[rdfweb-dev] Labelling stability of FOAF vocabulary properties

[Dan Brickley]

Hi Leigh,

Having just been talking to you about this, I was inspired to  ship a 
draft namespace on this that merges my original design with your
suggestions below.

It should have semweb applicability way beyond FOAF, so I'm doing it
with a dayjob hat on, see http://www.w3.org/2003/06/sw-vocab-status/ns

the main property is 'vs:term_status', defined as:

<rdf:Property rdf:ID="term_status">
  <rdfs:label>term status</rdfs:label>
    <rdfs:comment>the status of a vocabulary term, one of
    'stable','unstable','testing'.</rdfs:comment>
      <vs:term_status>unstable</vs:term_status>
</rdf:Property>

Never let it be said that RDF people can resist an opportunity for
cutesy self-reference! This definition is the crux of the schema (I've 
gone with a literal-valued property as you suggest, rather than using an
RDF class). The schema at http://www.w3.org/2003/06/sw-vocab-status/ns 
includes an example of its own use, in this case we state that the 
term_status property is 'unstable' (using the unstable term_status
property to do so).

The other properties I have included are 'moreinfo' and 'userdocs',
which nearly map to your 'issues' and 'examples'. In practice I would
expect it likely for the document at the namespace to point off to a hub 
document, eg. a wiki page, using 'issues', and other RDF written
elsewhere would keep track of user oriented documentation (examples
etc). Since editing the ns doc can be costly, so the ns is unlikely to 
offer a complete catalogue of all relevant user documentation.

A fancier version of this vocab might try to model more explicitly the 
states a term can be in, and the email announcement messages which
document their transitions (w.r.t. some user community). For now, I
think we can go a long way with this very simple vocab.

This should be an interesting exercise. I invite rdfweb-dev listmembers
to all have a think about the status of various properties in the 
FOAF namespace. I'm not sure how to gather the data; let's try it using
Wiki rather than email. http://rdfweb.org/topic/FoafVocab has a 
== Vocab Status Documentation Experiment == section now, although there
is work to do in figuring out how to use such a shared space.

A more ambitious version of this approach is written up at 
http://ilrt.org/discovery/2001/06/process/index.html though I think for
now we have our work cut out trying to explain what 'stable' etc might
mean.

cheers,

Dan

* Leigh Dodds <ldodds at ingenta.com> [2003-06-27 16:00+0100]
> Dan Brickley wrote:
> 
> > Re (ii), vocabulary stability and confusion, I have a plan(!). Rather than 
> > worry whether FOAF itself is 'stable' or not, I intend to tag each vocabulary 
> > item with 'stable','unstable','testing'. For eg., 'foaf:homepage' is stable;
> > 'foaf:surname' is unstable; much of the rest is 'testing'. More on this in due
> > course. I hope this should clear up some concerns folk have w.r.t. deployability
> > of different bits of FOAF vocab.
> 
> I got to thinking about this some over lunch today, and thought that perhaps 
> a nice way to do this would be to label up the FOAF schema to include some 
> extra information about each rdfs:Class and rdf:Property.
> 
> I was thinking that the following information would be useful to know about 
> each:
> 
> - status: literal value (stable, unstable, testing)
> - creation date: date property/class was added
> - issues: pointing to a resource describing issues/discussion of this property relating 
>   to its move from testing -> stable
> - examples: pointing to documentation about using this property, e.g: FoafExamples
> 
> Any advances on those?
> 
> I've made up some elements in a custom vocab and have begun whipping up an XSLT stylesheet 
> that takes the FOAF schema and produces some friendly documentation including 
> some of the above. Will circulate when its done (mainly formatting issues to resolve now).
> 
> I just wondered whether there were other properties that people thought might be 
> useful? Or whether there's related work elsewhere -- some time in front of Google didn't 
> pull out anything obvious.
> 
> Cheers,
> 
> L.
> 
> _______________________________________________
> rdfweb-dev mailing list
> rdfweb-dev at vapours.rdfweb.org
> wiki: http://rdfweb.org/topic/FoafProject
> http://rdfweb.org/mailman/listinfo/rdfweb-dev