Research career: Highlights

Performing fundamental hypertext and hypermedia research

Illustration of a hypertext systemThe late eighties and early nineties of the 20th century can be considered as the golden age of fundamental hypertext & hypermedia research, until the emergence of the Internet and the World-Wide Web (in itself still just a very simple and rudimentary hypertext system). From then on everybody became a lot more interested in making a quick ".com" buck, until the Internet bubble burst.

When I joined the Materials Information Processing Systems (MIPS) research group at the K.U.Leuven in 1988, I originally started out doing A.I. (Artificial Intelligence) and knowledge engineering research. At first sight it may seem pretty strange that a research group in the department of Metallurgy and Materials Engineering (MTM) would get involved in such types of research. Actually, if an engineer wants to be successful in materials selection and corrosion prevention, he or she needs to be able to understand and use vast amounts of intricately interrelated information. To achieve this, the MIPS group built advanced expert systems such as Prime (PRocess Industries’ Materials Expert) that assisted engineers in performing materials selection. When I was asked to built a hypertext-based help system for Prime, I soon came to realize that hypertext and hypermedia as a technology could become very powerful tools for materials engineering and corrosion prevention. From that moment on I re-oriented my research interests, and started working full-time on hypertext and hypermedia systems. Using the insights I had gained during my research into advanced knowledge engineering techniques, I was able to introduce new concepts and novel approaches for enhancing and extending the capabilities of hypertext and hypermedia systems. From 1991 onwards I published a series of scientific articles on knowledge-based hypertext and hypermedia systems that are still being read and cited today.

Knowledge-based hypermedia systems: research prototypes

From 1989 to 1995, I designed and developed IKON (Intelligent Knowledge Objects Navigator), a research prototype of a so-called "third-order" or knowledge-based hypermedia system. IKON was based on the MMVP (Model-Map-View-Praxis) conceptual architecture for third-order hypermedia systems I had first suggested in an 1991 article in the Hypermedia journal. IKON was a demonstration testbed for two important new concepts I had introduced in that article: hypermedia nodes and links had to be handled as objects of equal rank, and knowledge about both the nodes and the links could be represented and manipulated explicitly. IKON also supported a novel browsing mechanism, "link navigation through message passing", which allowed the system to actively change the presentation look and traversal feel of its hypermedia information. When the World-Wide Web emerged as a promising new technology in the early nineties, our MIPS research group immediately used it to develop innovative on-line hypermedia systems. One such research prototype was the GRACE system, which provided Web-based access to a large body of expertise on materials selection & corrosion engineering problems. Using information about the user's profile and his tasks, it generated hypermedia nodes and links "on the fly", incorporating materials and corrosion domain knowledge. GRACE also contained an intelligent Failure Analysis Assistant, that allowed a corrosion engineer to step through a complete failure analysis process, with each step automatically cross-linked to relevant corrosion information. For more details on the GRACE system, see the presentation "Intelligent hypermedia corrosion information tools" [PDF 730 KB].

Screen shot of application         Sample screenshot of GRACE system
IKON
(Intelligent Knowledge
Objects Navigator)
 
        GRACE
(Global Reading & Accessing
of Corrosion Expertise)
 

Knowledge-based hypermedia systems: commercial products

In 1991, our MIPS group (together with Elsevier Science Publishers) released one of the world's very first commercial hypermedia systems, the Active Library® on Corrosion. This was a CD-ROM-based hypermedia information system for corrosion engineers, consisting of a reference library with background information, a set of corrosion case studies and a series of densely interlinked guidelines for corrosion prevention. It had a novel, 3-dimensional navigational interface with built-in intelligence (the so-called "Cube of Contents"), that allowed any corrosion engineer to quickly pinpoint relevant corrosion information. In 1994, I had acquired enough experience in designing and developing advanced knowledge-based hypermedia systems to start with the development of a first commercial product. IKEM™ (Interactive knowledge-based system for materials properties) was a knowledge-based document management and annotation system, that supported intelligent navigation & retrieval of materials knowledge and tight co-operation between groups of materials experts working within a company. To achieve this, IKEM™ contained a domain knowledge thesaurus organized as a semantic network of hyperlinked concepts, and an intelligent indexing engine that automatically mapped keywords found in the multimedia documents to concepts in this thesaurus. IKEM™ was succesfully sold untill 2002.

