[GeoNetwork-devel] Thesaurus / SKOS for keyword in GN

Thanks a lot for the link Jo, I'll have a closer look to sioc.

Using SKOS, it could be quite easy to improve the region list for example.
In the current GN region list we could identified maybe 3 main concepts
which are :
-- country (eg. Algeria, ...)
-- ~ continent (eg. Eastern Africa, ...)
-- sea area (eg. All fishing areas, ...)

Then we could link countries to the "country" concept, same for the "sea
area" concept.
Once done, we could provide a more user friendly interface to user for
keyword selection (place and theme keyword). Users could browse to get into
the list of concepts and choose the appropriate term.

Thesaurus like Agrovoc or Gemet have a lot more relationships than the GN
region list we have.
You could also provide "related", "broader" and "narrower" term in the
interface ...

For example, user could select
-- sea area
--- Mediteranean sea
And access also to all "related" countries (Algeria, Spain, Italy, ...).

Cheers. Francois

-----Message d'origine-----
De : Jo Walsh [mailto:jo@anonymised.com]
Envoyé : jeudi 21 septembre 2006 15:14
À : François Prunayre
Cc : geonetwork-devel@lists.sourceforge.net
Objet : Re: [GeoNetwork-devel] Thesaurus / SKOS for keyword in GN

dear Francois, all,
On Thu, Sep 21, 2006 at 12:08:59PM +0200, François Prunayre wrote:

Hi list, on the sourceforge feature requests Roberto Ledda post the
integration of thesaurus interface in order to manage keywords in GN
in february 2006.
We plan to develop such an interface using SKOS
(http://www.w3.org/2004/02/skos/core/#data
<http://www.w3.org/2004/02/skos/core/#data).> ).

Have you looked at SIOC? I think the two are complementary; SIOC is in the
same kind of space as FOAF - an attempt to connect the harder core ontology
work to real-world use cases and casual developers. So it's a bit more
'alive' than SKOS is, many of the same protagonists.

http://rdfs.org/sioc/spec/
http://sioc-project.org/
http://sioc-project.org/node/113 <- some sioc/skos integration.

This would be for stage 2 of the discovery cycle that goes, "exchange
metadata .... who are you ... exchange data."

There's so much to be gained in terms of ontology mapping, language
translation, by using URIs instead of strings for keyword labels...

At this point I would like to travel back in time and explain why I am here
and put in a couple of hours getting the metadata model we've been designing
for internal use at OSGeo back in sync with itself, show you the code and
ask sensible questions about being able to plug a PostGIS based backend
(which has simple OO python interfaces) into GeoNetwork and 'leverage'
GeoNetwork for its standards compliant interfaces... but this is going to
have to wait a little while! :confused:

cheers,

jo

--
Ce message a ete verifie par MailScanner pour des virus ou des polluriels et
rien de suspect n'a ete trouve.

Les donnees et renseignements contenus dans ce message sont personnels,
confidentiels et prives. Toute publication, utilisation ou diffusion, meme
partielle, doit etre autorisee.

Any data and information contained in this electronic mail is personal,
confidential and secret. Any total or partial publication, use or
distribution must be authorized.

--
Ce message a ete verifie par MailScanner pour des virus ou des polluriels et rien de suspect n'a ete trouve.

Les donnees et renseignements contenus dans ce message sont personnels, confidentiels et prives. Toute publication, utilisation ou diffusion, meme partielle, doit etre autorisee.

Any data and information contained in this electronic mail is personal, confidential and secret. Any total or partial publication, use or distribution must be authorized.