HCI Bibliography Home | HCI Conferences | HYPER Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
HYPER Tables of Contents: 87899191Z9393X93Y93Z969797X98990001

Proceedings of ACM Hypertext'93 1993-11-14

Fullname:ACM Hypertext'93 -- Posters
Editors:P. David Stotts; Richard Furuta; William Jones; Roland Faragher-Horwell; Gary Perlman
Location:Seattle, Washington
Dates:1993-Nov-14 to 1993-Nov-18
Publisher:ACM
Standard No:ACM ISBN 0-89791-624-7; ACM Order Number 614930; ACM DL: Table of Contents hcibib: HYPER93; hcibib: HYPER93X; hcibib: HYPER93Y ; hcibib: HYPER93Z
Papers:36; 38; 32; 26
Pages:308; 3-18; 22-34

Proceedings of ACM Hypertext'93 -- Unpublished Program 1993-11-14

Fullname:ACM Hypertext'93 -- Posters
Editors:P. David Stotts; Richard Furuta; William Jones; Roland Faragher-Horwell; Gary Perlman
Location:Seattle, Washington
Dates:1993-Nov-14 to 1993-Nov-18
Publisher:ACM
Standard No:ACM ISBN 0-89791-624-7; ACM Order Number 614930; ACM DL: Table of Contents hcibib: HYPER93; hcibib: HYPER93X; hcibib: HYPER93Y ; hcibib: HYPER93Z
Papers:36; 38; 32; 26
Pages:308; 3-18; 22-34

Proceedings of ACM Hypertext'93 -- Demonstrations 1993-11-14

Fullname:ACM Hypertext'93 -- Posters
Editors:P. David Stotts; Richard Furuta; William Jones; Roland Faragher-Horwell; Gary Perlman
Location:Seattle, Washington
Dates:1993-Nov-14 to 1993-Nov-18
Publisher:ACM
Standard No:ACM ISBN 0-89791-624-7; ACM Order Number 614930; ACM DL: Table of Contents hcibib: HYPER93; hcibib: HYPER93X; hcibib: HYPER93Y ; hcibib: HYPER93Z
Papers:36; 38; 32; 26
Pages:308; 3-18; 22-34

Proceedings of ACM Hypertext'93 -- Posters 1993-11-14

Fullname:ACM Hypertext'93 -- Posters
Editors:P. David Stotts; Richard Furuta; William Jones; Roland Faragher-Horwell; Gary Perlman
Location:Seattle, Washington
Dates:1993-Nov-14 to 1993-Nov-18
Publisher:ACM
Standard No:ACM ISBN 0-89791-624-7; ACM Order Number 614930; ACM DL: Table of Contents hcibib: HYPER93; hcibib: HYPER93X; hcibib: HYPER93Y ; hcibib: HYPER93Z
Papers:36; 38; 32; 26
Pages:308; 3-18; 22-34
  1. Papers
  2. Videos
  3. Technical Briefings
  4. Panels
  5. Workshops
  6. Courses
  7. Commercial Symposium
  8. Reception
  9. Keynote

Papers

The Architecture and Implementation of a Distributed Hypermedia Storage System BIBAKPDF 1-13
  Douglas E. Shackelford; John B. Smith; F. Donelson Smith
Our project is studying the process by which groups of individuals work together to build large, complex structures of ideas and is developing a distributed hypermedia collaboration environment (called ABC) to support that process. This paper focuses on the architecture and implementation of the Distributed Graph Storage (DGS) component of ABC. The DGS supports a graph-based data model, conservatively extended to meet hypermedia requirements. Some important issues addressed in the system include scale, performance, concurrency semantics, access protection, location independence, and replication (for fault tolerance).
Keywords: Distributed data, Computer-supported cooperative work (CSCW), Distributed file systems, Performance, Scalability, Hypertext
Concurrency Control in Collaborative Hypertext Systems BIBAKPDF 14-24
  Uffe Kock Wiil; John J. Leggett
Traditional concurrency control techniques for database systems (transaction management based on locking protocols) have been successful in many multiuser settings, but these techniques are inadequate in open, extensible and distributed hypertext systems supporting multiple collaborating users. The term "multiple collaborating users" covers a group setting in which two or more users are engaged in a shared task. Group members can work simultaneously in the same computing environment, use the same set of tools and share a network of hypertext objects. Hyperbase (hypertext database) systems must provide special support for collaborative work, requiring adjustments and extensions to normal concurrency control techniques. Based on the experiences of two collaborative hypertext authoring systems, this paper identifies and discusses six concurrency control requirements that distinguish collaborative hypertext systems from multiuser hypertext systems. Approaches to the major issues (locking, notification control and transaction management) are examined from a supporting technologies point of view. Finally, we discuss how existing hyperbase systems fare with respect to the identified set of requirements. Many of the issues discussed in the paper are not limited to hypertext systems and apply to other collaborative systems as well.
Keywords: Collaborative work, Distributed hypertext systems, Concurrency control, Hyperbases, Open architectures, Extensibility, Supporting technologies, User-controlled locking, Events, Transaction management, Version control
Designing Dexter-Based Cooperative Hypermedia Systems BIBAKPDF 25-38
  Kaj Grønbæk; Jens A. Hem; Ole L. Madsen; Lennert Sloth
This paper discusses issues for the design of a Dexter-based cooperative hypermedia architecture and a specific system, DeVise Hypermedia (DHM), developed from this architecture. The Dexter Hypertext Reference Model [Hala90] was used as basis for designing the architecture. The Dexter model provides a general and solid foundation for designing a general hypermedia architecture. It introduces central concepts and proposes a layering of the architecture. However, to handle cooperative work aspects, such as sharing material and cooperative authoring, we have to go beyond the Dexter model concepts. To deal with such aspects we have extended our implementation of the Dexter concepts with support for long-term transactions, locking and event notification as called for by Halasz [Hala88]. The result is a platform independent architecture for developing cooperative hypermedia systems. The architecture consists of a portable kernel that constitutes an object oriented framework for developing Dexter compliant hypermedia systems. It is a client/server architecture including an object oriented database (OODB) to store the objects implementing the Dexter Storage Layer. We use a general OODB being co-developed to support long term transactions, flexible locking, and event notification. The transaction and locking mechanism support several modes of cooperation on shared hypermedia materials, and the notification mechanism supports the users in maintaining awareness of each others' activity. The portable kernel was used to implement the DHM system on two quite different platforms: UNIX/X-windows and Apple Macintosh.
Keywords: Dexter model, Open hypermedia, CSCW, Shared materials, Object oriented database
MORE: Multimedia Object Retrieval Environment BIBAKPDF 39-50
  Dario Lucarella; Stefano Parisotto; Antonella Zanzi
In this paper, we present a graph-based object model that will be used as a uniform framework for direct manipulation of multimedia information.
   Essentially, we propose a graph representation of the conceptual schema, the object instances, and the user queries. The resulting uniform approach is well suited to a visual interface in which the user, supported by appropriate tools, manipulates directly on the screen object graphs for different purposes as schema definition, querying, browsing, and viewing. Next, we give formal definitions of such operations along with examples concerning a real multimedia application we have developed in order to experiment with the proposed approach.
   Finally, design and implementation issues are discussed for MORE, a prototype system that, combining effectively browsing and querying techniques, provides a visual environment for multimedia information retrieval.
Keywords: Graph-oriented models, Hypertext querying, Information retrieval, Visual interface, Direct object manipulation, Multimedia databases
Should Anchors be Typed Too? An Experiment with MacWeb BIBAKPDF 51-62
  Jocelyne Nanard; Marc Nanard
Macweb is a hypertext system that uses types to incorporate knowledge in hypertext. In this paper, we examine an application developed with the system for accessing a technical document base in the context of a task. This application gives us the opportunity to discuss the extension of typing to anchors. We show that attaching knowledge to anchors through types must take into account the context of use of the anchored text. Thus, we introduce the notion of semantic anchoring of concepts within documents. We show how Macweb makes it possible to implement this approach without adding any new features and how it provides an answer to the famous sentence "Don't link me in". Beyond the experiment itself, the foundations of the approach and its connection with hypertext systems modeling is presented.
Keywords: Knowledge-based hypertext, Anchoring, Dynamic links, Virtual documents
Another Dimension to Hypermedia Access BIBAKPDF 63-72
  Satoshi Ichimura; Yutaka Matsushita
The OpenBook system using a book metaphor allows a user to leaf through a set of nodes retrieved from a hypermedia just like a book. While leafing through the book, the reader does not need to pay attention to the detailed description written in pages, but impressive information such as chapter titles and figures catches reader's eyes. In other words, leafing through an electronic book takes advantage of a cognitive capability to skim the outlines of the contents. Moreover, the system supports a query-based access mechanism which supports a structure search mechanism for the purpose of finding potentially useful nodes. Furthermore, this paper describes a method of linearizing complex hyper-networked nodes to facilitate high speed browsing, which is a unique aspect of OpenBook.
Keywords: Hypermedia, Information retrieval, Book metaphor, ISM-CSA
Hypercubes Grow on Trees (and Other Observations from the Land of Hypersets) BIBAKPDF 73-81
  H. Van Dyke Parunak
Much of the power of hypermedia comes from the development of techniques for information management that closely match natural cognitive processes. HyperSet, a hypermedia environment tailored for taxonomic reasoning [Parunak 91], is an example of this philosophy. People perform taxonomic reasoning when they classify, store, and retrieve a number of similar information objects (such as biological specimens, or linguistic constructions, or research projects). The process is essentially set-based. The user sorts objects into sets based on their characteristics; looks together at members of a single set to search for correlations or discernible subsets among them; examines the different sets of which one item is a member to see whether there are relations among them; and generates new sets from old ones.
   Two years of experience in using HyperSet has led to a deeper understanding of the patterns and processes of taxonomic reasoning and the kind of computer methods that can effectively support it. This paper reports on three of these insights:
  • 1. The set of sets that develops as classification takes place is not flat, but
        hierarchical. Analysis of this hierarchy yields a representation that
        combines the flexibility of a directed acyclic graph with the navigational
        properties of strict trees.
  • 2. It is useful for a taxonomic information system to support a simple dualism
        between sets and their elements, permitting one to do set operations on
        artifacts as well as on sets.
  • 3. Similarity measures among different sets are most usefully computed for a
        hypercube of such sets, a hypercube that emerges naturally from the
        hierarchical structure of sets.
    Keywords: Information structures, Taxonomic reasoning, Classification, Authoring, Retrieval
  • Dynamic Hypertext and Knowledge Agent Systems for Multimedia Information Networks BIBAKPDF 82-93
      Yoshitaka Shibata; Michiaki Katsumoto
    In this paper, a new intelligent human interface which can provide simple and flexible user access capabilities, based on the concept of dynamic hypertext System is introduced for multimedia information networks. In this dynamic hypertext system, Metanodes and Metalinks are defined as abstract nodes and flexible links, and organize a dynamic information space where user can easily retrieve the desired information objects by asking to the knowledge agent. The knowledge agent based on the knowledge-base can decide the link from the current reference point to the suitable Metanodes among the multimedia databases distributed over the network. The knowledge agent also performs media format conversion of the original information units to adjust to the users workstation capabilities and temporal synchronization among different media. In order to evaluate the functions of the suggested human interface, two applications are introduced and developed on the prototyped multimedia information network.
    Keywords: Multimedia, Hypermedia, Human interface, Agent, Knowledge-base
    Applying AI Models to the Design of Exploratory Hypermedia Systems BIBAKPDF 94-105
      Ray Bareiss; Richard Osgood
    Hypermedia systems offer great promise for capturing expertise and subsequently providing multifaceted access in support of a user engaged in a complex task. A primary issue in building such systems is how to structure the knowledge contained in them such that a user with a problem can find the most appropriate knowledge easily and naturally. Our artificial intelligence-based research has produced two approaches to structuring knowledge that show promise for organizing hypermedia knowledge bases. The first is the use of abstract models of a user's problem solving task to provide a global structure for the knowledge base and to reflect that structure in a meaningful human interface to the system. The second is the use of a conversational model to provide local coherence of the links among pieces of information in the knowledge base. These two types of models have been applied in the construction of Trans-ASK, a large hypermedia system in the domain of military transportation planning. This paper discusses the theory underlying the models, their application within Trans-ASK, and finally provides a preliminary evaluation of the resulting system.
    Keywords: Hypermedia, Task models, Indexing, Exploratory systems
    The Knowledge Weasel Hypermedia Annotation System BIBAKPDF 106-117
      Daryl T. Lawton; Ian E. Smith
    We describe the organization and implementation of the Knowledge Weasel (KW) Hypermedia Annotation System which we are using to explore knowledge structuring by collaborative annotation. Knowledge Weasel incorporates many useful features: a common record format for representing annotations in different media for uniform access; dynamic user control of the presentation of annotations as a navigational aid; global navigation using queries and local navigation using link following; support for collecting related sets of annotations into groups for contextual reference and communication. KW purposely leverages off of free, publicly available software so it doesn't require building specialized tools and also so it can be freely available. We discuss some of the issues involved with annotating non-textual material such as images and sound and conclude with a brief discussion of ongoing and future work.
    Keywords: Hypermedia, Collaborative annotation, Query-based navigation, Link-based navigation
    Hypertext by Link-Resolving Components BIBAKPDF 118-130
      Frank Wm. Tompa; G. Elizabeth Blake; Darrell R. Raymond
    In conventional hypertexts, links are explicit connections between specific regions of a text. We describe an architecture that treats links as the outcome of responses to user pokes. In this architecture, a hypertext is a collection of link-resolving components, each interpreting a user's request according to its own resolution algorithm. Each link-resolving component is a set of cooperative processes, communicating with a central network manager. When a user points at some location within a window, each link-resolving component is given a key derived according to a previously-stored algorithm; the link-resolving components concurrently update their displays according to their algorithms for resolving the keys. Multiple applications can easily share a common source and be invoked simultaneously, providing a concurrent browsing mechanism. Two example hypertexts employing this architecture are described.
    Keywords: Hypertext system architecture, Dynamic linking, Link resolution, Database keys
    Selective Text Utilization and Text Traversal BIBAKPDF 131-144
      Gerard Salton; James Allan
    Many large collections of full-text documents are currently stored in machine-readable form and processed automatically in various ways. These collections may include different types of documents, such as messages, research articles, and books, and the subject matter may vary widely. To process such collections, robust text analysis methods must be used, capable of handling materials in arbitrary subject areas, and flexible access must be provided to texts and text excerpts of varying size.
       In this study, global text comparison methods are used to identify similarities between text elements, followed by local context-checking operations that resolve ambiguities and distinguish superficially similar texts from texts that actually cover identical topics. A linked text structure is then created that relates similar texts at various levels of detail. In particular, text links are available for full texts, as well as text sections, paragraphs, and sentence groups. The linked structures are usable to identify important text passages, to traverse texts selectively both within particular documents and between documents, and to provide flexible text access to large text collections in response to various kinds of user needs. An automated 29-volume encyclopedia is used as an example to illustrate the text accessing and traversal operations.
    Keywords: Full-text access, Information retrieval, Passage retrieval, Text analysis, Global text comparisons, Local context checking, Automatic text linking, Selective text reading, Text summarization
    HieNet: A User-Centered Approach for Automatic Link Generation BIBAKPDF 145-158
      Daniel T. Chang
    Most hypertext systems facilitate one-at-a-time link creation, but only a few support automatic link generation. In systems that do support automatic link generation, user interests are either ignored or explicit user actions are required to enter a set of keywords and queries.
       In this paper we present a new linking mechanism called HieNet, implemented on top of a commercial hypertext system, that generates new links based on previously-created user links. Our approach allows users direct control over similarity thresholds, node granularity, and the extent of linking in composite nodes. Our work derives from Salton's work on automatic generation of links using term vectors, and extends the ideas incorporated in Bernstein's Link Apprentice and Lotus SmarText. Preliminary tests show that HieNet generates plausible links with acceptable performance and that users can learn to control the link generation parameters.
    Keywords: Links, Link generation, Vector space model, Link apprentice, SGML
    Media-Based Navigation for Hypermedia Systems BIBAKPDF 159-173
      Kyoji Hirata; Yoshinori Hara; Naoki Shibata; Fusako Hirabayashi
    In this paper, we present the concept and the general framework of a new navigation style for hypermedia systems, the media-based navigation. The user browses through a hypermedia system using the specific clues such as shape, color, construction for still image, motion for movie, and tone or melody for auditory data. In this navigation, the user and the system interact with each other without translating the textual representation. We describe the visual-based navigation and show its algorithms. The algorithms are implemented on an experimental hypermedia database system called "Miyabi." We show some experimental results and our current evaluation. We also describe the implementation of the algorithms for large scale hypermedia systems and show that these algorithms can effectively apply to the system which have more than 10000 images. We also describe the auditory media-based navigation. The media-based navigation is a useful interface for hypermedia systems to improve human-machine interactive interfaces.
    Keywords: Hypermedia database, Media-based navigation, Pattern matching, Information retrieval, Indexing
    Hypertext and the Author/Reader Dialogue BIBAKPDF 174-182
      Susan Michalak; Mary Coney
    Hypertext theorists tend to approach the hypertext concept from radically different philosophical positions. Some theorists stress hypertext's utility as an information storage and retrieval device; others praise hypertext's ability to free the reader from linear media; still others applaud hypertext's connectivity and its ability to create a basis for the communal creation of knowledge. The hypertext documents that these theorists envision (and create) are quite different from one another because they are based on each theorist's particular perspective on hypertext. Recently, many hypertext theorists have acknowledged a need for hypertext authors to develop a better rhetorical understanding of their readers; however, the reader roles that most hypertext theorists have thus far anticipated do not encompass all of the reader roles that hypertext can accommodate. Coney (Cone92a) has offered a comprehensive "taxonomy of readers" that -- although it was originally conceived as a taxonomy of readers of conventional print -- provides deeper insight into hypertext reader roles. In this paper, we will discuss the philosophical traditions invoked by various hypertext theorists and the reader roles that are accommodated or required by those traditions. Finally, we will discuss the hypertext author's virtual presence, the implied author, as a corollary of reader role.
    Keywords: Hypertext, Literary theory, Implied author, Rhetoric, Reader roles, Reader-response criticism
    Links in Hypermedia: The Requirement for Context BIBAKPDF 183-191
      Lynda Hardman; Dick C. A. Bulterman; Guido van Rossum
    Taking the concept of a link from hypertext and adding to it the rich collection of information formats found in multimedia systems provides an extension to hypertext that is often called hypermedia. Unfortunately, the implicit assumptions under which hypertext links work do not extend well to time-based presentations that consist of a number of simultaneously active media items. It is not obvious where links should lead and there are no standard rules that indicate what should happen to other parts of the presentation that are active.
       This paper addresses the problems associated with links in hypermedia. In order to provide a solution, we introduce the notion of context for the source and the destination of a link. A context makes explicit which part of a presentation is affected when a link is followed from an anchor in the presentation. Given explicit source and destination contexts for a link, an author is able to state the desired presentation characteristics for following a link, including whether the presentation currently playing should continue playing or be replaced.
       We first give an intuitive description of contexts for links, then present a structure-based approach. We go on to describe the implementation of contexts in our hypermedia authoring system CMIFed.
    Keywords: Hypermedia links, Context for links, Structured multimedia, Hypermedia presentation
    Exploring Large Hyperdocuments: Fisheye Views of Nested Networks BIBAKPDF 192-205
      Emanuel G. Noik
    Browsing large hypertexts by following links can be difficult and confusing, especially if links span distant nodes. Often, a user would like to explore several regions of a network simultaneously, when studying the end points of one or more links, for example. Although this can be achieved by displaying each area of interest in a separate zoomed-in window, the union of such views is not always meaningful. In particular, valuable context showing the relationships between the views is lost. By balancing local detail and global context, fisheye views display information at several levels of abstraction simultaneously. We have devised a novel technique for generating fisheye views of hierarchically nested graphs with multiple variable magnification focal points. In this paper we demonstrate its feasibility as a tool for exploring large nested hypertext networks.
    Keywords: Nested hypertext networks, Graphs, Navigation, Overviews, Fisheye views
    Browsing Through Querying: Designing for Electronic Books BIBAKPDF 206-216
      Nipon Charoenkitkarn; Jim Tam; Mark H. Chignell; Gene Golovchinsky
    The emerging technologies of pen-based navigation and hand-held computing pose a number of challenges for hypertext and the development of electronic books. In this paper we explore methods of query-based browsing that meet some of these challenges. We describe an existing prototype (Queries-R-Links) that we have developed and we then discuss an enhanced version of query-based browsing that uses methods of text analysis and related approaches to provide a more focused set of hits (links) during browsing.
    Keywords: Information retrieval, Pen-based navigation, Electronic books, Hand-held computing, Relevance feedback, Text analysis
    Searching for the Missing Link: Discovering Implicit Structure in Spatial Hypertext BIBAKPDF 217-230
      Catherine C. Marshall; Frank M., III Shipman
    Hypertexts may be implicitly structured, based on either node content or context. In this paper, we examine implicit structures that rely on the interpretation of node's spatial context. Hypertext authors and readers can perceive and understand these idiosyncratic structures, but, because they are implicit, they cannot be used by the system to support users' activities. We have explored spatially structured hypertext authored in three different systems, and have developed heuristic recognition algorithms based on the results of our analyses of the kinds of structures that people build. Our results indicate that (1) recognition of implicit structures in spatial hypertext is feasible, (2) interaction will be important in guiding such recognition, and (3) the hypertext system can provide layout facilities that will render later systematic interpretation much easier. Found structures can be used as a basis for supporting information management, as a straightforward way of promoting knowledge-base evolution, as a way of solving representational problems endemic to many hypertext systems, or as a basis for collaboration or interaction.
    Keywords: Implicit structure, Spatial hypertext, Link automation

