HCI Bibliography Home | HCI Conferences | ECHT Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
ECHT Tables of Contents: 90929494X

Proceedings of the ECHT'94 European Conference on Hypermedia Technologies

Fullname:Proceedings of ECHT'94 the ACM European Conference on Hypermedia TechnologyECHT94XACM ECHT'94 -- Demonstrations, Posters, Videos
Location:Edinburgh, UK
Dates:1994-Sep-18 to 1994-Sep-23
Standard No:ACM ISBN 0-89791-640-9 ACM Order Number 614940; ACM DL: Table of Contents hcibib: ECHT94 ; hcibib: ECHT94X
Papers:30; 36
Pages:280; NA
  1. Papers
  2. Technical Briefings
  3. Panels
  4. Demonstrations
  5. Demonstrations/Posters
  6. Posters
  7. Videos


Coexistence and Transformation of Informal and Formal Structures: Requirements for More Flexible Hypermedia Systems BIBAKPDF 1-12
  Jorg M. Haake; Christine M. Neuwirth; Norbert A. Streitz
In this paper, we argue that some tasks (e.g., meeting support) require more flexible hypermedia systems and we describe a prototype hypermedia system, DOLPHIN, that implements more flexibility. As part of the argument, we present a theoretical design space for information structuring systems and locate existing hypertext systems within it. The dimensions of the space highlight a system's internal representation of structure and the user's actions in creating structure. Second, we describe an empirically derived range of activities connected to conducting group meetings, including the pre- and post-preparation phases, and argue that hypertext systems need to be more flexible in order to support this range of activities. Finally, we describe a hypermedia prototype, DOLPHIN, which implements this kind of flexible support for meetings. DOLPHIN supports different degrees of formality (e.g., handwriting and sketches as well as typed nodes and links are supported), coexistence of different structures (e.g., handwriting and sketches as well as typed nodes and links are supported), coexistence of different structures (e.g., handwriting and nodes can exist on the same page) and mutual transformations between them (e.g., handwriting can be turned into nodes and vice versa).
Keywords: Hypermedia interface, Information structures, Meeting support, Design space for hypermedia systems, Flexibility
VIKI: Spatial Hypertext Supporting Emergent Structure BIBAKPDF 13-23
  Catherine C. Marshall; Frank M., III Shipman; James H. Coombs
The emergent nature of structure is a crucial, but often ignored, constraint on authoring hypertexts. VIKI is a spatial hypertext system that supports the emergent qualities of structure and the abstractions that guide its creation. We have found that a visual/spatial metaphor for hypertext allows people to express the nuances of structure, especially ambiguous, partial, or emerging structure, more easily. VIKI supports interpretation of a collected body of materials, a task that becomes increasingly important with the availability of on-line information sources. The tool's data model includes semi-structured objects, collections that provide the basis for spatial navigation, and object composites, all of which may evolve into types. A spatial parser supports this evolution and enhances user interaction with changing, visually apparent organizations.
Keywords: Spatial hypertext, Emergent structure, Interpretation, Visual structure recognition, Composites
Fixed or Fluid? Document Stability and New Media BIBAKPDF 24-31
  David M. Levy
One of the crucial properties of documents through the ages has been their fixity. The ability to mark surfaces in relatively stable ways has made it possible for people distributed across space and time to see the same images and thereby to have access to the same meanings or communicative intent. Today, however, with the increasing use of digital technologies, it is often asserted that we are moving from the fixed world of paper documents to the fluid world of digital documents. In this paper I challenge this assertion, arguing instead that all documents, regardless of medium, are fixed and fluid. Thus, although paper documents do fix aspects of communication, they do (and must) also change; and although digital documents are easily changeable, they must also be capable of remaining fixed. I make use of this analysis in two ways: first, to examine the fixity and fluidity of hypertext; and second, to critique Bolter's argument in Writing Space concerning the movement from "fixed to fluid."
Keywords: Hypertext, Documents, Fixity, Fluidity
Extending the Microcosm Model to a Distributed Environment BIBAKPDF 32-40
  Gary Hill; Wendy Hall
In recent years, there has been significant growth in the use of computer networks to support electronic delivery of information. As the volume of available information has grown, a need for powerful tools that can manage access has arisen. It has been suggested that hypertext techniques can provide such a facility.
   The Microcosm system is a hypertext link service developed at the University of Southampton. The system is based upon a modular architecture which allows the functionality of the system to be easily and dynamically extended. This paper describes the development of a distributed version of Microcosm based upon this modular design.
   The distributed system described utilises the fine granularity of the Microcosm model to support a wide range of possible configurations. The system also extends the document management facilities of Microcosm to allow information stored by other information services to be incorporated. The result is a system that can apply Microcosm's open linking services to a wide range of networked information.
Keywords: Open, Distributed, Hypertext, Microcosm
Light Hypermedia Link Services: A Study of Third Party Application Integration BIBAKPDF 41-50
  Hugh C. Davis; Simon Knight; Wendy Hall
Recently there has been a tendency for the research community to move away from closed hypermedia systems, towards open hypermedia link services which allow third parties to produce applications so that they are hypertext-enabled. This paper explores the frontiers of this trend by examining the minimum responsibility of an application to co-operate with the underlying link service, and, in the limiting case where the application has not been enabled in any way, it explores the properties and qualities of hypermedia systems that can be produced. A tool, the Universal Viewer, which allows the Microcosm Hypermedia System to co-operate with applications which have not been enabled is introduced and a case study is presented which demonstrates the functionality that may be achieved using entirely third party applications, most of which have not been enabled.
Keywords: Open hypermedia, Hypermedia link services, Integration, Microcosm
Adding Networking to Hypertext: Can it be Done Transparently? BIBAKPDF 51-58
  Peter Brown
Networks are becoming increasingly available and hypertext systems with networking capabilities are currently enjoying exponential growth. The vast majority of hypertext systems were not, however, designed to cater for networking. This paper examines whether it is possible to add networking to such systems and, if so, whether it can be done without upsetting existing hyperdocuments, existing authors and existing readers. The examination is done using one specific hypertext system, UNIX Guide, but the lessons are, I hope, more general.
Keywords: Wide-area network, Storage, Link, File, Active document, Distributed hyperdocument, GUIDE
Composites in a Dexter-Based Hypermedia Framework BIBAKPDF 59-69
  Kaj Grønbæk
This paper discusses the design and use of a generic composite mechanism in the object oriented DEVISE Hypermedia (DHM) development framework. The DHM framework is based on the Dexter Hypertext Reference Model, which introduces a notion of composite to model editors with complex or multiple types of contents. The original Dexter notion of composites is, however, insufficient to cover structural composites including or referencing other components. Thus the DHM framework has been extended with generic composite classes suited to support structures within the hypermedia network itself. The paper presents and discusses the design of the generic composite classes belonging to the STORAGE and RUNTIME layers of the framework. A central aspect of the design is that the structuring mechanism is a true composite with a collection of components as its contents rather than an atomic component with links to other components as in the classical systems such as NoteCards, Intermedia, and KMS. It is also shown how the powerful generic classes can be used to implement a variety of useful hypermedia concepts such as: hierarchy by inclusion, hierarchy by reference, virtual and computed browsers, TableTops and GuidedTours.
Keywords: Composites, Structure, Hierarchies, GuidedTour, Dexter model, Object oriented framework
Adding Multimedia Collections to the Dexter Model BIBAKPDF 70-80
  Franca Garzotto; Luca Mainetti; Paolo Paolini
The Dexter Model defines the notion of atomic components and composite components, but it does not prescribe, nor it suggests, any particular structure for composite components. This paper proposes a specific type of composite component, called "collection".
   A collection is a container holding several members. Collections can contain other collections (nested collections). Collections can be regarded as sets, but they can also have an inner structure. Collections can be created in several ways: manually, through queries, by operations on other collections, by exploiting links, etc.
   Collections introduce a navigational pattern, based on their structure, that is different from the standard node&link navigation.
   If active media are considered, collections allow the design and implementation of complex synchronisation strategies, difficult to obtain otherwise.
   The paper describes the motivations for using collections, their structure, their navigational capabilities and a number of possible authoring mechanisms. It also examines the interplay between standard navigation and collection navigation, possible synchronization strategies for collections, as well as the requirements for the definition of a runtime support (which could be used to extend the runtime layer of the Dexter Model).
Keywords: Dexter model, Composite, Hypermedia design, Collection, Guided tour, Active media
Under CoVer: The Implementation of a Contextual Version Server for Hypertext Applications BIBAKPDF 81-93
  Anja Haake
At GMD-IPSI we are developing CoVer, a contextual version server for hypertext applications [11]. Another characterization of CoVer is that CoVer integrates state-oriented versioning concepts with task-oriented versioning concepts. While other version models in general support only one of these groups of concepts, we argue that the explicit composition of versions of complex hypertext networks has to be complemented by automatic version creation in the context of tasks or jobs performed while manipulating the hypertext network and vice versa. Regarding the implementation of version models, it turns out that the state-oriented implementation approach -- representing every legal state of a hyperdocument explicitly -- and the task-oriented implementation approach -- computing versions of complex hypertext networks due to changes executed during a task or job -- are interchangeable. While the separation of state- and task-oriented concepts at the conceptual level of the version model is desireable to support version creation and selection for different hypertext applications, the implementation of such a dual model can be based on a single implementation approach. This paper discusses both types of implementation with an emphasis to cope with alternative versions that are in particular meaningful for hypertext publishing applications.
