purl, vocabularies and ivoa.net

Frederic V. Hessman hessman at astro.physik.uni-goettingen.de
Thu Jun 1 08:04:55 CEST 2017


Hi everyone!

A few thoughts about central curation…

> On 1 Jun 2017, at 00:18, Mireille Louys <mireille.louys at unistra.fr> wrote:
> 
> There is also a need to discover more easily where to find the various vocabularies proposed by the Ivoa .
> I would appreciate to have a link from the IVOA document repository page directly , to a IVOA Vocabulary page like what we have for XML schemata, for instance.
> 
> Various vocabularies already exist , in various formats and are used in DAL( datalink, RDF, html), Registry (VOResource litterals, CSV), Theory ( theory vocabulary RDF/SKOS )
> So this page would actually redistribute the link access to the corresponding locations .
> Each of these vocabularies should be maintained in close contact to their WG and users , and I see the role of Semantics more as a coordinator and curator for the long term. 

Exactly.

> How to organize the vocabulary tree , is still to be decided .
> ivoa.net/rdf  does not mean every file in this directory should be in rdf format but this is somewhat misleading.
> In order to allow to gather all representation formats for some particular vocabulary, I would prefer to have one branch for each vocabulary and the various formats below like : 
> ivoa.net/vocabularies/datalink/rdf/datalink-1.0-20151107.rdf
> ivoa.net/vocabularies/datalink/csv/datalink-1.0-20151107.txt
> ...
> ivoa.net/vocabularies/theory/skos/vocabForTheory-1.0-20160101.rdf
> ivoa.net/vocabularies/theory/skos/vocabForTheory-1.0-20170509.rdf
> etc...
> and the same for registry vocabularies if needed. 
> Each of them could be redirected to another site as purl.obspm.fr or anyother.

The network should be intelligent enough to reduce the number of directories to a minimum, since the user presumably knows what format it’s looking for, e.g.

	ivoa.net/vocabularies/{vocabulary-root-name}/*.rdf, *.txt, *.xml,…..

The external vocabularies will presumably have their own naming conventions, but it would make it easier to use the central collection if these are all labelled the same, e.g. using Mireille’s notation...

	{vocabulary-root-name-without-hyphens}-{version#}.{sub-version#}-{yyyymmdd}.{format-suffix}

That way customers can easily look for the latest version, knowing how to parse the filenames.

It would also be good to know that the vocabularies will always be available in one or two standard formats, no matter how the external parties created them.

Rick






More information about the semantics mailing list