    Videos

    The Microcosm Link Service: An Integrating Technology BIBPDF 231-232
      Wendy Hall; Hugh Davis; Adrian Pickering; Gerard Hutchings
    Miyabi: A Hypertext Database with Media-Based Navigation BIBPDF 233-234
      Kyoji Hirata; Hajime Takano; Yoshinori Hara
    The SEPIA Hypermedia System as Part of the POLIKOM Telecooperation Scenario BIBPDF 235-237
      Jorg Haake; Thomas Knopik; Norbert Streitz
    WAX or the Discovery of Television Among the Bees BIBPDF 238-239
      David Blair
    FRESS Hypertext System BIBPDF 240
      David G. Durand; Steven J. DeRose
    The Perseus Project BIBPDF 241
      Elli Mylonas

    Technical Briefings

    Enactment in Information Farming BIBAKPDF 242-249
      Mark Bernstein
    Information farming views the cultivation of information as a continuing, collaborative activity performed by groups of people working together to achieve changing individual and common goals. Failure to differentiate information farming from related but distinct activities like information mining and data factories has been a fruitful source of misunderstanding and discord in the hypertext literature and in the design of hypertext environments. Dramatic enactment and visual salience -- not recall, precision, or usability -- assume primary roles in design for information gardening. In this technical briefing, we examine how enactment contribute to the success and failure of a variety of Hypergate and Storyspace features.
    Keywords: Design, Rhetoric, Enactment, Collaboration, Information farming
    Technologically Assisted Focussing in Psychotherapy with Couples: A Hypertext Application for Clients, Clinicians & Researchers BIBAPDF 250-255
      Den' Raphaely
    Video, controlled by computer, can be used in psychotherapy to give a couple feedback on how they interact. Organized hypertextually, these clips can be cross-referenced with other video-records (e.g. other sessions with this couple or another couple; movies; documentaries) and with other kinds of texts (e.g. therapist notes, scholarly references, genograms, cartoons). But how are these to be organized practically? And on what theoretical basis? This demonstration will show how a hypertext shell (the nWAYZ Project) built on HyperCard and VideoToolkit can be used to construct a hypertext for use by clients, therapists and researchers.
    The Microcosm Link Service BIBKPDF 256-259
      Wendy Hall; Gary Hill; Hugh Davis
    Keywords: Open, Integrated, Link service
    Information Retrieval Techniques for Hypertext in the Semi-Structured Toolkit BIBAKPDF 260-267
      Gary Perlman
    The Semi-Structured Toolkit (SST) is a C library that provides universal functions based on abstractions for storage format- and data type-dependencies of semi-structured/frame-based information units. The SST provides searching, sorting, viewing, and linking operations for data stored in its native formats, without requiring proprietary formats or conversion. Hypertext capabilities such as linking and outlining are implemented in the SST with inverted indices for each of the fields in semi-structured records. This paper describes the implementation of hypertext capabilities in the SST.
    Keywords: Information retrieval, Underlying technologies, Hypertext, Indexing, Linking, Outlining, Open systems
    Design of Hypermedia Script Languages: The KMS Experience BIBAPDF 268-269
      Robert M. Akscyn; Donald L. McCracken
    This briefing describes the design of the KMS script language and some of the lessons learned from experience using it. The language -- the result of over 20 years of ZOG/KMS development -- is a procedural, block-structured language characterized by a simple 'command line' syntax, a large number of intrinsic commands (approximately 800), and the use of nodes and links as a central aspect of the syntax and semantics of the language. The intrinsic use of nodes and links in the script language provides interesting opportunities, not only for the design of other aspects of the language such as control structures, but also for the use of hypermedia as a programming environment to facilitate development and maintenance of scripts. In addition to designing the language, we have used it extensively to develop many hypermedia-based applications. Our experience, and that of end-user organizations, strongly reinforces our general belief that a script language is a valuable adjunct to hypermedia systems and instrumental to the utility of hypermedia for real world task environments.
    The Perseus Project: Developing Version 2.0 BIBAKPDF 270-273
      Elli Mylonas
    A discussion of the improvements and changes from Perseus 1.0 to Perseus 2.0, and the motivations for these changes.
    Keywords: Hypertext systems, Classics, Evaluation, Perseus, Education

    Panels

    Hypertext Fiction: Structure and Narrative BIBA --
      Robert Coover; Howard Becker; Jane Douglas; Michael Joyce; Stuart Moulthrop; John McDaid; Mary-Kim Arnold
    Serious literary hypertexts (notably the work of the so-called "Eastgate School") have emerged in recent years as a growing influence on both mainstream hypertext writing and on the larger literary community. By exploring the interaction between structure and narrative, and the hypertextual interplay between reader and writer, these works break ground of equal importance to literature, scholarship, and technical communication. The panel will create a constructive, collaborative hypertext live and onstage.
    Designing and Building Structure BIBA --
      Daniel M. Russell; George P. Landow; Norbert Streitz; Stuart Moulthrop; Jay David Bolter
    Does a priori structure lead to well designed hypertexts? Or merely dull hypertexts? Moderator Daniel Russell begins with a comparative technical briefing, exploring structure-building in Storyspace, IDE, and MacWeb. A free-wheeling panel discussion follows, exploring shifting opinion on this vital issue.
    Argumentation in Action BIBAPDF 274-275
      Mark Bernstein; Catherine C. Marshall; Norbert Streitz
    Can explicit hypertext structure shed light on complicated arguments? Three hypertext systems -- Storyspace, Aquanet, and SEPIA -- will be used to explore and to represent issues from the Hypertext 93 panel, "Hypertext Fiction: Structure and Narrative." Through a realistic experiment in capturing a particularly challenging exchange of views, this panel seeks to illuminate different approaches to hypertext argumentation.
    The Business of Hypertext BIBA --
      Bob Glushko; Dale Dougherty; Dennis Egan; Eliot Kimber; Peter Martin; Elli Mylonas; Bruce Warren
    An expert and diverse panel, drawn from business and academe, and from large corporations and entrepreneurial small businesses, discusses the business and management of serious hypertext projects. Topics include the transition from research to practice, the role of standards, identifying markets and customers, and organizations and mechanisms that ensure editorial quality, successful deployment, and commercial viability.
    Index BIB 276-294
      Rosemary Simpson

    Workshops

    Hypertext in Engineering BIBA
      Kasper Østerbye
    On a number of occasions it has been stated that hypertext is the perfect technology for supporting engineering teams in organizing information about ongoing and past projects. In general, hypertext has been proposed for interrelating such things as project diaries, schedules, documentation, blueprints, test requirements, standards materials, etc. In software engineering, program-related information is envisioned to be managed through a hypertext system that supports collaborative work. This workshop will bring together researchers working toward these visions, with workshop discussions concentrating on how these visions can be realized and supported.
       The workshop will be organized as a number of discussions around specific topics. Topics of interest include, but are not limited to:
  • Experiences of using hypertext in engineering
  • Requirements for using hypertext systems in engineering
  • Concurrent engineering, version and configuration control
  • Integration of documentation and specifications of design
  • Use of hypertext for software engineering
  • Hyperbase Systems BIBA
      John J. Leggett
    The hyperbase (hypermedia database) system architectures to be discussed are large-scale (tens of terabytes), open, distributed, heterogeneous and extensible with capabilities for computation and collaboration. Typical application environments are archival digital libraries, large-scale collaborative systems and large engineering enterprises. Discussion will concentrate on hyperbase systems and not hypermedia environments or user-interface issues except as they might impact the hyperbase.
       The areas listed below may serve as focal points for discussions. Participants are encouraged to identify additional areas of critical importance.
  • Models and Architectures
  • Node, Link and Structure Management
  • Browsing/Search and Query
  • Version Control
  • Concurrency Control, Transaction Management and Notification Control
  • Impacts of Application Environments
  • Courses