Keywords: Versioning, Alternatives, State-oriented versioning, Task-oriented versioning, Implementation techniques, Publishing applications
Chimera: Hypertext for Heterogeneous Software Environments BIBAKPDF 94-107
  Kenneth M. Anderson; Richard N. Taylor; E. James, Jr. Whitehead
Emerging software development environments are characterized by heterogeneity: they are composed of diverse object stores, user interfaces, and tools. This paper presents an approach for providing hypertext services in this heterogeneous setting. Central notions of the approach include the following. Anchors are established with respect to interactive views of objects, rather than the objects themselves. Composable, n-ary links can be established between anchors on different views of objects stored in distinct object bases. Viewers (and objects) may be implemented in different programming languages afforded by a client-server architecture. Multiple, concurrently active viewers enable multimedia hypertext services. The paper describes the approach and presents an architecture which supports it. Experience with the Chimera prototype and its relationship to other systems is described.
Keywords: Multimedia information systems, Software engineering, Tools and techniques, Document preparation, Hypertext/hypermedia, Design, Experimentation, Heterogeneous hypertext, Hypertext system architectures, Link servers, Separation of concerns, Software development environments
SIROG -- A Responsive Hypertext Manual BIBAKPDF 108-116
  Lothar Simon; Jochen Erdmann
Power plant operation and control in modern screen-based control rooms takes place using computer displays which are directly coupled to the plant state. However, operators are provided with operational instructions and background information by means of paper manuals or at best hypertext manuals with fixed structure and contents. Thus, information presentation is independent of the current situation.
   To improve information accessibility we developed a situation-dependent information medium: responsive manuals. A responsive manual consists of a "standard" hypertext-based operational manual and a task description. It monitors the changing situation and based on this is able to point to relevant information.
   To show the advantages of the responsive manual approach in the domain of power plant operation we implemented the SIROG (situation-related operational guidance) system in close cooperation with Siemens. It covers all parts of an operational manual for accidents in a Siemens nuclear power plant, and is coupled directly to the plant state. The article discusses the basics of the responsive manuals approach and the role of "responsiveness" in SIROG.
Keywords: Hypertext manual, Situation-dependence, Task model, Process monitoring, Process control
Repertory Hypergrids: An Application to Clinical Practice Guidelines BIBAKPDF 117-125
  David Madigan; C. Richard Chapman; Jonathan Gavrin; Ole Villumsen; John Boose
Creation and maintenance of links in large hypermedia documents is difficult. Motivated by an application to a federal clinical practice guideline for cancer pain management, we have developed and evaluated a repertory grid-based linking scheme we call repertory hypergrids. Harnessing established knowledge acquisition techniques, the repertory hypergrid assigns each "knowledge chunk" a location in "context space". A chunk links to another chunk if they are both close in context space.
   To evaluate the scheme, we conducted a protocol analysis. Six users of the guideline addressing typical cancer pain management tasks made 30 explicit links. The repertory hypergrid using a neighborhood size of 16 captures 24 of these links. With optimization, the repertory hypergrid captures 27 of the links with a neighborhood size of 13.
Keywords: Implicit linking, Repertory grid, Clinical practice guidelines, Link maintenance, Evaluation
Accessing Hyperdocuments through Interactive Dynamic Maps BIBAKPDF 126-135
  Mountaz Zizi; Michel Beaudouin-Lafon
We propose a new navigation paradigm based on a spatial metaphor to help users access and navigate within large sets of documents. This metaphor is implemented by a computer artifact called an Interactive Dynamic Map (IDM). An IDM plays a role similar to the role of a real map with respect to physical space. Two types of IDMs are computed from the documents: Topic IDMs represent the semantic contents of a set of documents while Document IDMs visualize a subset of documents such as those resulting from a query. IDMs can be used for navigating, browsing, and querying. They can be made active, they can be customized and they can be shared among users. The article presents the SHADOCS document retrieval system and describes the role, use and generation of IDMs in SHADOCS.
Keywords: Navigation, Maps, Information retrieval, Visualization, Interaction paradigm
Interactive Clustering for Navigating in Hypermedia Systems BIBAKPDF 136-145
  Sougata Mukherjea; James D. Foley; Scott E. Hudson
This paper talks about clustering related nodes of an overview diagram to reduce its complexity and size. This is because although overview diagrams are useful for helping the user to navigate in a hypermedia system, for any real-world system these become too complicated and large to be really useful. Both structure-based and content-based clustering are used. Since the nodes can be related to each other in different ways, depending on the situation different clustered views will be useful. Hence, it should be possible to interactively specify the clustering conditions and examine the resulting views. We present efficient clustering algorithms which can cluster the information space in real-time. We talk about the Navigational View Builder, a tool that allows the interactive development of overview diagrams. Finally, we propose a 3-dimensional approach for visualizing these abstracted views.
Keywords: Navigation, Overview diagrams, Clustering, Information visualization
Frame-Axis Model for Automatic Information Organizing and Spatial Navigation BIBAKPDF 146-157
  Yoshihiro Masuda; Yasuhiro Ishitobi; Manabu Ueda
In taxonomic reasoning tasks, such as scientific research or decision making, people gain insight and find new ideas through analysis of large numbers of factual data or material documents, which are generally disorganized and unstructured. Hypermedia technology provides effective means of organizing and browsing information with such nature. However, for large amounts of information, the conventional node-link model makes linking or browsing operations be complicated because their relationship have to be represented as binary relations. In this paper, we propose a hypermedia data model call Frame-Axis Model, which represents relationship between information as N-ary relations on mapped space. Also, the automatic information organizing mechanism which is based on this data model and the browsing interface HyperCharts which employ spatial layout are provided. Finally, we show some browsing examples on our working prototype system, CastingNet.
Keywords: Data model, Automatic linking, Spatial hypertext, Browsing, Visualization
Backtracking in a Multiple-Window Hypertext Environment BIBAKPDF 158-166
  Michael Bieber; Jiangling Wan
Multi-window interfaces allow users to work on logically independent tasks simultaneously in different sets of windows and to move among these logical tasks at will (e.g., through selecting a window in a different task). Hypertext backtracking should be able to treat each logical task separately. Combining all traversals in a single chronological history log would violate the user's mental model and cause disorientation. In this paper we introduce task-based backtracking, a technique for backtracking within the various logical tasks a user may be working on at any given time. We present a preliminary algorithm for its implementation. We also discuss several ramifications of multi-window backtracking including the types of events history logs must record, deleting nodes from history logs that appear in multiple logical tasks, and in general the choices hypermedia designers face in multi-window environments.
Keywords: Hypertext, Hypermedia, Backtracking, Multiple window, History log, Session log, Multiple pane
An Interaction Engine for Rich Hypertexts BIBAKPDF 167-176
  Kasper Østerbye; Kurt Normark
In semantically rich hypertexts it is attractive to enable presentation of a network of nodes and links at different levels of abstraction. It is also important that the user can interact with the hypertext using a command repertoire that reflects the chosen abstraction level. Based on a characterization of rich hypertext we introduce the concept of an interaction engine that governs the separation between internal hypertext representation and external screen presentation. This separation is the key principle of the HyperPro system. The HyperPro interaction engine is based on simple rules for presentation, interpretation of events, and menu set up. Much of the power of the interaction engine framework comes from the organization of these rules relative to the type hierarchy of nodes and links, and relative to a hierarchy of so-called interaction schemes. The primary application domain discussed in the paper is program development and program documentation.
Keywords: Tailorability, Interaction engine, Aggregated views, Event control, Program development
The Hypermedia Authoring Research Toolkit (HART) BIBAKPDF 177-185
  John Robertson; Erik Merkus; Athula Ginige
A major obstacle hindering the advancement and commercial acceptance of hypermedia is the cost of converting paper based information into hypermedia form. The Hypermedia Authoring Research Toolkit (HART) was developed to support the human editor during this media-to-hypermedia conversion process. The tool's goal is to help improve the correctness and completeness of the hypermedia database, as well as reduce the media-to-hypermedia conversion cost.
   We believe it is not possible to properly convert media to hypermedia without the participation of a human editor during the transformation. It is therefore necessary to develop tools to assist the human during this process. By reducing the overhead associated with the physical management of the hyper-database construction, the subject specialist is better able to concentrate on the information content.
   Support is provided in two basic ways:
  • By providing procedural guidance. From our experience constructing
       hypermedia systems we have developed an efficient process for this
       media-to-hypermedia transformation.
  • By providing intelligent assistance. At each phase in the transformation the
       system can suggest likely nodes, key phrases, index values, anchors, and
       links to the editor. The project's research focus is to identify the most effective methodologies to assist the human editor transform linear text, images and video into hypermedia structure.
    Keywords: Hypertext, Hypermedia, Media-to-hypermedia authoring
  • Querying Structured Documents with Hypertext Links using OODBMS BIBAKPDF 186-197
      V. Christophides; A. Rizk
    Hierarchical logical structure and hypertext links are complementary and can be combined to build more powerful document management systems [28,25,24,13]. Previous work exploits this complementarity for building better document processors, browsers and editing tools, but not for building sophisticated querying mechanisms. Querying in hypertext has been a requirement since [19] and has already been elaborated in many hypertext systems [11,7,4,21], but has not yet been used for hypertext systems superimposed on an underlying hierarchical logical structure.
