A. (1984) The Human Interface. Belmont, CA: Lifetime Learning Publications Bonime, W. (1982) The Clinical Use of Dreams, DaCapo Davis, P.J. and Hersh, R. (1981)77ie Mathematical Experience, Boston, Houghton Mifflin Foley, J. D. (1987), Interfaces for Advanced Computer, Scientific American, October 1987, 127-132 Greif, I., ed. (1988) Computer-Supported Cooperative Work: A Book of Readings San Mateo, CA: Morgan Kaufmann Publishers, Inc. Horn, R. E. (1967) Language and Change and Communication, Chicago, S. R.A. Horn, R. E. (1976) The Guide to Simulations! Games for Education and Training (Ed) Cambridge: I.R.I. Horn, R. E., editor, (1983) Trialectics: Toward a Practical Logic of Unity, Lexington, MA I.R.I. (now distributed by the Lexington, Institute, 80 Marrett Road, Lexington, MA, 02173) Lanier, J. Virtual Reality — An Interview with Jaron Lanier, Whole Earth Review, Fall 1989, 108-1 19 Macdonald- Ross, M. and Smith, E. (1977) Graphics in Text: A Bibliography, Milton Keynes, Open Univ. Institute of Educational Technology, Monograph No. 6 Miller, G. A. (1956) "The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information" Psych. Rev., 63, 2, March 1956, 81-96. Pask, G. (1976) Conversation Theory. Amsterdam: Elsevier Reid, F. and Wright, P.Written Information: Some Alternatives to Prose for Expressing the Outcomes Complex Contingencies Journal of Applied Psychology 1973, Vol. 57, No. 2, 160-166 Root-Bemstein, R.S. (1982) The Problem of Problems J. Theor. Biol. 99, 193-201 Simon, H. A. (1979) "How Big is a Chunk?" in Models of Thought New Haven: Yale University Press pp. 50-61 Smith, S.L., and Mosier, J. N. (1986) Guidelines for Designing User Interface Software Bedford, MA: Mitre Walker, J. (1988) Through the Looking Glass: Beyond "User Interfaces" Sausalito, Sept 1, 1988 280 Acknowledgments A book like this could not have been written without the help of many people. First, I want to acknowledge the encouragement and support of the people of Information Mapping, Inc., who have by their quality work, made the company and the methodology what it is today. Especially, I want to thank the old timers there, especially Nancy Fohl, Tim Burke, George Coufos, Mary Ann Cluggish and Jerry Paradis for their spirit and their excellence. Discussions with Doug Gorman, President of Information Mapping, have always been challenging and useful. And I have also learned a lot from Barb Ross, Vice President, who has pioneered in applying Information Mapping's approach to on-line text, and from Carol Vallone, Vice President, who is now leading Information Mapping in its computer-based applications. Specific acknowledgment and thanks is given to the company for permission to use copyrighted materials. I have learned something that eventually found its way into this book almost every time I got together with my friends David Sibbet, Bob Weber, Jim Channon, Bill Verplank, Michael Cone, and Paul Foraker. For reading earlier versions of the book, certain chapters, or offering suggestions on particular aspects of the book, I want to thank Carl Binder, Michael J. Steinbach, Bob Weber, John Kelly, Scott Kim, Aaron Marcus, Paul Bellerive, Doug Gorman, Don Cook, Barbara Ross, Jeff Beegle, and Jan Walker. And thanks to my typist, Gail Sheehan for putting up with my experiments and my many revisions and to Jeanne Beegle and Ming Kendall for proofreading the book and a second round of thanks to Ming Kendall for doing the index. Thanks also go to Patricia D'Andrade for preparing initial drafts of a few of the abstracts in Chapter 8 and for insightful discussions on improving the usefulness of abstracts. And also thanks to Mrs. Betty Anne Cross and Mrs. Vicky Feteris of the reference section of the Lexington Library without whose help, especially in the inter-library loan area, this book would have taken much longer to get out. I also want to acknowledge the inspiration of my long time acquaintanceship with Doug Engelbart since 1970, and to Ted Nelson, whom I didn't meet until recently, but whose ideas and visions have always sparked my imagination. Robert E. Horn Lexington, Massachusetts December 1989 281 Index Accessible detail principle, 197 Administration mode, in hypertext systems, 29 Akscyn, Robert M., 262 Analysis mode, 35 Analysis process, Information Mapping's method, how it aids, 162-163 stages of developing documents, 102-103 