Introduction
The Semantic Web
is much more than a simple buzzword to entertain specialized medias.
The Article
published in Scientific American (may 2001) in particular by
Tim Berners Lee
(the famous WWW father), is considered as the Semantic Web founder's article.
(I you want to know more about the Semantic Web Goals, I recommend you
to read it)
Semantic Web Technologies (at least until the "logic" part of the "layer cake" (The two last part are very ambitious and shouldn't be available before a couple of years)) are today enough mature to be used by everyone.
My prototype
I conceived and developped an prototype allowing to do dynamic
Web Services Match-Making by only using semantic informations.
The match-making permits to discover dynamically a coherent Web
Services Workflow. By "Workflow", I mean something like
BPEL.
So the set of Web Services discovered are compatibles between them
and can be executed sequentially.
If you want to know more, you can download
my paper.
(in french but you'll find some source code and diagrams)
This dynamic discovery is only possible because we're doing
a "semantic enrichment" of the Web Services.
This enrichment is made with the knowledge domains formalized inside
Ontologies.
Here, Ontolologies I've developped with Protégé
was OWL (Lite)
compliant.
OWL(Web Ontology Language) Lite was sufficient and the development of
a OWL DL Ontology will have took to much time and didn't add something more
to the demonstration.
By the way, I think Web Semantic won't be a reality until
open-source (or not) Ontologies covering a large range of topics
will be available to everyone.
In the future, we will certainly see the emergence same of the same
system as the one use by the root DNS servers but dedicated to Ontologies.
The prototype architecture :
