1. Use (dcterms) dc:subject as the way of pointing from a document to it’s SKOS subject.
Jack
> --
> You received this message because you are subscribed to the Google Groups
> "SIOC-Dev" group.
> To post to this group, send email to sioc...@googlegroups.com.
> To unsubscribe from this group, send email to
> sioc-dev+u...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/sioc-dev?hl=en.
>
>
only q i'd have is dc or dcterms
regards,
Jack Park wrote:
> +1
> My limited view is that proliferation of property types seems worrisome.
>
> Jack
>
> On Thu, Jan 14, 2010 at 1:41 PM, Stephane Corlosquet
> <scorl...@gmail.com> wrote:
>> Hi,
>>
>> I wish I had received danbri's feedback on RDF in Drupal earlier, but here
>> it is http://danbri.org/words/2010/01/14/549
>>
>> Given that we're one day before alpha release, there isn't much we can do.
>> The only quick change I can imagine possible is:
>>
>>> 1. Use (dcterms) dc:subject as the way of pointing from a document to it�s
+1 as well; personal view being use what people will sparql at a later
date; and due to the widespread usage of dc:subject odds are when
querying the cloud it'll be used more often.
only q i'd have is dc or dcterms
Hi,
I wish I had received danbri's feedback on RDF in Drupal earlier, but here it is http://danbri.org/words/2010/01/14/549
Given that we're one day before alpha release, there isn't much we can do. The only quick change I can imagine possible is:
1. Use (dcterms) dc:subject as the way of pointing from a document to it’s SKOS subject.
sioc:topic has been on the RDF schema proposal since March 2008: http://groups.drupal.org/node/9311 and I wish someone had suggested this change earlier :( I can quickly roll a patch, but before I go ahead, let me ask: sioc:topic being a subproperty of dc:subject, what is the difference between the two? Why creating sioc:topic at the first place?
If I can have a few +1/-1 for this change with maybe some explanation of the difference, we can hopefully get the patch committed.
Steph.
you could use both.. triples are pretty lightweight in a quad store
--
You received this message because you are subscribed to the Google Groups "SIOC-Dev" group.
To post to this group, send email to sioc...@googlegroups.com.
To unsubscribe from this group, send email to sioc-dev+u...@googlegroups.com.
Jack
Just my 2 cents,
Axel
--
--
"For what it's worth, I always aim to please..." (Placebo)
Consensus seems to be for dc:subject...
When we created sioc:topic we did so because dc:subject was being used
ambiguously to refer to a literal or a resource depending on who used
it.
Often times, dc:subject referred to a (literal) keyword like a tag or
category name. We wanted sioc:topic to always point to a resource and
not a literal. Assuming that dc:subject could be anything (resource,
literal) we made sioc:topic a sub-type so we would always point to a
resource, in particular, a tag or category as you would have in social
software systems.
Regarding Person vs. User - we still need to vote as to whether to
change User to UserAccount to clarify things. Am happy to do this if
others still agree.
John.
I'm finally OK with that.
We need to ensure that both can still live together during the transition between both.
owl:equivalentClass + deprecation of sioc:User might work, I'm just wondering it the semantics of owl:equivalentClass will imply that sioc:UserAccount is also deprecated (as with an owl:sameAs) or if that's different ?
Alex.
>
> John.
> --
> You received this message because you are subscribed to the Google Groups "SIOC-Dev" group.
> To post to this group, send email to sioc...@googlegroups.com.
> To unsubscribe from this group, send email to sioc-dev+u...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/sioc-dev?hl=en.
>
>
--
Dr. Alexandre Passant
Digital Enterprise Research Institute
National University of Ireland, Galway
:me owl:sameAs <http://apassant.net/alex> .
I've actually developed a rich set of predicates for internal use in
the next product I'm building. Although my data ~can~ be described w/
SIOC, I'm planning on exposing a subset of my internal vocabulary to
the outside because it describes my content more precisely. I suppose I
could assert that my properties are sub-properties of SIOC (and other
properties) with OWL, but I want to make things easy and do the
inference for you. If you don't know what to do w/ my properties you
can just ignore them.
As a "linked data consumer", I take the converse approach.
Just as we're never going to get people to use owl:sameAs correctly,
we can't assume complete standardization: a consumer ought to accept
dc:Subject, sioc:Topic and other competing predicates and do what it
takes to make sense of what comes in.
I'd add that this distinction is really important, and that if it can be modelled later in the beta or beyond, it would be a boon!
Question: in our cck approach
We exported both site vocabulary
properties and there mapped super-properties from external vocabs...
Would that be an option here as well (expose both dc:subject and sioc:topic)?
you could use both.. triples are pretty lightweight in a quad store
--
You received this message because you are subscribed to the Google Groups "SIOC-Dev" group.
To post to this group, send email to sioc...@googlegroups.com.
To unsubscribe from this group, send email to sioc-dev+u...@googlegroups.com.