Re: ANN: SKOS implementation of the ACM Communication Classification System

On 6/2/11 4:27 AM, Christoph Lange wrote:
> Dear all,
>
> in the context of our Planetary social semantic environment for 
> scientific, technical, and mathematical documents 
> (http://trac.mathweb.org/planetary), we needed a SKOS implementation 
> of the ACM Computing Classification System 
> (http://www.acm.org/about/class/ccs98-html).  This is now public at 
> its namespace URI
>
> http://oaff.info/ontology/acm-ccs#
>
> ... and has been validated with the poolparty SKOS consistency checker 
> (http://demo.semantic-web.at:8080/SkosServices/check -- thanks for 
> providing that!).  For now it's just one static RDF/XML file, but we 
> will soon also serve it through a SPARQL endpoint.  Let me know what 
> you need, and we might be able to reprioritize these things.  I just 
> thought it makes sense to already announce it now, as there is so much 
> redundant work done on this due to not being aware of existing 
> implementations.
>
> Note that we have not yet considered copyright issues -- but at least 
> preserved the original ACM copyright statement, which permits 
> "personal or classroom use".  That's probably not enough for 
> reasonable Linked Data applications.  I would be glad if someone 
> familiar with the subject could point out what to do.  What did 
> previous publishers of RDF versions of the ACM CCS do?
>
> Some technical notes:
> * We started with the RKB Explorer RDFS implementation of the ACM CCS 
> (which used rdfs:subClassOf, see 
> http://acm.rkbexplorer.com/ontologies/acm#)
> * Besides the pure concept hierarchy, we (manually) added further 
> information from the original HTML source:
>   * examples for certain concepts (skos:example)
>   * cross-references to related concepts (skos:related)
>   * notes on deprecated concepts (skos:historyNote)
>
> While not having verified it formally, we are 99.9% sure that our 
> implementation completely covers the ACM CCS.
>
> BTW, to see how _we_ are using this, I have, for now, to point you to 
> something similar: We serve the content of the PlanetMath.org 
> encyclopedia through our system and enable by-topic navigation powered 
> by a SKOS implementation of the MSC (Mathematical Subject 
> Classification): see 
> http://trac.mathweb.org/planetary/wiki/Demo_PlanetMath for a 
> description and http://alpha.planetmath.org/article/msc for the 
> system.  The association of PlanetMath articles to their MSC classes 
> is made by translating the LaTeX sources of the articles, which have a 
> custom MSC metadata macro, to XHTML+RDFa, using dct:subject.  Soon 
> there will be a similar navigation through arXiv.org articles served 
> via our system (http://trac.mathweb.org/planetary/wiki/Demo_arXiv), as 
> well as our own "dog-fooded" computer science lecture notes 
> (http://trac.mathweb.org/planetary/wiki/Demo_PlanetBox).  Note that 
> arXiv articles may have both MSC and ACM classifications, so aligning 
> those two SKOS concept schemes is also planned.
>
> Note that our MSC SKOS implementation is a preliminary and unofficial 
> one (and therefore not yet available under a stable URI).  It is 
> incomplete and just good enough to drive the Planetary navigation.  In 
> collaboration with the AMS, the maintainer of the MSC, we are working 
> on a complete and official SKOS reference implementation, obtained by 
> translating the original plain-TeX sources to RDF, which will soon be 
> available from http://www.msc2010.org.
>
> Cheers,
>
> Christoph
>
Christoph,

Great stuff!

See: 
http://linkeddata.uriburner.com/describe/?url=http%3A%2F%2Facm.rkbexplorer.com%2Fontologies%2Facm%23K.8.1 
.

Would be nice if you increase navigability (of the follow-your-nose 
variety) by adding skos:inScheme relations to your dataset .

-- 

Regards,

Kingsley Idehen 
President&  CEO
OpenLink Software
Web: http://www.openlinksw.com
Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca: kidehen

Received on Thursday, 2 June 2011 18:27:58 UTC