    Educational Uses of Hypermedia: From Design to the Classroom BIBA
      Daniel M. Russell; George P. Landow
    Objective: To provide a broad survey of educational uses of hypermedia along with specific advice for designing and deploying effective educational applications.
       Content: Hypermedia creates rich possibilities to dramatically expand on traditional educational practices. In this course we'll examine the state-of-the-art in educational hypermedia. At the same time, we'll be grounded in reality. Based on the instructors'real-world uses of hypermedia in a variety of educational settings, many stories will be told, examples will be given, and heuristics for selecting, using, and creating effective educational hypermedia will be illustrated. This is both a broad survey of the field and a course that can be directly applicable in your use of hypermedia for learning.
    The HyTime Interchange Standard BIBA
      Steven J. DeRose; David G. Durand
    Objective: This course will provide the essential background required to read the HyTime standard, understand its structure, and apply it in practical situations.
       Content: After a brief review of SGML, this course introduces and describes HyTime (ISO 10744) an SGML-based standard for representing hypermedia and multimedia documents. HyTime provides flexible ways of representing links, without constraining the kinds of data objects that can be connected by those links. HyTime also includes methods for handling both temporal and spatial rendering and synchronization. We will describe HyTime mechanics in some detail and will also include recommendations about its use.
    Introduction to Hypertext and Hypermedia BIBA
      Jakob Nielsen
    Objective: The goals of this course are to give attendees an introduction to the concepts of hypertext (non-sequential writing) and hypermedia (multimedia hypertext), to provide them with the background necessary to take more advanced courses at Hypertext'93, and to participate in the conference even if they are not hypertext specialists.
       Content:
  • Definition of hypertext and hypermedia.
  • Survey of the state-of-the art of hypertext with several demonstrations.
  • Hypertext architectures and system-level hypertext.
  • User interface issues and problems in hypertext.
  • Empirical tests of the usability of hypermedia systems and documents.
  • Navigating large information spaces.
  • Putting Documents Online: A Manager's Guide BIBA
      Ann Rockley
    Objective: Participants will be able to evaluate information, identify key factors affecting decisions, and develop criteria for decision making in projects to put documents online.
       Content: There are a number of decisions to be made before beginning to put a document online. Making the correct decisions up front can save you months of work and assist you in avoiding pitfalls. This workshop will show participants how to identify criteria for document design and software selection based on the results of audience and information analysis. The pros and cons of different online document formats (e.g., document database, SGML, proprietary) and distribution methods (e.g. CD-ROM/WAN/diskette) will be examined. We will look at the issues of documentation management (link management, revision control, and multiple authors). Some of the pitfalls of conversion will also be reviewed. This course does not review issues of writing for online. A sample exercise scenario will be used throughout to assist participants in applying the information and skills.
    Becoming a CD-ROM Publisher BIBA
      Bruce Winters
    Objective: This course describes the current environment for CD-ROM publishing and outlines a general and cost-effective path to follow.
       Content: This course is a comprehensive introduction for the author, publisher, and user of electronic information delivered on CD-ROM. The course addresses both technology "nuts and bolts" issues and the management issues for successful CD-ROM publishing. Topics include user types, authoring tools, retrieval software, desktop production, effective use of hyper- and multimedia, platforms, CD drives, standards, distribution, security, allocation of resources, intellectual property rights, cost-benefit analysis, and project management.
    SGML for Writers: An Introduction to Document Structure and Analysis BIBA
      Elizabeth Gilmore
    Objective: The SGML concept that most affects writers is that SGML documents have a hierarchical or tree structure that is rigorously defined as a Document Type Definition or DTD. The primary objective of this course is for writers to gain sufficient knowledge of SGML to understand the role of DTDs in an SGML-based publication system. The course is designed to teach writers how to think effectively about SGML, not simply to understand SGML syntax.
       Content: After an overview of SGML-based electronic publishing, the instructor uses a sample document to introduce DTDs, SGML markup and the process of analyzing documents to identify and describe their structure. Following the instructor's example, participants analyze some short documents, describe their document's hierarchical structure in a simple graphic tree notation, and begin learning how to use SGML to describe document structures.
       The course is designed to give writers first hand experience with the decisions and trade-offs that are inherent in DTD development.
    Automatic Information Retrieval and Text Utilization BIBA
      Gerard Salton
    Objective: This course covers basic methodologies for dealing effectively with large full-text information files consisting of documents of different type and covering many different subject areas.
       Content: The course provides a brief introduction to basic text storage and file organization methods. Procedures are described for carrying-out a viable analysis of text content (text indexing) applicable to collections in arbitrary subject areas, for efficiently comparing query statements with stored items, and for optimizing the query formulations during the search process.
       Since long book-size documents are often processed, it is important to provide access to smaller text excerpts rather than only to full documents. Strategies are described for accessing text passages of arbitrary size, and for automatically linking related text passages. The manipulation of collections of linked text passages leads to procedures for constructing text summaries, identifying the main themes apparent in the texts, and implementing selective text reading and text traversal strategies. Such applications and other similar ones are briefly described.
    Designing Electronic Publications: How We Do It BIBA
      Paul Kahn; Krzysztof Lenk
    Objective: The purpose of this course is to teach the design of electronic publications by leading participants through the basic principles of visual rhetoric and logic, highlighting those universal elements of visual design that apply to design for the computer. We will focus on issues of information design, symbolic and representational language in icons, the challenges of type on the computer screen, and methods for representing hypertext links.
       Content: The lecture will be illustrated with slides of information design examples, computer animations of effective typography on the computer screen, and examples of electronic publications produced by the instructors. We will illustrate the methodology used to develop visual language for electronic publications such as online software and hardware documentation, college-level text books, and consumer reference books.
    Visual Language: Analyzing the Emerging New Language Combining Words, Images, and Shapes BIBA
      Robert Horn
    Objective: Visual language is an emerging new language composed of words, images and shapes, which are tightly coupled. So closely are words and visual elements integrated that, in many communication situations, we can not do without either. This course will focus on approaches to analysis of visual language as a language and such topics as how to integrate words and images. The outcomes for participants will be a greater understanding of visual language and a greater propensity to use it in their own communication.
       Content: This course will present evidence for why we should begin to treat visual language as a language that most of us will be "speaking" more frequently in the next 10 years. We will analyze visual language to sort out what words do best and what the images and other shapes contribute. We will then explore what is known about the "systematics" (the grammar and semantics) of the language. Finally, a wrap-up group discussion about the hypermedia implications.
    Interactive Learning Environments: Where They've Come from and Where They're Going BIBA
      Elliot Soloway
    Objective: To enable attendees to better understand the alternative computing technologies that are available for learning, teaching and training, i.e., strengths/weaknesses, technology & classroom requirements, domain/task applicability.
       Content: A historical survey of the various teaching and training technologies will be given. The architectures of computer-assisted instruction (CAI) systems, simulations, intelligent tutoring systems (ITS), microworlds, and interactive learning environments (ILE) will be described. Emphasis will be placed on how the 90's computing infrastructure (e.g., high-MIP/GIP computation, high-bandwidth networks, multimedia, portable computing) will impact teaching and learning.
    Hypermedia Documents for Regulatory and ISO 9000 Compliance BIBA
      Bruce A. Warren; Neil W. Blahut
    Objective: You will be able to decide if you need electronic document management and delivery to meet OSHA and ISO 9000 requirements. You will be able to identify, specify, and satisfy the architectural and user interface requirements that distinguish a document management/workflow system suitable for white collar document creators, and a plant floor hypermedia tool that will be accepted and used by production workers.
       Content: Electronic at-the-job information is almost the only way to meet OSHA and/or ISO 9000 compliance requirements without large staff increases to handle paperwork. Case histories from three large chemical companies that have implemented OSHA/ISO hypermedia projects will be used as examples and project planning templates. The case studies will illustrate how to identify the information needs of production workers. We will explain and demonstrate user interface designs for packaging and delivering information in ways that plant floor workers can absorb while on the job.
    Forging the Business of Hypertext Publishing BIBA
      Dale Dougherty
    Objective: This course provides a basic understanding of how publishing works as a business. It reviews current academic and commercial experimentation with distributed hypertext publishing to explain how technological innovations in distribution are changing the activities and economics of publishing.
       Content: Traditional publishing consists of acquiring or developing content, manufacturing the content as a tangible product, and marketing and distributing those products in volume. Large computer networks such as the Internet are making it possible for a new kind of information publishing that incorporates hypertext technology on a global scale. We will look at existing Internet-based models for distributed information servers. In particular, we will look at the WorldWideWeb, a hypertext project originating at CERN in Switzerland. We will examine how information servers provide a useful framework for satisfying the requirements of the owners of information as well as providing transparent access to users.
    Text to Hypertext Conversion: A Practical Engineering Approach BIBA
      Mark Chignell; Bernd Nordhausen
    Objective: This course is a comprehensive survey of how people are solving the problem of converting text to hypertext, and it focuses on practical engineering approaches. Participants will gain the required knowledge to convert existing text to hypertext.
       Content: This course begins with an overview of the conversion problem, and an introduction of the two basic problems, node segmentation and automatic linking. We then show how SGML can simplify the conversion process, and improve the quality of the resulting hypertext documents. In addition, we touch upon several advanced topics of hypertext conversion, such as lexical and linguistic approaches for automatic linking, and signature methods. The course also includes an evaluation of different conversion methods and a demonstration of conversion tools.
    Hypermedia in Museums BIBA
      David Bearman
    Objective: This course surveys the use of hypermedia in museum exhibitions and reviews the unique requirements for hypermedia applications in the museum context. Participants will be better able to understand issues of design, copyright, production, staffing, standards, and distribution that determine the success of museum applications.
       Content: The workshop will explore the museum setting as an environment, museum audiences, and the design consequences for interactive multimedia. It will demonstrate and discuss examples of the best hypermedia applications installed in or developed by museums worldwide.
       Requirements specific to the museum setting will be matched to capabilities and features of hypermedia. Production related issues, including contracting with museums, copyright concerns of museums, hardware and software requirements for delivery in museum environments will be discussed.
    Introduction to Virtual Environments and Advanced Interfaces BIBA
      Daria Bergen; James K. Hahn; Robert J. K. Jacob
    Objective: To introduce basic concepts of virtual environments and other advanced user interfaces. Attendees will learn about current virtual environment research, implementation techniques for virtual environments, hardware and software for virtual environments, and some advanced interaction techniques, including the use of eye movements.
       Content: Topics: human-computer interaction styles; hardware, software, and application considerations for getting started in virtual environments; building a laboratory; creating virtual environments; image, sound, and motion synthesis; advanced interfaces using eye movements; designing interaction techniques for eye movements.
    Evaluating Hypermedia Usability BIBA
      Gary Perlman
    Objective: To introduce cost effective methods of evaluating hypermedia systems, particularly early in the development process, when redesign is least expensive. To provide participants with enough experience during the course so that they are able to apply the methods on their own.
       Content: The course introduces a development methodology for hypermedia applications designed to enable cost-effective evaluation and usability testing. The course will focus on three methods for evaluating systems: Observational Skills and Video; Program Instrumentation; and Questionnaires and Checklists. All three methods are designed to be used by a broad base of evaluators, minimizing skill and equipment requirements. Participants will be given hands-on experience gathering and interpreting each kind of evaluation information.
    Hypertext in Legal Education and Law Practice BIBA
      Ronald W. Staudt; William R. Andersen; David E. Kiefer
    Objective: To provide a survey of current hypertext applications in law schools and law firms, including basic legal education for law students, training for new associates in law firms, continuing legal education for experienced attorneys, legal research, and litigation support.
       Content: The course will provide an overview of law as an inherently hypertextual domain. The instructors will describe specific hypertext and multimedia applications that are being used in the teaching and practice of law around the world. The instructors will discuss design and implementation methods and review lessons they have learned from extensive use of hypertext in legal applications.
    Metaphor Design in Hypertext Systems: Managing Expectations and Surprise BIBA
      Aaron Marcus
    Objective: This course will help participants manage the fundamental terms, images, and concepts in complex systems and displays that they develop for hypertext systems to make it easier for users to understand complex structures and processes. Participants will learn terminology, theory, and practical techniques that will improve their use of metaphors in products and provide guidance for research.
       Content: This course will begin with a discussion of user interface components, classical semiotics and rhetoric, and the history of metaphors in computer systems. Then we'll look at what is happening in the industry as the metaphors of computer systems are changing. We'll close with a pen-and-paper design exercise to explore the basic nouns and verbs of a hypermedia system.
    Cooperative Hypermedia Systems BIBA
      Norbert A. Streitz
    Objective: This course will help participants to evaluate the innovative potential of hypertext/hypermedia for CSCW scenarios and the need to expand hypermedia systems into cooperative systems. Learning about the roles of hypermedia, i.e. to be content of and medium for cooperative work, participants will gain a better understanding of the opportunities of cooperative hypermedia.
       Content: The course will begin by introducing the basic concepts of hypertext/hypermedia and selected scenarios of CSCW applications (e.g., document preparation, decision making and argumentation). These scenarios are used to discuss requirements for supporting synchronous as well as asynchronous cooperative work using hypermedia features. Examples of multi-user hypermedia systems are presented. This includes design decisions, system features and experiences. In addition, the integration of additional CSCW support (e.g. audio/video conferencing, shared screens) will be discussed.
    SGML for Implementers BIBA
      Steven DeRose; Darrell Raymond
    Objective: This course will help participants apply ISO 8879 Standard Generalized Markup Language to hypermedia applications. SGML's document model can greatly enhance portability and effectiveness of electronic documents. This course will demonstrate how to use SGML effectively, and the fundamental database principles underlying it.
       Content: We will focus on using SGML well (not merely "validly"); how to workaround certain rough edges of SGML; and how to think analytically about document models, structural design, and electronic documents. Attendees will learn principles of structured documents, means by which SGML supports them, and when to consider alternative techniques.
    Intellectual Property Protection BIBA
      Pamela Samuelson
    Objective: This course introduces the intellectual property issues faced by software developers, including the specific issues that are raised by the unique character of hypertext, multimedia, and digital libraries. The course is not designed to provide specific legal advice, but people taking the course will be much better able to seek and understand legal advice about intellectual property concerns.
       Content: This course will begin with an overview of the intellectual property system and the basic principles of copyright law. The course will explain why the special character of digital media poses problems for traditional legal models of intellectual property. The course will review and critique key judicial decisions on copyright protection for software, graphics, and multimedia.
    Presenting Information Visually: Graphic Design Principles for Use-Oriented Designers BIBA
      Suzanne Watzman
    Objective: This "device-independent" course introduces graphic and information design principles to professionals with no formal visual design training. Understanding and appropriate application of visual design principles are the criteria for successful communication of ideas. Today's challenge is as much an issue of creating good technology as making it intuitively useful. This course will demonstrate and instruct in the value of quality visual design when integrated into the user interface design process.
       Content: This is a highly-illustrated overview of principles and concepts, including: terminology, process, perceptual concepts, typography, page/screen composition and organization, forms/diagrams, graphics, color, style guides. Case studies and examples demonstrate principles in context and a guidelines section will summarize principles. This course WILL NOT turn participants into graphic designers! IT WILL demonstrate how understanding and appropriate application of these principles can make products easier to use and understand and, ultimately more successful.
    Resource-Based Hypermedia for Education and Training BIBA
      Wendy Hall; Hugh Davis; Gerard Hutchings
    Objective: This course will examine the potential for resource-based learning in education and training and show how open hypermedia systems can be used to create instructional material from large sets of multimedia resources. Based on examples drawn from a number of case studies, participants will be shown how such material can be customised and tailored to meet a variety of educational and training needs.
       Content: The course will show how hypermedia systems can be used to guide students through core instructional material whilst having full access to large databases of multimedia resources. Such environments can be customised and tailored to meet individual needs, authoring effort is reduced, and instructional material can be integrated with existing working environments. The course will present the authoring, navigation and interface issues of such environments through a number of case studies, drawn from both industry and education.
    Interactive Electronic Technical Manuals and the Content Data Model BIBA
      Bryan K. Caporlette
    Objective: The course explains the information requirements and user interface capabilities needed for Interactive Electronic Technical Manuals (IETMs), especially those based on the U.S. Department of Defense's Content Data Model (CDM). It will enable participants to plan an evolution from page-oriented technical manuals to database-oriented ones in which "neutral" information can be reused for different presentation devices and user tasks.
       Content: The CDM is an SGML-based specification that is required for defense contractors and which is well-suited for interactive manuals in air transport, automotive, and other applications involving maintenance of complex systems. The CDM supports context-dependent filtering, branching based on user interaction, dynamic generation of information, and other complex presentation techniques that can be exploited in IETMs.
       The course begins with an introduction and rationale for this model-based approach to technical manuals. It explains how the CDM distinguishes application-specific information from a "generic layer" and emphasizes how generic templates for information development can be applied in IETMs in a variety of industries.

