See also: IRC log
 -> previous meeting 2005-04-05
-> Last Meeting's action summary
ACTION: DanBri RDF schema for new XHTML2 namespace elements [CONTINUES]
Ben: I spoke with DanBri and he said he hoped to do this action before the end of the month
ACTION: Tom Baker and Gavin to get feedback about use of RDF in XHTML in their respective communities [CONTINUES]
Ben: more email sent to Gavin
Ralph: Tom reported at last
week's WG telecon that he had sent feedback
... but couldn't find it in email
... Tom will regenerate it if necessary
Ben: neither could I; Tom though
he'd sent feedback but we're not able to find
it
ACTION: Ben to move bnode discussion to email list. [DONE]
ACTION: Steven to send email about latest draft of RDF/A included in XHTML 2 [CONTINUES]
Ben: there is not yet a public URI for the XHTML2 draft, correct?
Steven: correct, the current document is Member-only
<Steven> http://www.w3.org/MarkUp/Group/2005/WD-xhtml2-20050411/
Ralph: it's only a problem if the HTML WG doesn't want those restricted links named in our public forum
Steven: posting those URIs is not a problem for the HTML WG
ACTION: Ben to ask Tom, Gavin and CC about opinion on GRDDL and pre-XHTML2 [DONE]
ACTION: Jeremy to ask HP about need for pre-XHTML 2 solution to RDF-in-HTML problem [DONE]
<benadida> http://lists.w3.org/Archives/Member/w3c-archive/2005Apr/0032
<benadida> http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2005Apr/0004.html
Ben: action done, as question was asked
ACTION: Ben to talk to Jeremy about HP feedback
ACTION: Ben to Provide RDF/XML examples and english description to Steven and Mark (use cases) [CONTINUES]
Ben: I have a work-in-progress
Ben: I hope to add English
descriptions of the goal of the community for each case
... I will provide RDF/XML examples
... hoping that Steven and Mark will provide the XHTML2
serialization
Mark: yep
Mark: Steven and I have a number
of sample documents we have been using
... we'd like someone else's documents to check as well
<Steven> (e.g. RSS in XHTML)
Ben: it would be nice to have some tests that combine RSS, FOAF, and Dublin Core
Steven: Last Call WD now hinges
on getting the schema up-to-date
... specifically, we discovered our schema for XForms was out
of date
Ben: will the RDF Schema hold you up?
Steven: no, that can be done as a Note
Mark: I started to reply to some of last week's bnode comments
<MarkB_> http://www.w3.org/MarkUp/2004/rdf-a.html#div248219168
<Steven> (Note that the rdf/a draft is now in W3C space)
Mark: see Creative
Commons example in rdf-a.html
... why would this example (section 6.1) mean anything but an
anonymous node?
... the only thing you can't do is refer to this node from
outside the document
... from the HTML side, you'd argue that id is creating
anonymous nodes
... from the RDF side, what we want is a way to express
anonymity
... we already have about
... and html:id
... so why do we need another mechanism?
... if you are really creating an external identifier use
about, otherwise use id
... this makes most things be 'anonymous' nodes unless you use
about
... I've tried this for a few examples and it seems to work
Ben: is there no longer subject inheritance; does about no longer inherit from parent?
Mark: Steven and I have gone back
and forth about this, not yet reflected in drafts
... we've concluded that anything that has only a property and
object should always be about the document itself
... so you no longer get nodes being bnodes by default as in
the previous draft
... see example in 5.2
... the link inside the blockquote would now have the document
as subject, not the blockquote
... Jeremy had commented that the inheritance rules were
complicated
Steven: another advantage is that this is more natural for the HTML community
Ralph: I suspect this is a much
cleaner design
... it likely solves Jeremy's observation that the October
draft appeared to create many more triples than the prose
implied
Ben: I think most document authors will not be explicitly interested in anonymous nodes, so this design feels better
Steven: this design is reflected
in the current XHTML2 editor's draft
... the rdf-a document is not in sync
Mark: if you wanted to talk about
the person 'Ben' in example 6.1, you could add <p
about='
http://ben.adida.net'>
<Steven> Oh, I see; OK I agree
Mark: anonymous nodes don't have any significance to HTML authors
Steven: I posted an example to
the HTML WG that I thought explained this
... e.g. referring to a table of contents document that does
not yet exist
Ralph: I can't imagine a case where it would be bad for nodes to have identifiers
Mark: I'm not suggesting that all nodes should have identifiers
Steven: this might undermine the URL mechanism
Mark: the difference is that in HTML the fragment identifier refers to a part of a document but in RDF the interpretation would be a node
ACTION: MarkB to email TF with his bnode proposal -- that (1) any node with @id is an anonymous node; (2) that an @href with a fragment refers to these nodes (like @rdf:nodeID)
Ralph: Ralph: What's the schedule for this new XHTML document in W3 space?
Mark: Mark: it requires some rewriting
Ralph: is http://www.w3.org/MarkUp/2004/rdf-a.html a copy of the October RDF-A document?
Mark: yes
Ralph: can the RDF-A module be made part of XHTML2 in such a way that it can be normatively referenced within XHTML2 spec and still reusable by other non-HTML vocabularies?
Mark: good idea; I will think about it
Ralph: Another possible path is
to combine RDF/A and metainformation module prose
... an HTML author needs this to make sense to them
... pieces of RDF/A formalism interspersed can be included
Mark: Mark: earlier drafts did
just that (beginner's guide to triples)
... HTML authors have been using LINK and META for years. This
new approach is easy to understand.
Ralph: Ralph: we're so close; some formalisms are important and we can educate people
Ben: we'll continue to look for
feedback from TomB and Gavin
... Creative Commons really wants people to use the RDF/A
mechanism in HTML4 and XHTML1
... but we don't want this to delay the XHTML2 work at all
Mark: the only missing feature
for the Creative Commons example is qnames
... the about attribute is new and 'name' became 'property'
within meta but the CC examples don't use these
Ben: if XHTML1 with the RDF/A
Creative Commons example can be validated, that's what we'd
like
... Creative Commons would like an HTML4 solution that looks a
lot like the XHTML2 solution
Ralph: the whole of GRDDL may not
necessary for a number of use cases
... if we can't find anyone who needs the whole thing
Ralph: if we cannot find a community who wants the full generalization of GRDDL perhaps the specific subset of the RDF-A/XHTML2 solution that can be made to work in HTML4 is an approach more acceptable to implementors?
Ben: even if GRDDL is determined not to be essential to RDF-in-XHTML, I believe it could be useful to other Semantic Web applications and the SWBPD WG may still wish to consider it
next meeting 26 April