Analytical labels, 92 Anthologies, in hypertext, 31 Applications overviews, in hypertext, 72 Applications, hypermedia, 20-21 for Shakespeare, 22-23 hypertext, current, 29 early, 12-13 Argumentation analysis, definition, 186 history, 186-187 hypertext, 207 Information Mapping's method, framework for, 194-197 problems, useful for presenting, 198-205 Atkinson, Bill, inventor of HyperCard, 270-171 Authoring mode, in hypertext systems, 29 Block, see Information block Block label, kinds of, basic, analytical, 92 display, 92 guidelines, 93 using, 93 Bolt, Richard A., 264-265 Bottom up analysis, 116-117 Branching stories, in hypertext, 32 Briefing (presentation) mode, 35 Brown, Peter, 266-267 Brown University, commentary: definition of hypertext, 7 history of hypertext, 3 hypertext systems for university instruction, 260-261 Browsing, mode, in hypertext, 34 rapid, 10 Bush, Vannevar, commentary: memory is associative and it fades, 4 history of hypertext, 2 inventor of concept of hypertext, 7, 252-253 Buttons, definition, 9 design issue, 150-151 development issue, 62 example, 9 solution of issues, 151 user interface issue, 46-47 Camegie-Mellon University, 262-263 Category filter, multiplicity of links, 44 Chunking principle, argumentation analysis framework, 194 constraint, 86 definition, 92 examples, 88 guidelines for managing size for modularity, 102-103 levels of, 108-109 Miller's paper on concept, 218-219 principle, 83 Simon's test, 220-221 Chronological hypertrails, 132-133 Clark, R., on metacognition, 54 Classification hypertrails, 130-131 Classification, as information type, 1 1 1 Cluster link, in multiplicity of links, 43 Cognitive overload issue, 156-157 Combination labels, 92 Commentaries, in hypertext, 30 Communication, integrating with computing media, 236-237 Computer metaphor, 32-33 Computer-based training, 68 Computing media, integrating with communication, 236-237 Concept, as information type, 111 Consistency principle, argumentation analysis framework, 195 implication, 93 principle, 85 Controlled vocabulary, 69 Cross-references, in hypertext, 31 283 Dataland, 264-265 Decision hypertrails, 140-141 Definition/definitions, analytical label, 92 author-created links, 27 backing, 191 bottom up analysis, 117 branching stories, 32 buttons, 9 claims, 188 classification, 110-111 classification hypertrail, 134 combination label, 92 computer-based training, 68 concept, 110-111 controlled vocabulary, 69 decision hypertrail, 140 definition hypertrail, 142 discourse cues, 48 display label, 92 domain of experimental knowledge, 211 domains of discourse, 104 example hypertrail, 144 fact, 110-111 full text search, 68 granularity, 40 grounds (data), 189 holist learners, 52 hypermedia, 18 hypertext, 6-7 hypertext systems, 15 hypertrails, 126 ill-structured problems, 198 information sharing, 28 links, 8 metacognition, 54 modes, 29 multiplicity of links, 45 nodes, 9 on-line help messages, 69 one-way directionality, 44 paragraph, 90 prerequisite hypertrail, 128 principle, 110-111 procedure, 110-111 process, 110-111 project hypertrail, 136 qualifiers, 193 re-coding, 83 rebuttal, 192 relational databases, 33 reference-based training, 1 1 8 relatively stable subject matter, 106 rhetorics, 98 semantic nets, 32 serialist learners, 52 simulations, 33 structure, 110-111 Definition/definitions, continued structure hypertrails subject matter independent label, 92 subject matter label, 92 system-supplied links, 26 top down analysis, 1 16 two-way directionality, 44 user-created links, 27 versions, 70 warrants, 190 well-structured problems, 198 Design issue, in hypertext, 24, 38-45 links, 42-45 nodes, 40-41 Development issue, in hypertext, 24, 39, 62-73 computer-based training, 68 controlled vocabulary, 69 converting text, applications overviews, 73 full text search, 73 hierarchical structured indexes, 73 keyword indexing, 73 cost tradeoff, 72 full text search, 68 indexing, 65 keywords, 69 labor-intensive, 62-65 on-line help messages, 69 pruning and filtering, 65 skills needed, 66-67 versions, 70-71 Directionality of links, 44 Discourse cues, 148-149 Discourse domains, see Domains of discourse Display labels, 92 Domains of discourse, definition, 104 discourse analysis, 106-107 disputed discourse: argumentation analysis, 185-207 backing, 191 claims, 188 grounds (data), 189 qualifiers, 193 rebuttal, 192 warrants, 190 documentation