Re: fun with embedded RDF

On Thu, 2006-02-23 at 08:59 +0000, Ian Davis wrote:
> On 23/02/2006 08:39, Dan Connolly wrote:
[...]
> > I've got some schemas that I want to re-do using XHTML/GRDDL...
> > I started reviving HyperRDF
> >  http://www.w3.org/2000/07/hs78/
> 
> Interesting - a first glance appears to show a lot of similarities. I 
> need to read this properly. What's the best way to move forward on 
> reconciling the various approaches?

I dunno... in fits and starts, I work on projects like...

 - noodling on my homepage and foaf file every once in a while,
  especially in preparation for trips like the TP
 - noodling on my list of past events, publications, and the like
   (I owe MIT an update to my CV)
 - formalizing DAWG records
 - the /TR/ page stuff
 - Ivan's work on /Talks/
 - timbl's roadmap diagrams
 - webarch formal descriptions

and across all those are all kinds of best-practices
issues like:

 - http://esw.w3.org/topic/PropertiesForNaming
   when to use rdfs:label vs cyc:nameOfAgent etc.
   This shows up in the tabulator, which knows rdfs:label
   and foaf:name and dc:title and maybe a few others,
   but isn't extensible (yet)

 - schema size. cyc and wordnet are nice, but they
   don't seem to be published in the web as RDF in
   stable, manageable chunks. SUMO is promising. See
    http://www.w3.org/2000/10/swap/util/split_sumo.py

 - http://esw.w3.org/topic/BuildOrBuyTerms
   when to use something out there, and when to
   make your own schema

 - schema documentation and testing. The tabulator
   will tell you when you're using (directly or indirectly)
   a schema that's 404. It doesn't catch foaf:Preson errors
   yet. We have some N3 rules that do, but they're too slow
   to be usable. Pellet is quite good for that sort of thing.

Some organizational possibilities include
 - random #swig progress
 - the SemWeb BP WG
 - the TAG
 - the recently started HCLSIG
 - PAW and TAMI research projects
 - a possible GRDDL WG-to-be

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

Received on Thursday, 23 February 2006 15:12:19 UTC