See also: IRC log, previous 2008-07-10
<markbirbeck> I'll be on IRC for a little while, but have to leave early to get kids.
<ShaneM> Have people seen umbel.org ?
ACTION: Ben to talk to Allan about CR ISSUE-122 [recorded in http://www.w3.org/2008/07/10-rdfa-minutes.html#action15] [CONTINUES]
ACTION: Jeremy to write up TQ's contribution for impl report and wiki [recorded in http://www.w3.org/2008/07/10-rdfa-minutes.html#action16] [CONTINUES]
Jeremy: I've written the report, will post soon
<JeremyCarroll> current draft report
ACTION: [DONE] Manu to write up current issues with information resource vs. actual resource with regard to Audio RDF vocabulary. [recorded in http://www.w3.org/2008/07/10-rdfa-minutes.html#action14]
<msporny> uri-vs-resource-ambiguity-problem
ACTION: Michael to rewrite and extend http://rdfa.info/wiki/Tutorials for RSS (maybe with seq) [recorded in http://www.w3.org/2008/07/10-rdfa-minutes.html#action08] [CONTINUES]
ACTION: [DONE] Shane to make ISSUE-121 editorial change and report back [recorded in http://www.w3.org/2008/07/10-rdfa-minutes.html#action13]
ACTION: Shane respond to commentor for ISSUE-121 [recorded in http://www.w3.org/2008/07/17-rdfa-minutes.html#action06]
ACTION: Shane to write home page for SPREAD. [recorded in http://www.w3.org/2008/07/03-rdfa-minutes.html#action12] [CONTINUES]
ACTION: Ben followup with Fabien on getting his RDFa GRDDL transform transferred to W3C [recorded in http://www.w3.org/2007/11/15-rdfa-minutes.html#action01] [CONTINUES]
ACTION: Ben to follow up with Elias on JavaScript test harness + EARL [recorded in http://www.w3.org/2008/07/03-rdfa-minutes.html#action13] [CONTINUES]
Ralph: do we consider this important enough to keep at this point?
Manu: they wanted to get it for the
implementation report
... it was going to make Michael and Ben's life easier to create the
implementation report for Javascript
<Zakim> JeremyCarroll, you wanted to mention TQ's conformance statement
Jeremy: while writing the report for
TopQuadrant, I noted the following:
... any advice on anything better to say?
<JeremyCarroll> The conformance of the RDFa input depends on the conformance of the RDFa parser being used, currently Fabien Gandon's transform. Therefore we do not submit a separate report.
<JeremyCarroll> It is not clear what conformance tests we should do on the RDFa editing functionality. As is, the normal process of software maintenance should ensure that all output from the RDFa editor is legal RDFa input, with the expected triples, and most reasonable RDFa input can be produced with the editors. The autocomplete facility will currently generate qnames rather than CURIEs, since the other RDF formats currently do not use CURIEs. If and when other RDF
Shane: QNames are technically a subset of
CURIEs
... so you _are_ generating CURIEs
Jeremy: just a question about providing an implementation report for RDFa consumers
Manu: for the implementation report, we have a
set of implementations that run against the RDFa test harness
... if an RDFa parser is setup correctly it can be [automatically] run
against the test harness
... so the expectation is to say whether or not you pass each test
Jeremy: the existing tests are for RDFa
parsers. TopQuadrant integrates parsers written by others
... for the editing functionality it's hard to know what to say
Manu: my understanding is that the implementation report planned to only cover parser tests
Ralph: perhaps TopQuadrant could show how to use its tools to generate the test examples (or their equivalent)
Jeremy: nice idea
Manu: yes, would be nice to describe a full round-trip; generating then parsing RDFa
Shane: in the spec we only define conformance
for _parsers_
... in the implementation report we can have sections that talk about other
supporting tools
... so, e.g., user agents; IE, Opera, FireFox
Ralph: yes, for example it would be good to mention SearchMonkey, even though they don't yet support all of RDFa
Shane: it's also good to document that browsers
don't break when they're fed RDFa
... though we don't actually have any user agent conformance clauses
ACTION: [DONE] Manu write haudio examples for wiki [recorded in http://www.w3.org/2008/06/26-rdfa-minutes.html#action06]
Manu: these are 1-1 mappable to the microformat
vocabulary
... Toby Inkster has written a tool to intersperce RDFa in haudio
microformat
ACTION: Manu write the perl code for Slashdot [recorded in http://www.w3.org/2008/06/26-rdfa-minutes.html#action03] [CONTINUES]
ACTION: Mark create base wizard suitable for cloning [recorded in http://www.w3.org/2008/06/26-rdfa-minutes.html#action12] [CONTINUES]
ACTION: Mark write foaf examples for wiki [recorded in http://www.w3.org/2008/06/26-rdfa-minutes.html#action07] [CONTINUES]
ACTION: Michael to create 'RDFa for uF users' on RDFa Wiki [recorded in http://www.w3.org/2008/03/13-rdfa-minutes.html#action12] [CONTINUES]
Ralph: I hand-wrote a bunch of comments on that last week, never got to enter them :(
ACTION: Ralph think about RSS+RDFa [recorded in http://www.w3.org/2008/06/26-rdfa-minutes.html#action11] [CONTINUES]
Manu: there's been a lot of work in the SIOC
vocabulary that has to do with posts and the authors of posts
... SIOC is meant for bloggers and folks who post news
Shane: regarding haudio, Manu mentioned generalized media RDF. Is there a video RDF?
Manu: yes, there's a base vocabulary for media
on which the audio and video vocabularies build
... media builds off of Dublin Core
Shane: for example, how might one annotate IMDB with RDFa
Manu: this vocabulary is intended to help with
that; marking up television episodes, all the actors, grips, etc.
