RE: Results of survey "Do controlled vocabularies matter?"

We'd also be interested in discussing potential for standardised SKOS APIs and agree that a standard API could potentially help reduce development effort for SKOS applications.



This has been a recurring theme over the last few years - the SWAD Europe project designed a SKOS API in the early days. We reviewed some of this work recently, along with some of our own developments in a paper for Knowledge Organization 37(4), 2010.
There's a preprint at
http://hypermedia.research.glam.ac.uk/media/files/documents/2010-12-16/cbinding_dtudhope_KO_rev7.pdf

regards

Doug


Douglas Tudhope
Professor, Faculty of Advanced Technology
University of Glamorgan
Pontypridd CF37 1DL
Wales, UK
Tel +44 (0) 1443-483609
Fax +44 (0) 1443-482715
dstudhope@glam.ac.uk
http://hypermedia.research.glam.ac.uk/people/tudhope/
Editor : The New Review of Hypermedia and Multimedia

________________________________________
From: public-esw-thes-request@w3.org [public-esw-thes-request@w3.org] on behalf of Christophe Dupriez [christophe.dupriez@destin.be]
Sent: 27 June 2011 09:43
To: Andreas Blumauer (Semantic Web Company)
Cc: Simon Cox; holger@topquadrant.com; public-esw-thes@w3.org
Subject: Re: Results of survey "Do controlled vocabularies matter?"

Hi Andreas,

I am definitively willing to implement this kind of "shortcuts" in
ASKOSI.org

We must then also standardize the structured result formats
     (RDF, JSON : SparQL results? Which columns are included in the result?)
... and respect REST API principles ...

I would also prefer to avoid "all" and stick to SKOS terminology:
http://vocabulary.semantic-web.at/semweb/131/narrower

Have a nice day!

Christophe


Le 27/06/2011 10:19, Andreas Blumauer (Semantic Web Company) a écrit :
> Same with PoolParty Thesaurus Management [1]:
> We provide "stored procedures" and can offer a variety of pre-defined SPARQL queries,
> which can execute queries mentioned by Simon.
>
> But this is not enough: If developers could execute at any SPARQL endpoint, e.g. [2]
> queries like
>
> http://vocabulary.semantic-web.at/PoolParty/sparql/semweb/allConceptSchemes
>
> or for any given concept, e.g. [3]
>
> queries like
>
> http://vocabulary.semantic-web.at/semweb/131/allNarrowers
> http://vocabulary.semantic-web.at/semweb/131/allRelateds/prefLabel="foo"
>
> in a STANDARDIZED way, then developers don´t have to care about anymore
> if the underlying technology is PoolParty, TopBraid or something else AND
> they don´t have to learn Sparql but rather could concentrate on a vocabulary´s inherent scheme&  logic,
>
> and exactly this is necessary to get things going also for non-SemWebber in my opinion.
>
> If anyone is interested in defining such queries, please let me know!
>
> Cheers,
> Andreas
>
> [1] http://poolparty.punkt.at/
> [2] http://vocabulary.semantic-web.at/PoolParty/sparql/semweb
> [3] http://vocabulary.semantic-web.at/semweb/131
>
>
> ----- Ursprüngliche Mail -----
>
>> Von: "Simon Cox"<Simon.Cox@csiro.au>
>> An: holger@topquadrant.com, public-esw-thes@w3.org
>> Gesendet: Montag, 27. Juni 2011 04:49:38
>> Betreff: RE: Results of survey "Do controlled vocabularies matter?"
>> E.g.
>> 1. Get concept (all triples with concept as subject or object -
>> DESCRIBE) with prefLabel = “foo”
>> 2. Get concept with prefLabel in Fonglish = “foo”
>> 3. Get concept http://example.org/bar and all narrower concepts
>> 4. Get concept http://example.org/bar and all broader concepts
>> 5. Get concept with prefLabel in Fonglish = “foo” and all narrower
>> concepts
>> 6. Etc ...
>> i.e. pre-rolled queries which use the SKOS properties.
>> Reusable SPARQL queries might be part of a solution.
>> From: public-esw-thes-request@w3.org
>> [mailto:public-esw-thes-request@w3.org] On Behalf Of Holger
>> Knublauch
>> Sent: Monday, 27 June 2011 9:11 AM
>> To: public-esw-thes@w3.org
>> Subject: Re: Results of survey "Do controlled vocabularies matter?"
>> What aspects of SPARQL would such a standard layer include? We and
>> our customers are heavily using SPARQL in the context of SKOS, and
>> we have published a SPARQL-based constraint checking library [1]
>> using the W3C Member Submission SPIN [2]. This will validate the
>> various SKOS-specific integrity constraints and has been integrated
>> with TopQuadrant's Enterprise Vocabulary Net (EVN) solution [3].
>> SPIN also makes it easy to encapsulate reusable SPARQL queries into
>> so-called templates. These templates are basically parameterized
>> queries that have a URI and arguments. Compare to stored procedures
>> in SQL. It would be straight-forward to build and publish such a
>> template library for SKOS, if someone here has a list of queries
>> that they like to ask repeatedly.
>> Cheers,
>> Holger
>> [1] http://topbraid.org/spin/skosspin
>> [2] http://spinrdf.org/
>> [3] http://www.topquadrant.com/solutions/ent_vocab_net.html
>> On Jun 27, 2011, at 10:34 AM,<  Simon.Cox@csiro.au>  wrote:
>> “ a "standard" layer encapsulating Sparql especially designed for
>> SKOS based systems
>> might help in this situation. ”
>> +1
>> From: public-esw-thes-request@w3.org
>> [mailto:public-esw-thes-request@w3.org] On Behalf Of Andreas
>> Blumauer (Semantic Web Company)
>> Sent: Sunday, 26 June 2011 4:36 PM
>> To: public-esw-thes@w3.org
>> Subject: Results of survey "Do controlled vocabularies matter?"
>> Hello all,
>> we are happy to announce that the results of the survey titled
>> "Do controlled vocabularies matter?" are available now.
>> A short write-up and also the link to the report can be found here:
>> http://blog.semantic-web.at/2011/06/26/thesaurus-based-search-engines-will-become-main-stream-in-the-near-future/
>> The report itself was published on issuu:
>> http://issuu.com/andreas_blumauer/docs/survey_do_controlled_vocabularies_matter_2011_june
>> We would be happy to receive comments from the community on this.
>> As an interesting outcome it can be stated, that people working on
>> controlled vocabularies
>> seem to have put aside scepticism about the semantic web finally.
>> On the other side the whole semantic web cake has not been digested
>> yet:
>> E.g. only a few people from the vocab-community deal with Sparql,
>> which brings me to the conclusion
>> that a "standard" layer encapsulating Sparql especially designed for
>> SKOS based systems
>> might help in this situation.
>> Any activities going on in this area?
>> Best wishes,
>> Andreas
>> ** Semantic Search Online Demo: http://bit.ly/semantic_search
>> ** LOD2 - Creating Knowledge out of Interlinked Data -
>> http://lod2.eu/
>> | Andreas Blumauer, Managing Director
>> | Semantic Web Company GmbH
>> | Lerchenfelder Guertel 43/5
>> | A - 1160 Wien, Austria
>> | Tel +43 1 402 12 35 27
>> | Mob +43 676 9 234 16 2
>> COMPANY INFORMATION
>> | http://www.semantic-web.at/ | http://www.i-semantics.at/ |
>> | http://blog.semantic-web.at/
>> PERSONAL INFORMATION
>> | mailto:a.blumauer@semantic-web.at | skype:blumauerpunkt | Blog:
>> | http://ablvienna.wordpress.com
>
>