and training, 168-183 examples, 104-105 experimental discourse, 210-231 experimental knowledge, 21 1 importance, 105 relatively stable discourse, 104-105, 168-183 examples, 170-173 product knowledge case study, 174-175 284 Index Editing mode, in hypertext systems, 29 Englebert, Douglas C., builder of hypertext system, 6, 254-255 personal computer, 256-257 Example hypertrails, 144-145 Experimental discourse, 210-231 Expert filter, in multiplicity of links, 45 Fact, as information type, 111 Fairness to learners principle, 129 Filters, category of, category, 44 expert, 45 menu of links, 45 voting, 44 development issue, pruning and filtering, 65 Footnotes, in hypertext, 30 Full text search, in hypertext, 68, 73 Geographic hypertrails, 134-135 Graphic skills, for hypertext authoring, 67 Grazing, 34 Groups, working together, 162-163 Guaranteed access to prerequisite principle, 129 Guide, 266-267 Hartley, J., 222-223 Help mode, 34 Hierarchical link, 42 Hierarchical structured index, in hypertext, 73 Hierarchy of chunking and labeling principle, argumentation analysis framework, 197 principle, 94 rationale, 64 Holist learner, 82 HyperCard, Atkinson, Bill, developer, 270-271 history of hypertext, 3 Hypermedia, application, 20-21 for Shakespeare, 22-23 integration of communication and computing media, 236-237 introduction to, 18-23 Mellon, T.H., coiner of term, 7 virtual reality, 244-247 Hyperspace, addressing lost in hyperspace issue, 156-157 navigating through, 16-17 Nelson, T., 16 user issue, 56-57 Hypertext, applications, see Applications of hypertext Atkinson, B, 270-271 authoring, skills needed, 66-67 basics of, 2-17 Brown University, 260-261 Hypertext, continued Bush, V., inventor, 252- 253 browsing mode, 34 characteristics, 10-11 computer metaphors, branching stories, 32 linked note cards, 32 linked screens or windows, 33 popup notes, 32 relational databases, 33 semantic nets, 32 simulations, 33 stretched text (outlines), 33 concepts, current, 24-35 browsing, rapid, 10 definition, 6-7 dimensions of, 24-25 applications, 29 information sharing, 28 modes, 28-29 administration, 29 authoring, 29 editing, 29 using, 29 Englebert, D.C., 254-255 features of, 10-11 highlighting deficiencies, 230-231 history of, 2-3, 250-271 HyperCard, 270-271 hypertext systems, dimensions of, 28-29 usage contexts, 14-15 hypertrails, 126-147 Information Mapping's method, addressing problems in hypertext, 150-165 connection with hypertext, 79, 211 information sharing, 28 integrating communication and computer media, 236-237 introduction to, 2-35 issues, 24, 38-73 addressing of, cognitive overload, 156-157 design, 150-151 lost in hyperspace, 156-157 cognitive overload, 156-157 design, see Design issue development, see Development issues implementation, 24, 62-65 indexing, 65 labor-intensive creation and maintenance, 160-161 lost in hyperspace, 56-57, 156-157 solutions, clustering blocks, 154-155 design, 150-151 structured hypertext, 152-153 user, see User issues 285 Hypertext, continued menu driven interfaces, 262-263 metaphors, see Metaphors navigation, options, 164-165 through information space metaphor, 16-17 through whole subject matters, 238-243 Nelson, T.H., 3, 7, 258-259 non-linear discourse, 10 on-line documentation and training, relating with, 68-69 Owl's Guide, 266-267 paper metaphors, 30-3 1 anthologies, 31 commentaries, 30 cross-references, 3 1 footnotes, 30 indexes, 30 library card catalogs, 30 quotes, 3 1 sticky notes, 31 personalized, feature of, 1 1 pointing way toward advances, 230-231 private (or in-house), usage context, 14 problems, identifying, 228, 229 multiple representational ecologies, 60 resolving of, 150-165 product knowledge case studies, 174-183 public access, usage context, 15 read only, feature of, 1 1 reading, multiple ways of, 3 skills needed, 66-67 software, essence of, 8-9 sources of, 4 structured hypertext, 152-153 university instructions, using for, 260-261 using, ways of, 34-35 Van Dam, A., 260-261 virtual reality, 244-247 Zog, 262-263 Hypertrails, chronological, 132-133 classification, 130-131 decision, 140-141 definition, 126 example, 144-145 geographic, 134-135 linearized sequences, 146-147 navigating through, 126-147, 238-243 prerequisite, 128-129 project, 136-137 structure, 138-139 Implementation issues, in hypertext, 24, 62- 65 Indexes, in hypertext, 30 Indexing issues, in hypertext, 65 