W3C home > Mailing lists > Public > public-lld@w3.org > October 2010

Re: VIAF contributor model

From: Richard Light <richard@light.demon.co.uk>
Date: Fri, 29 Oct 2010 17:21:45 +0100
Message-ID: <4vXRdYIZSvyMFwv1@light.demon.co.uk>
To: Antoine Isaac <aisaac@few.vu.nl>
Cc: public-lld@w3.org
In message <4CCAE706.6020200@few.vu.nl>, Antoine Isaac 
<aisaac@few.vu.nl> writes

>> (Somewhat answering my own question, I think that given this 
>>explanation, foaf:person would be used outside of the library data 
>>environment, so library authority data might link to other resources 
>>that focus on the person rather than the name. But I can't think of a 
>>place in library data would make reference to foaf:Person except, 
>>perhaps, in the administrative fields of the authority record.)
>I think all this is about being pragmatic about which stuff can be 
>re-used across different audiences/communities. Not every data in VIAF 
>can apply to a foaf:Person, but some of it can. From this, it seems a 
>bit a pity not to output that data that can be mapped to FOAF. Many 
>data consumers in the SW environment using FOAF could make use of it.
>Imagine that I maintain a database linking a foaf:Person Leonardo da 
>Vinci which connects him to many places he visited/lived in. But I 
>don't have the many versions of Leonardo's name which are in VIAF. In 
>such case I'd be very interested in importing the VIAF data with the 
>many names, or linking my data to the foaf:Person in VIAF via an 
>owl:sameAs. Of course if VIAF produces already bits of FOAF that would 
>make my task way easier.

I would strongly support this approach.  If VIAF mints a URI for the 
real-world person, and records what it knows about that person (which 
includes e.g. birth and death dates in many cases) in an RDF format 
which is widely understood, then there is potential added Linked Data 
value in both directions.  If the names are just recorded as names, and 
not linked to the concept of a real-world person, that opportunity is 
lost.  An obvious example of cross-domain linking would be pictures of 
Jane Austen in museums and galleries.

One thing puzzles me, though (and maybe I've just not been paying 
attention at the back of the class ...).  The 'SKOS' entry for Jane 
Austen (viaf/102333412/#skos:Concept) has no skos:preferredLabel, but 
includes copies of the three EstablishedHeading rdfs:label's as 
skos:altLabel's.  Is this because you can only have one 
skos:preferredLabel per language, and no attempt is made to specify 
language in VIAF?


Richard Light
Received on Friday, 29 October 2010 16:22:20 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:27:39 UTC