See also: IRC log
<jeremy> AOB item: #fragIDs in RDF/XML and HTML
-> QName production in SPARQL grammar [Jeremy 2005-12-13]
Jeremy: in my message to DAWG I only noted some of the QName syntax restrictions, not all the restrictions that we care about in RDF/A. SPARQL appears to have taken the syntax for QNames from XML 1.1 spec and mangled it a little. I suggested to them that they mangle it more, though we [in CURIEs] mangle it a lot more. There is a larger question about "what is a namespace URI"; the reference to namespace names in SPARQL spec is clearly meant to be informative, not normative. I don't really care whether SPARQL chooses to allow a digit after ':' but it's chipping away at the issue. SPARQL appears to be the first normative use of QNames as derived from the Turtle and N3 heritage; it is thus an opportunity to raise the issue formally. DanC did ask whether the [RDF-in-XHTML] Task Force has a position
Ralph: WG discussion yesterday, during TF update for RDF-in-XHTML I told the WG where 3 documents would end up, with primer the only published doc by SWBPD. Guus pushing all TFs to publish their docs before end of january. I said I expected that the Primer will include more rationale for CURIEs and when that was ready for review we will be asking the WG for opinion on CURIEs. Guus was skeptical and requested that CURIE piece be modular (so it could be removed?) He may have somewhat backed off of that after further discussion from Jeremy and me. More importantly, the Primer should be written for readers who are not experts. This is in agreement with our previous discussions within the TF. This led to a longer discussion about CURIEs, about how important CURIEs are. Jeremy and I cited the number of features that would be lost without CURIEs. There was some discussion within the WG, and we should anticipate more of this discussion from the WG.
Jeremy: in Guus's role as chair, it's appropriate for him to express skepticism about including material that may be controversial
Ben: I checked-in a new RDF/A Primer draft yesterday.
$Id: 2006-01-15-rdfa-primer.xml,v 1.5 2005/12/13 14:57:55 adida Exp $
Ben: I've tried to follow one thread through the examples. If we don't mention bnodes or reification then the justification for CURIEs is not as clear
Jeremy: The new Table of Contents appears to address Guus' concern that the advanced material, e.g. CURIEs, be later in the document
Mark: we could refer to "compact URI" in the early part of the document before we introduce the term "CURIE"
Ben: I took special care to keep it simple, specifically not referring to jargon such as "qname"
Mark: it's good to keep things simple in the first part of the document, however anyone who is XML-literate will assume those abbreviations are QNames. The current version is still very RDF-oriented; another motivation of RDF/A is to encourage authors to mark-up HTML documents so that we can extract more triples from them. Perhaps adding another scenario to illustrate this rather than overloading the Shutr scenario. I have hundreds of examples from HTML that I'd like to see people do and am willing to write another section. Perhaps a much slower build-up, not introducing N3 results as quickly. Perhaps a person's home page that gets linked into Shutr later
Jeremy: The current document is an easy read for those who live and breathe triples. For others who understand something about semantics and are happy to add more semantics but aren't thinking in triples, it would be nice to introduce this in a way that says we're showing how to tell what a document is about
Ben: I'd like to keep the write baton until the
end of the week then turn it over to Mark for 2 weeks.
The editing form is the
xmlspec form.
Ralph: as we all can see the xmlspec version nicely formatted, we can trash the HTML version while we're in editing mode
Jeremy: Gary NG and David Booth have agreed to review the primer if we send it to the WG by 9 January
next meeting: 3 Jan
<MarkB> Main thing is that next week will probably be hectic due to trying finish off things for customers before Xmas.
Ralph and Mark will communicate via email
Ben: want to review the issues list and propose status for each of them, possibly deferring some
(scribe note: all these are copied from the previous meeting record. These were not explicitly discussed during this meeting.)
[CONTINUES] ACTION: Ben start separate
mail threads on remaining discussion topics
[CONTINUES] ACTION: Jeremy followup on
<head about=...> edge case
[CONTINUES] ACTION: Jeremy followup with
Mark on the question of multiple triples from nested meta and add to issues
list
[CONTINUES] ACTION: Jeremy propose wording
on reification
[End of minutes]