Preface

"...A year here, and he still dreamed of cyberspace, hope fading nightly. All the speed he took, all the turns he `d taken and the corners he `d cut in Night City, and still he `d see the matrix in his sleep, bright lattices of logic unfolding across that colorless void...The Sprawl was a long strange way home over the Pacific now, and he was no console man, no cyberspace cowboy..."

from "Neuromancer", William Gibson, 1984, p.4-5 [Gib84]

Cyberspace, as envisioned by William Gibson, can be accessed through any computer linked into the system; it is one limitless place which can be entered equally from a basement in Boston, a mountain in California, or a bar in Zurich. It is a new universe, created and sustained by the world's computers and communication lines. The cyberspace world of Gibson is also an unhappy world which is tied to a desperate vision of corporate hegemony, urban decay, and neural implants, of a life in paranoia and pain.

Cyberspace as described by William Gibson, fortunately, does not exist. But with the advances of hypertext, multimedia, and virtual reality, the technological possibilities are finally here, today. At least partly, the dream described above has come true. Today there are vast collections of computer-searchable information available. On-line systems like the Internet, in particular the World-Wide Web, but also off-line systems such as databases on CD-ROMs offer direct access to huge quantities of information. But how do users know where to find the particular information for which they are searching? How can we make sure that the dream of getting all knowledge of mankind at one's fingertips does not become the nightmare of being lost in cyberspace?

The hypermedia authoring process has been vividly described in a special issue of the Economist as a combination of writing a book, a play, a film, and a radio or television show. A hypermedia document combines all these elements and adds some of its own. The author's first job is to structure and explain all of the information. The author then must distill the information into brief, descriptive nodes. Each node has to contain a list of the ingredients, and instructions on how the ingredients are mixed together to the greatest advantage. The structure of the material provided is translated into an architectural metaphor of some kind; much of the designer's work is the creation of this imaginary space. Then, the designers must chart the details of what to animate, what to film, who to interview, and how to arrange the information in the space to be built. (Adapted from [Eco95]).

It is the goal of this book to present guidelines, tools and techniques for prospective authors such that they can design better usable hypermedia documents and applications. The book tries to convey an overall approach to hypermedia design. When designing a hypermedia product, three different design elements have to be considered: content, structure, and presentation (figure 1).


Figure 1 Hypermedia Design Elements and corresponding book parts

This book addresses all three elements: (1) It gives in-depth coverage to how to structure information for optimal navigation and access. (2) It describes how to present and visualize complex concepts by introducing the reader to algorithm animation. (3) It also outlines a novel approach for authoring and editing of multimedia contents. And finally, it illustrates how these three design elements can be put together to publish user-friendly hypermedia documents.

The book consists of four parts:
I--Structuring Information

After a careful introduction into information retrieval, user modeling and the world wide web, part I introduces the seven concepts for navigation in cyberspace: linking, searching, sequentialization, hierarchy, similarity, mapping, and agents, and discusses a broad palette of existing systems, tools, and techniques to implement these concepts. The third section of part I presents our own tools and techniques: Navigation Diamond, Viewfinder, Hiermap, CYBERMAP, and Cybertrees, that have been developed by the author and his group to address various aspects of the "navigation in cyberspace" problem.

II--Visualization

Part II introduces a rarely mentioned aspect of visualization, namely algorithm animation, a concept that has been around for some decades, but has only sporadically been used outside of the computer science domain. Using algorithm animation to its best offers an author a unique way to explain complex concepts that are very hard to convey by other means. After a short survey, part II focuses on experiences collected when implementing Animated Algorithms, a large scale project that combined a hypertext version of the textbook "Introduction to Algorithms" [Cor90] with animations of the most important algorithms. Later on, a blueprint for algorithm animation as well as a concept called "algorithm animation by scripting" are presented.

III--Multimedia Editing

Until now, most of the digital multimedia authoring tools are based upon visual, easy-to-use direct-manipulation interfaces. Part III describes a complementary language-based multimedia authoring system called VideoScheme that melds the common direct-manipulation interface with a programming language enhanced to manipulate digital audio and video. The resulting system can automate routine audio and video editing tasks as well as perform jobs based on sophisticated media recognition algorithms.

IV--Hypermedia Publishing

Part IV puts together the elements of hypermedia design by discussing aspects of hypermedia publishing. As an exemplary application covering many facets of the hypermedia creation and authoring process, issues and obstacles encountered while producing a series of multimedia conference proceedings are introduced. The DAGS multimedia proceedings have been produced first on CD-ROM, while the latest proceeding volume has been published on the web.

Original contributions of this book are in four areas: Many of the existing tools and techniques introduced in the introductory part of the book are based on some sort of manual preprocessing or are even constructed by hand. As the content of the information spaces grows in scale and dimensions, it will become increasingly harder to manually construct tools to navigate and explore this information space. This means that their generation will have to be supported by the computer to a much larger extent than is the case today. Techniques based on automatic recognition of structure and contents of the information space will become increasingly important. Unfortunately, today's implementations of these systems still need too much human intervention. Our vision of the ideal system encompasses a toolset for navigating in cyberspace that operates on "raw", unstructured documents, is capable of giving users an overview of their field of interest, offers guidance of what to do next and gives readers a graphical overview of their local and global context. We hope to convince the reader in this book that we offer solutions that do, even if they do not yet fully reach the vision, illustrate promising approaches for possible solutions.