Information blocks, analysis, applied to scientific papers, 214-217 clustering, kinds of, 95 to form maps, 155 construction, 86-87 definition, 84-85 examples, 88-89 grouping into maps, 95 guidelines, developing of, 92-93 history of, 76-77 identifying domains of discourse, 105-107, 114-115 key blocks, 112-113 kinds of block content, 88-89 labels, see Block label managing completeness, 100-101 mapping metaphor, 121 modularity, 1 02- 1 03 navigational options, 164-165 principles of constructing, 85 replacing paragraphs, 90-91 sequencing, 95 standards, developing of, 92-93 structured hypertext, 152-153 types of, 84 using, most frequently, 109 Information map, blocks, clustering of, 154-155 definition, 94 guidelines, 94-95 history, 76-77 mapping metaphor, 121 navigational options, 164-165 organization and structure of documents, 98-99 structured hypertext, 152-153 Information Mapping's method, applying to, on-line text, 168-183 scientific papers, 210-21 1 approach, 80-81 argumentation analysis as framework, 194-197 definition of, 80 goal, 77 groups, working together, 162-163 history, 76-77 hypertext, connection with, 79, 21 1 resolving problems, 150-161 cognitive overload, 156-157 hypertext design problems, 150-151 labor-intensive creation and maintenance, 160-161 lost in hyperspace, 156-157 reading cues problems, 158-159 structured, 152-153 hypertrails, navigating through, 126 information blocks, see Information blocks 286 Index Information Mapping's method, continued introduction to structured writing, 76-123 intuitive chunking vs. precision modularity, 85 labeling standards, commentary, 60 mapping metaphor, 120-121 navigational options, 164-165 problems, solving, 78-79 reference-based training, 118-119 replacing paragraph with precision modularity, 91 results, 114-123 Shaffer's study, performance advantages, 224-255 trends, integration of communication with computing media, 236-237 Information sharing, in hypertext systems, 28 Information space, mapping future, 234-249 metaphor, 16-17 trends, integrating communication with computing media, 236-237 Information types, key blocks for, 112-113 kinds of, classification, 110-111 concept, 110-111 definition, 1 10 fact, 110-111 principle, 110-111 procedure, 110-111 process, 110-111 structure, 110-111 use, 110-111 InfoSpace, see Information space Integrated graphics principle, 196 Integration of communication and computing data, 236-237 Issues, in hypertext, addressing of, cognitive overload, 156-157 design, 150-151 lost in hyperspace, 156-157 cognitive overload, 156-157 design, see Design issue development, see Development issues implementation, 24, 62-65 indexing, 65 labor-intensive creation and maintenance, 160-161 lost in hyperspace, 56-57, 156-157 solutions, clustering blocks, 154-155 design, 150-151 structured hypertext, 152-153 user, see User issues Key blocks, 112-113 Keyword indexes, 73 Keyword links, 42 Keywords, 69 Knowledge based management skills, for hypertext authoring, 66 Knowledge navigator, 268-269 Label standards, 60 Labeling principle, argumentation analysis framework, 194 block labels, see Block label constraint, 86 guidelines and standards for developing, 92 principle, 85 Labor intensive creation and maintenance issue, in hypertext, 160-161 Learning mode, 35 Library card catalogs, in hypertext, 30 Linearized sequences, for hypertrails, 146-147 Link directionality, 44 Linked note cards, in hypertext, 32 Linked screens or windows, in hypertext, 32 Links, characteristics of, directionality, 44 multiplicity, 45 computer metaphor, 32-33 definition, 8 design issue, 38, 42-45, 150-151 development issue, 62,63, 65 examples of, 8 filters, see Filters issues with, directionality, 44 multiplicity, 45 kinds of, cluster, 43 hierarchical, 42 keyword, 42 referential, 43 link-making options, 1 1 pruning and filtering, 65 types of, author-created, 27 system-supplied, 26 user-created, 27 paper metaphors, 30-31 personal, 253 solution of issues, 150 trails, retrieving of, 253 user issue, 59 Long term memory, 82-83 Lost in hyperspace, solution to problem, 156-157 user issue, 56-57 MIT Architecture Machine Group, 264-265 Maps, see Information maps McCracken, Donald L., 262 Memory, association, 4 Memory, long term and short term, 82-83 287 Menu of links, 45 Metacognition, 54-55 Metaphor/metaphors, computer, 32-33 information space, 16-17 mapping, 120-121 navigation, 