Re: struggling to understand conformance in POWDER formal semantics

On Tue, 2008-12-02 at 16:34 +0000, Phil Archer wrote:
> Dan,
> 
> Thanks for taking time to review our documents.

Well, I started reading them; here's hoping I find time to keep going...

>  I'll work through your 
> comments, starting with this one.
> 
> Dan Connolly wrote:
> > I'm puzzled by this...
> > 
> > "Conformance with this document means that Description Resources may be
> > transformed from POWDER to POWDER-BASE to POWDER-S as set out below
> > without any change to their semantics. When querying or inferring from
> > the RDF/OWL graph, conformant implementations of this document will also
> > implement the semantic extension defined below."
> >  -- http://www.w3.org/TR/powder-formal/#conformancestatement
> > 
> > What product classes are being defined here?
> > I see "conformant implementations" but I don't really see
> > a definition of the form
> > 
> >   A conforming POWDER implementation is ...
> > 
> > as I'd expect.
> 
> Yes this is tricky as what the formal doc defines is:
> 1. The semantic extension
> 2. The transformation through the 3 species of POWDER.
>
> Also, for various reasons (some historical, some practical), 'POWDER' is 
> defined in 3 separate documents so we can't say "A conforming POWDER 
> implementation is ..." in this document. We do say something similar in 
> the definition of a POWDER Processor, however [1].
> 
> OK, but taking your comment on board, how about:
> 
> "A conformant implementation of this Recommendation will transform 
> POWDER to POWDER-BASE to POWDER-S as set out below without any change to 
> the semantics at each stage. When querying or inferring from a POWDER-S 
> document (an RDF/OWL graph), conformant implementations of this document 
> will also implement the semantic extension defined in Section 4.3."

I got the impression (from... the primer, maybe?) that this is something
of a theoretical process, but if you create such a conformance clause,
you'll be expected to exhibit software that does it as a CR exit
criterion.

I don't see any strong motivation for a conformance section in
the formal semantics document. My impression is that the
definions of 1 and 2 above are clear enough without defining
a class of software products (though I haven't actually
read closely enough to be sure).


> 
> [1] http://www.w3.org/TR/2008/WD-powder-dr-20081114/#conformance
> 
> > I'd also expect to see something about conformance of documents,
> > though perhaps that's in one of the other documents? In which
> > case a cross-reference seems in order.
> 
> Agreed. So I'll add to the previous sentence: A separate conformance 
> statement relating to POWDER and POWDER-S documents is given in Section 
> 2.9 of the the Description Resources document [@ref].
> 
> OK with you?
> 
> Phil.
> 
-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
gpg D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

Received on Tuesday, 2 December 2008 17:37:36 UTC