W3C home > Mailing lists > Public > public-lld@w3.org > September 2010

RE: Library data diagram

From: Pete Johnston <Pete.Johnston@eduserv.org.uk>
Date: Wed, 1 Sep 2010 17:03:46 +0100
To: "Young,Jeff (OR)" <jyoung@oclc.org>, Thomas Baker <tbaker@tbaker.de>, Karen Coyle <kcoyle@kcoyle.net>
CC: "public-lld@w3.org" <public-lld@w3.org>
Message-ID: <59E2F1A55E23D64DB03813F9B680F9D924015E2F0D@edu-vmw-eml-l01.edu2000.com>
Hi Jeff,

> I have the feeling that "application profiles" aren't necessary for OWL
> vocabularies because they are relatively self-explanatory. DC
> vocabularies don't use OWL so they have to be explained with
> "application profiles". IMO, Linked Data is easier to understand if it
> the resources are based on OWL. Comments?

I think the DCMI community's notion of an "application profile" really comes from a fairly "document-centric" perspective, with a fairly strong notion of "structural validation". 

Leaving aside for a moment some of the contextual information and the human-readable guidelines that are typically present, the "core" part of a DC "application profile" is a set of "structural constraints", or patterns, against which some bounded chunk of data can be "tested".

There's a post here by Dave Reynolds

http://www.amberdown.net/2009/09/faq-using-rdfs-or-owl-as-a-schema-language-for-validating-rdf/

which, IIRC, I first encountered in its initial form on a SWAD-E blog or wiki page a long time ago, and which explains why OWL typically isn't very good for doing (what I think are) the sort of things the DCMI community has aspired to do with DC application profiles.

Having said that, if some of those assumptions are changed (e.g. one sets aside the open world assumption and considers just "the graph at hand") then things change slightly, and Dave has a more recent post which touches on that

http://www.amberdown.net/2009/10/owl2-for-rdf-vocabs/

I was at the Bristol VoCamp meeting that Dave mentions in his post

http://vocamp.org/wiki/VoCampBristol2009 

and talked a little bit about DCMI's approach, and others talked about approaches based on query patterns, and Dave mentioned OWL 2. I wrote a little bit here

http://efoundations.typepad.com/efoundations/2009/09/vocamp-bristol.html

but sadly, I haven't had time to follow up any of those discussions since then.

There's a comment on Dave's post from Kendall Clark pointing to further work of theirs in that vein, but the URI now seems to redirect to the home page of their new blog, but I think the references were to e.g.

http://weblog.clarkparsia.com/2009/02/11/integrity-constraints-for-owl/
http://weblog.clarkparsia.com/2009/03/30/integrity-constraints-reasoning-and-a-preview-release/
http://weblog.clarkparsia.com/2009/05/07/owl-integrity-constraints-serql-and-new-sparql/
http://weblog.clarkparsia.com/2010/02/02/using-owl-to-validate-rdf-pellet-integrity-constraints/

Pete 
---
Pete Johnston
Technical Researcher
Eduserv
E: pete.johnston@eduserv.org.uk
T: +44 (0)1225 474323
F: +44 (0)1225 474301
http://www.eduserv.org.uk/
http://efoundations.typepad.com/

Eduserv is a company limited by guarantee (registered in England & Wales, company number: 3763109) and a charity (charity number 1079456), whose registered office is at Royal Mead, Railway Place, Bath, BA1 1SR.
Received on Wednesday, 1 September 2010 16:04:29 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 1 September 2010 16:04:29 GMT