Re: ANN: Sudoc bibliographic ans authority data

Le 11/07/2011 11:12, Dan Brickley a écrit :
> On 7 July 2011 23:17, Yann NICOLAS<nicolas@abes.fr>  wrote:
>> Bonjour,
>>
>> Sudoc [1], the French academic union catalogue maintained by ABES [2], has
>> just been released as linked open data.
>>
>> 10 million bibliographic records are now available as RDF/XML.
>>
>> Examples for the Sudoc record whose internal id is 132133520 :
>> . Resource URI : http://www.sudoc.fr/132133520/id
>> . Generic document : http://www.sudoc.fr/132133520 (content negotiation is
>> supported)
>> . RDF/XML page : http://www.sudoc.fr/132133520.rdf
>> . HTML pages with schema.org microdata [3] for search engines :
>> http://www.sudoc.fr/132133520.html . The users are not supposed to visit
>> these microdata pages : they are redirected to the standard UI :
>> http://www.sudoc.abes.fr/xslt/DB=2.1/SRCH?IKT=12&TRM=132133520
>>
>> Sudoc RDF data are linked to http://lexvo.org and http://dewey.info/ .
>>
>> They are also linked to IdRef [4], ie the Sudoc authority file that ABES
>> considers as a separate and open application.
>> 2 million IdRef records are also available as RDF data (since October 2010).
>> The links between Sudoc and IdRef are bidirectional.
>> For example, http://www.sudoc.fr/110404416/id ( Rethinking symbolism by Dan
>> Sperber ) links to D. Sperber's IdRef URI: http://www.idref.fr/027146030/id
>> .
>> But, in the other direction, http://www.idref.fr/027146030/id links to *all*
>> the Sudoc documents that are linked to this authority.
>>
>> In next months, we hope to add more links to our data, to OCLC and BnF
>> resources among others.
>>
>> More info (in French) here : http://punktokomo.abes.fr/
> Congratulations, this is fantastic nes.

Hi Dan,

Thanks !

> And I think also a very timely
> test-case for how community-maintained and consortium-based standards
> (schema.org) can be deployed alongside each other.
As this thread shows, this bi-face publishing of data is (legitimately) 
controversial.
Someone has to experiment...
We'll see how it goes...


>
> Could you say a little more about the subject classification aspects
> of this data? I don't know a lot about French cataloguing.  In the
> sample URIs you give above, I find only Rameau. You mention also
> Dewey.info, so I guess there's Dewey in there.
Here is an example with Dewey.info :
http://www.sudoc.fr/004471245.rdf

Some more in Sindice (http://dewey.info/class/330/):
http://sindice.com/search?q=http%3A%2F%2Fdewey.info%2Fclass%2F330%2F


>   And Rameau also has
> some mappings to LCSH.

Rameau is maintained by BnF. [1]
Sudoc loads and uses a copy of Rameau.
IdRef [2], our application dedicated to authority data, publishes HTML 
and RDF pages for Rameau :

    Generic document : http://www.idref.fr/035798734
    RDF document : http://www.idref.fr/035798734.rdf
    <http://www.idref.fr/035798734>

We are eager to add owl:sameAs links to an official and up-to-date 
Rameau-in-SKOS dataset. By now, such an official dataset does not exist, 
despite Antoine Isaac's bright experimental work [3]. I hope it will. I 
am sure it will (just my feeling) but i don't know when :)

[1] http://rameau.bnf.fr/
[2] http://www.idref.fr
[3] http://www.cs.vu.nl/STITCH/rameau/

>   Are there other schemes?
I should have mentioned that we link to FMeSH as well. [4]
FMeSH is the French translation of MeSH.

Example    Example        Example :
             http://www.idref.fr/040697916.rdf

This record links to http://purl.org/commons/record/mesh/D004454
But these MeSH PURLS seem to be experimental.
Are there official MeSH URIs maintained bu NLM (US National Library of 
medecine) ?

[4] 
http://extranet.inserm.fr/information-scientifique-et-technique/le-mesh-bilingue

>   e.g. I'm interested in
> particular to find instance data for UDC and for Library of Congress
> Classification (LCC), but also anything else that has a SKOS

Even if Rameau is preferred in our Sudoc context, we have many MARC 
records containing UDC codes or LCC terms.
We should add this information in our RDF bibliographic data.

I hope it helps

Cheers
Yann

> expression.
>
> Thanks for any more info,
>
> cheers,
>
> Dan
>
> ps. some Gremlin examples follow (see
> http://danbri.org/words/2011/05/10/675 ) ... it uses the Linked Data
> Sail to pull in pages on demand from the Web, as you explore into the
> graph.
>
> g = new LinkedDataSailGraph(new MemoryStoreSailGraph())
> i1 = g.v('http://www.sudoc.fr/132133520/id')
>
> gremlin>  i1.out('dcterms:subject').out('skos:inScheme')
> ==>v[http://stitch.cs.vu.nl/vocabularies/rameau/autorites_matieres]
> ==>v[http://stitch.cs.vu.nl/vocabularies/rameau/autorites_matieres]
> ==>v[http://stitch.cs.vu.nl/vocabularies/rameau/autorites_matieres]
> ==>v[http://stitch.cs.vu.nl/vocabularies/rameau/autorites_matieres]
> gremlin>  i1.out('dcterms:subject').out('skos:prefLabel')
> ==>v["T?l?communications"@fr]
> ==>v["Th?ses et ?crits acad?miques"@fr]
> ==>v["Nouvelles technologies de l'information et de la communication"@fr]
> ==>v["Internet"@fr]
> ==>v[""]
>
>
> gremlin>  i2=g.v('http://www.sudoc.fr/110404416/id')
> gremlin>   i2.out('dcterms:subject').out('skos:inScheme')
> ==>v[http://stitch.cs.vu.nl/vocabularies/rameau/autorites_matieres]
> ==>v[http://stitch.cs.vu.nl/vocabularies/rameau/autorites_matieres]
> gremlin>
> gremlin>  i2.out('dcterms:subject').out('skos:prefLabel')
> ==>v["Signes et symboles"@fr]
> ==>v["Anthropologie"@fr]
>
>
>
>
>
>> [1] http://www.sudoc.abes.fr
>> [2] http://www.abes.fr
>> [3] Shame on us ;) (twice)
>> [4] http://www.idref.fr


-- 
--
Yann NICOLAS
Etudes&  Projets
ABES