       In this paper we use the model and the SQL-like query language of [10] in order to manage structured documents with hypertext links. The model represents a structured document with typed links as a complex object, and uses paths through the document structure, as first class citizens in formulating queries. Several examples of queries illustrate, from a practical point of view, the expressive power of the language to retrieve documents, even without exact knowledge of their structure in a simple and homogeneous fashion. It must be stressed that the proposed model and language implement the equivalent HyTime [1] Location Address Module. In fact, the language is more powerful than the corresponding HyQ query facilities. The implementation and the description throughout the paper use the SGML standard [2] to represent the document structure and the object-oriented DBMS O2 [12] to implement the query language and the storage module.
    Keywords: Structured documents, Hypertexts, Object oriented databases, Information retrieval, Query languages, Path expressions
    Querying Typed Hypertexts in Multicard/O2 BIBAKPDF 198-205
      Bernd Amann; Michel Scholl; Antoine Rizk
    Due to the growing complexity of modern hypertext applications, current hypertext systems require new mechanisms to support authoring and user navigation through large sets of documents connected by links. A general solution is to extend hypertext systems to cater for semantics of application domains. This requires new hypertext models providing strongly typed documents and links. Such models have been proposed and put to use in systems such as HDM and MacWeb to facilitate authoring of large hypertexts. In addition, Gram and MORE use typing and graph-based hypertext schemas for querying hyperdocuments. In this paper, we will show how query languages could be further exploited for designing sophisticated general query-based navigation mechanisms. We illustrate our examples using the Gram model and describe an implementation with the hypermedia system Multicard connected to the object-oriented database management system O2.
    Keywords: Hypertext querying, Browsing, Hypertext schema, Visual interface
    Where No Mind Has Gone Before: Ontological Design for Virtual Spaces BIBAKPDF 206-216
      Nancy Kaplan; Stuart Moulthrop
    Hypermedia designers have tried to move beyond the directed graph concept, which defines hypermedia structures as aggregations of nodes and links. A substantial body of work attempts to describe hypertexts in terms of extended or global spaces. According to this approach, nodes and links acquire meaning in relation to the space in which they are deployed. Some theory of space thus becomes essential for any advance in hypermedia design; but the type of space implied by electronic information systems, from hyperdocuments to "consensual hallucinations," requires careful analysis. Familiar metaphors drawn from physics, architecture, and everyday experience have only limited descriptive or explanatory value for this type of space. As theorists of virtual reality point out, new information systems demand an internal rather than an external perspective. This shift demands a more sophisticated approach to hypermedia space, one that accounts both for stable design properties (architectonic space) and for unforeseen outcomes, or what Winograd and Flores call "breakdowns." Following Wexelblat in cyberspace theory and Dillon, McKnight, and Richardson in hypermedia theory, we call the domain of these outcomes semantic space. In two thought experiments, or brief exercises in interface design, we attempt to reconcile these divergent notions of space within the conceptual system of hypermedia.
    Keywords: Spatial hypertext, Interface design, Information mapping, Navigation
    Aesthetic and Rhetorical Aspects of Linking Video in Hypermedia BIBAKPDF 217-223
      Gunnar Liestol
    This paper reports on the development of a hypermedia environment for public access in a museum. It discusses problems encountered when making video interactive and multilinear and when linking video and text in the creation of the system. Through the exchange of properties between print and video, media approaches to linking and continuity are presented. Visual examples are used to illustrate this and related to the need to further develop aesthetic and rhetorical aspects of linking video in hypermedia.
    Keywords: Aesthetic, Rhetoric, Digital video, Media integration

    Technical Briefings

    Music in Time-Based Hypermedia BIBAKPDF 224-227
      Jacco van Ossenbruggen; Anton Eliens
    The paper describes the extension of a hypermedia class library with music as a new component type, but will focus on the development of a software wrapper object as an application programmers interface to the Csound software sound synthesis program. This wrapper provides the flexible, interactive and object oriented interface needed by a hypermedia system. Additionally, some consequences of the fundamental difference between static and time-based media will be discussed.
    Keywords: Time-based hypermedia, Object oriented programming, Software sound synthesis
    Experience with the Use of Acrobat in the CAJUN Publishing Project BIBAPDF 228-232
      David F. Brailsford
    Adobe's Acrobat software, released in June 1993, is based around a new Portable Document Format (PDF) which offers the possibility of being able to view and exchange electronic documents, independent of the originating software, across a wide variety of supported hardware platforms (PC, Macintosh, Sun UNIX etc.).
       The fact that Acrobat's imageable objects are rendered with full use of Level 2 PostScript means that the most demanding requirements can be met in terms of high-quality typography and device-independent colour. These qualities will be very desirable components in future multimedia and hypermedia systems. The current capabilities of Acrobat and PDF are described; in particular the presence of hypertext links, bookmarks, and 'yellow sticker' annotations (in release 1.0) together with article threads and multimedia 'plug-ins' in version 2.0.
       This article also describes the CAJUN project (CD-ROM Acrobat Journals Using Networks) which has been investigating the automated placement of PDF hypertextual features from various front-end text processing systems. CAJUN has also been experimenting with the dissemination of PDF over e-mail, via World Wide Web and on CD-ROM.
    An Editor's Workbench for an Art History Reference Work BIBAKPDF 233-238
      Lothar Rostek; Wiebke Mohr
    The architecture and some of the realized functionality of a prototype Editor's Workbench that supports the creation and maintenance of an object-oriented publisher's knowledge base is presented. The knowledge base is the repository not only for the actual publication content, but for all the information needed to manage and control the publication process. The concrete application context is an art history reference work. We discuss content acquisition and data modelling aspects of the underlying object network.
    Keywords: Editor's workbench, Hypermedia reference work, Knowledge-based system, Object-oriented data modelling
    Representation and Manipulation of Conceptual, Temporal and Geographical Knowledge in a Museum Hypermedia System BIBAKPDF 239-244
      Carl Taylor; Douglas Tudhope; Paul Beynon-Davies
    This paper discusses a semantic database approach to museum hypermedia systems based upon binary relations, with a restricted set of abstraction relationships. We describe examples of schema, queries and navigation aids for a prototype system designed as a social history museum exhibit, with around one hundred historical photographs. Media items are classified according to conceptual, temporal and geographical schema which attempt to model the changing nature of geography over time. The application yields a sparse information space with densely populated clusters. Implementations of notions of semantic closeness, term generalisation, best fit solutions, media density and media similarity show potential to assist the exploration of such information spaces.
    Keywords: Knowledge based hypermedia, Museum information systems, Semantic modelling, Temporal & geographical knowledge, Navigational aids, Semantic closeness


    HTML -- Poison or Panacea? BIBAPDF 245-246
      Robert Glushko; Dale Dougherty; Eliot Kimber; Antoine Rizk; Daniel Russell; Kent Summers
    Many people are having their first experience with a distributed hypertext system by using Mosaic or some other viewer based on HTML, the HyperText Markup Language of the World Wide Web. HTML's simplicity allows it to be created without special authoring tools or expertise, and the ubiquity of free WWW viewers like Mosaic removes one of the cost barriers. Because HTML is an application of SGML, the Standard Generalized Markup Language, it has also introduced many people to the concepts and syntax of application-independent markup. The explosive growth of the WWW makes it undeniable that HTML and Mosaic will serve as the reference point for much future thinking about hypertext and SGML outside of the academic and research community.
       But if HTML is to some people a democratizing force for hypertext authoring and publishing, to others its lack of structure and validation is a substantial step backwards for authors, readers, and their organizations. HTML has proven the basic premise of SGML, that a standard, application-independent data representation can enable blind interchange among disparate and even unknown or unpredicted applications. But to many SGML advocates, HTML is too primitive, and today's HTML documents will end up as tomorrow's cast-off legacy data format. HTML viewers may seem appealing for network publishing, but to some they are merely seductive and superficial, undermining years of careful research on usability and business models.
       The goal of this panel is to de-balkanize the bi-polar HTML camps and seek a productive role for all points of view. HTML and the WWW are simply too important for the academic and research community to ignore. HTML will not go away. We must participate in the evolution of HTML and find ways to become part of the network publishing revolution it represents.
    Does Multimedia Make a Difference? BIBAPDF 247
      Paolo Paolini; Robert Glushko; Dale Dougherty; Eliot Kimber; Antoine Rizk; Daniel Russell; Kent Summers
    Hypertext applications and tools in general deal with "passive" media. Values of passive media are static, in the sense that as time progresses their presentation (state) does not change. Values of active media have the property of changing presentation (evolution of the state), as time progresses. Active media such as video, animation and sound are becoming a "necessity" in most recent Hypertext applications.
       The panel discusses the impact of the introduction of active media in the area of hypertext. The impact of active media upon hypertext can involve several aspects: presentation of the content, design of the applications, structuring techniques of the applications, interplay between synchronization and links, run-time support, communication performances and techniques for LAN or WAN architectures.
       Panellists argue that:
  • Hypertext should remain well separated from Multimedia, being a separated
       field, with a different set of notions and needs;
  • A limited number of changes will be enough in order to adapt most of the
       Hypertext notions and techniques to multimedia;
  • A revolution is needed.
