| What Do Readers Do to Pop-Ups, and Pop-Ups Do to Readers? | | BIBA | 2-9 | |
| Heather A. Stark | |||
| This study examines how readers of an electronic document are affected by alternative types of window giving additional information. People read descriptions of properties in order to recommend one for a client. These descriptions contained new details that were presented either in a "pop-up" or a "replacement" window. In post-task interviews readers having replacement windows complained mostly about details of the interface whereas those with pop-up windows suggested improvements to the content, suggesting that the pop-up windows were less cognitively demanding. Window type also affected use of other navigation facilities, and people having replacement windows claimed difficulty in remembering clients' requirements. Window type did not affect total time on a subsidiary task of judging whether the new detail was consistent with the main text, but the distribution of time between the detail and the main text was significantly different for the two window types. | |||
| A Comparison of Linear and Hypertext Formats in Information Retrieval | | BIBA | 10-19 | |
| Cliff McKnight; Andrew Dillon; John Richardson | |||
| An exploratory study is described in which the same text was presented to subjects in one of four formats, of which two were hypertext (TIES and HyperCard) and two were linear (Word Processor and paper). Subjects were required to use the text to answer 12 questions. Measurement was made of their time and accuracy and their movement through the document was recorded, in addition to a variety of subjective data being collected. Although there was no significant difference between conditions for task completion time, subjects performed more accurately with linear formats. The implications of these findings and the other data collected are discussed. | |||
| Getting to Known Locations in a Hypertext | | BIBA | 20-27 | |
| Andrew F. Monk | |||
| A distinction is drawn between directed navigation where the user knows the
location of the information being sought because that location has been visited
regularly in the past and exploratory navigation where the user cannot specify
precisely where the information might be within the hypertext system. Possible
ways of facilitating directed navigation are considered. Maps or "browsers"
will provide direct access to nodes but with large hypertexts they become
unwieldy. The same problem arises when using a key word index with a large
hypertext. The solution is to allow users to select frequently visited nodes
for inclusion in a "Personal Browser".
The idea is illustrated within HyperCard (TM Apple Computers). The user is asked whether a card should be placed in the Personal Browser after it has been visited some criterion number of times. The Personal Browser can be revealed with a single keystroke and contains a button for each location recorded. | |||
| Intelligent Navigation for Semistructured Hypertext Documents | | BIBA | 28-42 | |
| Craig D. B. Boyle; James R. Snell | |||
| Navigation around a complex hypertext system is a serious problem because of the largely unstructured nature of hypertext. The SINS (Semistructured Intelligent Navigation System) system shows how knowledge-based techniques, based on the Information Lens concept, can be used to ease navigation difficulties through the creation of virtual links. The power of SINS comes from the richly interwoven semantic network defined by the hypertext links. SINS has been applied to a Veterinary Science domain. A reference manual with a high degree of intra-connectivity was used as a prototype. Domains with a high degree of networking, connectivity and modularity are suitable for a SINS type intelligent navigation system. | |||
| Creating and Viewing the Elastic Charles -- A Hypermedia Journal | | BIBA | 43-51 | |
| Hans Peter Brondmo; Glorianna Davenport | |||
| We describe a collaborative effort involving approximately 15 people creating material and tools for the Elastic Charles hypermedia journal. The tools are used to create an additional structural layer upon video, sound and text material which has already been edited and is in a final format. This structural layer -- hyperlayer -- is used to link related portions of the journal's stories together to create the hypermedia environment. Our paradigm for linking in temporal media is presented. | |||
| A Hypertext System with Controlled Hype | | BIBA | 52-63 | |
| Ian D. Benest | |||
| Hypertext systems appear to have complemented the window managed display to produce a chaotic work environment. There are two conflicting models. On the one hand there is an allegiance to the desk-top metaphor (Smith et al, 1982), and on the other there is the determination to break away from the metaphors of real world objects to provide access to non-linear text. When, in addition, the system supports concurrent windows that can be moved, resized, closed, opened, etc, on a display surface many times smaller than a real desk, the manipulation of the graphics environment must interfere significantly with the tasks that the professional person has to perform during a normal working day at the workstation. This paper describes a display object which handles text, graphics and animation and which is beginning to be linked symbiotically into an environment that attempts to overcome the disadvantages expressed above. It may be argued that this object, called the Book Emulator, has many of the hallmarks of a hypertext system, but it is one where linear text provides the base with which the user can maintain a sense of position and openness, and where navigation links provide for a controlled non-linear capability. | |||
| Two Field Studies of Hypermedia Usability | | BIBA | 64-72 | |
| Jakob Nielsen; Uffe Lyngbaek | |||
| We present the results of two empirical studies of the usability of hypermedia systems. One of adult professionals reading a technical report and one of children in a kindergarten playing with a non-verbal hypermedia interactive fiction. Both studies are field studies in the sense that they address the usability of hypertext in the users' own environment rather than in a laboratory setting. | |||
| Navigation in Hypertext: Structural Cues and Mental Maps | | BIBA | 73-83 | |
| Annette Simpson; Cliff McKnight | |||
| This chapter describes a study examining the influence of three types of cue
on subjects' navigation through, and structural knowledge of, a hierarchical
hypertext document. The three types of cue were: (i) an alphabetical index
versus a hierarchical contents list; (ii) topographical cues; (iii) the
provision of a "footprint" on the contents card indicating the last text card
seen. An additional measure of interest was the relationship between subjects'
ability to navigate within a hypertext document and their ability to produce a
map of its structure.