    Commercial Symposium

    Multimedia Authoring and Viewing Tools BIBA
      Daryl Savell; Clay Vervalen; Jeff Day; Steve Schmitt
    Multimedia technologies offer tremendous new opportunities to developers of interactive information systems. The experiential nature of multimedia is fueling its growing use in electronic documentation systems, computer-based instruction, and other application areas. Speakers from Microsoft, Asymetrix, and Silicon Graphics will describe different approaches to developing and delivering applications that integrate digital media objects with traditional text, graphics, and fixed forms of data. Speakers will demonstrate their latest development and delivery tools and present practical, finished applications.
    Corporate-Wide Hypertext Systems BIBA
      Ken Kershner; Dennis Egan; Tom Rearick; Mike Judson
    Hypertext applications in corporate environments are subject to frequent data changes, simultaneous access by multiple users, heterogeneous computing environments, and requirements for personalization. Practical issues in building and deploying large-scale, commercial hypertexts will be addressed by representatives of Bellcore, Lotus Development, and Folio Corporation. They will share their experiences and recommendations for developing robust hypertext systems in corporate environments. Demonstrations will include systems that have been fielded and used in distributed corporate settings.
    Hypertext Writing and Publishing BIBA
      Rosemary Simpson; Rob Akscyn; Mark Bernstein
    Hypertext publishing can provide new forms of interaction for readers, but it imposes new requirements on authors and computing systems. This session will focus on present and future tools and systems to support hypertext publications. Representatives of Eastgate Systems and Knowledge Systems will discuss the major issues in creating and distributing literary and scholarly hypertexts. Presentations will include examples of published works and projections concerning the future of electronic publishing.
    Corporate Conversion Strategies and Methodologies BIBA
      Daryl Savell; Robert Glushko; John Johnson
    Many organizations recognize the benefits of converting to on-line publishing and SGML source files. However, the transition from paper-based methods and proprietary authoring tools can be risky and expensive. Speakers from Passage Systems and Communitec will describe their experiences helping organizations convert to on-line publishing while preserving their sizable investment in technology, process, and legacy documents. Presentations will include a review of the benefits of SGML and on-line publishing, typical costs and problems in making the transition, and step-by-step guides to help ensure success.
    Building Embedded Hypertext Applications BIBA
      Dennis Egan; Bruce Warren; Dan Cunningham; Surend Dayal
    Hypertext as a technology is finding its way into many different application areas. As an integral part of applications in industrial process control, computer aided design, legal research, and many other systems, hypertext has become an embedded information-access technology. Warren Forthought, Intelligent Controls, and SoftLaw Corporation will describe their experiences in building and fielding embedded hypertext applications in industrial, engineering, and legal settings.
    SGML Authoring and Viewing Tools BIBA
      Victor Riley; Faye Merrideth; Jolana Leinson; Bill Smith
    SGML is an international standard for text representation that enables a single document source to take on multiple forms for viewing and printing across different platforms. Speakers from Arbortext, Frame Technology, and Electronic Book Technologies will discuss their tools for creating and delivering SGML-based data, and will demonstrate how their tools support the process. Both complementary and competing approaches among the vendors may emerge as they discuss and demonstrate their individual tools.
    Converting Legacy Data to SGML BIBA
      Robert Glushko; John McFadden; Harry Summerfield; David Silverman
    Electronic publishing and neutral data formats are attractive visions, but many organizations have large volumes of data in several different proprietary formats. How can these legacy printed and word-processor documents be converted to a single form that enables both hypertext and paper delivery? Representatives of Exoterica, Zandar, and Data Conversion Laboratories compare and contrast their approaches to converting legacy data. Typical benefits, costs, and problems associated with data conversion will be discussed, along with presentations of vendor tools and services for aiding the process.
    Minimum Platform Hypertext Tools for Technical Publications BIBA
      Phill Murray; Scott Johnson; Ronald Turner
    Hypertext is viewed by some as an expensive technology, requiring pricey software tools and high-end graphical workstations, but tools are available for creating and deploying high quality, cost effective hypertexts on MS-DOS computers. Representatives from Ntergaid and Sophware Supplements will present electronic technical manuals created and deployed on low-end PCs, and will discuss the creation and conversion processes. The emphasis will be on delivering low cost, high quality, electronic technical documents. Demonstrations will address problems such as automatic link-creation, approaches to handling graphics, and deployment strategies.

    Reception

    Above and Beyond Publishing: The Inexorable Logic of Metamedia Publishing BIBA
      Ted Nelson
    THE TRAGEDY OF CLOSED MEDIA.
  • Closed media, unusable together, unable to quote from or link to one another,
       are creating a Balkanized world of electronic media. Whereas what we need
       is on-line media with comparable connections -- for scholarship,
       controversy, anthologizing, understanding, the reworking of ideas, and
       freedom. THEY DON'T UNDERSTAND WHAT WE NEED --
  • A Point-and-click universe. We want to be able to jump from document to
       document, like a squirrel from branch to branch, passing through one
       document on the way to another, purchasing just that moment's portion.
  • Networked interwoven documents. We need to be able to treat the network as a
       seamless whole, and the contents as a unified docuverse.
  • Royalty. Owners need to be paid. Copyright is not an imposition; it is the
       compromise under which things are made available. The new system must
       extend it inconspicuously and innocuously.
  • Open hypermedia publishing. Everyone must be free and equal to publish, to
       comment, to publish links, to quote. Quotation can work within the
       copyright system, by purchasing the quotation from the document of origin.
  • Compatibility. Enough of this hardware incompatibility hogwash! We need
       compatible data, to which the hardware is a window.
  • Connectedness. Any document must be able to connect to any other.
  • Stability. We need to know the material will be addressable in the same form
       and the same way, decades from now.
  • A Populist System. This cannot just be for the wireheads, early adopters,
       elite; it must be accessible to all.
  • THEREFORE WE START OUR OWN UNIVERSE. CONCLUSION: THESE PREMISES LEAD TO A NEW PLAYING FIELD:
  • Open-ended technology, capped by trademark: a variety of database methods can
       implement this publishing model, in a mix of assured upward compatibility.
  • Defined by contract. Participants agree to unusual terms of sale, etc. No
       rights are relinquished by participants except within this closed system.
  • Promise of stability. Publisher and vendor agree to keep material in the
       repository.
  • Minute sale. Sale is interactive and on request, with royalty even on tiny
       portions.
  • Clean royalty. Proportional on every piece bought by user.
  • Arbitrary links. Any type of connection, from an expansible list of link
       types.
  • Quotation. Transclusion, or virtual instance pointer, says "buy from there." TECHNICALITIES --
  • Harpoon table. Inbound links and transclusion pointers are stored also with
       the destination document; this storage is paid for by those who point.
  • Local caching of transclusions. Quoted material may be stored with a new
       document, still owned by original publisher, but storage paid for by new
       publisher.
    Note: Handout, Xanadu On-Line Publishing, Sausalito, California
  • Keynote

