On 11/012 6:07 PM, Martin Hepp wrote:
> Effective immediately, the full GoodRelations vocabulary for e-commerce
> (http://purl.org/goodrelations/) is now directly available from schema.org,
> the official library of data schemas maintained and promoted by the four
> biggest Web search engines, i.e. Google, Bing, Yahoo, and Yandex. (01)
Congratulations! This is an important step toward bringing ontologies
into the mainstream of IT. (02)
I'd also like to point out that the representation of GoodRelations
in OWL is still available, but that Schema.org supports only the tiny
subset of OWL that most people actually use: (03)
From http://wiki.goodrelations-vocabulary.org/Cookbook/Schema.org
> GoodRelations will include mapping axioms so that an RDF environment
> with minimal reasoning support (rdfs:subclassOf, rdfs:subPropertyOf,
> owl:equivalentClass, owl:equivalentProperty, owl:sameAs) can operate
> on the data purely from the GoodRelations namespace. (04)
This is a minimal amount of logic that is compatible with almost
anything, including the legacy systems that have been successfully
interoperating for the past 50 years. (05)
Fundamental principles: (06)
1. Schema.org hits a very simple "sweet spot" in knowledge
representation that programmers and other IT professionals
can immediately understand without any special training. (07)
2. It is sufficiently underspecified that it does not create
any conflicts with more specialized languages and systems,
either declarative or procedural. (08)
3. Instead of URIs or IRIs that point to unreadable and unread
definitions somewhere on the WWW, Schema.org uses readable
names that are self-defining (or nearly so). (09)
By comparison, OWL hits a "sour spot" in knowledge representation: (010)
1. Nobody can use OWL without a training course or an extended
amount of self study. Even the very limited subset that most
people use has stumbling blocks that are inconsistent with
the conventions that programmers and webmasters know and use. (011)
2. Going beyond that limited subset requires a great deal
of further study, and anything specified in the superset
will not be automatically mapped into the mainstream
programming languages. In other words, it's useless. (012)
3. Using unreadable URIs or IRIs destroys readability. Users do
not want to and will never go to some remote location to study
a cryptic definition in some unreadable notation with some
unnecessarily complicated syntactic conventions. (013)
4. Finally, OWL is so limited in its expressive power that it
cannot be used by itself to implement applications. For any
practical application, OWL is always used in conjunction with
a more expressive, undecidable, Turing-complete language.
Forcing OWL itself to be decidable is of no practical value. (014)
There is a need for more expressive power in logic-based languages.
The SQL WHERE-clause shows that programmers are willing and able
to learn and use a query language with the full expressive power
of first-order logic. Various kinds of rule-based languages
show that people are able to learn and use the Horn-clause
subset of logic in practical applications. (015)
Recommendation: Schema.org is the wave of the future. Any work on
formal ontologies and the Semantic Web should embrace and build on
the simple versions specified in Schema.org. I recommend that the
W3C should work with Schema.org to make it an integral part of the
SW strategy. (016)
John Sowa (017)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (018)
|