It was found that navigation through the document, in both an initial reading and an information-location task, was more efficient using a hierarchical contents list than an alphabetical index. In the initial reading performance was enhanced by the provision of a "footprint". Maps produced by subjects provided with a hierarchical contents list were more accurate than those produced by subjects using an alphabetical index. The addition of typographical cues had no significant influence on any of the measures. Finally, navigation efficiency in both the initial reading and item location tasks was positively correlated with the accuracy of the maps produced by subjects. | |||
| An Empirical Comparison of Two Navigation Systems for Two Hypertexts | | BIBA | 84-93 | |
| Patricia Wright; Ann Lickorish | |||
| Adults unfamiliar with computers answered questions by consulting information in two hypertexts using either a navigation system that involved jumping to and from an Index separate from the text display, or a system that gave many navigation options directly on the text page. Readers preferred Index navigation for a book-like hypertext, with a modular information structure, but for a less book-like text whose structure was a symmetrical hierarchy (matrix), page navigation was preferred and gave better performance on some kinds of questions. These data show that authors' choice of navigation system will depend on the information structure of the hypertext material, and also on details of the tasks undertaken by readers. Combining navigation systems could widen the range of tasks for which hypertexts can be easily used. | |||
| Hypertext Documents for the Learning of Procedures | | BIBA | 96-104 | |
| Aude Dufresne; Nicolas Jolin; Alain Senteni | |||
| The present research is an application of hypertext to the design of a tutorial system for the learning of procedures. Direct manipulation interfaces pose problems for the learning of complex applications like spreadsheets, which require planning and the co-ordination of many operations. Hypertext interactive possibilities can be used not only to offer an exploration environment for knowledge, but also to illustrate iconic and dynamic concepts that are very hard to grasp in ordinary text. Our protocol analysis of subjects reading the text manual of Excel and interacting with the system has revealed the cognitive limits associated with different semantic, syntactic and tutorial strategies. A HyperCard application was developed to create a better training environment, simulating aspects of a spreadsheet with animation and exercises to explain complex concepts and sequences of operations. Though the system covers only a subset of operations (database functions in Excel), it offers an interesting view of the way procedural knowledge can be explained as hierarchies of goals on hypertext. | |||
| Using Hypertext for Educational "Help" Facilities | | BIBA | 105-113 | |
| Christopher J. Colbourn; Tracey Cockerton-Turner | |||
| An initial empirical study looked at children's use of an on-line "help" facility in solving a computer-presented physics problem. Children worked alone or in pairs. Analyses of performance, video and questionnaire data revealed that the low performance of the help users related to their misunderstanding of the nature of "help", interface design problems, and the learning ethos of the classroom. A subsequent investigation utilising hypertext principles for the design of the task and help facility showed that a computer-initiated, context-sensitive, help facility benefited the performance of both pairs and individuals, relative to a comparable request help group. The difficulty of knowing when help is needed may explain this effect and clearer navigation in the hypertext-based task may relate to improved success. | |||
| Learning Styles in a Non-Linear Training Environment | | BIBA | 114-120 | |
| Neville A. Stanton; R. B. Stammers | |||
| In this study aspects of learning styles in a non-linear learning environment were investigated. From previous work three self-reported broad-band learning styles were proposed: top-down; bottom-up; and sequential. Further consideration was given to why non-linear training was superior to a linear environment. It was suggested that it allows for greater individuality on behalf of the learner. | |||
| Signposts for Conceptual Orientation: Some Requirements for Learning from Hypertext | | BIBA | 121-129 | |
| Terry Mayes; Michael R. Kibby; Tony Anderson | |||
| This chapter examines two very different approaches to the use of hypertext for learning. It is concluded that the problem of disorientation in hypertext can be capitalized on in the phase of learning that requires new structures to be created by the learner. A spatial map may be inappropriate for navigating in conceptual space. We discuss an alternative and briefly consider how we can go about deciding what is appropriate for the development of such interactive learning systems. | |||
| An Examination of Hypertext as an Authoring Tool in Art and Design Education | | BIBA | 130-135 | |
| Alan Dyer | |||