  • CEC Briefing: Information Engineering and Telematics BIBPDF 248-249
      Geoffrey Stevenson; Lindsey Holman; Chris Adie; Roberto Minio
    Private Sector Perspectives on Advances in Hypermedia BIBAPDF 250
      W. G. Nisen; Jeff von Limback; Scott Johnson; Kent Summers; Maurice Shephard
    Numerous market and industry segments are becoming increasingly interested in hypermedia. For example, hypermedia is a fundamental cornerstone in electronic publishing, electronic performance support systems, and interactive entertainment. Most of the current state of the art in hypermedia is being defined by research which is conducted in universities and research institutions. Yet many of the hypermedia systems that are in use today by industry are first generation systems, often at least five years old. This begs the question: just how important is leading edge hypermedia technology to the private sector?
       The areas that this panel addresses are:
  • Is there commercial rationalization for incorporating advanced hypermedia
       technology, and what is it?
  • How important is basic research and development in hypermedia technology to a
  • Is the current state of the art in hypermedia sufficient to successfully
       commercialize hypermedia products and services?
  • What are the best ways to transfer the results from hypermedia research from
       the academic and research communities to the private sector?
  • Demonstrations

    DHM -- Dexter-Based Hypermedia Systems BIBA
      Kaj Grønbæk; Jens A. Hem
    DHM (DEVISE HyperMedia) includes an object oriented application framework for developing Dexter-based hypermedia systems as well as a number of example applications. Hypermedia systems built with the framework fulfills the Dexter Hypertext Reference Model, including: Multi-headed (n-ary) links with bi- and uni-directional traversal. In addition, a variety of composites are applied to implement GuidedTours, TableTops, browsers and to collect results from queries. DHM systems also supports integration of third party applications. DHM systems can also be configured to support cooperative work, such as users' navigation and editing of shared materials. The cooperation support is based on Object-Oriented Database (OODB) technology, and it provides long term transactions, flexible locking, and awareness notifications. The DHM framework is platform independent, and DHM system variants exists for Unix (Sun Sparc and HP), Macintosh and Windows/NT. The DHM framework provides generic classes for developing Dexter-based hypermedia systems, and it also supports tailoring systems already developed using the framework.
       A system based on the DHM framework may provide the following features:
  • Atomic components for media such as: text, graphics, video;
  • A variety of composites used for GuidedTours TableTops, browsers and
       collecting results from simple queries;
  • Integration with third party applications such as Microsoft Excel on
  • Span-to-span links for text components, object-to-object links for graphics
  • Multiheaded (n-arity) links with bi- and uni-directional traversal;
  • Cooperation support including long term transactions, flexible locking and
       awareness notifications based on augmented OODB technology; and
  • Sharing of hypermedia networks ('hypertexts' in Dexter terms) across
       hardware/OS platforms.
  • Hypermedia Applications for Children Educators BIBA
      Lucia Amante; Lina Morgado
    Part I -- The Evolution of Children Drawing: The main goal of this application is to support adult learning in self-learning contexts. The target group of adults which can benefit from this application includes psychologists, students of psychology and general educational scientists, as well as professionals in artistic fields. It will also be useful to Pre-school and Primary school teachers. The application has 4 main topics: General Introduction to Graphical Activity in Children; Scribbling; The Development of the Representation of the Human Figure; The General Developmental Stages of Children's Drawing.
       Part 2 -- Early Mathematics -- The Child's Construction of Number: This application aims at illustrating some of the main works of Jean Piaget, namely his theoretical approach, methodological procedures and several of his classic experiments in this field.
       The information is organized in two main Topics: About Jean Piaget, containing biographical and bibliographical data, methodological principles and a brief glossary of the key concepts of his general theory of the child's development: The Child's Construction of Number, subdivided in the following subtopics: Conservation of Quantities, One-to-One Correspondence, Seriation, Class Inclusion. This Topic is organized around the classic experiments of Jean Piaget in this field, with a description of each experiment supported by visual material, and the conclusions drawn from each of them.
    DarkStar -- Studying Hypertext Creation by Naive Authors BIBA
      Margit Pohl; Andreas Dieberger; Peter Purgathofer
    The goal of the DarkStar project is to support naive users in the creation of hypertexts and to study the process of hypertext creation. DarkStar documents are small-scale hypertexts but hypertext generation problems can be studied well in such small documents. The present authoring system, STEP2, incorporates many lessons learnt in previous versions. One part of STEP2 is the structure editor where nodes and links can be defined or deleted at any point during design of the document. As we want to study the temporal development of the hypertext this creation process is monitored by logging functions. An external tool converts those logs to animations that visualize the process of hypertext creation. The system also logs hypertext reading to determine major navigational paths and where people have to backup. Another tools serves to visualize those logs. A third evaluation tool is a statistical tool used to determine general statistics about documents. DarkStar is a Client-Server concept supporting networked group-work. The envelope concept allows authors to create links to other hypertexts very soon in the design process without risking to create dangling links. Envelopes provides standard entry points to a hypertext and valid link destinations even if the nodes in the envelopes are not filled with data yet or if nodes in the envelope are deleted. At the ECHT94 we will present a single computer version of our system. The focus of the demo will be on the editor STEP2, the evaluation tools and lessons learnt in the iterative redesign of the authoring system.
    CastingNet: A Hypermedia System Based on Frame-Axis Model BIBA
      Yasuhiro Ishitobi; Yoshihiro Masuda; Manabu Ueda
    CastingNet is a hypermedia system based on the novel hypermedia data model "Frame-Axis Model" which is suitable for organizing and browsing a large amount of, various structure of, and disorganized information. The fundamental elements of the data model are frame and axis. Each frame is a unit of information which consists of one or more properties, and each axis has a classificatory criterion of frames and a result of a classification as mapped space for frames. Axes represent conceptual relationship among frames as spatial N-ary relations. Each frame is automatically mapped onto axes with their own mapping rule. Consequently, the total number of objects as relations or links can be eliminated compared with binary relations of the node-link model.
       The mapped or organized frames can be browsed through spatial overviews called "Hypercharts", each of which visualizes one or combined axes as one chart. Also, the browsing is accomplished by repetition of both way access between frames and hypercharts. Each hyperchart is automatically updated according to user operations to frames or axes.
       In the demonstration we will present a browsing example where various kinds of e-mail forms and news forms are mapped onto the axes such as date, priority, persons, or topics.
       This study was performed as a part of the project "Self Organizing Information-Base Systems for Creative Research and Development" through the Special Coordination Funds of the Science and Technology Agency of the Japanese Government.
    CD-ROMA Project BIBA
      Roberto Fasciani
    CD-ROMA project is a large project born to explore new frontiers of man-machine interface of hypermedia techniques. The field of application is the area of PC, since we are interested in application of wide diffusion. The accent is in the way to give instruments for navigation to the reader, in order to reduce the risks of getting lost. To improve the sense of orientation navigating in a large quantity of data, we use a strong metaphor for the interface: a real city, with famous and historic places, following the tricks of ancient rhetoric schools: to associate concepts with real places. To make that, we started developing a large number of pages of text (we needed that to show the power of the metaphor), focusing 7 areas of interest in human communication: Hypertext, Multimedia, Virtual reality, Telecommunication etc., and we associate them with areas of Rome, trying to match place and concepts in a subjective but meaningful way. More of that, the navigation from pages (collected in buildings or other place of interest), and areas, is devoted to 3D animation in order to emphasize the moving from a place to another. The result is a sort of encyclopedia (2500 pages planned, half of that completed) of human communication, nested in a 3D reconstruction of the historical town of Rome that can be an interesting technique just to explore the maze of Rome for touristic purposes. In addition, every pages is associated with seven variables (one for each context). The author settles the variables using some tools, in order to place, in a seven dimension world coordinates, then page in relation with others in the database. In this way, the reader could ask the system for a list of pages conceptually "near" to the pages where he/she is. This could allow not to link manually the nodes (pages) in the hypertext giving more degrees of freedom in writing hypertext.
    Intelligent Navigation in a Hypertext Network BIBA
      Martin Subbotin; Dmitry Subbotin
    IntelText is a hypertext system oriented to access and arrange loosely-structured information. This task is performed by the navigation process itself. The navigation is automatic and is based on a set of heuristic rules. Besides the automatic navigation IntelText also has the manual navigation mode and the combined mode.
       The algorithms of navigation presuppose that linking of nodes was performed basing on their direct closeness by content. The linking can be manual or automatic. The latter is based on the similar sets of key terms. The built-in rules of navigation permit to choose the next node on each step of navigation upon criteria of its closeness by content to the nodes already included in the navigation path. The result of navigation (the navigation path) corresponds to the user-set topic.
       Depending on the kind of hypertext nodes, the path can be interpreted variously, yet its coherence is high. The coherence is understood here as the orderliness of text fragments by content and not as the correspondence to a particular style of discourse. Each subsequent text fragment is based on the preceding ones, as if arises from them. Consequently, the user can perceive the material through these judicious sequences of nodes.
       In many works, the mechanism of linearization uses the preliminary structuring of the hyperbase. Unlike this, IntelText uses the topology of the network as it is.
    Contexts for Hypermedia Links BIBA
      Lynda Hardman; Guido van Rossum; Jack Jansen; Sjoerd Mullender; Dick Bulterman
    In conventional hypertext, the model of a link is clear: when a link is followed, one 'leaves' the information in the source node and 'goes to' the information in the destination node. In most cases the source node is replaced, or a new window for the destination information is created. The option is, however, not generally given to the author, but determined by the destination of the link.
       When following links in a hypermedia presentation, composed of structured collections of static and dynamic media items, it is necessary to define which part of the presentation will be affected on following a link. For example the complete presentation may be replaced, or perhaps only one picture is changed.
