W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > July 2002

RE: datatypes discussion on other lists

From: Brian McBride <bwm@hplb.hpl.hp.com>
Date: Fri, 12 Jul 2002 10:36:39 +0100
Message-Id: <5.1.0.14.0.20020712101911.06203890@15.144.25.13>
To: "Jeremy Carroll" <jjc@hplb.hpl.hp.com>, "RDF Core" <w3c-rdfcore-wg@w3.org>

At 09:54 12/07/2002 +0100, Jeremy Carroll wrote:

>I have some sympathy with Peter's position, and I am sorry that Brian did not
>give us a chance to comment on his text before sending it.

I've been feeling the time pressure.


>In particular I think the fact that:
>
>
>Test A*:
>
>    <Jenny> <ageInYears> "10" .
>    <John>  <ageInYears> "10" .
>    <ageInYears> rdfs:range xsd:decimal .
>
>holds in all cases is relevant and was omitted.

Hmm,

[[
These test cases only relates to the situation where there are no range 
constraints on the properties.
]]

wasn't enough.  I'm happy to send an amendment, but I'm worried about the 
form you suggest, as in the tidy case its a contradiction.  I think we need 
to use drange, which could open another can of worms.

I could send:

[[

 > These test cases only relates to the situation where there are no range
 > constraints on the properties.

I perhaps did not emphasize this point enough.  Where the appropriate range 
constraints are present, all solutions produce the "expected" answer, e.g.

>    <Jenny>      <ageInYears> "10" .
>    <John>       <ageInYears> "10" .
>    <ageInYears> rdfd:drange  xsd:decimal .

an RDF processor could conclude that Jenny and John have the same age.

Note however, that there may be issues handling very general properties, 
such as the container membership properties and rdf:object where it isn't 
really feasible to use a range contraint.

]]

Brian




Brian










>Jeremy
>
>
> > -----Original Message-----
> > From: w3c-rdfcore-wg-request@w3.org
> > [mailto:w3c-rdfcore-wg-request@w3.org]On Behalf Of Brian McBride
> > Sent: 11 July 2002 20:04
> > To: RDF Core
> > Subject: datatypes discussion on other lists
> >
> >
> >
> > I have sent, at last, the datatypes message to rdf interest and rdf logic
> > and await the explosion.
> >
> > I have modified it from the draft approved by the WG in the light of recent
> > discussion.  I trust the WG will approve of the modification.
> >
> > At the f2f and elsewhere, concern has been raised the we might simply
> > repeat the discussion, in all its confusion and volume, that we have had in
> > RDFCore, in a wider forum.  It would be a disaster if that were to happen.
> >
> > I am not sure what other WG's have done in the past, and would welcome
> > guidance from those who have been here before.
> >
> > However, I strongly suggest that members of the WG refrain from any form of
> > advocacy for their preferred position on the other lists.  We are seeking
> > input from the community on this issue, not a debate.
> >
> > I recognise that what is advocacy and what is helpful explanation is a
> > difficult line to draw.  I am loathe to suggest gagging people, but I also
> > suggest that folks leave it to me to answer for the WG on any questions or
> > issues that arise on the other lists.  If anyone is dissatisfied with
> > anything I say, then raise that with me first and we will correct it.
> >
> > Brian
> >
> >
Received on Friday, 12 July 2002 05:37:59 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:49:50 EDT