Cover page of ALC cd-rom         Screen shot of web page
ALC
(Active Library®
on Corrosion)
 
        IKEM™
(Interaktief kennissysteem
voor materiaaleigenschappen)
systeem architectuur [PDF 109 KB]
 

Being present at the very birth of the World-Wide Web

In 1990, I was briefly touched by History (with a capital "H") when I talked to Tim Berners-Lee at ECHT'90, the first European Conference on HyperText which took place at the Palais de Congres de Versailles (Paris, France). Tim Berners-Lee and Robert Cailliau, both working at CERN (the European Laboratory for Particle Physics) were visiting the conference and were talking excitedly to anybody who would listen about their plans to develop an "on-line hypertext system" for the CERN researchers. They were going to call it the "WorldWideWeb" (shortened to WWW), and they were now looking around for good ideas on how to build such a system. I remember talking at length with Tim Berners-Lee about the challenges of building a robust hypertext system and about the benefits of using markup languages such as SGML. Unfortunately, I didn't follow up on our conversation afterwards (if I had only known ...), and when they released the first Web browser and Web server in 1991, I still didn't realize they had just given the starting shot for the Internet revolution. Although for some unexplained reason, I did keep the business card Tim then gave me.

Scanned image of front of business card         Scanned image of back of business card
Business card (front) of Sir Tim Berners-Lee,
one of the fathers of the Web
 
        Business card (back) of Sir Tim Berners-Lee,
one of the fathers of the Web
 

I consider myself to be very fortunate to have personally witnessed the first babysteps of what would later become the World-Wide Web. I still remember compiling one of the earliest versions of the original NCSA Mosaic web browser (which would later turn into the first version of Netscape's Navigator) on the Sun workstations at the university. I also remember setting up NCSA HTTPd, one of the very first web servers at the university (and probably one of the first web servers in Belgium), in order to experiment with the possibilities of HTML. Realizing quite early how quickly the Web was growing in size and popularity, and how important it was going to be to effectively search through this vast information space, in 1995 I set up a small web site on Internet search tools. This web site contained a list of various search tools available on the Internet, which I rated on the basis of their usability and effectiveness. The site quickly became very popular, and when I left the university it was taken over by the Universiteitsbibliotheek Gent (no longer active).

Screen shot of web page
"A selection of
Internet search tools" web site
(original 1997 version)
 

Evangelizing the use of the Internet and the World-Wide Web

As the excitement around the Internet grew in 1994, I.T. Works organized the very first seminars in Belgium on the business use of the Internet. These seminars dealt primarily with Internet basics (what is the Internet, how to use e-mail, FTP, USENET newsgroups, Gopher, etc.). When the World-Wide Web itself became more widely known I was invited to talk about this new and promising technology. This series of talks at the end of 1994, beginning of 1995, was one of the first times in Belgium where the Web was introduced to a non-technical audience. In my conclusions I already called the Web the "new frontier for software and services". In order to raise awareness, in 1994 I also wrote a report "World-Wide Web: A hypermedia information universe being woven across the global Internet" [PDF 77 KB]. This report quickly turned into the most widely read and most frequently requested report of our MIPS research group.

Sample slide of presentation
World-Wide Web: A hypermedia infospace
built on top of the Internet
[PDF 412 KB]
 

In subsequent talks in early 1995, I talked in more technical detail about the coming Internet revolution, stressing that the World-Wide Web was going to be the "killer application" that would turn the Internet into a technology that would transform society. I explained how HTML was the "esperanto" of the Internet, the markup language that allowed us to link up information and systems wherever they were located on the globe. It was still early days, but the excitement (and greed) that would later turn into the Internet boom (and bust) of the late 1990's was already in the air.

Sample slide of presentation         Sample slide of presentation
The Internet (r)evolution [PDF 628 KB]
 
        HTML: The language of
the World-Wide Web
[PDF 330 KB]
 
Coded in valid XHTML 1.0 Creative Commons License Made with Cascading Style Sheets