       We have introduced the notion of context for the source and destination of a link in order to specify the presentation's behaviour on following a link. Tools are needed for specifying this information, along with presentation parameters for whether the information at the source of the link continues playing or not, and where the destination information should be displayed.
       We present our implementation of contexts for links in CMIFed, an editing and viewing environment for hypermedia presentations. This demonstration is based on the theoretical work presented in a paper at Hypertext '93: "Links in Hypermedia: the Requirement for Context".
    Hypermedia Authoring Research Toolkit (HART) Demonstration BIBA
      John Robertson
    This demonstration will highlight the distinctive nature of the HART system. HART is a MS-Windows based software system which provides a human editor with an interactive hypermedia editing environment. HART aids the human editor by providing ACTIVE computer assisted support during the process of transforming paper based media into a structure that is amenable to hypermedia delivery systems.
       HART provides the human operator with two forms of support during the hyper-database development process. Both of these features reduce the complexity of the authoring process:
  • (1) Procedural Guidance -- The system guides the human editor through the
        various phrases of the conversion process. This ensures that the most
        effective and efficient processing is achieved.
  • (2) Intelligent Assistance -- The quality, completeness, and correctness of the
        resulting hypermedia system is fundamentally dependent upon the ability of
        the human editor to mentally manage both the conceptual and physical
        contents of the hyper-database. The cognitive overload commonly
        experienced during this process is a principle reason why it is so
        difficult to develop medium or large size hyper-databases. HART actively
        supports the human editor manage the hyper-components during the
        conversion process. At each phase in the transformation, HART is able to
        suggest likely key terms, index values, node titles, anchors, and links to
        the human editor. John will demonstrate the features of the system by converting a series of documents into a hypertext database, and producing a Microsoft Multimedia Viewer2 project from the HART process.
  • Legal Documentation and Hypermedia BIBA
      Roberto Colotti; Rosa Maria DiGiorgi; Roberta Nannucci
    Hyperlaw2, developed at the Istituto per la Documentazione Giuridica of Italian National Research Council, Florence, in collaboration with the Department of Computer Science of the University of Padua, Italy, manages a collection of unharmonized documents (legal norms, case law documents and texts based on legal authority), within a specific environment law domain (noise pollution). The aim is to provide an exhaustive legal database which is user-friendly for the lawyer or public administrator as well as for the citizen. The database should also combine the hypertext features with those of the traditional information retrieval systems: the legal documents have been indexed with keywords organized according to a special classification table, enabling the user to have direct semantic-type access and to navigate through a wide range of pre-established links (citations, legislative references, keywords, classification codes). Hyperlaw2 is based on a model called EXPLICIT, using a two-level structure to make the different parts constituting the body of data explicit (and hence the name), at the document level, called the hyperdocument, and at the auxiliary data level (the semantic structure according to which the indexing terms are organized), called the hyperconcept. The two levels are linked by the relations between the concepts and the documents the concepts describe. At the same time, the single elements contained on each of the two levels are interconnected: the documents are linked by references or citations, while the links among the auxiliary data are made up of the semantic structure in which the terms are placed.
    Content-Based Navigation within Microcosm for Multimedia Documents BIBA
      Robert J. Wilkins; S. R. Griffiths; P. H. Lewis; Wendy Hall; Hugh Davis
    Link authoring within a hypermedia application is an arduous and time consuming process. As the size of the hypermedia application increases the number of links needed to relate the disparate chunks of information (or documents) will also increase. Authoring all links by hand soon ceases to be feasible.
       The X-Windows version of Microcosm has been extended, allowing links from non-textual media to include the content of their source anchors. This extension enhances the overall link model, allowing the implementation of link mechanisms such as generic links and content based retrieval for non-textual documents. Both mechanisms have the potential to ease the authoring load by reducing the number of links that need to be manually authored.
       The extensions take the form of a system of modules that exists under the link database filter, the content retrieval filter (CRF) and the indexing process responsible for creating the inverted indexes used by the CRF. Each module (called a signature module) is responsible for maintaining an index for a particular signature type (for example colour distribution, shape representation or texture statistics).
       The same suite of modules can be used to pre-index collections of documents. After pre-indexing the user can make a selection from within a document and the system will return a list of the most similar documents from its indexes. For example, should the user have an image of a sunny day they would be able to use the colour distribution information to locate similar images in the application.
      Ray McGuigan
    MIPS, the Multimedia Information Presentation System, is an interactive multimedia presentation system under development within a CEC-funded ESPRIT project. It is being developed by a European consortium with partners from Greece, Spain, Denmark, Belgium, Republic of Ireland and the United Kingdom.
       The MIPS architecture is modular in nature and encompasses: a Presentation component to allow multimedia presentations within seamless windows; a Selection and Retrieval component, for accessing remote databases; a HyTime Engine and Web Builder component to enable hyperlinking; and a Knowledge Base System to enable customisation and user specific profiles.
       This presentation is an interactive poster rather than a live demonstration since presenting a version of the MIPS system from CD-ROM which includes pre-canned queries to remote databases with mock answers rather than using the real system with communications and remote databases. The demonstration includes a digitised video animation outlining the problem addressed by MIPS and the approach taken. It also includes digitised video discussions of many aspects of the system (e.g. architecture, potential, queries, use of HyTime) by developers and members of the project team. The CD-ROM was authored by the Multimedia Centre at the Department of Computer Science, Trinity College Dublin under the direction of the MIPS consortium.
    Microcosm: The Next Generation BIBA
      Nick Beitner; Simon Knight; Hugh Davis; Wendy Hall
    In this demonstration we will show new features of and extensions to the Microcosm model that represent future developments for the next generation of Microcosm. This will include demonstrating the capabilities of the Universal Viewer which is the subject of one of the papers to be presented at the conference. This shows the research that has been undertaken into integrating third party applications into a Microcosm hypermedia web. It raises questions that debate the minimum responsibility of an application to co-operate with the underlying link service, and the properties and qualities of a hypermedia system that can be produced using light hypermedia link services. The demonstration will show how a hypermedia integrated desktop would be of use in the industrial workplace.
       A second feature of the demonstration concerns a more fundamental rethink in the underlying Microcosm architecture and has grown out of the consideration of how multimedia data can be fully integrated into the model. Under Microcosm++, an object-oriented extension to the basic Microcosm model, object support services are a means for tools to describe functional behaviour for specific media without encoding the support within the service application. The scope for link abstraction made possible by the Object Services Hierarchy within the new model leads to a consistent and uniform management of inter-media relationships. This promotes links to a status independent of the media and the documents involved, allowing minimal authoring to describe maximum connectivity of data. The ease of authoring that this permits will be demonstrated using a large multimedia dataset.
    Hypermedia Othello and Otello: A Study and Comparison of Shakespeare's Play with Verdi's Opera BIBA
      Albert O. Cordell
    This hypermedia demonstration uses text, graphics, sound, and video to help cultivate an understanding of Shakespeare's play Othello, to familiarize the user with Verdi's operatic masterpiece Otello, and to compare the two great works -- a comparison only possible through the use of hypermedia. The complete text of the play is augmented with annotation links, links to the various media, and links to the libretto of the opera. The play itself is accessible as readable text or as digitized sound, and the opera can be cued from the musical notation, from the libretto or from parallel situations in the play. The texts of both the play and the libretto of the opera have numerous fixed navigational aids, and the hypermedia reader is given the option of creating new links, annotation notes and routes of perusing the hyperdocument.
       In the late nineteenth century Verdi and his librettist, Arrigo Boito, read Othello and made their collaborative hypermedia event, Otello. "Hypermedia Othello and Otello" gives the electronic reader a chance to study the Verdi-Boito reading of Othello and to create new pathways for a totally new and personal understanding of these two masterworks. Chronologically the play came first, but the significance of this hypermedia product is that it allows an infinite number approaches to the subject of the play and opera -- approaches that are not chronological, not sequential and some even beyond the imagination of the hypermedia author.
    Computer-Based Learning Resources for Instruction in Software Engineering BIBA
      John Wilson; Andrew McGettrick; Alan Spence
    Student-centred approaches to learning can provide a learning environment in which students are able to tailor the instructional process to their own requirements. In addition, adoption of teaching methods other than the traditional lecture/tutorial/practical paradigm offers potential for facilitating professional development. The purpose of developing computer-aided learning systems for software engineering was to explore the potential for innovative use of multimedia in technical re-training. Prototype course modules have been developed to provide instruction on human-computer interfaces, project management, object oriented development and database systems. This demonstration focuses on the database systems materials.
       In common with many aspects of computer based tutoring systems, the subject of database systems lends itself to presentation by interactive use of module content. This provides the necessary underpinning of theoretical material and allows the student to gauge his/her understanding of the content. The module content is supported by associated written material as well as by student assessed exercises which form part of each topic.
       The instructional material is embedded in a template which provides navigational facilities to allow the user to redisplay material or select new material for viewing. In addition a Note Pad feature allows the user to record personal notes which can later be edited and used to generate a hard copy listing. Help facilities are provided to explain the function of elements within the template and students are also provided with a Bookmark facility which allows the resumption of study from the point reached during earlier work.
       The prototype systems we have produced demonstrate that it is feasible to provide significant technical learning experience via hypermedia systems. Such systems could be used by hardware or software engineers who wish to develop their expertise in particular branches of software engineering. The multi-sensory nature of the material means that it is particularly attractive to those with special learning needs.