| An important feature of current Art and Design education at undergraduate and postgraduate level is the integration of theoretical studies with studio practice. Computer-based hypertext systems are providing new and creative techniques for manipulating and authoring research work in the Art and Design context. The paper summarizes some of the questions which the introduction of hypertext authoring has raised at Coventry Polytechnic Faculty of Art and Design. | |||
| Hypertext Meets Interactive Fiction: New Vistas in Creative Writing | | BIBA | 136-141 | |
| Gordon Howell | |||
| This chapter overviews some of the current thinking of the possible uses of hypertext in the field of creative writing. This synthesis is called "interactive fiction". The fundamental difference between conventional fiction (also called linear fiction) and interactive fiction is that the reader becomes a part of the story and partially controls the direction and experience of the art. This paper describes some of the fundamental issues and concepts in this area. | |||
| Semantic Network Elicitation: Tools for Structuring Hypertext | | BIBA | 142-152 | |
| David H. Jonassen | |||
| Many hypertext researchers and designers claim that hypertext mimics
semantic networks, that is, hypertext information structures may reflect the
structures of human memory (Fiderio, 1988). The untested assumption of this
claim is that hypertext knowledge-bases which model an expert's knowledge
structure can be used to map that structure more directly or explicitly onto
the learner's knowledge structure. This further presumes that learners can
assimilate that knowledge structure and use it to accommodate information
contained in the knowledge base. Indirect or implicit mapping results from
traditional instruction. That is, the knowledge structures of learners
increasingly resemble the knowledge structures of their teachers over time
(Shavleson, 1972 and 1974). So, if hypertext knowledge-bases are organized to
reflect the semantic structure of the expert or the teacher, then there is good
reason to hypothesize that learners will reflect these over time.
This paper reviews the development of associative network theory, which is the conceptual basis for hypertext, and compares it to the logical relationships that define other forms of software systems. The next section reviews both a rationale and a set of techniques for ascertaining an expert's knowledge structure in the hypertext. Finally, some hypertext mechanisms for reflecting the expert's knowledge structure are described. | |||
| On the Design of a Frame-Based Hypermedia System | | BIBA | 154-165 | |
| Toh-Tzu Koh; Peing Ling Loo; Tat-Seng Chua | |||
| While the evolution of hypermedia started as efficient machine-supported ad-hoc links among pieces of data, the researchers at ISS have observed and agreed with the trend towards the support for some structure imposition upon the hypermedia network. The benefits are two-fold: a suitable amount of structuring could enhance creativity among the idea-processing users; and system intelligence thus derived could provide contextual cues for the information-seekers' search/query/browse activities. To stimulate thinking along this direction we present our conception of an information model to augment the node and link model with the structuring and deductive power of frames. The structuring capability helps us sort out from among the massively interlinked data of the target domain, those pieces that pertain to the users' information-seeking goals. Deductive mechanisms would be part of the frame support to allow us to infer missing pieces of information from associated ones and to filter relevant pieces of information among overwhelming possibilities. We are treading on the common ground shared between the Information Retrieval (IR) researchers and the Artificial Intelligence (AI) enthusiasts. We share their common interest -- using frames intelligently upon a huge data store -- and are committed to implement this upon the hypermedia foundation consisting of a node and link model and an interactive multimedia interface. | |||
| Towards the Third Generation: The Case for IKBH (Intelligent Knowledge-Based Hypermedia) Environments | | BIBA | 166-173 | |
| Roderick I. Nicolson | |||
| Despite its unparalleled support for prototyping, screen design, and user interaction via browsing, HyperCard lacks high-level authoring support in terms of program design, debugging and maintenance. Furthermore, its capability for answering direct questions is negligible unless the question type has been foreseen by the program designer. These limitations (added to problems of bulk and speed) severely reduce the viability of HyperCard for the construction of large scale "information resource" types of application. The solution explored in this paper is the use of a combined Prolog/HyperCard environment in which HyperCard is used for prototyping, for creation of program "content" and as the "foreground" information display medium, whereas the system is driven from a parallel database in Prolog. In addition to its use for answering direct queries and for problem solving, the Prolog database is used as a "master" for generating the HyperCard database thus providing a principled methodology for system maintenance and validation. Two further advantages of this design methodology are that knowledge-based authoring techniques may be used to systematize and speed up program design, and that natural language interface techniques may be used to support unconstrained natural language queries. A prototype IKBH system -- USHIR, the University of Sheffield Information Resource -- has been created to explore the potential of the methodology. | |||