17 paper, 30-3 1 project hypertrails, 136 Miller, George A., concept of short term memory limits and chunking, 218-219 on chunking, 82-83 spatial information management principle, 264 Modes, definition, 29 kinds of, authoring, 29 briefing, 35 browsing, 34 editing, 29 help, 34 learning and analysis, 35 referencing, 35 training, 34 using, 29 Multiple representational ecologies, 60 Multiplicity of links, 45 Navigation, options, 164-165 through hypertrails, 126-147, 238-243 through information space metaphor, 16-17 through subject matters, 238-243 Negroponte, Nicholas, history of hypertext, 3 managing information spatially in Dataland, 264-265 Nelson, Theodor (Ted) Holm, coiner of terms hypertext and hypermedia, 7, 258-259 commentary, another definition of hypertext, 7 multiple ways to read hypertext, 3 history of hypertext, 2 hypertext system, 16, 29 navigation through hyperspace, 16 Newell, Allen, 262 Nodes, definition, 9 design issue, 38, 40-41, 150-151 development issue, 62-63 granularity, 40 solution of issues, 150 structured hypertext, 152-153 user issue, 59 Non-Linear discourse, 10 On-line help messages, 69 One-way directionality, of links, 44 Overchoice, hypertext user issue, 58-59 Overload, cognitive, hypertext user issue, 58-59 Owl's Guide, first commercial hypertext system, 266-277 history of hypertext, 3 Paper metaphor, for hypertext linkages, 30-31 Personalized hypertext, 1 1 Popup notes, in hypertext, 32 Precision modularity, 85-87, 91 Prerequisite hypertrails, 128-129 Principle, as information type, 1 1 1 Principle/principles, accessible detail, 197 chunking, see Chunking principle consistency, see Consistency principle fairness to learners, 129 guaranteed access to prerequisites, 129 hierarchy of chunking and labeling, see Hierarchy of chunking and labeling principle information blocks, 84-85 integrated graphics, 196 labeling, see Labeling principle relevance, see Relevance principle Private (or in-house) hypertext systems, 14 Procedure, as information type, 1 1 1 Process, as information type, 1 1 1 Project hypertrails, 136-137 Public access hypertext systems, 15 Quotes, in hypertext, 31 Read-only hypertext, 1 1 Reader behavior, 50 Reading cues, 48, 158-159 Reference-based training, 118-119 Referencing mode, 35 Referential link, 43 Relational databases, in hypertext, 33 Relevance principle, argumentation analysis framework, 194 constraint, 86 implication, 93 principle, 85 Reid, F„ 226-227 Rhetorical skills, for hypertext authoring, 66 Robertson, George G., 262 Science information system, 212-213 Sculley, John, 268-269 Semantic nets, in hypertext, 32 Serialist learner, 52 Shaffer, Eric M., 224-225 Short term memory, 82-83 Miller's concept, 218-219 Simon's test, 220-221 288 Index Simon, Herbert A., 220-221 Simulation, in hypertext, 33 Skills, needed for hypertext authoring, graphic, 67 interface design, 66 knowledge base management, 66 rhetoric, 66 Sticky notes, in hypertext, 31 Stretch text (outline), 33 Structure, as information type, 1 1 1 Structure hypertrails, 138-139 Subject matter independent label, 92 Subject matter label, 92 Titling documents, seven commandments for, 61 Top down analysis, 116-117 Training mode, 35 Trends, integration of communication and computing media, 236-237 Trueman, M., 222-223 Two-way directionality, of links, 44 University instructions, using hypertext systems for, 260-261 User issues, in hypertext, 39, 46-61 buttons, 46-47 cognitive overload, 58-59 lost in hyperspace, 56-57, 156-157 overchoice, 58-59 metacognition skills, 54-55 reader behavior, 50-5 1 reading cues, 48-49 serialist readers, 52-53 title for documents and their parts, 38-39, 60-61 Using modes, in hypertext systems, 29 Van Dam, Andries, 260-261 Versions, in hypertext, 70-71 Virtual reality, 243-244 Voting filter, multiplicity of links, 44 Walker, Jan, 34 Wright, P„ 226-227 Writing, non-sequential, 7 Zog Group, 262-263 289 Mapping Hypertext The Analysis, Organization, and Display of Knowledge for the Next Generation of On-Line Text and Graphics a ne w book by Robert E. Horn Contents Part 1. Hypertext — Hypermedia, New Opportunities 1 . Introduction to Hypertext and Hypermedia 2. Current Issues with Hypertext Part 2. The Method of Information Mapping 3. Introduction to Information Mapping's Method of Structured Writing 4. Navigating Structured Hypertrails 5. Resolving Some Hypertext Problems Part 3. Some Applications of Structured Hypertext 6. Relatively Stable Discourse: Documentation and Training 7. Disputed Discourse: Argumentation Analysis 8. Experimental Discourse: Scientific Information Part 4. So What? What Next? 9. Mapping Future InfoSpace: Summary and Trends Appendix A. Some Historical Notes ® Information Mapping and Info-Map are registered trademarks of Information Mapping, Inc. JWA\SNWW%V.