       Future developments of the database systems prototype include the incorporation of query interpreters together with an example database to allow students to apply the principles introduced by the tutoring material. The Structured Query Language (SQL) database which provides part of this facility supports the core query language facilities which would be found in a fully functional database system. In this way students are provided with direct feedback of their understanding of the course material.
       Acknowledgment: The hypermedia tutoring material has been developed as a joint venture by The Institution of Electrical Engineers, Talkback Training Ltd., and Glasgow, Heriot-Watt, Stirling and Strathclyde Universities. The project is funded by the European Social Fund and the Institution of Electrical Engineers.
    HyperScape: The Hypertext and Information Management Environment for the Macintosh BIBA
      Adrian Vanzyl
    This demonstration shows how a hypertext system functions as a component within an open information management environment. The system is specifically aimed at meeting the requirements of researchers who deal with large amounts of unstructured information. We feel that an environment aimed at dealing with this type of data requires three core components: Searching, Linking, and Structuring.
       The HyperScape environment addresses these requirements with four separate, but tightly integrated and intercommunicating software components.
       Searching -- via a freetext retrieval engine;
       Linking -- via HyperTED, a hypertext editor;
       Database Structuring -- record based bidirectional extraction via database templates for common database systems; and
       Outline Structuring -- hierarchical outline style organisation via the HyperBrowser.
       In order for a hypertext system to successfully cooperate with these other components, it is required to adhere to certain requirements of any open hypermedia system. The HyperTED system demonstrates some key functional requirements. These include:
  • (1) Linking to any media or file type available on the Macintosh;
  • (2) Robust links. Links created within the system remain valid even if a file
        is moved, renamed or edited. Since the system is 'open', users are free
        to edit any file in this environment with a word processor of their
        choice, and move or rename files as desired. The HyperTED system
        cooperates with the operating system in maintaining the links despite such
        changes; and
  • (3) Invisible storage of link information. All linkage information is held
        invisibly within each file, with no markup required, no embedding of
        control codes, and no central monolithic links database. Certain components of the system (such as the search engine and databases) are also available for user access through wide area networks such as the Internet. The system was developed by Adrian Vanzyl and Chris Priestley.
  • Demonstrations/Posters

    Glasgow University Library TILT Project BIBA
      Linda Creanor
    The Glasgow University Teaching with Independent Technology (TILT) project is a three year project funded by the Higher Education Council's Teaching and Learning Technology Program (TLTP). The Library is developing a number of information skills modules dealing with areas such as literature searching, evaluating library material and using electronic databases. They are intended mainly for first year undergraduates, and by developing generic material rather than concentrating on specific subject areas it is hoped that they will be widely applicable. Although a number of libraries have used hypermedia software to produce library guides, its use for reader instruction rather than orientation is an innovation.
       The packages developed to date have been produced for a PC platform, with GUIDE and Toolbook software, both of which run under Windows and include: How to Choose Books and Journals; Library Search Skills; Computer Sources; and a Tutorial for the Biological Abstracts on CD-ROM Database.
       In order to enliven what is often seen as a rather 'dry' subject and also to encourage students to make use of packages which are not (as yet) a compulsory part of their course work, much use has been made of graphics, often with a touch of humour, to attract and hold the user's interest.
       It is hoped that the project will demonstrate that students can be encouraged to become more independent learners, and that by promoting the usefulness of information skills, the knowledge gained will extend beyond the requirements of university life.
    Active Information Seeking BIBA
      Fay Mark; Ellen Campbell
    The "Active Information Seeking" demo shows evolving alternatives to current (keyword search, table of contents, back-of-the-book index, hypertext links) online information search and navigation paradigms. With the overwhelming surge of online information it becomes increasingly difficult to efficiently find and productively use these resources.
       Our solution is to provide two alternatives: a graphical navigational feature called the Topic/Task Navigator (T/TN) and interactive presentations that reuse content from the electronic libraries to: discriminantly retrieve topic or task specific information; and facilitate learning and problem resolution through the integration of multimedia and hyperlinking.
       The Topic/Task Navigator employs a hierarchical organization scheme where a user traverses progressively more specific topic nodes. While T/TN helps to address a number of searching issues, it is limited to movement in only two directions, up or down one level.
       The task-oriented interactive presentations relationally link audio, and visual media to enhance learning and problem resolution. These presentations provide a means to move away from the linear, text-base approach of an electronic book to a new method of accessing, navigating and retrieving digital information.
       The Topic/Task Navigator and interactive presentations run on the IRIX OS using the following technologies and applications: Electronic Book Technology (EBT) information search and retrieval engine, SGI -- Showcase, and Aimtech -- IconAuthor.
    A Large-Scale Industrial Application of an Open Hypermedia System BIBA
      R. Bernard; R. Crowder; Ian Heath; Wendy Hall
    A cornerstone of a successful industrial maintenance, and engineering policy is to have the ability to integrate many forms of information, from a number of sources, and allow the user to access them with ease. The traditional computer, or paper-based systems used for the presentation and distribution of text and a limited range of graphical information, in a linear or branching fashion are no longer considered sufficient. If the stored information is expanded to include high quality still or animated graphics, and still or moving video, and provided with an access system that permits the user to move easily between these different items of information in a structured fashion, then the management of the engineering data resource for a number of applications can be optimised.
       With the increasing complexity of production machinery in the modern automated factory, there is a corresponding increase in the supporting maintenance and production documentation. Industry is therefore looking for advanced informational retrieval systems that can operate both in the office environment, and more importantly on the shop floor. In collaboration with Pirelli Cables at Aberdare, South Wales, a hypermedia system for the provision of process information, maintenance and operator training is being developed, for a sophisticated process machine. The application uses the Microcosm open hypermedia system developed at Southampton. The resource base for this application contains electronic versions of all existing paper documents (i.e., procedures, diagrams and electrical and mechanical drawings), together with additional documents that have been produced specifically for the hypermedia application, these are primarily to allow the user to navigate the resource base.
    HOME: Hypermedia Object Management Environment BIBA
      Erik Duval; Henk Olivie
    We will demonstrate our hypermedia DataBase Management System (DBMS), called HOME (Hypermedia Object Management Environment). The system supports facilities for the development of open and flexible hypermedia servers that can be accessed by clients. HOME is based on a layered architecture with four levels.
  • (1) A distributed raw data storage level relies on digital and analogue
        multimedia data stores, accessible over electronic networks.
  • (2) A relational DBMS takes care of the ACID properties (Atomicity,
        Consistency, Isolation and Durability) of transactions, concurrency,
        access control, backup, recovery and accounting. The higher layers rely
        on the query engine of this layer for dynamic links and query based
  • (3) A multimedia DBMS we have developed supports creation, deletion and
        modification of multimedia objects, as well as object retrieval based on
        search criteria. This layer is responsible for node content.
  • (4) The fourth layer is our hypermedia DBMS, based on the HM hypermedia data
        model, that structures data as S-collections with a content and a set of
        other S-collections, called members, related by navigable links. Using HOME, we have developed HOED, a Hypermedia On-line Educational Database, which incorporates ca. 5900 images on biomedicine, linked to a set of characteristics and a full-text description. We have also developed a gateway between HOME and the World-Wide Web (WWW), so that HOME servers can be accessed by WWW clients.
       We are currently planning a number of projects for development of HOME servers. These include a bibliographical database, a hypermedia server on life art, an annotated slide collection on architecture, a distributed European network of topical courseware databases and an information resource on job opportunities.
  • Hypermedia Based Learning Environments in Mathematics: Authoring and Learning BIBA
      Jari Multisilta; Seppo Pohjolainen
    It has not been easy to present mathematical information (formulae) in current hypermedia authoring systems. They have not supported the use of a mathematical tool program in a hypermedia environment either. In general it is required that the hypermedia system for mathematics learning should: integrate hypertext, computer aided exercises, graphics, videos, and sound; import mathematical text files created with for example Microsoft Word; be easily available to the students (possibly as a shareware); be able to record the actions of the students in order to examine different study styles in hypermedia learning environments; support the process of learning mathematics; and be easy to maintain and update.
       We have developed software tools for authoring hypermedia based learning environments (HMLE) for mathematics education. The main achievement is not only a single hypermedia course on mathematics, but also a set of software tools for translating lecture notes in mathematical sciences into hypermedia and supporting learning process. Authoring tools help to import mathematical text (super- and subscripts, formulae, pictures) from Microsoft Word to HyperCard and create hypertext links between text files and create hypertext links to almost any application program.
       Learning tools in HMLE are mathematical tool programs, concept maps, interactive exercises and digitised videos. Mathematical tool programs (Matlab, Mathematica or Maple) help students to perform numerically or algebraically complex computations in a short time and so they help students to concentrate on the subject matter, not on the computations. Concept maps help students to conceptualise the subject matter and see how different concepts and properties inherit itself to other concepts. Videos can be used to motivate the student and to give an visualisation of a mathematical idea or concept.
       Interactive exercises are important element in HMLE. Exercise Maker is a tool that presents the exercises generated with Mathematica. The student answers are also checked with Mathematica.