| CASE: A Case Analysis Support Environment | | BIBA | 174-182 | |
| Robert W. Lawler | |||
| The synergy between hypertext tools for organizing large, heterogeneous databases and functioning models as explanations of processes may permit us to address a class of problems remaining largely bypassed and undervalued in the study of human learning. If these "power tools for the mind" permit us to better manage and model complexity, they may bring within our grasp a series of problems long considered beyond the reach of well articulated understanding. One such cluster of problems centres on how cognitive development of the individual relates to particular interactions. Case study has long been an effective tool in unravelling the more intricate patterns of human behaviour and development. Its focus on individual behaviour leads to the detailed observations necessary to understand an individual's performance. As contrasted with methods which attempt to "hold all else equal", case study traces in detail the path of change, and so it is the method of choice in studying learning. Technology has enhanced the dependability of case study materials -- videotape permits capture of enough of a context to permit later interpretation in detail. This blessing is a burden in disguise, for there is no balanced development of techniques of analysis. Computers can facilitate the analysis of case material to redress that imbalance. One may hope their use will introduce a new period in the study of intelligence, one where cognitive scientists will have at last the tools to study the development of knowledge in its full particularity. | |||
| Creating Multimedia Documents: Hypertext Processing | | BIBA | 183-192 | |
| Sebastian Rahtz; Les Carr; Wendy Hall | |||
| Much of the work undertaken on hypermedia has been concerned with creating
interfaces that allow an author to create documents in a hypertext format,
incorporating buttons and links in the ways permitted by the particular
hypertext system. It is vital, however, that hypertext systems provide
interfaces to existing text-processing systems so that documents created in
this way can easily and perhaps automatically be incorporated into a hypertext
format.
This chapter describes a project being undertaken in Southampton University to provide a hypertext interface for texts written using a genetic markup language, such as LaTeX. The author creates an ordinary LaTeX document using normal facilities for cross-referencing, indexing etc, and this generic information is used to generate tags for a hypertext interface, which allows the document to be viewed section by section and provides buttons for accessing tables, diagrams, figures, references and cross-references to other documents. | |||
| A Hypertext Approach to Browsing and Documenting Software | | BIBA | 193-204 | |
| Nigel T. Fletton | |||
| Software documentation should be produced as a by-product of the development
process and handed over as a complete package along with the source-code to the
team that will maintain the program. However, this is rarely the situation in
practice. A large proportion of software reaches completion without useful
documentation for the people who have to maintain it. Redocumentation is an
activity that many maintenance teams are forced into because the software
documentation that is supplied with the program they have to maintain is
inadequate or non-existent. It involves the creation of documentation from
analysis of the source-code by experienced programmers and the recovery of
useful information from the original documentation where available.
This chapter describes a hypertext-based technique for browsing and documenting source-code using a purpose-built prototype. The technique described was originally aimed at providing software maintainers with a system which would allow them to redocument a program during their day to day maintenance tasks. However, there are no reasons why the same technique could not equally be applied to the production of maintenance documentation during development. The system uses hypertext links to allow programmers to locate areas of interest rapidly and efficiently in source-code and to examine and update related documentation. The hypertext structure of the source-code, documentation and source-code cross-referencing information is described along with how it is used. | |||
| Separating Hypertext Content from Structure in Trellis | | BIBA | 205-213 | |
| Richard Furuta; P. David Stotts | |||
| We have defined a Petri-net-based model of hypertext (the Trellis hypertext
model). This model distinguishes the hypertext's structure, content, and
context. The structure of the hypertext describes the relationships that tie
together elements of the content (eg the adjacencies or links). The
hypertext's reader examines the content, which may be textual, graphical, or
perhaps audible. The association of content with the structure defines the
context in which the content is presented to the reader.