    Hypertext and Group-Enabling: Lessons from the Desktop BIBA
      Irene Greif
    Desktop products for individual use have been richly developed in the past decade. A generation of "group-enabled" products is now emerging to meet users' needs to share documents, graphics and models from the desktop. Key technologies for this group-enabling include messaging, versioning and distributed data. We show how Hypertext can be another valuable group-enabling technology in such application as spreadsheets, worked processing and presentation graphics.
    Writing Toward the Future: What Avant-Garde Writing Has to Tell Hypertext Designers BIBA
      George P. Landow
    In an important sense all writing in hypertext is experimental since the medium is taking form as we read and write. Electronic linking, one of the defining features of this technology, can reconfigure notions of author, text, reader, writer, intellectual property, and other matters of immediate concern to those who design hypertext systems or author documents with them. Because hypertext fiction -- writing at and over the edge -- sets out to probe the limits of the medium itself, it acts as a laboratory to test our paradigms and our fundamental assumptions. A sample of experimental hypertexts shows the ways they illuminate issues ranging from reader disorientation and authorial property to the nature of hypertext genres and the rules of electronic writing.
    Reengineering the Field: Hypertext in the 21st Century BIBA
      Robert Akscyn
    What goals (if any) should the field of Hypertext have and how should those goals be pursued? This talk will assess the current 'state of the field' and present some proposals for future directions.
    A Hypermedia Approach to James Joyce's "Finnegans Wake" BIBAK 3
      Risto Miilumaki
    The application consists of the "Anna Livia Plurabelle" chapter of James Joyce's last novel "Finnegans Wake" (1939) and the manuscripts and French and Italian translations of that text published in Joyce's lifetime. Included is also the gramophone reading by Joyce of the last three pages of the chapter, and facsimiles of the corresponding manuscript pages. A hypertextual and hypermedia approach is well suited to demonstrate the development and the complexities of the text and help the reader better grasp the subtleties of Joycean art. An electronic medium would also be a flexible alternative to traditional, heavy editions of classic texts.
    Keywords: Electronic publishing, Literature on-line
    DynaText at Hypertext'93 BIBA 3
      Gregory Lloyd
    The Problem: Navigating information online: providing effective features and visual cues for navigation information in a new medium that takes advantage of familiarities with traditional printed information access paradigms.
       What Was Done: Automatic hypertext Table of Contents generation from document hierarchy elements. Inline "hot-text" navigation elements defined to replace printed footnotes. On-the-fly text formatting. Navigation history journals to provide sense of document "locality" online and to support task-oriented navigation (CBT).
       Why the Work is Important: DynaText is the only online information retrieval product to take ISO-compliant SGML data from arbitrary DTDs as the direct input stream. Thorough knowledge of document structure semantics provides superior information delivery capabilities which far outpace "page-flipper" models.
       Demo Illustration: The DynaText Demos (2) show several interesting and powerful features:
  • - Fast indexing of raw SGML data for online retrieval
  • - Full-text search including boolean, proximity, and document "structure aware"
       searching
  • - Inline and Table of Contents hypertext generation and navigation.
  • - Extensive GUI customization capability
  • A CD-ROM Version of "Who Built America" BIBA 4
      Alexandra Fischer
    In the past, American history has been the exclusive domain of a small and specialized group. Primary documents are often seemingly impossible to access, making it difficult for the non-professional historian to analyze source material and draw their own conclusions. Voyager's CD-ROM version of Who Built America directly confronts this problem by removing the filter of the historian and allowing the user to actually participate in the process of writing history.
       Voyager, with the full cooperation of the authors, has turned one section of the original Who Built America textbook by Steve Brier and Roy Rosenzweig into an Expanded Book. Voyager's version allows the user to not only read the original text (covering the period from 1874-1914), take notes and mark pages (as one would with a traditional book), but it also gives the reader access to the primary documents used by the authors when writing Who Built America. Excursions take the reader to a group of primary documents attached at relevant points in the text, with new texts for each excursion that provides additional context for the documents. The CD-ROM includes 5000 text documents, 600 photographs and illustrations, 75 audio clips (including many archival recordings), and 20 archival film segments (including an entire ten-minute film, The Great Train Robbery).
       The increased indexing capabilities of the CD-ROM version of Who Built America allows the user easy access to the text and attached documents. For example, throughout the text the user can easily find out the number of times a word appears as well as locate all of its occurrences. The excursions can be accessed either through a reference in the text or directly through a primary document index. Town names are annotated with an interactive map attached which allows the user to visualize the patterns of American settlement. And there is a timeline organized according by year and category (e.g. politics, arts, economics, etc.) which provides the user with a more general perspective of American history.
       Who Built America is important because it provides a non-traditional approach to American history, both in the original text and the CD-ROM format in which Voyager presents it. This is the first time that primary documents have been fully integrated with a historical analysis.
    Chimera: Hypertext for Heterogeneous Software Development Environments BIBAK 4
      Ken Anderson; Jim Whitehead
    A major characteristic of modern software development environments (SDEs) is heterogeneity. SDEs are composed of diverse object stores, user interfaces, and tools. The benefits of hypertext in organizing and browsing the immense amount of information stored in SDEs appear obvious. Yet these benefits cannot be realized by hypertext systems that attempt to "control" the environment they are placed in. This control manifests itself in various restrictions, such as requiring all data (both application and hypertext) to be stored in the same database, or requiring a single user interface for hypertext services across all tools.
       Chimera is a hypertext system which attempts to provide hypertext functionality for heterogeneous SDEs in a minimally intrusive fashion. In Chimera, anchors are associated with views of objects, rather than directly with the objects. This has the potential of allowing a complete separation of hypertext information from application data, as well as having anchors be view-specific (consider, e.g., anchors on a Petri-net, which has both a textual and graphical representation).
       Chimera hypertext anchors and links can be created on views of objects displayed by viewers in separate processes. Links can be established on two or more anchors on the same or different views. Since our hypertext information leverages off the concept of a view, links between objects stored in different databases are easily supported.
       The context of this demonstration is the support of a software maintenance process for a flight simulator. The Chimera demonstration illustrates the traversal of links between anchors in graphical views (the flight simulator's instrument panel), word processing documents (as supported by FrameMaker), and requirements documents (as supported by a graph-based tool), all implemented as separate viewers. The demonstration also illustrates the creation of anchors in each of the available viewers, and the creation of links between these anchors.
    Keywords: Heterogeneity, Hypertext, Software development environments
    PassageWays: SGML Production & Document Management BIBAK 5
      Robert J. Glushko; Bryan Caporlette; Daniel Chang
    PassageWays is an SGML-based document management and production system that supports the automated conversion of information into SGML and the subsequent assembly of SGML into hardcopy and online documents. It provides object-oriented configuration and version management, automated SGML conversion and build environments, graphics conversion, verification and validation utilities, workflow and workgroup management, graphical and command line user interfaces, and integration with other software applications.
       PassageWays abstracts the complexity of SGML, file systems, format conversion tools, and other technology to simplify the creation, management, and production of structured information, while its X/Motif GUI gives information producers an automated user-friendly desktop.
       PassageWays is completely open, and does not assume any single set of development or delivery tools. It can be used to produce online books in a variety of formats, such as Silicon Graphics' IRIS InSight, Electronic Book Technologies' DynaText, and Bellcore's SuperBook. It can incorporate conversion software from Avalanche Technology, Data Conversion Laboratory, or other source. Native SGML authoring using ArborText and conversion to SGML from FrameMaker can peacefully co-exist. The conversion of non-SGML formats to SGML is aided by a "Document Debugger", which invokes the appropriate non-SGML word processor at the position in the source file that causes an conversion error.
    Keywords: SGML, Document management, Format conversion, Workflow, Document "debugging"
    An Environment for Hypertext Application Engineering (ENHANCE) Based on Experiences Gained in ESPRIT Project HYTEA BIBAK 5
      Klaus Meusel
    Technical approach: The demonstration shows a hypertext generator plus targeting tool for electronic catalogs and point-of-sales / point-of-information systems. Markups are added to existing linear material, so that text and graphics can be compiled into a system-independent intermediate format. Standard systems like FrameMaker or ToolBook then can be used for automatic targeting. During the demonstration, we use the target system ToolBook, together with MS Word for adding the markups. The compiler is written in C++. ToolBook is hooked to the open ENHANCE environment by a few scripts implemented in OpenScript.
       Our benchmarks: Within one or two days, authors can design and implement high-quality hyperdocuments of about 500 ToolBook pages. The compilation and automatic targeting process for 500 pages with text, graphics and links takes about 5 minutes.
       The theoretical background: The approach is based on HDM and the ESPRIT project HYTEA. The idea is to keep hyperdocuments consistent and to reduce authoring costs by authoring on an abstract, global level (authoring-in-the-large). By focusing on a special class of applications, the general concepts of HYTEA were transformed into our industrial environment. After instance specification for a predefined schema, the following HDM objects are created automatically:
  • - entities
  • - applicative links
  • - structural links To improve the quality of the applications, an extra effort was put into the professional design of the user interfaces. The demonstration will show the authoring process, compilation and final hyperdocuments. Issues on hypertext interface design (corporate identity, screen layout, use of a grid, typography, etc.) can be discussed by navigating through the running applications.
    Keywords: Model-based hypertext, Structural approach, Cost-result ratio, ToolBook targeting, HDM
  • The Microcosm Link Service: An Integrating Technology BIBAK 6
      Wendy Hall; Hugh Davis
    Microcosm has been used extensively over the last two years to develop a number of applications that demonstrate its capabilities as an open hypermedia system and its ability to provide a link service across different types of application packages. This demonstration will show some of these applications, chosen to illustrate the capabilities of the system. In particular we will be demonstrating the use of Microcosm with the 3D-modelling package Autocad, to show how the link service can be applied to packages of this nature and also how a 3D-model can be used as an interface to a set of hypermedia documents.
    Keywords: Link service, Open system, Multimedia, Integration, 3D
    DHM -- A Dexter-Based Hypermedia System BIBAK 6
      Kaj Grønbæk; Jens A. Hem; Lennert Sloth
    DHM is a cooperative hypermedia system supporting users' navigation in and editing of shared materials. Cooperation support includes long term transactions, flexible locking and awareness notifications based on OODB technology. The hypermedia functionality fulfils the Dexter Hypertext Reference Model, including: Multi-headed (n-arity) links with bi- and uni-directional traversal. A variety of composites being used for TableTops, browsers and collecting results from simple queries. DHM exists in several variants, including a combined Unix(Sun)/Mac configuration supporting cross platform sharing of hypertexts. DHM is based on a platform independent object oriented framework for building Dexter model compliant hypermedia systems. DHM is also an open hypermedia system allowing integration of third party applications.
    Keywords: Cooperative hypermedia, Dexter model, Open architecture, Object oriented development Framework
    The Artifact-Based Collaboration (ABC) System BIBAK 7
      Kevin Jeffay; John B. Smith
    ABC is a multi-user hypermedia system intended for distributed collaborative groups. Novel features include the following:
  • a data model based on graph theory, conservatively extended to meet
       hypermedia requirements,
  • a scalable data store that is implemented in a distributed architecture
       allowing data to be stored in multiple locations and moved among locations,
  • an open architecture that permits users to incorporate existing applications
       into an ABC environment,
  • a general layer of infrastructure that permits any application to be
       conferenced (shared) over the network. We will demonstrate several users working on the same hypermedia structure both independently and collaboratively.
    Keywords: Distributed hypermedia, Computer-supported collaborative work
  • Navigational Search in the World-Wide Web BIBAK 7
      Reinier Post
    The X-Mosaic World-Wide Web browser, developed at NCSA, is quickly becoming the most popular interface for WWW hypertext documents, Gopher and WAIS databases and other sources of information, available on the Internet.
       X-Mosaic offers a point-and-click interface for hypertext documents and a keyword search interface for Gopher and WAIS databases. Keyword searching is not possible in WWW documents, while following hypertext links is not possible in the searchable (index) nodes of the databases.
       We augmented X-Mosaic with a navigational query algorithm for hypertext, providing keyword search at the client side. This is seamlessly integrated with the server-based keyword search for Gopher and WAIS databases. The new interface makes X-Mosaic a much more effective tool for finding information in the Internet.
       Regular expression search and the search for multiple keywords are currently available. Searching is time consuming and can therefore be limited by means of several parameters. Future plans include client-caching to reduce the network load caused by the retrieval of documents over the Internet.
    Keywords: Navigational search, Search-by-browsing, World-Wide Web, Information retrieval, Hypertext search
    Using World-Wide Web Hypertext as a Generic User Interface BIBAK 8
      Steve Putz
    I have been experimenting with World-Wide Web (WWW) at Xerox PARC to provide a generic user interface to a variety of information sources and services. One example is a World Map Viewer which utilizes the ability of WWW hypertext documents to include images along with formatted text and hypertext links.
       For the Map Viewer application, links embedded in an HTML document are used as controls allowing the user to change map rendering options (e.g. pan, zoom, level of detail). Additional user input can be obtained via the search keyword capability provided by WWW browsers. The HTML format and the HTTP protocol are in effect used as a generic user interface tool kit to provide not only document retrieval but a complete custom user interface.
       Links in each Map Viewer document (with labels such as "Zoom In") have URLs corresponding to different map rendering parameters, allowing the user to modify the map image by selecting the links. The view can also be modified by selecting on the map image itself.
       Users may interact with the Map Viewer using any WWW browser that supports in-line images in HTML documents, such as the NCSA Mosaic client developed at the University of Illinois at Urbana-Champaign.
       The significance of this application lies in the ease with which the server was created and the ease with which it is made accessible as a network service to users throughout the world. In the first four months it has been available on the Internet, the Map Viewer server at Xerox PARC has received over 100,000 requests from 5,600 users from around the world.
    Keywords: WWW, Hypertext, User-interface, Map
    HyperWriter 4.0: A Hypermedia Architecture for Diverse Applications BIBAK 8
      J. Scott Johnson; Brian C. Giedt
    NTERGAID will be demonstrating the HyperWriter family of hypermedia authoring tools including HyperWriter 4.0, HyperWriter for Training and the HyperWriter AutoLinker. Three key aspects of HyperWriter 4.0 that will be demonstrated are the HW-Basic internal scripting language, the HyperWriter 4.0 text retrieval engine, and a new system for formatting and graphically presenting both rich text and tables. The HyperWriter 4.0 text retrieval engine integrates full text retrieval capabilities with HyperWriter's native hypertext linking as well as extending the text retrieval facilities into the hypertext arena. Additionally, several real world applications of HyperWriter will be demonstrated including online documentation, interactive training and using HyperWriter as a hypertext help system. Also demonstrated will be an early version of HyperWriter for SGML -- a SGML to hypertext delivery environment using the Omnimark SGML parsing engine. The demonstrations will be conducted by J. Scott Johnson and Brian C. Giedt, the designers of the HyperWriter system.
    Keywords: Hypermedia authoring system
    The Philadelphia Project: A Hypermedia Application in Theatre History BIBAK 9
      John R. Wolcott
    The Philadelphia Project is a hypermedia application which manages complex textual and visual data relating to the development of the city of Philadelphia and the growth of the professional theatre in the United States between 1794 and 1835. What distinguishes The Philadelphia Project is its design to accommodate various learning styles and needs: it addresses both student users whose goal is to learn generally about the Philadelphia theatre and its role in the city's cultural life, and experienced scholar-researchers who wish to examine directly the primary textual and iconographic data. The centerpiece of the database is a to-scale 3D computer model of America's first professional theatre, the Chestnut Street, which serves as a gateway to associated textual and pictorial information in the database.
       In conjunction with the Henry Art Gallery at the University of Washington, a laser disk has been prepared with hundreds of examples of historic clothing and accessories. The disk also contains performances in the style of the early Federal period, and examples of actors performing in 19th century settings, created through the use of video chroma-key and a large theatre model from the period.
    Keywords: Hypermedia, Higher education
    A CAI System for Linear Algebra Based on "World Wide Web", Xmosaic and a Few Special Purpose Programs BIBAK 9
      S. Arnesen; H. Bjar; T. Gulliksen
    Undergraduate mathematics education is one of the factors limiting the number of students our university can accept. Demand for higher education, is both large and increasing, and only limited resources are available for building new universities. Several alternative paths are being investigated in order to increase the productivity of the existing universities. One of these paths is Computer Aided Instruction.
       Our objective was to see if it was possible to create a CAI-system that actually helps students while using open standards and free software making it possible to give a large number of students access to the system at a relatively low cost.
       The tools we chose to use was the SGML based HTML markup language for hypertext documents, the Xmosaic browser for HTML-based World Wide Web documents developed by the NCSA at the University of Illinois at Urbana-Champaign.
       We extended HTML to allow us to include formulas from TeX and plots and animations from Mathematica. Perhaps more interesting is the special markup introduced for pedagogical elements such as explanation of concepts, revealing of hints, and gradual revealing a document.
       The initial courseware is structured as reviews of previous exams. Since reviews of previous exams is a widely used pedagogical method, it lessens the number of new concepts the teachers has to learn. A teacher can directly use her pedagogical skills instead of first having to develop into a interface designer.
       A relatively simple extension to the HTML language can be used as a pedagogic markup language, encoding concepts such as hints and stepwise disclosure of a solution. The courseware produced using this tool is in a format easily accessed by the World Wide Web protocol, thus making it simultaneously available to a large number of users in an heterogeneous internet environment.
    Keywords: WWW, Linear algebra
    World-Wide-Web Hypertext in Physics Research BIBAK 10
      Vladimir Chaloupka
    Two large physics Collaborations are experimenting with hypertext webs built with the World-Wide-Web (WWW) technology. The projects are the Study of Charm Baryons (Experiment E781) at Fermilab, and the Deep Underwater Muon and Neutrino Detection (DUMAND) off the coast of Hawaii. The webs are publicly accessible from WWW. The best way to access the WEB is with a graphic browser such as Mosaic. The most rudimentary, but universally available, method of test -- browsing is
  • "telnet info.cern.ch"
  • "go http://web.phys.washington.edu/" (once on the WEB, it should be easy to quickly find out how to get Mosaic or other advanced browsers)
       The goal is to build webs capable of providing a prompt answer to (almost) any question about the experiments, serving a broad range of users (from the members of the two Collaborations needing a particular technical detail, all the way to the outsiders interested in a popular introduction). Both projects are at a stage where the real usefulness of this approach can be evaluated.
    Keywords: WWW, World-Wide-Web, Physics research, Collaborative hypertext
  • A Semantic Database Approach to Hypermedia Systems for the Museum of Local History BIBAK 10
      Carl Taylor; Douglas Tudhope; Paul Beynon-Davies