    Studio Tour -- Capturing Requirements for Large Multimedia Systems BIBA
      David England; John Patterson
    This project aims to look at an extended cinematic metaphor is an appropriate way of supporting authors of large, interactive, multimedia systems. We are building a 3-Dimensional Sketch Pad system for multimedia authors. The sketch pad system will use the cinematic metaphors of "Story Boards" and "Leica Tests" (or Lecia Reels) as a means of providing a high-level, conceptual view of a multimedia production. This will be akin to making a studio tour of ideas. This system will enable multimedia designers to express their initial ideas and either demonstrate them to colleagues or to a multimedia technical author for more detailed design and implementation. The environment will be evaluated by existing multimedia authors as outlined below. This project will be an example of the type of work that will be carried out in the proposed University Interactive Media Centre.
       When writing a book there are established conventions for organising the material to be included. For example, contents pages, chapter and section headings, help authors and editors to work together in organising the team effort. For multimedia authors there are, as yet, no established conventions for document production. What is required are techniques, and supporting tools, that help the author to express their ideas at a higher level, i.e., which matches the way they conceptualise their designs, and then present them as a requirements specification to an experienced multimedia author.
    Automatic Generation of Electronic Books BIBA
      Jean-Louis Vuldy
    Our purpose is to automatically generate hypertext webs from structured documents according to rules defined in our Research and development Project For Mechanical Software. The original model specified the physical layout, sequencing and referring rules for the documentation. The sequencing mechanism applied to the documentation allows a unique identification of the smallest parts of the document. The objective of the application is to enhance structured documentation by providing access methods specifically devised for hyperdocuments. Formalization of the document model, used for the documentation, was considered essential in order to obtain an abstract model which would be both transposable into any document format and suitable for the design and/or the development of automatic processing tools for automated segmentation, automated tagging grammatical analysis for identification of cross-references links, etc. SGML standard was selected. Practically, the documents are produced with Microsoft Word software. They are converted into SGML form according to our DTD with the recognition of cross-references. Then the browser allows the reader to navigate through the documentation structure or the cross-references. A terminological index of relevant terms was added to the documentation. LEXTER, an extraction software produces such index, based on the analysis of grammatical structure of French sentences. The identification method of each component of our documents allows LEXTER to deliver the relevant terms linked plus their location in the documents. This method automatically creates the links between terms and document occurrences. So readers can browse documentation using linguistic links.
    Information Gardening with Hypertext: A Working Environment for Knowledge Elicitation BIBA
      Sylvain Fraisse; Thi-Lien-Nga Duong; Marion Jaeger Amieux; Lysiane Randazzo; Jose R. Dos Santos; Marc Nanard; Jocelyne Nanard
    The presented hypertext environment is aimed at dealing with hypertext structure emergence to model collective knowledge from a large collection of multimedia documents. Automatic learning tools (INNE) are coupled to the hypertext platform MacWeb in order to support the structure emergence by both experts and machine. The hypertext is used as a common repository for knowledge representation as well as the human interface to the entire system and as the place for cooperation.
       The application corpus is a collection of about 2000 pages of interviews that concern vernacular knowledge about plants and their use for health care from country people living in the mountains of southern France (Cevennes). The environment helps to exhibit explicit or implicit regularities in the descriptions and to determine collective knowledge that is specific to a given region. Such regularities can be correlated to social, ethnic, or geographical data. The whole process is managed as information gardening.
       The poster illustrates the application context and the reason of our approach. It shows how the two processes of emergence of meaning by experts and by machine are coupled. It also focuses on mechanisms that provide adaptive interface based on the result of cooperation.
       The hypertext structure is the result of a two steps loop: experts do incremental information gardening, then, learning tools do farming. The gardening results in MacWeb structures corresponding to a formal frame-based-like representation of facts and the farming operate on the partially elicited structure according to the experts control in order to automatically compute regularities and make inferences.
    Navigation in Spatial Information Environments: User Interface Design Issues for Hypertext and VR Systems BIBA
      Andreas Dieberger
    The Information City project (presented in a poster at Hypertext 93) uses the spatial user interface metaphor of a city to organize and navigate large collections of hypertextual information. As we are used to navigate real life cities the city metaphor -- enriched with magic features -- should help to navigate information structures. A first implementation of the Information City was started in a MUD system. MUDs are networked multi-user text-adventure games which usually make use of a house / city metaphor. MUDs are conceptually similar to hypertext systems and navigational findings in those systems are therefore relevant also to hypertext. While implementing the first parts of the city research into navigation in MUDs was found necessary. This poster presents some results of this navigational study and describes how knowledge in the domains of architecture and city-planning can be used to design an easy to navigate virtual city. Highlights of the results concern magic features and collaboration. Magic features extend the spatial metaphor beyond typical properties of space. An example is the hypertext link which allows tunneling through the spatial structure. Other results concern the richness of spaces (or space-descriptions) and communication between users. It seems the chief benefit of the spatial metaphor of the city is in communication about spatial relationships of information. The findings probably are valuable in designing any information system using spatial metaphors. They are especially useful for hypertext systems realized in some virtual environment -- be it a MUD or an immerse virtual reality system.
    Applying Discourse Theory to Aid Hypertext Navigation BIBA
      Robert Inder; Jon Oberlander
    We discuss ways of improving navigation facilities in hypertext systems, considering theoretical and implementation issues, from a the perspective of natural language processing research. The key claim is that certain ideas from the theory of discourse structure can be exploited to improve the context-sensitivity of navigation facilities. From among the competing theories in the field, we choose one of the least complicated (Grosz and Sidner). This lets us identify <> and <> node links with discourse coordinations, and links for hyperjumps with digressions, a form of discourse subordination. On this basis, (i) after a node is reached by a jump, links to other nodes can by dynamically suppressed where these are rendered irrelevant by the user's mode of arrival; and (ii) at any point after a jump (or sequence of jumps), the user can easily return to a limited set of structurally accessible nodes.
       The approach therefore combines certain advantages of navigation based on document structure with those based on individual interaction histories. The ideas have been tested by altering the implementation of INFO, the hypertextual help system built into the Emacs text editor, which runs on Unix, Macintosh and PC systems. Evaluation studies are currently being piloted, in which subjects use a version of INFO to answer a number of questions about Emacs. We are collecting both simple task performance data and user evaluations. If the results are positive, we see further potential for exploiting notions from discourse structure to inform the design of navigation aids.
    Cooperative Working in X-Cosm BIBA
      Mylene Melly
    Microcosm is an open hypermedia research tool developed at Southampton University, originally for the MS-Windows environment.
       Currently, the X-Windows version is essentially a single user application. Our challenge was to integrate CSCW facilities with this version, in order to support co-authoring facilities. To accomplish this, we built a cooperative server, introduced a cooperative filter inside Microcosm and specified a communication protocol between instances of Microcosm and the cooperative server. This architecture provides awareness information and allows users to exchange ideas.
       Our CSCW approach uses information retrieval mechanisms to facilitate the generation of links between the nodes. Basically, during node authoring, authors supply keywords describing the nodes, called global keywords. Suppose that another author wants to create a link from a selected part of a document. As it is difficult for an author to know the contents and even the names of the possible destination nodes, we suggest that the link author supplies some local keywords that summarise the source anchor. Combining local keywords with global keywords we create a local context to be matched against global keywords of all others nodes to find out possible destinations.
       Version control is something essential for cooperative work. Exodus, the object oriented database toolkit we are using for the Microcosm linkbases already has some version control for the objects it stores. Our model creates a version control mechanism for link using Exodus facilities, combined with RCS (Revision Control System) to control the version of the node.
    GraphCards: Hypertext for Graph Theory BIBA
      Rojal Pradhan
    The main goal of this research is towards developing a graph theory information base for learning and referencing, integrated with tools designed to create and manipulate graphs as well as to illustrate execution of graph algorithms and their applications. Graph Theory has had a tremendous impact on various fields such as Communication Networks, Parallel Computing, Mathematics, Computer Science and many more. This research dedicates to helping those who are not experts in the field of graph theory, in learning and referencing while empowering experts with more tools.
       The development of this application is focused on using NoteCards, a hypertext system. It provides a variety of tools for collecting, representing, managing, interrelating, and communicating ideas. It provides the user with a network of electronic NoteCards interconnected by typed links, which serves as a medium in which the user can represent collection of related ideas. This facility is the basis for the development of graph theory information base for GraphCards. In addition, NoteCards' provision of adding new card types will support GraphCards' requirements for creating new card types for graphs, algorithm execution and animation.
       The application can be classified into two major components:
  • (1) Graph theory information base which will cover most of the graph theory
        topics and graph algorithms; and
  • (2) Graph theory experiment tool set with facilities like the Graph editor for
        viewing, creating or manipulating graphs and, Graph algorithm execution
        and animation, search and annotation utilities, testing and assessment
        utility, etc.
  • Computer-Assisted Generation of Hypermedia Documents in a HyperCard Environment BIBA
      M. Miralles; J. J. Sancho; O. Garcia; F. Sanz
    OBJECTIVES: Development of a tool for the creation of hypermedia documents. The target users of the generating tool -- authors -- would be the teachers of a Faculty of Medicine. The users of the final hypermedia documents -- readers -- would be medical students and physicians. Using the tool should require only basic computer skills.
       MATERIAL: Macintosh IIsi 8/80, Epson GT-6000 scanner and HyperCard v2.1.
       REQUIREMENTS: Macintosh LC, 4 MB of RAM, PICT image files. ASCII text files containing heading marks. The mark unit is an ASCII character that doesn't appear in the text (i.e., $ or #). Levels are indicated by recurrences of the character.
