Information Technology Reference
In-Depth Information
on the Semantic Web, ranging from pictures to its height, encoded in a knowledge
representation language available via a URI for the Eiffel Tower.
As the previous exposition of Web architecture explained in detail, resources on
the Web are given by a URI that identifies the same content on the Web across
different encodings. What drives the Semantic Web is the realization that at least
some of the information on the Web is representational, i.e. information about distal
content. Then instead of HTML, which is mainly concerned with the presentation
and linking of natural language for humans, the Web needs a knowledge repre-
sentation language which describes the represented content as fully as possible
without regard to presentation for humans. The mixture of content and encodings
for presentation forces web-spiders to “scrape” valuable content out of hypertext.
In theory, encoding information directly in a knowledge representation language
gives a spider more reliable and direct access to the information. As Berners-Lee
puts it, “most information on the Web is designed for human consumption, and
even if it was derived from a database with well defined meanings (in at least
some terms) for its columns...the structure of the data is not evident to a robot
browsing the Web” (1998b). This has led him to consider the Semantic Web to be
a Web “for expressing information in a machine processable form” and so making
the Web “machine-understandable” (Berners-Lee 1998b). This leads to the contrast
between the Semantic Web as a 'web of data' as opposed to the hypertext 'web
of documents.' W3C standards such as XML were originally created, albeit rarely
used, precisely in order to separate content and presentation (Connolly 1998).
Furthermore, the purpose of the Semantic Web is to expand the scope of
the Web itself. Most of the world's digital information is not natively stored in
hypertext. Instead, it is stored in databases and other non-hypertext documents
and spreadsheets. While this information is slowly but surely migrating towards
the Web, as more and more of this information is being exposed to the Web via
scripts that automatically and dynamically convert data from databases into HTML,
the Semantic Web imagines that by having a common knowledge representation
language across the entire Web, all sorts of information that previously were not on
the Web can become part of the Web. This makes the Semantic Web not a different
and parallel Web to the hypertext Web, but an extension of the current Web, where
hypertext serves as just one possible language.
3.1
A Brief History of Knowledge Representation
The creation of the Semantic Web then depends on the creation of at least one
(if not multiple!) knowledge representation language for the Web, and so the
Semantic Web inherits both the successes and failures of previous efforts to create
knowledge representation languages in artificial intelligence. The earliest work in
digital knowledge representations was spear-headed by John McCarthy's attempts
to formalize elements of human knowledge in first-order predicate logic, where
the primary vehicle of intelligence was to be considered some form of inference
Search WWH ::
Custom Search