As we have investigated the Trellis model, and have designed and implemented prototypes based upon it, we have come to realize that separating the content from structure increases the flexibility of the system for the hypertext's author. However, the separation has also required inclusion of special mechanisms in the prototype implementation to ensure that the presentation of the context to the hypertext's reader is easy to comprehend. In this chapter we describe and discuss the implementation and implications separating content from structure as reflected in our hypertext model. | |||
| Hypertext as a Programming Environment for the Authoring and Production of Interactive Videodisc Training Packages | | BIBA | 214-222 | |
| Philip J. Gartshore | |||
| The Design Information Research Team has produced training packages for the
construction industry professions using interactive videodiscs. These
stand-alone systems have been centred on a purpose-designed screen template and
an iconic interface linked to the control of the videodisc, a sophisticated
database storing details of videodisc sequences, and the generation of a
composite computer display integrated with the videodisc image.
Standard authoring environments required a long turn-round time to produce complete packages. Hypertext enabled a common environment to explore ideas about a training package at an early stage for later modification. This approach was implemented in an authoring system, IVA 1.1, using the DIRT iconic interface. The package can be played in student mode to simulate the interactions as they are developed. This chapter will describe the authoring package, with details of a videodisc package that has been produced. | |||
| AI Methods for Structuring Hypertext Information | | BIBA | 224-230 | |
| Thomas Knopik; Sigrid Ryser | |||
| In this chapter we present a new approach to building up complex structures of links between hypertext nodes. Our system supports the authors of large hypertext applications by automatically building up parts of the link structure of the application depending on the content and type of the nodes. We apply Artificial Intelligence methods such as natural language analysis and heuristic rules to find or create nodes that could be linked by the system. | |||
| Querying an Object-Oriented Hypermedia System | | BIBA | 231-238 | |
| John C. Chen; Thomas W. Ekberg; Craig Thompson | |||
| Browsing is the primary way to access information in current hypermedia systems. However, it is often easier for the user to describe what information he is looking for than to find it himself. This paper discusses Panorama, an object-oriented hypermedia system, focusing on its query facilities which incorporate the functionality of retrieval, display, and navigation. | |||
| Interacting with Hypertext: Functional Simplicity without Conversational Competence | | BIBA | 239-243 | |
| Peter J. Thomas; Michael A. Norman | |||
| Empirical studies have been carried out which show that the intended functional simplicity of hypermedia systems is belied in the interactions of users with those systems. A hypermedia system was used to develop educational materials by a novice user. A corpus of video-recordings were investigated within the analytic framework of "Conversation Analysis". This investigation indicated that users import fundamental knowledge of conversational structure and interactive process into their interactions with hypermedia systems. | |||
| Hypermedia as a Communication Design Software Support | | BIBA | 244-251 | |
| Yutaka Hirakawa; Shuichi Uchida; Michihiro Monden | |||
| This chapter discusses a deliberation support method for a communication software designer utilizing design history which takes the form of a hypertext model. First, an information-growth model is proposed for design history. Next we propose a deliberation-support method for the designer with timely presentation functions of early design information which is closely related to the current design stage. | |||
| User-Centred Hypertext Design: The Application of HCI Design Principles and Guidelines | | BIBA | 252-259 | |
| Lynda Hardman; Brian S. Sharratt | |||
| Current HCI principles and guidelines do not specifically address the design of hypertexts. This chapter presents a set of design principles and guidelines tailored to the hypertext design process. The guidelines are divided into four functional areas: user action; information display; dialogue design; and online assistance. The application of these guidelines to static (layout/presentation) or dynamic (navigation) design issues is noted. An example is used to illustrate these principles and guidelines. | |||
| The Designer's Notepad -- A Hypertext System Tailored for Design | | BIBA | 260-266 | |
| Ian Sommerville; Neil Haddley; John A. Mariani; Ronnie Thomson | |||
| This chapter describes a hypertext system tailored to support the early stages of the design process. The tool allows a design graph to be created, annotated and viewed in a number of different ways and supports the production of design reports. Although the initial version of the tool is intended to support software design, the system is sufficiently general that the design of any class of product may be supported. | |||