> Comments from Fundación CTIC.
> ________________________________

> From: Luis Polo
> Subject: SIOC specification (CTIC)

> Dear Uldis,

> I am Luis Polo from CTIC Foundation (Spain). We have received the invitation
> to participate in the SIOC Ontology specification to submit it to the W3C.
> We have analyzed the initiative and we consider it is a useful idea to
> interlink Web Communities. We would like to support the W3C member
> submission. We have two comments:

> 1) Fisrt of all, Sergio Fernandez, a researcher of our department, has
> developped an application, as a research project around the semantic web
> technologies, to publish mailing lists´s archives into an RDF format. This
> project is called SWAML. The SWAML vocabulary is built upon FOAF and Dublin
> Core and is completely compatible with SIOC specification, as you can see it
> as a subset of SIOC elements. If you consider it could be a useful
> application of SIOC, we can adapt quickly SWAML vocabulary to SIOC and
> include the software application as an actual implementation of SIOC. For
> more details, you can see http://swaml.berlios.de.

> 2) About SIOC specification, we have one comment. SIOC is a RDF vocabulary
> to describe Online Communities. It has a lot of links with FOAF (and Dublin
> Core), so the domain or range of some properties is for example foaf:Person.
> Also, the class sioc:User is subclass of foaf:Person. We think this is a
> good idea: reusing concepts for another vocabularies. The problem is that
> there are some SIOC elements that are semantically identical to some FOAF
> and Dublin Core elements but with another URI. So for example, SIOC rewrites
> the property dc:creator as sioc:has_creator or foaf:mbox_sha1sum as
> sioc:email_sha1. This could be a problem, so you have two syntactic
> different elements with the same associated semantics. In SIOC, as sioc:User
> is subclass of foaf:Agent, it is also a subclass of foaf:Agent (by
> transitivity), so sioc:User would have the property foaf:mbox_sha1sum. Of
> course, it will have also sioc:email_sha1. In RDF Universe it is not
> contradictory, but redundant (like reinventing the wheel).

> We suggest to identify these conflictive elements, eliminate them and then
> add their FOAF and Dublin Core counterparts. If you have any comments, feel
> free to contact us.

> Best Regards,

> p.

> Luis Polo Paredes
> Departamento I+D+i
> Fundación CTIC
> -Centro Tecnológico de la Información y la Comunicación-
> Parque Científico Tecnológico Gijón-Asturias-Spain www.fundacionctic.org