SIOC Changes and OWL-DL compliance
Alexandre Passant
We just made some changes to the SIOC Core ontology and to the related modules:
- Added OWL-DL compliance statements for SIOC Core and the Types /
Access / Services modules
- Edited owl:disjointWith statements for some classes of SIOC Core
- Removed domain of sioc:note
- Removed domain of sioc:has_owner and range of sioc:owner_of
- Defined sioc:account_of as inverse property of foaf:holdsAccount
- Defined sioc:avatar as a subproperty of foaf:depiction
So, SIOC is now OWL-DL !
This change was motivated by the current SWANSIOC integration project
[1] that will be introduced during the upcoming ISWC tutorial on
Semantic Web for Health Care and Life Sciences [2].
The SIOC Core Ontology Specification has been updated according to the
changes - http://rdfs.org/sioc/spec/
Best,
Alex.
[1] http://esw.w3.org/topic/HCLSIG/SWANSIOC
[2] http://iswc2008.semanticweb.org/tutorials/semantic-web-for-health-care-and-life-sciences/
Simon Reinhardt
On Oct 14, 12:42 pm, "Alexandre Passant" <a...@passant.org> wrote:
> - Defined sioc:account_of as inverse property of foaf:holdsAccount
instead of a foaf:OnlineAccount? Might as well make sioc:User and
foaf:OnlineAccount equivalent classes then, or?
> So, SIOC is now OWL-DL !
Simon
Matthias Samwald
> So, SIOC is now OWL-DL !
I hope that other important schemas/micro-ontologies such as SCOT,
SKOS, FOAF etc. could follow suit sooner or later...
-- Matthias
Sergio Fernández
> Doesn't that imply that foaf:holdsAccount always links to a sioc:User
> instead of a foaf:OnlineAccount? Might as well make sioc:User and
> foaf:OnlineAccount equivalent classes then, or?
Actually sioc:User is subClassOf foaf:OnlineAccount.
BTW, good work Alex!
--
Sergio Fernández - sergio.f...@fundacionctic.org
R&D Deparment
CTIC Foundation - www.fundacionctic.org
Phone: +34 984 29 12 12
Fax: +34 984 39 06 12
Edificio Centros Tecnológicos
Parque Científico Tecnológico
33203 Cabueñes - Gijón - Asturias - Spain
Simon Reinhardt
> On Tue, 2008-10-14 at 08:14 -0700, Simon Reinhardt wrote:
> > Doesn't that imply that foaf:holdsAccount always links to a sioc:User
> > instead of a foaf:OnlineAccount? Might as well make sioc:User and
> > foaf:OnlineAccount equivalent classes then, or?
>
> Actually sioc:User is subClassOf foaf:OnlineAccount.
foaf:OnlineAccount is a sioc:User.
Regards,
Simon
Sergio Fernández
> > Actually sioc:User is subClassOf foaf:OnlineAccount.
>
> That's exactly the problem, they're not equivalent. Not every
> foaf:OnlineAccount is a sioc:User.
That's right, but every sioc:User is a foaf:OnlineAccount ;-)
So, or I'm not understanding your doubts, or I can not see the problem
with this.
Cheers,
Simon Reinhardt
look at this example:
ex:Mary a foaf:Person ; foaf:holdsAccount ex:m .
Since the range of foaf:holdsAccount is a foaf:OnlineAccount, we can
infer:
ex:m a foaf:OnlineAccount .
But we also know that foaf:holdsAccount has the inverse property
sioc:account_of, so the following can also be inferred:
ex:m sioc:account_of ex:Mary .
Since the domain of sioc:account_of is sioc:User, ex:m is now also a
sioc:User.
So everytime someone uses the property foaf:holdsAccount it can be
inferred that the object of the statement is a sioc:User, if they want
or not. Basically this means making foaf:OnlineAccount and sioc:User
equivalent. So where's the point in having a sub-class of
foaf:OnlineAccount at all? How're they different? Inference-wise they
aren't really.
Simon
Olivier GENDRIN
<simon.r...@gmail.com> wrote:
> So everytime someone uses the property foaf:holdsAccount it can be
> inferred that the object of the statement is a sioc:User, if they want
> or not. Basically this means making foaf:OnlineAccount and sioc:User
> equivalent. So where's the point in having a sub-class of
> foaf:OnlineAccount at all? How're they different? Inference-wise they
> aren't really.
A foaf:OnlineAccount may not be part of an "Online Community", so not
relevant for SIOC...
--
Olivier G.
http://www.lespacedunmatin.info/blog/
Simon Reinhardt
wrote:
>
> > So everytime someone uses the property foaf:holdsAccount it can be
> > inferred that the object of the statement is a sioc:User, if they want
> > or not. Basically this means making foaf:OnlineAccount and sioc:User
> > equivalent. So where's the point in having a sub-class of
> > foaf:OnlineAccount at all? How're they different? Inference-wise they
> > aren't really.
>
> A foaf:OnlineAccount may not be part of an "Online Community", so not
> relevant for SIOC...
with the inferencing. Making sioc:account_of the inverse of
foaf:hasAccount practically merges the two classes.
Sergio Fernández
> A foaf:OnlineAccount may not be part of an "Online Community", so not
> relevant for SIOC...
Literally from [1]: "a foaf:OnlineAccount represents the provision of
some form of online service". And sioc:User specializes it for users of
online communities (a subset of online services).
Sorry, but I continue without understand your point... why is not
relevant for SIOC?
Cheers,
[1] http://xmlns.com/foaf/spec/#term_OnlineAccount
Olivier GENDRIN
<sergio.f...@fundacionctic.org> wrote:
> On Mon, 2008-10-20 at 16:26 +0200, Olivier GENDRIN wrote:
>> A foaf:OnlineAccount may not be part of an "Online Community", so not
>> relevant for SIOC...
>
> Literally from [1]: "a foaf:OnlineAccount represents the provision of
> some form of online service". And sioc:User specializes it for users of
> online communities (a subset of online services).
>
> Sorry, but I continue without understand your point... why is not
> relevant for SIOC?
>
Can't find an example right now, my answer was mostly theoretical...
Sorry...