       RESULTS: Hiperlibro (358 KB) creates a new document (the 'book') that contains an 'Author' palette with the tools: Import Text, Create Alphabetical Index, Create/Undo Text-to-Text Link, Create/Undo Text-to-Image Link, Create/Undo Text-to-Note Link, Modify Notes and Deactivate AuthorTools. General and Chapter Tables of Contents-in a collapsible outline format-are automatically created, each entry linked to the corresponding page. Links created by the author appear on the body text as particular styles. A single piece of text can contain the three kind of links. A separate index is maintained for each type of link. Horizontal scrolling bars allow browsing through pages or through chapters. Quit, Go To, Search, Help and History buttons provide further utilities. A pediatrics textbook was used to test Hiperlibro.
       CONCLUSIONS: Hiperlibro is a demonstration of the capability of modest platform and software tools in developing a hypertext generator equipped with the basic functionalities of superior systems.
    Distributed Hypermedia Link Service on WAN BIBA
      Antoine Rizk; Francis Malezieux; Alain Leger
    This poster shows the architecture of a distributed hypermedia link server for the wide area network. The proposed architecture defines a hypermedia protocol and uses the MHEG standard as a common information representation and exchange format. A prototype is currently implemented in C++ under Unix (server and client) and Windows (client only).
       MHEG is used for the exchange of multimedia content data as well as a means of coding more complex hyper objects. We address major issues such as data exchange of node contents and open access to hypermedia facilities. In this way, our architecture allows the real distribution of data "objects" as well as hypermedia "documents" and third party applications. By using hypermedia facilities such as following links, applications may have access both to remotely stored data objects and to other applications or service elements registered in specialised servers.
       The proposed architecture consists of five distinct components that communicate through a specially designed protocol across the WAN:
  • 1) The link server which provides a set of hypermedia primitives independently
        of the applications. The server stores hyper objects such as nodes and
        their definition attributes, links and groups of nodes in a hypergraph.
  • 2) The front-end which is what a telecommunication operator calls a "kiosk"
        service. The main function of the kiosk is to connect the user requesting
        a service to the application server that offers a hypermedia application
        service such as hypernews for example.
  • 3) The user station which typically consists of presentation/editing resources,
        such as graphical interface and monomedia decoders, storage and network
        resources. The user station receives MHEG objects, which are interpreted
        and executed using a MHEG engine. The MHEG engine resolves the object
        internal links and relies on the link server to resolve external links.
  • 4) An application server which defines hypermedia application services. The
        application server uses the link server facilities to store and access
        distributed content data or to call services from other application
  • 5) The content servers are responsible for the multimedia data management in
        close cooperation with the kiosk. Each server could be dedicated to a data
        type like video, sound... to give the optimal quality of service.
  • Videos

    Hypertext Campus Project BIBA
      David Slater; Vicki Brown; Hassan Khan
    In recent years there has been a significant growth in the use of hypertext and hypermedia technology in Higher Education. The Hypertext Campus Project at the University of Kent was established to help introduce this technology at institutional level.
       The video shows excerpts from interviews conducted with staff and students who have been actively involved in the use of hypertext and so offers first-hand accounts of the medium in use:
  • (1) Professor John Slater, Director of the Computing Laboratory, discusses the
        background to the project and the Hypertext Support Unit. He also
        suggests some of the steps necessary in encouraging staff of all
        disciplines to use technology in their teaching;
  • (2) Staff describe how they first became interested in hypertext [as an
        alternative to "reams of paper", an answer to diminishing resources, an
        interactive medium]; how they have made use of the Hypertext Support Unit;
        the effect hypertext has had on their teaching and their students'
        learning; the ways in which their efforts have been recognised; how they
        see the future of hypertext in Higher Education; and
  • (3) Professor John Craven, Pro-Vice Chancellor, gives an account of the ways in
        which staff respond to new initiatives, such as the Hypertext Campus
  • The SCHOLAR Project BIBA
      Su White; Wendy Hall
    This video describes the progress and methods of the TLTP funded Scholar Project at the University of Southampton.
  • The aim of the project is to shift the culture of the university in such a way that staff and students make greater use of hypermedia and multimedia for in their academic lives. The project makes extensive use of the Microcosm Hypermedia system, which originated at the University, to develop a resource-based approach to teaching and learning. Working from the Interactive Learning Centre, the project sponsors the development of teaching and learning materials, runs an associated staff development programme, and deals with the planning and progress of the supporting infrastructure. The video shows a range of sample applications and describes the approaches taken to develop materials which can be integrated into existing academic programmes.
  • Street Kid: Hypertext Branching in Cyberspace BIBA
      Sascha Becker; George Landow
    This video describes and demonstrates Street Kid, a virtual-reality hypertext project developed to explore issues of orientation and navigation in cyberspace. Virtual reality, by its very nature, presents a more concrete story environment than does traditional text or hypertext. The reader expects to experience the virtual world just like she experiences the physical world: she expects to remain oriented in space. In contrast, the experience of reading text on a monitor or in a book neither intrinsically presents nor requires much orientation; the reader's orientation is created inside her mind, by processing the language of narrative constructs and information organizations. Discrete or sudden jumps common to traditional hypertexts would be quite disorienting to the reader immersed in virtual reality. Street Kid makes two important contributions to VR- hypertext: First, it replaces "reading a story" with immersion into a virtual character's experience. It enhances that immersion with text: words representing the character's thoughts hang in the air. Second, it embellishes the traditional 2-d hypertext link with spatial and temporal properties that can only be created in a virtual reality; the reader traverses my VR links by moving from one view to another. These ideas can be applied to a wide range of virtual reality hypertexts; this particular hypertext centers on a homeless child prostitute, the Street Kid. The piece is her experience and thoughts during an afternoon's retreat into a quiet alley in a suspiciously empty city.
    CMIFed: a Transportable Hypermedia Authoring System BIBA
      Lynda Hardman
    We present "CMIFed", an authoring environment for constructing and playing hypermedia presentations. CMIFed supports a rich, platform-independent hypermedia document model, allowing structure-based composition of hypermedia presentations and the specification of synchronization constraints between constituent media items. An author constructs a presentation in terms of its structure and additional synchronization constraints, from which the CMIF document player derives the precise timing information for the presentation. We describe a subset of the facilities in the CMIF authoring environment. The two main authoring views of a hypermedia presentation are briefly described: (a) the hierarchy view for manipulating and viewing a presentation's hierarchical structure; and (b) the channel view for managing logical resources and specifying and viewing precise timing constraints.
       We present the authoring environment in terms of a short example -- a walking tour of Amsterdam -- and briefly demonstrate authoring from within the hierarchy view.
    Hypertext at Brown: Campaign for Libraries of the Future BIBA
      George Landow
    This video shows examples of three stages of using educational hypertext in George P. Landow's courses at Brown University. First, an extensive Storyspace document, the In Memoriam web, exemplifies how read-only hypermedia helps students acquire (a) information and (b) sophisticated critical thinking -- the habit of making connections. Second, the Victorian web provides instances of participatory (so-called read + write) hypermedia that empowers students by placing them within -- rather than outside -- the world of research and scholarly debate. Finally, examples of experimental hypermedia documents created entirely by students show how this new information technology enables them to explore and create new modes of discourse appropriate to the twenty-first century.
    CD-ROMA Project BIBA
      Roberto Fasciani
    CD-ROMA project is a large project born to explore new frontiers of man-machine interface of hypermedia techniques. The field of application is the area of PC, since we are interested in application of wide diffusion. The accent is in the way to give instruments for navigation to the reader, in order to reduce the risks of getting lost. To improve the sense of orientation navigating in a large quantity of data, we use a strong metaphor for the interface: a real city, with famous and historic places, following the tricks of ancient rhetoric schools: to associate concepts with real places. To make that, we started developing a large number of pages of text (we needed that to show the power of the metaphor), focusing 7 areas of interest inhuman communication: Hypertext, Multimedia, Virtual reality, Telecommunication etc., and we associate them with areas of Rome, trying to match place and concepts in a subjective but meaningful way. More of that, the navigation from pages (collected in buildings or other place of interest), and areas, is devoted to 3D animation in order to emphasise the moving from a place to another. The result is a sort of encyclopedia (2500 pages planned, half of that completed) of human communication, nested in a 3D reconstruction of the historical town of Rome that can be an interesting technique just to explore the maze of Rome for touristic purposes. In addition, every pages is associated with seven variables (one for each context). The authors settles the variables using some tools, in order to place, in a seven dimension world co-ordinates, the new page in relation with others in the database. In this way, the reader could ask the system for a list of pages conceptually "near" to the pages where he/she is. This could allow not to link manually the nodes (pages) in the hypertext giving more degrees of freedom in writing hypertext.
    At Home on the Web BIBA
      Borre Ludvigsen
    This video is an edited version of a presentation given at the first World Wide Web conference, WWW'94, which was held at CERN in May 1994. The abstract for the paper reads as follows:
       The aim of this paper is to give some insight into our experience with continual and for some of us immersive connectivity to what is commonly called Cyberspace. It will cover both technical aspects as they pertain to the practical installation, maintenance and administration of the network as a fully integrated subdomain of the Internet. It will also cover aspects of functionality, usage and perceived social impact focusing especially on the exposure and opportunity for publication provided by the simple functionality of a domestic World Wide Web server.
       The Ludvigsen WWW server can be accessed via http://www.ludvigsen.dhhalden.no