\V.V.'.V.V.V.VV.%V.V.V.V.V.V.V.V.V.V.V.V.V.VV.V.W.V.%%%V.V-V-^VA\%VAVAVAV.V.V.VAy.VV\WVLW.S\%%V^ I Early Comments on Mapping Hypertext * % \ I am convinced that the future of man’s knowledge production and utilization will be deeply emplanted in the structure, conventions I and methods associated with the descendants of today's hypertext. Bob Horn has produced a notable step toward that end. —Doug Engelbart, Bootstrap Project, Stanford University; first person to implement hypertext on a computer system Mapping Hypertext is a thoughtful and provocative overview of both hypertext and Information Mapping, full of useful advice and interesting bits of history. It is a must read for anyone concerned about how computers can become effective tools for human i communication. -Paul Sajfo, The Institute for the Future; columnist, Personal Computing I 1 This book will change the wav people think about their current information and the hypertext revolution. I —Ken Blanchard, co-author of the best selling The One Minute Manager •i | f Bob Horn suggests an antidote for the problem of disorientation that often comes with navigating through hypertext... 1 —Patricia Seybold, founder of Patricia Seybold's Office Computing Group and sponsor of the Seybold Office Computing | Conferences. From Paradigm Shift: Patricia Seybold's Guide to the Information Revolution £ | Mapping Hypertext by Robert E. Horn is a tour de force in several respects. First, it is an amazing example of "graphic language.. .Mapping Hypertext is a unique and seminal work, covering the history and conceptual underpinnings of hypertext, suggesting applications and design principles capable of stimulating hypertext and hypermedia design for years to come..." -Carl Binder, Performance and Instruction, October 1991 Boy, do I wish we'd had this book when we were designing the CD-ROM Electronic Whole Earth Catalog.... This book is the most thorough survey of solutions thus far. And it is organized in a highly visual hypertext-like format which effectively illustrates many of the principles being discussed. An absolutely first-rate work. -Keith Jordan, Whole Earth Review, Summer 1991 Send to: The Lexington Institute 300 Third Avenue Waltham, MA 02154 Now available via MASTERCARD or VISA! Call (617) 890-7003 to order □ Please send me _ copy(ies) of Mapping Hypertext at $34.50 each, plus $3.00 shipping and handling. I've enclosed a check. (Add $1.73 state tax per copy) □ Please use my Mastercard or VISA for payment. (Circle which one) Card #_ _ _ Name _ Title _ Exp. Date. OR CALL (617) 890-7003 Department . Company _ Address _ City, State_ Zip Business Phone J mmi. About the Book The technology of hypertext offers the very real potential of helping both business and society deal productively with the information explosion. Mapping Hypertext illuminates the promise and the reality of hypertext and information management, bringing hypertext together with a complementary methodology critical to its success: Information Mapping's method for analyzing, organizing, and presenting information. The book also breaks new ground in its highly graphic presentation, an intriguing visual simulation of hypertext. Mapping Hypertext will change forever the way people approach information organization and the hypertext revolution. About the Author Robert E. Horn is the inventor of Information Mapping's methodology and has spent his professional life applying the principles of cognitive science and learning theory to the solution of communications problems. He has taught on the graduate level at Harvard and Columbia universities. The company Horn founded, Information Mapping, Inc., the recognized leader in high performance communications, has helped many of the world's largest companies deal successfully with the management of large amounts of complex information. mm Analysis, Linkage, and Display of Knowledge for the Next Generation of On-Line Text and Graphics Robert E, Horn /