Welcome to the First International Workshop on Hypertext and the Semantic Web.
The last decade has seen considerable evolution in thinking in both the hypertext and the Web research communities. Hypertext systems researchers have designed and exercised a variety of systems which address the broad notions of linking upon which the work of the hypertext community is based, with models including navigational, spatial, taxonomic and temporal hypermedia. Meanwhile the Web has become established as the global platform for human access to information and, with the emergence of the Semantic Web and of Web Services as a service-oriented infrastructure, is now increasingly capable of processing content automatically - it provides a distributed fabric for information processing.
Semantic Web ideas are relevant to hypermedia systems for all the reasons that they are relevant to any contemporary interoperable information processing system. But we conjecture that the relationship is deeper than that. To put it simply, the Semantic Web provides a range of mechanisms to express relationships between things (in a machine-processable way) - and of course hypermedia is also about expressing relationships between things. So in this workshop we ask: what is the relationship between hypermedia and the Semantic Web?
The Open Hypermedia community, with its view of links as "first class citizens", represents a significant strand in the hypermedia developments. It has met at least annually for many years, developing (for example) the Open Hypermedia Protocol (OHP) and its successor the Fundamental Open Hypertext Model (FOHM). These workshops have explored the world where hypermedia links are expressed and managed separately from the documents to which they refer. The traditional Web linking model, in which links are expressed in HTML, was contrary to this notion of open hypermedia. However, we conjecture that the Semantic Web approach, in which relationships are expressed separately through the Resource Description Framework (RDF) and ontologies, is highly relevant to open hypermedia - and indeed vice versa. It is for this reason that we have positioned this workshop to build on the Open Hypermedia Systems workshop series.
Our belief in the fundamental role of linking in the Semantic Web vision provides a starting point for discussion. Surprisingly, this aspect is not always apparent to either community. The Semantic Web research community is seen to be engrossed in issues of ontologies and knowledge representation, but the vision is also about distributed metadata which is interconnected through the entities it describes - a situation which should give opportunity for network effects. This metadata can be seen as a hyperstructure. Thanks to the expressive power of the representation techniques, there is a richness in describing these associations which far exceeds the traditional linking models of the Web and provides a facility for the richness of hypermedia visions.
Some informative explorations in this space are immediately apparent. How does the expressive power of OHP and FOHM compare with the Semantic Web, and can they be represented with ontologies? How do Semantic Web Services facilitate the engineering of Open Hypermedia Systems? Does the Semantic Web itself satisfy the definition of Open Hypermedia systems? What is the relationship between the inferencing mechanisms of the Semantic Web and the functionality of hypermedia link resolution systems? What is best practice when it comes to designing a contemporary hypertext application? How do the Semantic Web and Open Hypermedia visions play into new application areas such as pervasive computing and grid computing? How do the interactive aspects of hypermedia relate to Semantic Web?
It is evident to a number of researchers who bridge these two communities that a flow of ideas between the two would be valuable - we feel a need to "put the Open into Open Hypermedia" and to "put the Web into the Semantic Web". In 2001 we held a Semantic Web Panel at the hypertext conference in Aarhus, which served the purpose of raising awareness of these issues and opening the debate. The First International Hypertext and the Semantic Web workshop is the next step in bringing these ideas together.
The organisers are very pleased with the set of papers submitted to the workshop, which address many of the issues raised above and others. We look forward to the sharing and discussion of these ideas. This workshop is just a step along the way and we also look forward to this research being carried forward into future events. One of our goals for this event is to sketch the agenda for future work in this space and to discuss how best to take it forward.
I am very grateful to Dave Millard for his help in organising the workshop, bringing his valuable experience of years of involvement in the OHS workshop series. I am also grateful to our international reviewers Jim Hendler, Jacco von Ossenbrugen and Sean Bechhofer for their help and support, and for bringing their considerable expert knowledge of both the Semantic Web and hypertext worlds.
David De Roure
Southampton, August
2003