See also: IRC log
Previous: 2006-02-13 http://www.w3.org/2006/02/13-htmltf-minutes
Ralph: I would like to talk about the implementation when Jeremy is here
ACTION: Ben talk off-line with Jeremy about a realistic implementation schedule [recorded in http://www.w3.org/2006/02/21-htmltf-minutes.html#action01]
Steven: Mark and I were discussing offering a lightning talk on RDF/A at the Tech Plenary. There is a session on microformats earlier in the day. We are thinking about two talks: one on RDF/A generally and a second on the relationship between RDF/A and microformats
Ben: I'm giving a talk at Semantic Technologies in 2 weeks on RDF/A but I am available for a telecon on 6 March
RESOLVED: next meeting 6 March
ACTION: Ben summarize options discussed for issue 12 [recorded in http://www.w3.org/2006/02/13-htmltf-minutes.html#action12] [DONE]
ACTION: Ben summarize the syntax options for issue 5. (Local) blank node identifiers [recorded in http://www.w3.org/2006/02/13-htmltf-minutes.html#action11] [DONE]
<benadida> RDF/A Primer 1.0 ($date 2006/02/19 01:31:56 $)
ACTION: Ben resolve the document encoding issue [recorded in http://www.w3.org/2006/02/13-htmltf-minutes.html#action10] [DONE]
Ben: I changed the XSLT, hope this fixed the problem.
Steven: http://www.w3.org/2001/sw/BestPractices/HTML/2006-01-24-rdfa-primer.html is being served to me as UTF-8 [and displays correctly in Opera]
ACTION: Jeremy followup on edge case [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action03] [CONTINUES]
ACTION: Jeremy followup with Mark on the question of multiple triples from nested meta and add to issues list [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action01] [CONTINUES]
ACTION: Jeremy propose wording on reification [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action02] [CONTINUES]
ACTION: once Steven sends editors' draft of XHTML2, all TF members take a look and comment on showstopper issues only [recorded in http://www.w3.org/2006/02/06-htmltf-minutes.html#action01] [CONTINUES]
Steven: had hoped to have editor's draft done last week. Editor's draft will be done this week because the HTML WG plans to discuss it next week face-to-face
ACTION: Ralph add a sentence to 2.2.3 pointing to a citation for the triples syntax [recorded in http://www.w3.org/2006/02/13-htmltf-minutes.html#action09] [CONTINUES]
ACTION: Ben update the editor's draft to add [missing triples] to section 2 [recorded in http://www.w3.org/2006/02/06-htmltf-minutes.html#action09] [DONE]
ACTION: Ben start separate mail threads on remaining discussion topics [recorded in http://www.w3.org/2005/12/06-swbp-minutes#action04] [CONTINUES]
Ben: I want to keep this action open as a reminder
ACTION: Ben to draft full response to Bjoern's 2004 email [recorded in http://www.w3.org/2006/01/24-swbp-minutes.html#action03] [CONTINUES]
Ben: I want Mark's input before closing this
<benadida> RDF/A Containers ($date 2006/02/12 20:08:25 $)
Ben: I had two different approaches as I
started to write:
... 1) modifying the triples produced by the previous rules
... decided this was a bad idea as the previous rules were created with the
idea that we were adding semantics where none existed before
... 2) make RDF/A containers add new triples in addition to the previous
ones
... all triples generated by the previous rules are still generated
... option 2 seemed best as we recognized that HTML lists do have
semantics
Steven: this is one place where we'd be specifying a tight binding with HTML. Do we want to be able to use RDF/A with other markup languages?
Ben: this is really just syntactic sugar for the HTML case; perhaps other languages would have similar sugar.Look at example in 2.4, for instance; the HTML behaviour would make the list items clickable and the additional RDF/A behaviour is to make triples. I believe the edge cases; e.g. an LI that specifies it's ordinal number, have been considered
<MarkB_> Mark would like to point out that then when I suggested this approach (;) ) I did point out that it is *one-directional*. I.e, we say that in HTML these items generate these triples. But we don't put these elements into RDF/A.
<MarkB_> Which also means that we don't say 'if you want RDF containers do this'. As I say, it only goes in one direction.
<MarkB_> if we had a 'core RDF/A spec' to be imported into other languages, I'm not sure that we would include ol/li/ul etc. in it.
<MarkB_> So HTML has list mark-up, but RDF/A doesn't.
