semantic web
Introduction
Why "semantic web" in lowercase? Very simply, it is my belief that this concept has been "hijacked" to mean a very narrow implementation of certain technologies, rather than a very useful and broad approach to adding knowledgeable navigation to networked information.The "Sematic Web" is defined - fairly I think - by Tim Berners-Lee of the World Wide Web Consortium (W3C) as "an extension of the current web in which information is given well-defined meaning, better enabling computers and people to work in cooperation." In the W3C, however, it is limted to a narrow set of activities and specifications, notably using the Resource Description Framework (RDF) and the emerging specifications of the Web Ontology Language.
Although the W3C contribution has no doubt been very valauble, there is a growing sense that they are losing the plot. Whereas HTML - the markup language that drove the massive growth of the World Wide Web in the 1990s - is easy to explain, understand, use and deploy, the latest W3C work seems to be drifting into academic theory and research, with little hope of anything practical that can be deployed by your averagely web-literate user or information provider.
Whose sematic web?
It is my contention that if the "semantic web" is to be anything more than back-room gimmickery or a massively complex information space reserved for the well-resourced and powerful, there needs to be some serious debate about basic concepts of information organisation and aggregation, methods of association and linking, trust and assertion mechanisms, all underpinned by generalisable principles of logic and independent of any particular technology.
That is why I have invested time in looking at policies for driving the semantic web, rather than technologies. Most of them are very simple axioms, but taken together can ensure that information systems are developed upon a solid architecture. Some of these principles are outlined in different papers on (or linked to) this site, including the concept of "logical object".
Logical Objects
The "Uniform Resource Identifiers" may well be the "axiom of Web architecture", as described by Berners-Lee, but they are not the axiom of information architecture: the web is a network of addressable information resources available via the Internet, but information resources exist well beyond the confines of the Web or the Internet, and yet still play an important part of our understanding of "networked knowledge". Building an architecture on the quicksands of volatile network endpoints offers no long-term stability, however much the W3C campaigns for persistent URIs.
In my work, we have developed the concept of "logical objects" to convey the semantic construct that an average user would have of, say, a document. All links on this web site use this concept, pointing users to the "abstract" concept of a particular work, and letting the system resolve that abstract reference to a specific "representation" (this html page, a PDF file, etc.) according to need. The "logical object" is persistent, representations much less so, so why does the W3C insist on an architecture based on a fundamental design flaw?
More follows....