Revisions to the SIOC Ontology - Impact

2006, May 12 - 04:34 — CaptSolo

Main impact of SIOC changes:

(1) SIOC Exporters

Introduction of foaf:Person - changes how information about
sioc:User(s) is represented.
Yet this change is a welcome one as this will allow to use SIOC data
with applications that already understand FOAF.

sioc:title to be instead of sioc:subject for post title - impact on
Drupal, .Clear.
sioc:type - used in .Clear exporter, but I can't tell if there are
exporter changes needed due to this
sioc:name - used in all 3 exporters for things that are not User or Usergroup

This sums up the impact noticeable on the first glance. You as
exporter developers will know better.

(2) Using SIOC Data (and "Smushing")

This is a more general question - how do we ensure, when RDF data from
different sources are mixed together, that we do not loose relation
between particular data of a person (name, surname, e-mail) and the
account that this information has been registered for.

While all the information is within one file or comes from one site
everything is fine - we know that there is no external information. So
one answer can be - while we keep the provenance (info about the
origin) of the data we are fine.

If not, then the link may be lost.

While this has not been solved we are keeping sioc:email and
sioc:email_sha1 though they could easily be replaced by appropriate
properties of foaf:Person.

Best regards,
Uldis Bojars

[ http://captsolo.net/info/ ]
[ CaptSolo @ #foaf and #swig ]

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "SIOC-Dev" group.
To post to this group, send email to sioc-dev@googlegroups.com
To unsubscribe from this group, send email to sioc-dev-unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sioc-dev
-~----------~----~----~----~------~----~------~--~---

Developers

2006, May 30 - 09:46 — cgoern@googlema...

Revisions to the SIOC Ontology - Impact

And here is another one...

Uldis Bojars wrote:
> Main impact of SIOC changes:
> (2) Using SIOC Data (and "Smushing")
> While this has not been solved we are keeping sioc:email and
> sioc:email_sha1 though they could easily be replaced by appropriate
> properties of foaf:Person.

Maybe it is an idea to have sioc mapping define sioc:email_sha1
owl:sameAs foaf:mbox_sha1 ? Or wouldnt that solve smushing problems?

Christoph

2006, May 30 - 09:46 — CaptSolo

Revisions to the SIOC Ontology - Impact

On 5/18/06, cgoern@googlemail.com wrote:
>
> > Main impact of SIOC changes:
> > (2) Using SIOC Data (and "Smushing")
> > While this has not been solved we are keeping sioc:email and
> > sioc:email_sha1 though they could easily be replaced by appropriate
> > properties of foaf:Person.
>
> Maybe it is an idea to have sioc mapping define sioc:email_sha1
> owl:sameAs foaf:mbox_sha1 ? Or wouldnt that solve smushing problems?

That would not help. Here's why:

1) According to the update to the SIOC ontology [ sioc:User
rdfs:subClassOf foaf:OnlineAccount ] because an online account is the
closes FOAF concept for a user on an online community account.

Hence you may not define [ sioc:email_sha1 owl:sameAs foaf:mbox_sha1 ]
because the domain of [ foaf:mbox_sha1 ] is foaf:Agent and
foaf:OnlineAccount (and sioc:User) is not a subclass of foaf:Agent.

2) But if you did define it, that would mean that sioc:User has an
inverse functional property [ sioc:email_sha1 ] (defined as IFP for
foaf:mbox_sha1) and when smushing all sioc:User(s) from different
sites would be merged together (if the person has provided same e-mail
when registering on these sites).

So we will end up with the same problem - that the data we do not want
to get mixed together will be mixed together.

That still leaves 2 options:

a) keep sioc:email (and email_sha1) properties of sioc:User and do not
define them as IFPs. this will ensure we will know the e-mail address
that this particular account was registered with.

b) assign all person's properties (including e-mail) to the
foaf:Person linked to sioc:User via foaf:holdsOnlineAccount. then it
is up to the users of SIOC data to keep the record of where the data
came from (provenance) in order to keep track of what personal
properties are accociated with a particular sioc:User.

I do not have a ready solution for this (but maybe it is not a big
problem after all?) since one may say that the same problem is true
for other properties that we will use to describe a person with and
for FOAF data in general.

Let's say - one site uses a first name / last name combination
'Captain' / 'Solo' and the other has 'Uldis' / 'Bojars'. After
smushing there will be a person with 2 first names and 2 last names
and no idea in what combination those shall be used.

(Hmm - if we do not export a FOAF property like foaf:mbox that the
data will be smushed against, then maybe there is nothing to worry
about?)

CC: RDFWeb-Dev [ http://rdfweb.org/pipermail/rdfweb-dev/ ]
CC: SIOC-Dev [ http://groups.google.com/group/SIOC-Dev/ ]

Uldis

[ http://captsolo.net/info/ ]

2006, May 30 - 09:45 — cgoern@googlema...

Revisions to the SIOC Ontology - Impact

Hello,

> (2) Using SIOC Data (and "Smushing")
> While this has not been solved we are keeping sioc:email and
> sioc:email_sha1 though they could easily be replaced by appropriate
> properties of foaf:Person.

I have been manually browsing throught what comes out of the sioc
plugin and have noticed that some information is not properly linked
(maybe, I am not that much of an RDF expert):

While retrieving data from
http://b4mad.net/datenbrei/index.php?sioc_type=post&sioc_id=291 it says
the Post has_containter f-1 (rdf:nodeID="f-1") f-1 is not definded in
that file, so I pulled in
http://b4mad.net/datenbrei/index.php?sioc_type=site which has a
sioc:Forum with rdf:nodeID="f-1"

Letting cwm get both URL the combined graph has no relation
(has_container) for any soic:Post

Shouldn't the plugin in post mode not refernce the plugin's URL in site
mode + "f-1" ? Something like

rdf:ID="http://b4mad.net/datenbrei/index.php?sioc_type=site#f-1"/>

That would also require that Forums are no longer blank nodes.

2006, May 30 - 09:46 — CaptSolo

Revisions to the SIOC Ontology - Impact

Fixed in v1.15.

sioc:Forum is now http://...?sioc_type=site#weblog
sioc:Usergroup is now http://...?sioc_type=site#authors

I would like to use nicer URLs for these, but can't find good
candidates.

One option would be to use http://blog/sioc/authors for the authors
list.
For the #weblog it might be natural to use the URL of the blog itself,
but this URI is already given to sioc:Site.

Uldis

2006, May 30 - 09:46 — CaptSolo

Revisions to the SIOC Ontology - Impact

cgoern@googlemail.com wrote:

> Shouldn't the plugin in post mode not refernce the plugin's URL in site
> mode + "f-1" ? Something like
>
> > rdf:ID="http://b4mad.net/datenbrei/index.php?sioc_type=site#f-1"/>

That's a bug in exporter. Added it to the bug / to-do list [1].

When we give it a URI we may as well it a real URI - for example - the
URI of main page of the blog. That, however, will create a URI "clash"
because sioc:Site already uses that URI.

[1] http://esw.w3.org/topic/SIOC/ToDoList

Best,
Uldis

[ http://captsolo.net/info/ ]

Revisions to the SIOC Ontology - Impact

Revisions to the SIOC Ontology - Impact

Revisions to the SIOC Ontology - Impact

Revisions to the SIOC Ontology - Impact

Revisions to the SIOC Ontology - Impact

Revisions to the SIOC Ontology - Impact

Primary links

Syndicate

User login