<RalphS> that's pretty much what Ben said, Mark; we don't expect other languages necessarily to have to follow this particular sugar
Steven: it doesn't look like it's dangerous; it doesn't override any other semantic, rather it makes those semantics explicit
Ben: the RSS example shows why this is particularly useful
Ralph: what about DL?
Ben: I expect that DL behaves in the same way but it doesn't use LI. It might be different triples
<MarkB_> DL is awkward...I think it it has numerous pairs of items, with no unique parent to each pair. I dont think it's like other lists.
Ralph: I think we should defer DL for the next version. Since the DT is clearly the list member but there's also a clear relationship between the DD and the DT and I think we'd want to be sure to get that relationship 'right' and make it explicit
RESOLVED: defer DL to a future version
ACTION: Ben note that DL is deferred to a future version [recorded in http://www.w3.org/2006/02/21-htmltf-minutes.html#action13]
ACTION: Ben add DL to the future issues list [recorded in http://www.w3.org/2006/02/21-htmltf-minutes.html#action14]
Ralph: does this add a new rule for determining subject?
Ben: yes, but only for the new triples; see,
e.g. 3.3.2. Even without the about= attribute the cc:license
triples are generated
ACTION: Ben note that about="" is redundant in 3.3.2 [recorded in http://www.w3.org/2006/02/21-htmltf-minutes.html#action15]
Ben: the argument in favor of including
about is to be completely clear; it doesn't matter what the
enclosing HTML does; the new Container triples depend only on the enclosing
element
Ralph: this subject question would come to light in the implementation phase; i.e. does this new rule interact in any inconvenient way with the other rules?
Ben: these are really just extra rules so should not interact with other rules
Ralph: another question: consider 3.3.1; this construct is common in HTML so with this container addition we're causing many more documents to generate triples. Is this OK?
Steven: I don't think the author consciously cares in the general case. But there is a semantic distinction between OL and UL; ordered vs. unordered so 3.3.1 probably ought to be unordered as grocery lists probably are not ordered. I think the semantics are good; whether you use the triples or not, [the semantics they imply are] sort-of there anyway so this gives an advantage for those who do care
Ben: this connects to the OPML crowd in an interesting way. The only other consideration might be NL. Does NL have any special rendering rules?
Steven: not really. NL has a label element, which OL and UL don't so one of the possible renderings is a menu
<MarkB_> The opposite :) It gives you nothing that you don't get from ul/li/@href
<MarkB_> (ul and ol should get labels too)
Steven: it is Mark's opinion that UL and OL should have optional LABEL element as well
<MarkB_> It's an HTML discussion...but I don't think that NL is needed now that @href is allowed everywhere.
<Steven> and role="navigation"
Steven: NL was introduced before we had role, and specifically before role='navigation' so NL is really a historic artifact. The HTML WG may need to review past decisions. Let's leave NL as Ben has it for now and see what happens with NL
Ralph: new question: there are interesting
semantics if the element has a start= attribute
Steven: also value=.
value gives the ordinal number of the LI; e.g.
value='30' on LI is the same as start='30' on the
list element
Ben: if there's an explicitly given ordinal then we should respect it and increment from there
Steven: yes, this is really nice for HTML; up to now this semantics has only been recorded in stylesheets
Ralph: this gives us a reason to keep Containers in RDF
Ben: there's a 2004 document from Eric Miller saying Containers and Collections are peers
<benadida> it's in the RDF Primer
Ralph: editorial; please change "subtype" to "subclass" in 2.2
Steven: there's a call for developers' track proposals for WWW2006. I think we should do something about RDF/A
Ralph: there is not a developer's _day_ -- WWW2006 has a "developer's track"
Ben: I don't currently have plans to attend, though I note the W3C AC meeting closeby
Steven: we should talk about RDF/A as the solution to the last problems that microformats have. Microformats do a lot of similar things but can only talk about the current document. We should present RDF/A as next generation microformats
Ben: expect debate from Tantek
Steven: Jeremy and Mark are the chairs of the Developer's Track. I don't want to do this presentation alone; would like to be joined by someone from SemWeb
Ben: I will consider whether I can attend
Ralph: I need to decide this week whether I can attend, but I'm willing to help with a presentation.
"Proposals for presentations are requested by the deadline
of March 5th 2006."
-- http://www2006.org/developers/
ACTION: Steven draft a WWW2006 Developer's Track proposal [recorded in http://www.w3.org/2006/02/21-htmltf-minutes.html#action16]
[adjourned]