Re: VIAF contributor model

On Mon, Nov 1, 2010 at 9:51 AM, Dan Brickley <danbri@danbri.org> wrote:
> On Mon, Nov 1, 2010 at 2:40 PM, Ross Singer <ross.singer@talis.com> wrote:
>> On Mon, Nov 1, 2010 at 6:31 AM, Haffner, Alexander <A.Haffner@d-nb.de> wrote:
>>
>>> However, back to the formats I don’t want to discuss J foaf doesn’t have the
>>> power to reflect our comprehensive data – I thought we want to make this
>>> high quality data available for the public –if so we should have a closer
>>> look modeling the data in FRBRer, FRAD and/or RDA in parallel to the SKOS
>>> representation.
>>>
>> I keep seeing this statement getting made: "FOAF/SKOS are not
>> expressive enough for our data" and I'm simply not buying it.
>>
>> Can somebody please back up this claim?  FOAF defines personal and
>> organizational entities.  SKOS defines concepts.
>>
>> Those are exactly the things we're describing.
>
> The design of RDF reflects this situation - typically no single RDF
> vocabulary captures all use cases and needs. If we can agree on the
> basic layout in terms of common classes, that gives a skeleton for
> interoperability, fleshed out (oh dear, excuse the metaphor) with more
> detailed precise properties from different application domains. So RDF
> is a design for sharing out the descriptive work...
>
> I'll repeat the earlier offer - if there are people/org/agent
> properties that are generally useful, and needed by several properties
> here, I'm happy getting them added to FOAF (if that's not treading on
> any FR** toes, of course).

+1

I imagine there will plenty of properties that would not be of general
interest outside of our community (although certainly the opposite
will probably be true, as well), but I think it makes *much* more
sense to align any vocabularies we need to mint ourselves to be built
on top of FOAF (or dcterms:Agent - as long as we're not creating new
classes, this should still give us interoperability with FOAF), SKOS,
dcterms, etc.

That is, I feel like we should be creating new classes (and
properties, even) *only* in scenarios where nothing exists that
defines the same sort of entity and, if there is nothing, only after
we have determined whether or not we're holding on to some legacy
worldview based on how we've always done it.

I would argue that treating the data in our 100/110/700/710/800/810
fields (I'm leaving 600/610 out of this, because there's plenty of
room for debate there) as some sort of bibliographic entity rather
than people or organizations gives us nothing and simply makes our
data hard for others to reuse.

-Ross.
>
> cheers,
>
> Dan
>

Received on Monday, 1 November 2010 14:07:37 UTC