W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > November 2001

Re: Proposed issue resolutions

From: Frank Manola <fmanola@mitre.org>
Date: Mon, 05 Nov 2001 15:39:56 -0500
Message-ID: <3BE6F91C.8F034BC1@mitre.org>
To: Brian McBride <bwm@hplb.hpl.hp.com>
CC: rdf core <w3c-rdfcore-wg@w3.org>
Brian McBride wrote:
> 
> Hi Frank,

Howdy!

>
> Frank Manola wrote:
> 
> [...]
> 
> >
> >>Propose postpone
> >>
> >>   http://www.w3.org/2000/03/rdf-tracking/#rdf-equivalent-uris
> >>
> >>on the grounds that it is out of scope of the charter.
> >>
> >
> > I agree with what I think the idea is, but I assume we're going to state
> > this somewhere public, so I'd like to see a little more in the actual
> > wording so those interested don't think we're sweeping this under the
> > rug (like, we agree that some way of stating equivalence is important,
> > we just don't think it fits in our "rewrite/clarification" of RDF-1;  it
> > could go in an RDF-2;  there's a capability for doing this in DAML+OIL
> > so there are layering questions;  etc.]
> 
> How about:
> 
> Whilst the WG recognises the importance of a mechanism for defining equivalence
> of URI's, the WG has decided it does not fit within the limited scope of its
> current charter.  The WG also notes that DAML+OIL has an equivalence mechanism
> which raises the question of which layer of the stack best suits such functionality.
> 
> Feel free to improve or suggest a replacement.

Your wording is fine.

> 
> >
> >
> >>Propose postpone
> >>
> >>   http://www.w3.org/2000/03/rdf-tracking/#rdfms-contexts
> >>
> >>on the grunds that it is out of scope of the charter.
> >>
> >
> > I'd suggest we see what we wind up doing with reification before
> > summarily disposing of this one.  There's a connection.
> 
> OK - this one is dropped from the list for now.
> 
> >
> >
> >>Propose postpone
> >>
> >>   http://www.w3.org/2000/03/rdf-tracking/#rdf-containers-otherapproaches
> >>
> >>on the grounds that it is out of scope of the charter.
> >>
> >>Propose close
> >>
> >>   http://www.w3.org/2000/03/rdf-tracking/#rdf-formal-semantics
> >>
> >>on the grounds that the model theory adequately addresses this issue.
> >>
> >
> > In terms of the technical problems to be solved, I agree with Peter PS
> > on this one, but it seems to me this might be recast to say that if we
> > deal with reification and containers (which have their own issues) then,
> > given the existence of the model theory, do we need a separate issue on
> > formal semantics in general?  (There are actually a lot of issues on our
> > list intertwined like this.)
> 
> Interesting approach.  How about
> 
> on the grounds that the model theory adequately addresses the general issue of a
> formal semantics for RDF.  New issues:
> 
>    http://www.w3.org/2000/03/rdf-tracking/#rdfms-reification-formal-semantics
>    http://www.w3.org/2000/03/rdf-tracking/#rdfms-containers-formal-semantics
> 
> have been raised to record more specific semantic issues.

Hmm!  What I actually had in mind was that (I thought) we were already
going to have to do some cleanup of reification and containers in the
course of addressing several other *existing* issues (not specifically
this formal semantics one), so we could basically say what your first
bit says, that the model theory addresses the general issues of formal
semantics, and then say that as we clean up reification and containers
<cite issues here> the model theory will be extended (if needed) to
reflect what we decide about them.  I wasn't thinking in terms of
resolving this one issue by adding two new ones (although that does make
things more precise, and you get some reuse out of it in dealing with
the next issue).

> 
> >
> >
> >>Propose close
> >>
> >>   http://www.w3.org/2000/03/rdf-tracking/#rdfms-logical-formalism
> >>
> >>on the graounds that the model theory adequately addressses this issue.
> >>
> >
> > I don't understand this issue anyway.  Dan?
> 
> This is the use/mention problem in reification.  Drop this from the current list
> and link in with the more general rdfms-reification-formal-semantics.  My mistake.
> 

OK

> >
> >
> >>Propose close
> >>
> >>   http://www.w3.org/2000/03/rdf-tracking/#rdfms-logical-terminololgy
> >>
> >>on the grounds that the new terminology introduced by the model theory
> >>adequately addresses this issue.
> >>
> >
> > Do these grounds mean that we're *only* going to use terminology from
> > the model theory?  For example, does it mean that the Primer shouldn't
> > talk about an "RDF Data Model"?  Lots of people know what a data model
> > is;  some of them can even manage to disambiguate that from a model in
> > logic!
> 
> No.  I didn't intend them to mean that.  I meant that in the model theory we
> have defined RDF in terms of terminology acceptable to logicians.  We have for
> example used the word 'graph', where in other contexts, such as API's, the term
> 'model' has been used.
> 
> Clearly we should use the model theory terminology where we can.  We MUST always
> be consistent with it.  But I see no reason why the primer cannot use other
> terms familiar to its audience in explanations.  In fact its probably essential.
> 
> I regard the terms 'model' and 'data model' as being different.
> 
> I suggest that the wording of proposal can stand.  The new terminology of the
> model theory adequately addresses the issue which was *inconsistency* between
> RDF and logical terminology.  The model theory provides an acceptable
> terminology consistent with logicians usage.  We must of course be careful not
> to introduce any inconsistencies in other terminology introduced elsewhere.
> 
> Note that DanC who raised the issue has seconded the proposal.

I don't mean to belabor this but:  The terminology in the model theory
addresses the issue of inconsistency between logical terminology and RDF
as long as "RDF terminology" is exclusively that used in the model
theory.  I wasn't sure what the scope of "RDF terminology" was intended
to be, and in particular whether terminology used in other "RDF
documents" (like the Primer, but possibly also syntax and schema
documents), might not possibly need to use terms in "nonlogical"
[hopefully not the same as illogical!] ways.   With the understanding
you've given above, I'm happy to see the issue disappear.

--Frank

-- 
Frank Manola                   The MITRE Corporation
202 Burlington Road, MS A345   Bedford, MA 01730-1420
mailto:fmanola@mitre.org       voice: 781-271-8147   FAX: 781-271-8752
Received on Monday, 5 November 2001 15:44:25 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:42:30 EDT