Foreword

My first contact with Hyper-G was through the "Hacker's Jargon Dictionary" in 1990, when that collection of Internet jokes became available on the Web through the Hyper-G server in Graz. WWW was itself then in a start-up phase, and Tim (Berners-Lee) and I worked all the time at its implemention. Tim had a browser-editor running on NeXTStep (the original NeXT-Cube, now blisfully retired, still squats under my desk at CERN with the original computer center dust sticking to it). When Mosaic made the Web explode, we lost some of the original functions of that prototype browser, but it was clear even from 1990, that authors needed much more computing support at their server site than just wysiwug HTML editing.

Our knowledge, as implemented by the Web, is grainy: There are things like srvers, which hold collections of documents, and around those servers are, geographically located, groups of authors. And it is just as well that the Web is not smoothly distributed: we need some objects to focus on, such as the local school, the tourist info of Munich, the art collection of the Louvre, the frogs of Australia, the armor museum of Graz.

As Hyper-G matured (and how often have I witnessed the briliant multimedia demonstration by Keith Andrews, until my neural network began to equate Hyper-G with fractal mountains and rock music) it began increasingly clear to me that it would make a very good authoring server: here were all the tools that we really wanted, but which the extremely distributed nature of WWW prevented us from even thinking about.

Hyper-G is the workbench of the Internet publishing workgroup: here we can author collections of documents, apply various rtansformations and semantics, experiment with multimedia documents. Yet at the end of the day, the community of users out there can use their ordinary Web browsers to look at what we have done. Anyone who has struggled with a flat set of HTML files in pure text will appreciate the structuring support that Hyper-G offers. Maintenance of structure in large numbers of documents became affordable. But there is much more: authors will find support for versions, multiple languages (a very European problem), cooperative work and so on.

This book is a tribute to the work of many dedicated developers, some I have met, some not, but I admire the efforts of all of them. It is my pleasure to recommend this volume to anyone who considers setting up "an Internet site". Enjoy reading this very informative and richly detailed description of the most advanced authoring server available!

Robert Cailliau

Geneva, CERN, Switzerland, December 1995