    The demonstration will show how a semantic database approach can be used to create hypermedia systems within the domain of local history museums. The demonstration will present an architecture based on an idea originally developed for database work -- the binary relational store. The store's power derives from its use of a single, uniform data structure. This structure can be used store both intensional and extensional information and is capable of representing abstraction in the application design. Through the presentation of two prototypes, Great Inclinations and HyperSHIC, it will be shown how the architecture might support automatic maintenance, different interface / navigation units such as general public browsing, more focusses activity, and authoring or classification by the curator.
    Keywords: Hypermedia, Semantic databases, Binary relational model, Museum information systems
    Balancing Ranking and Sequential Ordering in the Presentation of Search Results using Discussions BIBAK 11
      Alan J. Wecker
    Ranked lists of search results for in-book searches can lose sequence information which is valuable both to the user and to the retrieval/browsing system. We have conceived a unit of organization called the "discussion" to overcome this problem. A discussion is simply a contiguous region in a document which addresses a certain topic. Search results are presented as a ranked list of discussions, where the discussions are created dynamically based on the raw search results. Our demo, using a modified version of IBM BookManager READ/2, shows an implementation of discussion ordering, with parameters to interactively control discussion size and density. This technique is also relevant to hypertexts, information filtering, and other applications which combine the notions of sequence and relevance.
    Keywords: Ranking, Ordering, Tours, IR and hypertext
    CSILE: A Collaborative Educational Hypermedia System BIBAK 11
      Peter Rowley; Jim Hewitt
    The goal of the Computer-Supported Intentional Learning Environments (CSILE) Project at OISE is to build (a) an instructional theory driven by cognitive science research into collaborative learning processes and expert knowledge organization and (b) a collaborative educational hypermedia system to support educational practice inspired by the theory.
       We seek a new instructional theory to address the problems of inert knowledge and to encourage development of cognitive skills and attitudes associated with sustained inquiry. If students are merely exposed to curriculum content in a task-oriented context of doing projects, writing essays, taking tests and the like, there is a high likelihood of them remembering what they have been exposed to only as disconnected "inert" facts, if they remember them at all. In contrast, if students are actively engaged in establishing learning goals, stating their preconceptions about a phenomenon, seeking new information to fill in gaps, and reformulating their theories in conjunction with (and as a result of being challenged by) others, they tend to remember what they have learned more deeply and are better able to apply it in novel situations. They also begin to develop a genuine appreciation of the culture of sustained collaborative inquiry.
       The CSILE System's design is based on classroom experience and on the foundational theory, and in turn provides support for classroom experiments to test its predictions. It has been iteratively developed over the last seven years and is now in use in approximately a dozen schools across North America.
       We are demonstrating the latest version of CSILE, running on a small network of Macintosh computers, and will be showing CSILE's facilities for asynchronous integrated expression, discussion of, and reflection on knowledge in textual and graphical form. Our current research includes providing overviews of important links and other relationships within the database.
    Keywords: Education, Fine-grained collaboration, Asynchronous multi-user hypermedia
    INTELTEXT: Producing Coherent Textual Sequences While Navigating in Large Non-Hierarchical Hypertexts BIBAK 12
      Dmitry Subbotin
    IntelText is a software tool intended for accessing and arranging loosely structured information. The approach is based on a set of heuristic rules of navigation in a hypertext network.
       IntelText is capable of drawing out text items from a heap of information and representing them in a well-ordered sequence. Text items can be thoughts, facts, messages, abstracts etc. The user links the items (manually or automatically), forming a hypertext network. A link has no direction, weight, or other attributes. Its presence indicates just the semantic closeness of a pair of items. This implies the preference for mono-semantic text items.
       The automatic (algorithmic) navigation becomes possible in the network, based on the correspondence between topological and semantical connectedness of items. The algorithms provide for composing of well-ordered sequences of text items. The user sets the theme which he/she is interested in, by indicating the initial item. On each step of navigation the next item is chosen that is semantically (and topologically) close to the previous part of the navigation route, i.e. its links with already included items are considered.
       Depending on the kind of information, the resulting text sequence (the navigation route) can be either an ordered selection of information elaborating the given theme, or even a linear text. The application area of IntelText can be compilation of surveys, analysis of a flow of messages or a set of ideas, intelligent selection and arranging of information, authoring work, etc.
    Keywords: Intellectual navigation, Authoring, Coherence, Information organizers, Hypertext linearization
    Microsoft Access Cue Cards BIBAK 12
      Cyndi Bieniek
    Cue Cards are like an online coach. They allow people to do their own work as they learn Microsoft Access. Cue Cards provide step-by-step guidance through the most common Microsoft Access tasks. They communicate with Microsoft Access to determine what information to display, and can provide prerequisite information based on this communication. For example, if a user wants to create a form but doesn't have a database open, Cue Cards will provide the information on how to open a database first. For new Microsoft Access users, Cue Cards offer a structured yet personalized learning path.
       For Microsoft Access, Cue Cards replace the traditional tutorial, which provides instruction using "canned" examples, and follows a structured, linear path. Cue Cards are revolutionary because they break away from the rigid structure of a tutorial and allow the user follow a decision tree to create his or her own instructional path. Unlike a traditional tutorial, Cue Cards communicate with Microsoft Access and can "customize" instruction based on this communication. Finally, Cue Cards allow the user to work with his or her own data, rather than requiring practice with sample data.
    Keywords: "Do your own work as you learn," "a guided walk-through," "online coach," "What do you want to do?"
    Developing Global HyperMedia: The NCSA Mosaic System BIBAK 13
      Chris Wilson; Jon Mittelhauser
    NCSA is currently developing a new software tool, NCSA Mosaic, that will encompass all of the currently used global information systems and provide greatly increased functionality and ease of access to the Internet-based universal information space. NCSA Mosaic is a distributed hypermedia information system based on the World Wide Web technology originated by CERN. Mosaic provides a unified, coherent, hypermedia-based portal to the expanding Internet information space by enabling transparent access to all of the major information systems currently in use on the network (Gopher, WAIS, anonymous FTP, Archie, Usenet news, etc.). By virtue of its World Wide Web basis, Mosaic also provides a unique, flexible and important networked information functionality not available in other existing systems -- distributed hypermedia. In addition, Mosaic provides user- and community-level annotation and hyperlink support for collaborative work based on information accessible on the network. Mosaic is being developed across the X Window System, Macintosh and Microsoft Windows environments. Mosaic was originally conceived as an asynchronous collaboration system -- an environment for geographically distributed group or community members to operate on a common networked information base as part of their everyday work Current and future development will focus on enabling and expanding capabilities for information sharing, collaborative navigation and local information space construction across the global information space. Mosaic already supports extensive local information space customization methods, including text and audio annotations that can be attached transparently to any document available from any information source on the Internet. Future efforts will evolve this into a general system for sharing annotation, hyperlink, and document and information space construction activities across small and large groups and communities alike. As a result of these efforts, Mosaic and the Internet will become a flexible, malleable, and extensible information and collaboration system for a wide variety of uses by a large number of people, both independently and in groups and organizations.
    Keywords: Mosaic, World Wide Web, Distributed, Hypermedia, Information retrieval
    SuperCat in the SuperBook Document Browser System BIBAK 13
      Carol Lochbaum
    SuperCat is an automatically generated catalog of electronic documents that allows users to locate and access documents from a very large collection. SuperCat is simply added as another "document" to an existing SuperBook library, requiring no modification to the SuperBook browser itself.
       Finding a desired document in a full text database is a challenging problem. Typically catalog entries contain very limited information such as the title, author, and a few keywords, making it difficult to identify relevant documents. Moreover, different users may prefer different arrangements or views of a collection of documents, for example, chronologically vs. by author.
       SuperCat addresses these problems by providing one or more document catalogs for the SuperBook interface, taking advantage of the SuperBook search and navigation tools available for individual documents. Since SuperCat is automatically generated, the problem of frequent updates to the library can be dealt with efficiently.
       For alternative views of a document collection, several differently organized SuperCats can be constructed, without having to replicate the documents themselves.
       A SuperCat catalog contains the abstract or first 100 lines or so of text, the section headings, and hypertext links to the full text and graphics of each SuperBook document. Users first access SuperCat, then go to the actual documents using the links. Because SuperCat contains the complete heading structure of each document, which is fully indexed, it represents an improvement over traditional catalogs which generally contain only titles, keywords, etc.
       The demonstration will show how SuperCat is used to find documents in the SuperBook browser, and will also show how a new SuperCat is created when the SuperBook library is updated.
       SuperBook is a registered trademark of Bellcore.
    Keywords: SuperBook, Electronic library, Document browser, Library catalog, Hyperlinks
    Browz-All: A Hierarchical Hypertext System BIBAK 14
      Joel Remde
    Browz-All is a prototype hypertext user interface based on a hierarchical organization of information objects. Browz-All addresses the problem of efficient access -- any needed information should be available to the user, but frequently used items should require the least effort to access.
       A hierarchical design was chosen for several reasons: it has been successful in the SuperBook system; it scales up naturally; it allows flexibility in the amount of detail shown; and hierarchies are structures familiar to people.
       The user's own "information world", including both public and private information, is depicted in Browz-All as a hierarchy or tree of text icons. Detailed objects appear separately as tiled windows, and the user can easily switch between the two views.
       The tree can be expanded at points of interest until the desired information objects are found. The upper portion of the tree, which is always visible, provides context for the information to avoid the problem of "getting lost". Exploration and manipulation of the tree is done via a mouse gesture command interface. The tree structure can be customized according to the user's own needs or conceptual organization. Items may be placed under as many categories in the hierarchy as desired; color cues help identify them for navigation purposes.
       The system, not the user, manages screen layout, relieving the user of positioning and uncovering windows. This is supposed to improve efficiency of use. Browz-All attempts to keep frequently or recently used items visible and pushes such items toward the top of the tree where they can be found more readily.
       The demonstration will include a sample hierarchical "information space", and will show how it can be customized, navigated, and how new information can be incorporated.
       SuperBook is a registered trademark of Bellcore.
    Keywords: Hierarchy, Efficiency, Gesture commands, Color, Screen layout
    eText: An Interactive Hypermedia Publishing Environment BIBAK 14
      Rohit Khare
    The eText engine lays the foundation for a document-centric interface to a multimedia information space, affording an ease of construction and navigation heretofore absent in large-scale hypertext publishing systems. eText realizes its power from the innovative notion of a "self-aware" document, leveraging a portable file format (RTF) manipulated by an object-oriented graphical user interface under NeXTSTEP. The eText architecture abstracts and encapsulates conventional semantics for authoring systems into three classes of annotations to an underlying rich-text document: multimedia, interactive media, and hypermedia. The eText system has been developed for the Archetypes project, which aims to produce an online, extensible reference -- and text-book for teaching parallel programming. To that end, eText is designed to easily incorporate and cross-reference new documents, to support interactive educational material, and to adapt to individual student preferences. Soon, eText will also support the creation of an integrated development environment for writing parallel and distributed programs within the Archetype paradigm. This is possible because eText also supports an innovative ability to attach domain-specific Editors to individual documents, live objects which assist the user in the creation and editing of the document as a whole. For example, a source-code document can not only support hypermedia annotations to the code, but can also be attached to a code Editor, which offers syntax checking, version control, and debugging. Such Editors are the first step towards active agents in the construction of hypertexts. This demo is a work-in-progress, so interested parties are encouraged to contact the presenters directly. Public release of eText is projected for Q1 '94, tied to public release of the first chapters of the Archetypes Electronic Textbook. During 1994, we look forward to forming a consortium of parties interested in Archetypes and eText, as well as the development of document cross-compilation technology to allow viewing of eText documents on Macintosh and Windows platforms.
    Keywords: Hypermedia, Compound documents, Object-oriented analysis, Parallel programming education, User interface design
    Health Sciences Hypertexts at Columbia-Presbyterian Medical Center BIBAK 15
      Jeff Zucker; Robert M. Kahn; Narayanan Natarajan
    At the Columbia-Presbyterian Medical Center (CPMC) we have developed a number of hypertext and free text retrieval computer applications aimed at an extremely diverse audience which includes students and faculty in a university setting as well as health care providers and patients in hospital and clinic settings. Each application must be accessible to people from a variety of computer skill levels, from a variety of medical skill levels and from a variety of disciplines. In addition to the diversity of users, the CPMC applications encompass a diversity of text sources. Some sources such as the Physicians Desk Reference and the Guide to Clinical Preventive Services come from established print sources. Others such as the Medical Logic Modules Library and the Nursing Standards & Patient Education System were designed at CPMC specifically for use on computers. A third category of sources is represented by the Funding Opportunities database that we have adapted from an unwieldy relational database into a more flexible free-text retrieval system. Hypertext and free text systems offer features which make them ideal for presenting information from both computer and text sources in a manner that is individualized to the needs of users in a wide variety of learning contexts. We built traditional linear indexes which are familiar to those accustomed to the print sources. At the same time, we encourage a more freeform approach for those who wish to access the information in a more flexible manner. Ad hoc text searches and hard coded links allow users to combine information in ways not possible in the print equivalents of the applications. Hypertext links also allow for a more inclusive approach to user help and user feedback -- many of our applications allow several levels of help and have hot links which allow users to communicate directly with the developers and managers of the information.
    Keywords: Health, Medicine, Nursing, Education, Hospital, Library, Full text
    HyperGuide 2.0 for Windows BIBAK 15
      Beth Carter; Annie Breckenfeld; Dan Brown; Steven Wallace
    The demonstration illustrates how fast and easy it is to learn new material and find information with HyperGuide. The demonstration gives new users a quick, animated introduction to each component in Windows. The graphical presentation in each Feature Film lets users understand the purpose of the component and shows the basic techniques required to use each component. HyperGuide is an excellent example for prospective multimedia authors. The presenters can provide insight into the challenges of producing a multimedia document on the scale of HyperGuide. HyperGuide illustrates a variety of ways to access related information from the user's current reading location.
    Keywords: Online access to comprehensive documentation, Animated segments to illustrate and teach key concepts, Information available while applications are running
    ArborText's Hypertext Utilities for Authoring, Accessing, and Managing Information BIBAK 16
      Paul Klock
    The ArborText product demonstration gives ACM Hypertext '93 attendees hands-on experience with the ADEPT Series, a set of electronic publishing tools which support the creation and use of hypermedia documents.
       Underlying the ADEPT Series is a powerful programming language called the ADEPT Command Language. The Command Language gives hypermedia document developers the open system tools they need to implement hypertext applications. The Command Language is oriented to SGML information structures and it provides control of the rules, conditions, and actions of the hypermedia authoring system.
       For more tightly coupled hypermedia authoring applications, ArborText also provides an Application Programming Interface (API). The API allows an author to establish two-way communication between any document editing session and one or more other programs, e.g., a database system or full motion video application. All the programs are aware of events in the other programs and are able to exchange information freely through the API.
       ArborText uses Standard Generalized Markup Language (SGML) as its document structure. SGML is an International Standards Organization standard (ISO 8879) developed to facilitate document definition and interchange.
       SGML is a language for expressing documents in terms of their content and organization. In essence, the Document Type Definition (DTD) is the rule for defining and organizing Information Objects within a document. It provides the data structure against which the hypermedia document operates. The Information Objects can have as little or as much granularity as desired.
    Keywords: Authoring, Accessing and managing information
    Hardhat HyperText -- Interfaces that Work in Industrial Workplaces BIBAK 16
      Bruce A. Warren; Neil Blahut
    Popular hypertext user interfaces are designed by office workers, which is probably why they get rejected by operators in chemical plants. Missing are simplicity, robustness, scalability, customizability, speed, network document sharing, and navigators that make millions of pages quickly accessible with just a few mouse clicks. We will be demonstrating a user interface developed by trial and error during six years of in-plant use. The interface works in high pressure production environments to deliver documents and interactive training with no training required to use the terminal. The interface allows quick navigation through millions of documents using job-specific navigators. Documents are accessed by clicking on text, drawings, diagrams, and pictures. Industrial applications for hypertext and multimedia are here now -- electronic job aids, interactive training, document management. This massive market is being held back by consumer oriented hypertext interfaces being mis-applied to unforgiving industrial applications.
    Keywords: Manufacturing, Training, Network, Navigate, Multimedia, Hardhat
    Lust: A Hypertext Fiction BIBAK 17
      Mary-Kim Arnold
    "Lust" is a hypertext fiction of 36 nodes. It was developed in Robert Coover's experimental narrative workshop, and it addresses some of the issues specific to writing fiction in an electronic environment. These issues include repetition, multilinearity, multivocality, indeterminacy, narrative structure, and user interaction.
       The piece itself is an exploration of language and form. It begins with a prologue consisting of exactly thirty six words, and each word links to a particular node. Each node is linked to several others; some are text links, some are guarded space to space links. Because each reading is reader determined, each reading is unique, and the meaning and content of the story is changeable and transient. This feature of the piece calls attention to the indeterminate nature of language itself, and of hypertext fiction. Because there is no specific linear beginning, middle and end, the narrative is ongoing and changeable.
       Because each node uses one of an limitless number of points of view from which the "action" can be seen, the story is told and retold, with no one interpretation being superior to any other. It requires the reader to put the disparate pieces of information together, while also drawing the reader into the process of writing the story, since the reader chooses which words have texture to her, which determines the order in which the information is disclosed.
       In many ways, a piece like "Lust" embodies the nature of hypertext itself, in terms of fictional hypertext, at least. The structure of the piece (in this case, small, tight, limited) becomes part of the content, and perhaps in some cases, more important than the specific language used.
    Keywords: Hypertext fiction, Narrative structure, Repetition, Multilinearity, Multivocality, Indeterminacy
    DNE: An Internet Protocol for Hypertext Interoperability BIBAK 17
      Don McCracken; Robert Akscyn
    Present schemes for cross-application node/data linking tend to be operating system specific (e.g., OLE for DOS). They also tend to focus on linking services -- node and script services are sometimes not provided for. An alternative approach is to use an internetworking protocol that allows hypertext systems (and other systems wishing to provide node and link services) to interoperate (like WWW). This demonstration will illustrate a prototype internetworking protocol developed to explore interoperability of hypertext system and possibly provide an architectural foundation for large-scale hypermedia-oriented digital libraries.