... we have 80-85% coverage of the use cases we've found so far
... we've looked at a large number of Web sites that publish this sort of
data
... ties in with BBC comments on non-accessibility of microformats data, esp.
the date format
... the accessibility problem is that screen readers attempt to read out data
that's only intended for the machines
Jeremy: who's the BBC contact?
Manu:I've been chatting with Michael Smethurst at the BBC about Audio/Video/Media RDFa
Manu: we've been seeing more traffic on the
lists noting that it's nice to have XHTML support but what about HTML4 and
HTML5
... we've pointed out that the markup won't break HTML4/5 tools
... but the users want validation
Ralph:Well, HTML4 doesn't have a working
group
... HTML5 has it in their charter to support extensibility for RDFa plus some
others
<Zakim> JeremyCarroll, you wanted to note this issue should go through to ac review and to
Jeremy: this issue of the more extensive
ambition of the RDFa folk won't go away
... we do want RDFa used with HTML4
... the conversations on the mailing list show that some just want to get the
job done while others are concerned about conformance
... I think we have to acknowledge that the discussion continues though we've
made a particular choice and it's the right one
... we've specified what can be specified correctly while acknowledging that
some will use it with HTML4
... the tone of the primer does not rule out RDFa+HTML4 even though we can't
specify it
... we know that _in practice_ RDFa+HTML4 is usable even though it won't
validate
Manu: could we create something that would work with the W3C validator?
Shane: the W3C validator uses NSGMLS through a
perl interface (opensp)
... opensp in SGML mode
... validating a document against a DTD
... it will do this for any DTD that it is aware of
Manu: could we create a DTD for RDFa+HTML4?
Shane: SGML mode does not understand @xmlns
... so we'd need a different way to define prefixes or a way to persuade the
validator to ignore @xmlns errors in certain types of documents
Ralph: we don't actually have a charter to do
anything except XHTML
... so we're on weak ground to ask for any changes in the validator
Shane: what would be the best way for us to support RDFa in an HTML4 environment? Does it make sense to push @xmlns into this environment or should we create another prefixing mechanism?
Manu: it would be really nice to not have to
change anything
... or add a flag to parsers indicating whether they're processing HTML4/5
and find a different attribute than @xmlns
<markbirbeck> Hey guys...really sorry I'm not there. However, I think the easiest way to get out of the @xmlns bind is via CURIEs, and not RDFa.
<markbirbeck> Will have to explain more either next week, or on the list.
<markbirbeck> (Unless I work on a strawman with Shane....)
Manu: everything other than @xmlns maps cleanly
Shane: yes, everything else works fine
Jeremy: how many actual HTML4 tools will have a
problem with RDfa?
... the validator doesn't matter in the big picture
Manu: the validators matter for adoption
<ShaneM> I like the idea of @ns
Manu: authors don't want to hear complaints from their bosses about documents that don't validate
Jeremy: so just ask the validator to ignore XML namespaces
Shane: I think there is already a control for
this
... when it's validating in XML mode via a DTD it has to ignore namespace
errors
<msporny> Mark - it would be really great to hear your thoughts on this (on list, or via the wiki)
<ShaneM> @ns="prefix URI prefix2 URI..."
Shane: I already have the code to patch the W3C
validator to do this
... but other authoring tools also have 'validate me' buttons
... we'd want those other validators to also work
<markbirbeck> Manu...really sorry for absence. Speak to you all soon...gotta go. :)
Shane: but sure, we can tell people it's OK to put RDFa in HTML
Manu: we can't say there's backing from any W3C process but we can point to a starting place
Ralph: I think the HTML5 spec will have to
address something like namespaces
... we should not get out ahead of that
... whether they accept @xmlns or not, we need to wait
Shane: I disgree; we've overloaded @xmlns and
that's a mistake
... it's a convenience but in the real world @xmlns is not the way to extend
grammars
... if HTML5 does not define @xmlns we could still do something else and
HTML5 wouldn't care
... vocabulary extention needs something other than XML namespaces
Manu: the issue we're up against is how to stuff this into validators
Shane: I like the @ns= idea
Manu: it would be nice to have something that
shows people how to put RDFa into HTML4/5 and get it validated
... would the W3C object?
Ralph: no one moderates what we put in our wiki :)
Manu: it's becoming evident that there are
right ways and wrong ways to use RDFa vocabularies
... for example, I wasn't aware that the Dublin Core vocabulary is now a
legacy vocabulary and we should really be using DC Terms
... would be nice to have a tool that reminds people about current best
practices
... like a lint for RDFa
... suggestions for improving markup or flagging mistakes
<Zakim> ShaneM, you wanted to discuss lint for rdfa/xhtml+rdfa
Shane: there's a validator side-project that
has a framework for doing this sort of thing
... I'm certain that Olivier would welcome our help to do this
... I'll even volunteer
... the tricky part will be deciding the rules
... the tool is very good about suggesting how you should have done X or Y
... it was written to support the Appendix C validation of HTML
Manu: first thing is to come up with some best
practices
... document these in the wiki
... when we get 10 or 15 we could roll them into the tool
ACTION: Manu: Start Best Practices Section on the rdfa.info wiki [recorded in http://www.w3.org/2008/07/17-rdfa-minutes.html#action16]
Shane: makes sense
Shane: I'm especially interested in feedback on
my post "Re:
RDFa Primer comment"
... we need to have a consistent message from this Task Force
Manu: agree
... and we'll have to do more work on HTML4/5 before saying anything
[adjourned]