Last call comments from the I18N WG on RDF WDs from Dan Connolly on 2003-11-12 (www-rdf-comments@w3.org from October to December 2003) (original) (raw)

Dear RDF WG,

Here are the last call comments from the I18N WG on the RDF drafts. This is not necessarily by draft, but by feature.

Thanks for the careful review. The RDF Core Working Group noted several of these points in its 7 Nov teleconference. (The others are left for the editors to address).

In brief: In one case, we decided to add an issue and postpone it. In other cases, we found that the issues here have been considered in making earlier decisions, and we did not reopen them. While I don't expect this addresses your comments to your satisfaction, perhaps the time for exploring design alternatives for this version of RDF has come to an end. We look forward to discussing this with you and The Director.

Some details follow...

• Treatment of language information for XML Literals: We have already commented on this extensively. We think that the removal of language information from XML literals is a serious problem for internationalization. Details of our comments can be found at http://www.w3.org/2003/09/ri434.html.

  One example of how language information could easily be added to XML literals, with minimal impact on the overall design, (just a proposal, not intended to preclude any other solution)

Yes, we have explored designs along these lines but decided (9 May 2003) on a different design.

We have updated our issues list to elaborate on the various alternatives and the reasons we didn't choose designs like this. http://www.w3.org/2000/03/rdf-tracking/#rdfms-literal-is-xml-structure 1.217 2003/11/12 22:58:21

• XML Literals as typed literals: We have not objected to the treatment of XML Literals as typed literals in the previous last call, because there, it was possible to understand this treatment just as a technical convenience. However, with the change to the treatment of language information for XML literals, treating XML literals as datatyped literals becomes highly questionable. Unless the language information issue can be solved, we have to disagree with this treatment.

Yes, we were aware of these questions when we made the 9 May decision.

• XML Literals containing only text should be equivalent to the corresponding plain literals and to the corresponding string type literals. (Solving the language information issue for XML Literals seems to be a precondition for this, but once this is done, it does not seem to be too difficult, in the same way that it was possible to make strings and plain literals equivalent.)

The working group decided that this issue a new issue, but to postpone it:

[[[ add a new issue to the issue list: "relationship between XMLLiterals and plain literals," and postpone it.

Rationale:

The lack of semantic equivalence between XMLLiterals and plain literals has been clear since the first WD of RDF Concepts, and was arguable in RDF Model and Syntax.

The RDF Semantics does not preclude RDF applications using additional information to determine that two literals are equivalent, but does not mandate that they should be.

Hence, RDF applciations which require this equivalence may operate in such a mode, and so this issue is not a show stopper. ]] -- Minutes, 2003-11-07 (in progress) http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2003Nov/0063.html

• Examples/Primer: There is one important facility of RDF that is almost [...]

I'm leaving that point for the primer editor(s) to address.

• Alt container: Because of the special rule that the first element is the default or preferred value, this is a fake alternative. This should be changed, or a real alternative, without any preferences, should be provided.

Yes, we are aware that "The design of the RDF Model collection classes exhibit various awkward features." -- http://www.w3.org/2000/03/rdf-tracking/#rdf-containers-otherapproaches

We resolved, 15th February 2002, to postpone full treatment of that issue to a future working group. Perhaps that's sufficiently responsive to your comment at this point?

• Measures/weights: The primer in a very small number of instances uses 'weightInKg', and explains why, but for the rest, it always uses just 'weight', even when there is no reason for such an underspecified property.

I'm leaving that point for the primer editor(s) to address.

• The motivation for using xsd datatypes should clearly say that these are well established and should be used where appropriate.

I'm leaving that for the primer or concepts editor(s) to address.

-- Dan Connolly, W3C http://www.w3.org/People/Connolly/

Received on Wednesday, 12 November 2003 18:31:00 UTC