       This protocol, called "DNE" for "Dynamic Node Exchange", provides a means to experiment with cross-system node, link, and script services. Protocols such as DNE, if based on a sufficiently broad data model, could provide a simple way for hypertext systems to interoperate, possibly allowing very large hypertexts to be constructed by integrating multiple heterogeneous hypertext databases. The demonstration will illustrate the work accomplished to date by using an existing hypermedia system (KMS) and accessing nodes from a number of remote sites over the Internet. The demonstration will also contrast the approach with other internet-based schemes such as World-Wide Web and Gopher.
    Keywords: Wide-area hypertext, Interoperability of heterogeneous hypertext systems
    A Demonstration of MetaCard: "The World's Largest HyperCard Application" BIBAK 18
      David D. Sherertz
    MetaCard is a hypertext Metathesaurus browser implemented using HyperCard. The Metathesaurus is one of three Knowledge Sources released annually, since 1990, as part of the National Library of Medicine (NLM) Unified Medical Language (UMLS) Project. The current version of MetaCard contains over 400 MBs of information in 146 stacks and index files, about more than 152,000 biomedical concepts, including names, definitions, semantic types, and relationships. MetaCard supports both the display of biomedical information and navigation using the semantic relationships present. The principal metaphor is "click to get more information." Several hundred sites worldwide run MetaCard off the NLM-distributed CD-ROM.
       Virtually everything the user sees in the interface is "live", in the sense that almost anything can be clicked on, and it will do something. There are close to 10 million inter-card links amongst over 182,000 cards organized into 137 HyperCard stacks. The demonstration will also illustrate the notion of a "no-typing" interface by using only "pointing and clicking" to navigate to biomedical concepts of interest to members of the audience. Audience involvement as participants in "Stump the Metathesaurus" will also illustrate the breadth of coverage in the current release of MetaCard. In spite of its shortcomings, MetaCard remains a remarkable artifact. We know of no other low-end, point-and-click implementation of comparable scale. After some explorations with a few contacts at Apple we began claiming that MetaCard was the world's largest HyperCard application and then waited for someone to contradict us. We are still waiting.
    Keywords: Navigational browsing, Automatic hypertext, HyperCard, Semantic locality
    Hypertext on the Corporate Help Desk BIBAK 18
      Tom Rearick
    This presentation will demonstrate hypertext embedded in a corporate help desk application. The challenge is to empower the help desk professional by leveraging the collective resources in the corporation itself. The integration of Lotus Notes and Lotus SmarText provide a complete solution to in-house electronic publishing and collaborative problem-solving. Lotus SmarText automates the conversion of existing text files into hypertext-rich electronic books. Lotus Notes combines communication and database technologies to increase business productivity. A real-world help desk application will demonstrate how customer support professionals use Lotus Notes to:
  • Resolve customer problems
  • Verify product deficiencies and report them as part of a continuous program
       of product improvement
  • Simplify resource planning and management
  • Collect useful information and disseminate it to other help desk personnel.
  • Provide feedback to the documentation department Complementing Notes, Lotus SmarText is used to:
  • Add value to existing corporate documentation
  • Search a large corpus of text quickly and reliably
  • Deliver value-added product support information to user's sites. This demonstration will illustrate how these two products work in concert, not in a single department but across the entire Company and among its customers. Lotus SmarText and Notes are not simply alternatives to paper documentation, but a means for effecting positive change in the corporation.
    Keywords: SmarText, Notes, Help desk
  • The Virtual Curator: Multimedia Software as Discourse for the Visualisation of Material Culture BIBAK 22
      Suzette Worden; Colin Beardon
    The 'Virtual Curator' is a piece of experimental software, which challenges the concept of the Virtual Museum. It is an environment in which users can work with a 'museum' of objects that are all representations. There is a 'store' of information and an 'exhibition' where users can make statements. The museum is now a metaphor for the organisation of knowledges.
       A consideration of the use of collage/montage as a coherent way of organising visual space has informed this project. The work of artists who have used collage/montage suggests models for experimentation in the creation of a visual language that does not deny its own social construction, opposes any closure of meaning and offers the viewer an active engagement in the production of meaning. Definitions of objectivity / subjectivity or fact/opinion have to be taken on board
       We propose to illustrate these theoretical issues, which have been informed by post-structuralist and feminist work on discourse theory, power structures, subjectivity and authorship, through a description of the Virtual Curator.
    Keywords: Authoring, Education, Graphic design, Multimedia
    Hyper3D BIBAK 22
      John E. Williamson
    Hyper3D, currently in BETA, is an attempt to combine the beauty and significance of antique stereocards with the ease of use and organizational power of hypertext and the added dimension of stereoscopic 3D computer displays. Both the digitized stereocards and the hypertext are presented in stereoscopic 3D.
       The collection of technologies that have become known as "virtual reality" will incorporate text either directly as documents or indirectly as menus or organizational structures. Due to the very nature of the display technologies used, these text items will be presented in stereoscopic 3D. The effect of viewing, reading and comprehending text that is presented stereoscopically is a human factors question that has not yet been fully addressed. With rare exceptions, current head-mounted displays (HMDs) do not yet provide sufficient resolution to read text. In fact most current HMDs leave the user legally blind, with 20/200 to 20/400 vision. However, by using a non-immersive, stereoscopic computer display that is both affordable and readable, some of the initial human factors and human perception issues that relate to stereoscopic text can be addressed. Hyper3D, in addition to being a marketable product, is rich and flexible enough to serve as a test bed to examine these issues.
    Keywords: 3D, Virtual reality, Navigation, Education
    VirtualMedia: Virtual Reality as a Hypermedia Navigation Tool BIBAK 23
      R. J. Wilkins; M. J. Weal
    It is well known that navigation within existing hypermedia can be difficult. As the size and complexity of hypermedia increases such navigation problems will become more apparent. It is essential to provide the user with tools that enable them to maintain their spatial awareness.
       The desktop metaphor and the tools within do not readily present the user with the spatial and structural relationships which may exist within hypermedia. A new virtual space metaphor will provide the user with a consistent visual representation of any such structures.
       We have currently implemented two hypermedia applications which provide the user with a virtual space representations of the data and the relationships between that data. Both applications use Autocad to manipulate the virtual spaces and have been authored within the Microcosm open hypermedia system. The first application contains data about Bath (an English town). Links exist between a 3D model of the buildings and other data (photographs, descriptive text and rendered images/animations). The user can immediately see the spatial relationships between the buildings in the model and hence the other data in the system. The second application contains maintenance manuals for a cable packaging machine owned by Pirelli. The user locates a part of the machine in the 3D model and follows links to the relevant maintenance information.
       These two applications lead us to believe that such a spatial representation would benefit other, perhaps non-inherently spatial, hypermedia documents. A tool for producing and using such a generalised virtual space is the main subject of the poster.
    Keywords: 3D, CSCW, Metaphor/mental-models, Navigation, Open systems, Virtual reality
    Comparing Usability of Graphical and Textual Interfaces in Hypertext BIBAK 23
      Pawan R. Vora; Martin G. Helander; Valerie L. Shalin; Dennis Mike
    This study investigated usability of interface styles in hypertext for information search tasks. The four types of interfaces were: graphical with labeled links (GL), graphical with unlabeled links (GU), textual with embedded links (TE), and textual with a separate list of links (TS). An information database related to vitamins, food sources, and related disorders was used -- referred to as NutriText.
       Forty novices answered search questions using NutriText over two sessions. In the first session, the subjects in the GL group performed significantly faster and spent less time/node than others. All subjects performed significantly better in the second session; the subjects in the groups GL and TE, however, performed faster and spent less time/node than those in the GU and TS groups. Among the groups, there were no significant differences in the first session for the navigation measures: number of unique nodes accesses (NU), number of node visits (NV), and navigation redundancy (NU/NV). This confirmed that the subjects were equally inexperienced in the information domain. In the second session, however, the GL and TE group's navigation performance was superior than the GU and TS groups.
       Interestingly, despite statistical significance, the subjects using textual hypertexts spend only 3 seconds more per node than those using graphical hypertexts. This suggests that subjects did not actually read the text, but focused on the highlighted buttons. Therefore, the subjects may not have understood the relation between the nodes in hypertext. Such user-imposed "incoherence" may cause disorientation in larger textual hyperdocuments.
    Keywords: Authoring, Empirical/data, Navigation, Usability
    The Hypertext Campus Project BIBAK 24
      David Slater; Wilma Strang; Hassan Khan; Vicki Simpson; Sheila Hill
    There is currently a perceived need to introduce computer based learning into British higher educational institutions, with its promise of lightening teaching obligations and, as a response to a culture increasingly driven by technology, its ability to fulfill the needs and expectations of today's students.
       However, there is also a widespread reluctance on the part of academic staff to become involved in computer based teaching and learning strategies. Surveys suggest little awareness of what can be done, doubts over resources, and a lack of recognition accorded to courseware development as a research activity.
       Our work to date has addressed these issues. A widespread publicity programme has shed light for many on just what can be achieved with readily accessible technology, and a Hypertext Support Unit is available to undertake hypertext conversion of teaching material for staff without the resources to take it on themselves. The project is coordinated and managed at University level, with staff involvement actively encouraged and recognised by the Staff Development Office.
       While most UK research is focussed on developing discipline based courseware, we are taking hypertext to individuals. After one year, we have engaged over fifty academic staff and postgraduate students in hypertext related activities. We have started a shift in the university's cultural infrastructure, and while progress will undoubtedly be slow, we believe it is vital to the long term future of both the university and to higher education in the United Kingdom as a whole.
    Keywords: Authoring, Education
    Tale of a Text: Hypertext and the Eighteenth-Century Novel BIBAK 24
      Simon A. Rakov
    Hypertext fiction has an historical precedent in eighteenth-century printed literature. The early novels of Swift, Johnson, Richardson and Sterne vary widely in form, and self-consciously combine technology with narrative, much like the hypertext fictions of Michael Joyce, Carolyn Guyer, or Stuart Moulthrop. This year as the Maguire and Kirshner Fellow at the University of Oxford, I combined textual criticism, history of technology, literary criticism, images, and fiction to create "Tale of a Text," a hypertextual web that compares hypertext and eighteenth-century writing. The web follows three lines of inquiry: the physical production of narratives in print and hypertext, the relationships between authors and their respective technologies, and the tension between the demands of technology and the needs of a narrator. The poster will display extracts from "Tale of a Text," and propose the eighteenth-century novel as a useful lens for examining the progress of hypertext narrative. In their original editions many pioneering eighteenth-century novels helped to establish conventions of print typography, layout and linearity; today these works give historical perspective to efforts toward common writing practice in hypertext.
    Keywords: Authoring, Education, Graphic design, Literature, Metaphor/mental-models, Navigation, Rhetoric
    The HCI Bibliography -- It's Got Lots of HyperRefs, Too BIBAKWeb Page 25
      Gary Perlman
    The HCI Bibliography is a free-access online bibliographic database on Human-Computer Interaction. The basic goal of the project is to put an electronic bibliography for most of HCI on the screens of all researchers, developers, educators and students in the field through anonymous ftp access, mail servers, and Macintosh and DOS floppy disks. As of July 1993, through the efforts of over 100 volunteers from 13 nations, the bibliography has grown to a resource of over 8200 entries occupying over 9 megabytes, used by over 300 sites in 23 nations.
       Hypertext and hypermedia references make up a large subset of the bibliography; at least 600 entries and perhaps 10% of the database is relevant to hypertext. The bibliography is organized into modules of journal volumes, conference proceedings, and odds and ends. The journals abstracted in the bibliography include: Hypermedia and Transactions on (Office) Information Systems. The ACM Hypertext Conference and ECHT European Hypertext Conferences as are the ACM conferences on Systems Documentation (SIGDOC) and Information Retrieval (SIGIR), and the UK Hypertext conferences and the NIST Hypertext standardization workshop.
    Keywords: Education, Information retrieval, Link models
    Tailorable Interaction in a Language Independent Hyperstructure Programming Environment BIBAK 25
      Kasper Østerbye; Kurt Normark
    The aim of the Aalborg Hyperstructure Programming Environment project is to explore generic programming environments based on hypertext. The name of the environment, which we develop in the project, is HyperPro. The internal representation of programs as well as documentation is a network formed by nodes and links. The external appearance of the network is governed by an interaction engine, which is controlled by relatively simple interaction rules. The rules are formulated in a dedicated language and organized in a hierarchy of interaction schemes. Both presentation aspects and event responses are formulated in this language. We believe our approach is interesting in its attempt to provide tailorable interaction facilities of an underlying hyperstructure, thus emphasizing the separation between interface and data model.
       Using the rule language, we have created a default behavior of HyperPro, where a single node is presented in a single window and where links are presented using the name of their destination node. We have created specializations to support programming in the languages Eiffel and Smalltalk, and documentation in LaTeX. In some of the resulting environments, we are able to present and work on several nodes and links in a single window. For all of our example languages, we have created both an authoring mode to be used for interactive work, and compilation presentations, which linearize the appropriate parts of the hyperstructure network, so that information can be handed to a compiler or a document formatter.
    Keywords: Navigation, Open systems, Software engineering, Text conversion, Hypertext engineering
    Visualizing Human Performance Data -- A Hypermedia Approach BIBAK 26
      Don Monk
    The Computer Aided Systems Human Engineering: Performance Visualization System (CASHE:PVS) is an interactive, hypermedia system for human system designers, educators, and researchers. CASHE:PVS contains two complete and integrated human factors reference sources and two specialized visualization tools: the Perception & Performance Prototyper (P3) and DataViewer.
       The reference sources in the CASHE:PVS information base are the Engineering Data Compendium (EDC) and the Human Engineering Design Criteria for Military Systems, Equipment and Facilities (MIL-STD-1472D). Related materials are both inter- and intra-linked. Users may place notes and create their own fine-grained links between information found in text, figures, or tables. Directed searches of the information base as well as browsing the documents are supported. Full copyright information and protection is provided for the individual entry components in the reference documents.
       P3 is a unique, interactive simulation capability which allows the user to experience the technical data. Variables, from multiple entries of the reference documents, are consolidated onto direct manipulation "test benches." Using one of these behavioral phenomena test benches, users may interact with the variables which underlie a phenomena under investigation. For example, if the issue is the legibility of text under vibration, the user may either select a pre-defined spectrum or custom define a representative noise spectrum that will vibrate the subject text. By actually experiencing the phenomena first-hand, coupled with traditional means of reading a textual description and studying data graphs, the designer will be able to achieve a deeper understanding of the phenomena.
       DataViewer provides the user with a third party data analysis package so that data from multiple references may be scaled, transformed, replotted, and compared to gain further insight from the data.
    Keywords: Education, Empirical/Data, Graphic design, Hypertext engineering, Text-Conversion, Multimedia
    PECON: A System to Carry Out 'Search' in Distributed Hypertext Systems BIBAK 26
      Dieter Meiser; H. Scheidig
    The necessity to extend hypertext systems by "search functions" is well known. At the moment, more and more "distributed hypertext systems" are being developed and partly used in WANs. Distributed hypertext systems offer the possibility of building up large databases. Search functions in such large systems demand an efficient implementation. It has to be examined in how far the techniques, which have been developed in the field of information retrieval, can be used in hypertext systems.
       In our approach, search is implemented as spreading activation. In large hypertext systems, one search process can lead to a lot of hits. In the access to the results, the user should be supported by the system. In PECON, the result of the search is firstly described in the form of paths. Then Petri Nets overly this structure. The access to the elements found is ensured by running the Petri Net.
    Keywords: Information retrieval, Navigation
    Tailoring Hypertext Documents in Context: First User Testing Results BIBAK 27
      Nathalie Mathe
    The Computer Integrated Documentation (CID) system enables integration of various technical documents in a hypertext framework. It includes an intelligent context-sensitive indexing and retrieval mechanism, which utilizes on-line user relevance feedback and user profile information to refine its indexing links over time. This allows CID to provide helpful responses based on previous usage of the documentation. To evaluate CID learning capabilities, we conducted user testing experiments using a think-out aloud protocol. Two NASA domain experts used CID to answer queries in the Space Station Program Definition Requirements Document.
       Results showed that tailoring of hypertext documents during usage has been very well received by users. CID main advantage is to let users incorporate knowledge into a document over time, without disturbing the task they accomplish: CID only requires users to choose their user profile on log-on, then to click on Success or Failure button whenever they want to give feedback. Giving positive feedback not only marks a reference as successful, but also lets users retrieve references reinforced under the same or similar user profiles. This mechanism seems to be extremely useful for filtering information and building personalized views of large documents over time, as well as for sharing knowledge between users (access tailoring done by someone else). Users also requested more capabilities to make a better use of CID reinforcement mechanism: better visualization, filtering of, and access to previous reinforcements; being able to reinforce any hypertext link; and being able to save several reinforcements under one label.
    Keywords: Artificial intelligence, Empirical/data, Information retrieval, Link models, Navigation, Usability
    A Hypertext for Blind People BIBAK 27
      Mauricio F. Lumbreras
    Nowadays there are many works about powerful interfaces, which allow us appropriate access to hypertexts. The focus is generally placed on adequate graphic design, sometimes using video or computer animation as the media for carrying information.
       This choice does not take into account the gap which is produced between normal and blind users.
       The information media used by blind people, such as audio cassettes and Braille texts, are acceptable but they cause some difficulties, especially those related to the search of information.
       The advantages of hypertexts are widely known, but how can hypertexts be built considering the limitations that blind people have? The critical items to consider in their design are:
  • 1) Physical medium for information output.
  • 2) The conceptual model. A new technology allows us to create 3D sound with headphones, so that it is possible to create a world of objects that generate sound from different points in space. Additionally, each of these sounds has intuitive semantics. If the user can navigate in this environment and we provide a suitable metaphor to support a hypertextual model, the user will be able to access information in a hypertext more easily.
       Our work is focused on sound simulation inside a building. A mapping is done between one building and one hierarchical hypertext. By means of this metaphor, the user can navigate in a familiar environment, where each entity (corridor, elevator, room, etc.) has intuitive semantics.
    Keywords: 3D, Handicapped, Metaphor/mental-models, Multimedia, Navigation, Virtual reality
  • A Hypermedia Link Service as an Operating System Extension BIBAK 28
      A. J. Lewis
    Open hypermedia systems are becoming more popular amongst today's researchers and developers. A natural extension to the concept is a system that incorporates hypermedia support at an operating environment level for users of a system. The idea has been developed by such projects as the SUN Link Service, however, approaches like this require that applications that use that support be written specially with this aim in mind.
       Using Macrocosm, the Macintosh version of Microcosm currently under development at the University of Southampton, as a research platform, a different approach to this problem is being investigated. The intention is that this system will allow any application to access and use the available Macrocosm facilities, regardless of whether the applications were implemented with any knowledge of Macrocosm.
       Macrocosm consists of a set of communicating filters, at whose core resides the system-level filter manager. Key features of the Macrocosm system include a dynamic menu system, which acts as a universally available floating palette from which the currently available Macrocosm facilities are on offer. Macrocosm has been implemented with an extreme view to the openness of the system -- only the communications protocols are rigidly defined, while any hypermedia structures and linking mechanisms are entirely tailorable, and are attached to the system as modular additions.
    Keywords: Link models, Multimedia, Open systems, Text conversion, Hypertext engineering
    Icon and Diagram Use in Data Navigation: I Know I'm There, but Where is That BIBAK 28
      Geoffrey J. Kordell; Marc M. Sebrechts
    A popular instructional hypermedia system about conflict in the Middle East was used to test the hypothesis that diagrammatic information can improve hyperspace navigation. Three variants of the system provided alternative access to a fixed database. In the first "icon map" condition subjects used icons to access four high-level divisions of the database (history timeline, religion, geography, and bibliographic and documentary resources). The second "static schematic" condition augmented the icon map by displaying a static diagram of the database, including all major nodes and links. The final "active schematic" condition included an active version of the static schematic; subjects could access any node by selecting it directly from the diagram. Forty-two students spent approximately one hour using one of the three variants of the system to answer questions about the Middle East. Regardless of condition, the optimal strategy appeared to consist of an initial browsing phase followed by specific search. Subjects who failed to go through an overview of the materials often encountered the relevant data, but failed to include it in their summary report. Available diagrams were often ignored or misused. Content knowledge was far more important than knowledge about the structure of the database in setting expectations about information location. Navigating mental space may be very different from navigating a data space. This perspective gives some legitimacy to the seemingly paradoxical claim, "I know where I am but where is that."
    Keywords: Education, Empirical/data, Information retrieval, Link models, Metaphor/mental-models, Multimedia, Navigation
    AMHYRST: AutoMated HYpertext Reuse Search Tool BIBAK 29
      Tomas Isakowitz; Robert J. Kauffman
    The Problem: Searching for Reusable Software Objects: CASE tools that emphasize reuse are increasingly recognized by senior management for the value that they deliver in helping their firms to achieve higher levels of software development productivity and reduced software costs. Locating and retrieving software objects for reuse purposes, however, is so time consuming that developers tend to build from scratch instead of searching for objects to reuse. Thus, a cost-effective search mechanism is needed to support search for reuse.
       The Proposed Solution: Our design to support the search for reusable objects, AMHYRST, blends automated classification of software objects and hypertext. The software repository is automatically classified -- based on reuse criteria -- by an automated object reuse classifier. Software developers access the classified repository by formulating queries. Answers to these queries are automatically organized -- based on the classification schema -- into a hypertext system that facilitates their detailed inspection.
       Results: A key aspect of our approach lies in the specification of a classification schema that reflects the mental model that developers utilize when searching for reusable objects. This results in an intuitive, easy-to-use reuse search tool. We have conducted field studies that led us to construct such a classification schema. A prototype system that integrates automatic classification and hypertext navigation has been constructed. Initial experiences show that the tool enables higher levels of reuse.
    Keywords: Information retrieval, Metaphor/mental-models, Software engineering
    Palimpsest: A Data Model for Change Oriented Revision Control BIBAK 29
      David G. Durand
    Problem: The Palimpsest data model developed at Boston University is a data model for revision control that allows simultaneous editing in the presence of communication failures. The goal is to localize and reduce possible conflicts while allowing intelligent application resolution and examination of conflicts. It is not based on a particular implementation strategy but rather a set of assumptions about system environments for collaborative editing at a distance.
       Method: We have removed the traditional assumption that serializability is appropriate for human-controlled concurrent editing. Our solution is similar to some proposed in software engineering: the changes that create data are primary objects of the model. States of the data are derived from a generalized, non-serial set of changes (like an "editing history"). We assume that applications will need to be aware of collaborative editing, and that such awareness will be at the level of data representation, rather than user-interface or presentation.
       Results: The model meets our goals, and is very general (for instance traditional hypertext objects like nodes and links are derived objects not, primitives). We are now working on implementation strategies and test-applications. Palimpsest should serve well as an implementation model and a well-defined conceptual tool when application awareness and user-resolution of conflicting changes or variant versions are involved.
    Keywords: Authoring, CSCW, Software engineering, Versioning
    The Information City -- A Step towards Merging of Hypertext and Virtual Reality BIBAK 30
      Andreas Dieberger
    The Information-City is a three dimensional user interface for large hypertext collections. It uses the metaphor of a city to represent documents as buildings in a virtual environment the user navigates and makes heavy use of Jay Bolter's "writing on the world" concept to visualize as much information about the documents as possible without overloading the user (e.g. age, relevancy, accessibility). The city is restructured according to user interest profile and document similarity to create "districts of interest". The city metaphor shall help the user to build a conceptual map of the environment. As autonomous actor the system helps the user to locate information by providing light cues and transportation (taxi and subway metaphors). Significant information is pointed out by open doors for instance. Visualization of hypertext nets as house structures is done using hypertext abstractions, so called building blocks. These constructs define a hierarchical subset of the hypertext web that can be visualized different than the rest of the web -- namely as halls and rooms. A second, purely textual version of the information city is constructed in a MUD (Multi User Dungeon) system and will be available on the internet. This system is used to study collaborative navigational behaviour in a multi user environment (development of landmarks and navigation paths). Findings of this version will be used to improve the graphical version of the information city. At the time of writing an authoring tool using building blocks is under construction.
    Keywords: 3D, Metaphor/mental-models, Navigation, Virtual reality
    Authoring Intent-Based Video Documents BIBAK 30
      Andrew Csinger; Kellogg S. Booth
    Current technology restricts the "links" of the hypergraph representing an information space to what the author can foresee at "compile-time". Either the author supplies all of the content by linearizing the presentation, or the reader requires significant resources to infer missing content and often gets lost in hyperspace.
       The problems with traditional (hyper-)authoring approaches are particularly severe in the video domain, where the intrinsic, unstructured linearity of the medium calls for a dynamic, automatic approach that supports both the annotation and presentation of video records.
       In the presentation of video information, the need to generate user-tailored presentations (in the absence of the author) is particularly pressing; a knowledge-based approach permits automatic run-time edit-list preparation based upon explicit, consultable user models. A model of the reader is used to create a subgraph of the hypergraph; the complete graph is the information space created by the author(s), while the subgraph is the presentation space. The system gives the reader additional facilities to modify the subgraph and navigate through the presentation space at run-time.
       The emphasis in this work is the construction of an adequate model of the user; this model is used to fashion the right "links" in the hypergraph that ties related parts of the multimedium together.
       Using a minimal AI approach, we are implementing a video authoring prototype that partially overcomes the limitations of traditional approaches to authoring, with consequent savings in human effort in both the annotation and presentation phases of video authoring.
    Keywords: Artificial intelligence, Authoring, Graphic design, Information kiosks, Information retrieval, Metaphor/mental-models, Multimedia, Navigation, Open systems, Rhetoric, Video
    Measuring Hypertext Usability with Structure Recall BIBAK 31
      Edward Carter; Chris Mellish
    This poster describes the development of a quantitative measure of the usability of Hypertext documents.
       In the course of a larger project, a series of link-generation schemes were developed which linked a database of 'pre-formed' nodes into Hyperstructures. These various linking processes had to be evaluated, preferably with a measure independent of the domain of the document database and the expertise of a user.
       An experiment was devised in which subjects gained familiarity with a generated Hypertext by performing a series of trivial Query/Lookup tasks. Subjects were then required to reconstruct the Hypertext that they had been using, but without reference to the original texts. A correlation between the actual Hypertext and the recalled version provides a numerical score that, we argue, reflects the usability of the Hypertext within certain pragmatic limits. This 'usability score' appears to be significantly independent of both the domain of the test Hypertext and the general level of expertise of the user.
       Although the technique we describe is relatively limited in application, the apparent independence and reliability of the measure is a significant result and we suggest ways in which the recall measure may be extended for application to large databases or dynamic Hypertext systems.
    Keywords: Artificial intelligence, Information retrieval, Navigation, Text conversion, Hypertext engineering, Usability
    Biblical Information Systems BIBAK 31
      Theo J. D. Bothma
    Biblical information offers a very complex information environment. There are texts of varying lengths, e.g. full text documents of Hebrew and Greek manuscripts and translations of the Bible, dictionaries, grammars and exegetical commentaries, as well as the rich cultural-historical world of the Ancient Near East, supplemented by drawings, maps and photographs. The full spectrum of Biblical information is briefly outlined on the poster.
       Present systems dealing with Biblical information vary in sophistication from electronic concordances to integrated information systems. A brief overview of some existing systems is provided, indicating their strong points and limitations, especially regarding linguistic information.
       In any system the Biblical texts should be the main access point. Each of the other information entities, e.g. an online grammar, syntactic database, dictionary or history should be accessible separately, independent of the Biblical texts. This information should, however, be linked to the Biblical texts so that the user can obtain help on the language or interpretation of any specific verse. This can be accomplished by means of a hierarchical structure, with a rich network of cross references. A detailed flow diagram of the lay-out of a system is presented and the various technologies to be used in such a system are indicated.
       The poster therefore consists of two sections, viz a theoretical description, outlining the problems and possible solutions and a schematic layout of a comprehensive Biblical information system.
    Keywords: Education, Information retrieval
    Media Independent Hypermedia BIBAK 32
      Nick Beitner
    The use of explicit authoring, though effective and predictable, is not conducive to authoring large volume hypermedia with varied media types and ever increasing collections of data. The use of current implicit linking strategies goes only a small way towards solving this problem. A number of models exist that, through the use of a dictionaries containing a restricted set of keywords, allow descriptions of documents and, subsequently, implicit relationships to be derived. There are no links as such but the association of attributes with documents provide a link structure that allows wide access to similar documents. These links only apply where there is similarity between documents through association of shared attributes/keywords. While this model is useful it fails to provide us with any structure for describing paths through our data and there is no facility for relational structure with which we are familiar in hypermedia. The use of dictionaries as a means of allowing abstraction from the media allows potential for a media and representation independent authoring environment. With the addition of structure between the attributes in a dictionary a whole host of new relations can be expressed between documents. The expression of natural relations between attributes as well as arbitrary links at the attribute level allow minimum authoring to specify maximum connectivity throughout the data. This model provides separation between the authoring of links at an abstract level and the need for authors to handle data of multiple media types.
    Keywords: Authoring, Information retrieval, Multimedia, Navigation, Open systems
    A Study of the Relationship between Hypertext Authors and Readers BIBAK 32
      Lisa Baron; Edward Brown; Mark Chignell
    The process of authoring hypertext links is poorly understood. One remaining unresolved problem concerns how the needs of readers who have different task perspectives may be communicated to link authors.
       We carried out an experiment into how authoring and reading strategies change depending on the purpose (task) of the author and reader. In two conditions, authors and readers had either compatible or incompatible purposes with respect to a hypertext document. Five subjects participated as link authors while twenty-eight subjects participated as readers. Twenty screens of text on birds were used for the database content. Each author created two linked documents -- one multi-purpose and one special-purpose (anatomy of birds). Each reader used only one database (2 readers per database). Readers were given two sets of questions on bird anatomy and bird behavior. There was a nonsignificant trend for compatible hypertexts to produce faster and more accurate performance.
       Both semantic (similar, contrast, part-of, kind-of) and rhetorical (summary, explanation, continuation, illustration) link types were used in the hypertext documents created by authors. Three of the subjects used a preponderance of rhetorical link types while the remaining authors used a preponderance of semantic links. Overall more rhetorical links were authored (almost 60 percent of links). Despite this, readers tended to use the fewer available semantic links more frequently, so that the ratio of semantic to rhetorical links used in reading was closer to 50/50.
    Keywords: Authoring, Link models, Rhetoric, Task analysis
    User Interface Design Guidelines for a Hypertext Framework BIBAK 33
      V. Balasubramanian; Murray Turoff
    We believe that navigational tools alone are not sufficient to solve the problems of disorientation and cognitive overhead in hypertext systems. Hypertext systems tend to suffer from a lack of coherence due to ambiguity in meanings assigned to nodes and links. Although there have been earlier attempts to incorporate semantics into hypertext networks, a general framework for hypertext functionality wholly based on a cognitive model (Guilford's Structure of the Intellect Model) was first proposed recently. This framework classified nodes into six different semantic types and links into twelve different types. It was reported that all current hypertext systems fall under this framework, in their own limited ways, and that this semantic morphology could be extended to all future systems. Such a comprehensive framework would help designers develop better interface metaphors and implementation models for hypertext systems.
       A first step towards the implementation of a hypertext system based on this framework is to develop an appropriate user interface that would reduce the mismatch between the framework and the metaphor (functional opacity) and the mismatch between the metaphor and the implementation model (system opacity). Various guidelines and usability principles exist for the purpose of designing user interfaces for interactive systems. In this poster, we present our research efforts in applying a specific set of fourteen user interface design guidelines to the aforementioned hypertext framework. We have chosen these guidelines since they were found to be more systematic and comprehensive than other similar guidelines. We present some examples of a preliminary prototype. The next step is to implement a model incorporating this prototype and study the effectiveness of such an approach.
    Keywords: Graphic design, Link models, Metaphor/mental-models, Navigation, Usability
    Designing Hypermedia Applications BIBAK 33
      P. Balasubramanian; Tomas Isakowitz; Edward A. Stohr
    In this work we propose a methodology for designing and constructing organizational hypermedia applications. A good hypermedia system overcomes the inherent limitations of computer screens relative to paper. It also exploits the possibilities for rapid search and complex associations that can be provided in an electronic form. This is a time consuming process often requiring the linking of many chunks of information residing in vastly different media and connected in arbitrarily complex ways. A design methodology must focus on how these connections are represented, manipulated and stored. Finally, it is important to consider the dynamic interaction of the user with the hypertext (i.e. the "browsing semantics"). From our experiences in building hypermedia applications we concluded that:
  • a) Capturing the underlying information structure of the application domain is
        important.
  • b) A methodology that helps the developer through the design, screen-by-screen
        and link-by-link would be most helpful. The methodology that we propose is called Relationship Management Design methodology (RMD) utilizes tools from systems analysis and design together with entity-relationship (E-R) diagrams.
       The above methodology will facilitate the development process and help produce more complete and consistent designs of the application. We have developed the MBA handbook using this approach and this application is currently being used by the students. However, the development of an effective hypermedia design approach will require computerized tools that lead to computer-aided hypermedia design (CAHD). We wish to develop a RMD editor to assist users in drawing the RMD diagrams, an RMD compiler to apply the various rules and build the application and a screen design tool to customize the screens for each entity.
    Keywords: Authoring, Information retrieval, Multimedia, Navigation
  • The Functional Model of the Link BIBAK 34
      Helen Ashman; Janet Verbyla
    The Functional Model of the link has been developed to augment retrieval functionality and link creation in large, multiuser information collections with continually evolving underlying data. It extends the principle of externalisation which has been initiated by the open (hypermedia) systems. Open systems store their link computation parameters externally; the Functional Model system also stores the link computation instructions externally. This work also maximises the dynamism of links, permitting the dynamic computation of the source set of the link as well as the destination set. This allows automatic linking of new and modified information.
       The Functional Model represents every link specification as a pair of rules: the first describes the conditions an object must satisfy to be a source for the link, the second describes how to find the destination(s) from the selected source. Every link type can be represented in this way.
       A Functional Model system can support all the existing facilities available in open hypermedia systems. It has the added advantages of easy upgrading and exchange of link specifications, making it possible to customise the retrieval mechanisms of a retrieval system to suit site and individual user requirements, and to make use of new technologies as they become available. Furthermore, it has potentially unlimited functionality since the computation specifications can be in any language and can be sent to any specialised application for processing. Finally, existing retrieval systems can import all this functionality merely by establishing a communication channel with the hypermedia system.
    Keywords: Authoring, Information retrieval, Link models, Navigation, Open systems
    A CAI System for Linear Algebra Based on 'World Wide Web', Xmosaic and a Few Special Purpose Programs BIBAK 34
      S. Arnesen; H. Bjar; T. Gulliksen
    Our objective was to see if it was possible to create a CAI-system that actually helps students while using open standards and if possible free software making it possible to give a large number of students to the system at a relatively low cost.
       The course selected for study is the entry-level linear algebra course. This course is perceived as a difficult one where several rather abstract concepts are being introduced. The basic standards selected was the SGML based HTML markup language for hypertext documents, the Xmosaic browser for HTML-based World Wide Web documents developed at the National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign. These are all freely available tools that allow applications based on them to be run on multiple platforms. To facilitate the use of mathematical notation and markup based on pedagogical concepts we had to extend the basic tools. For this we extended HTML to allow us to include formulas from TeX and plots and animations from Mathematica. Perhaps more interesting is the special markup introduced for pedagogical elements such as explanation of concepts, revealing of hints, and gradual revealing a document. The initial courseware is structured as reviews of previous exams.
    Keywords: Education, Navigation