Previous: http://www.w3.org/2009/11/19-rdfa-minutes.html
See also: IRC log
<scribe> ACTION: Manu to ask somebody to draft errata text, clarifying that prefixes cannot be '_' character [recorded in http://www.w3.org/2009/11/12-rdfa-minutes.html#action10] [DONE]
<scribe> ACTION: Mark to author URIs in @about, @rel, @rev, @typeof and @datatype spec text [recorded in http://www.w3.org/2009/11/12-rdfa-minutes.html#action09] [DONE]
Mark: Yes, that's done... finding
link now.
<markbirbeck>
http://lists.w3.org/Archives/Public/public-rdf-in-xhtml-tf/2009Nov/0081.html
<scribe> ACTION: Someone let Ivan know what Open Document Format reference to use in the charter [recorded in http://www.w3.org/2009/11/19-rdfa-minutes.html#action08] [DONE]
<scribe> ACTION: Manu to aggressively push review of test cases via mailing list [recorded in http://www.w3.org/2009/10/29-rdfa-minutes.html#action08] [CONTINUES]
<scribe> ACTION: Manu to try and find other interested parties in RDFa WG. [recorded in http://www.w3.org/2009/10/22-rdfa-minutes.html#action08] [CONTINUES]
Manu: How is the process going?
Ivan: Formally, charter is under
review
... Waiting to hear back on comments.
... It is based on those reviews that the RDFa WG is formed or
not
... I think things are going well
... Want to get official vote request by the end of next
week... but it's up to the reviewers to get back to us.
... The charter starts the group on Feb 1st 2010
Manu: Should we meet in Jan?
Ivan: No reason why we shouldn't meet informally.
Shane: Actually, XHTML2 continues until we close the work we're doing.
Ivan: So, we can meet formally in January.
Manu: What are thoughts on moving
forward on this?
... Do we want to ask for proposals now?
Mark: Interesting that Toby and
Ivan are implementing this stuff straight away.
... Toby implemented a way of not requiring a colon
... That's good, but also comes with downside of solidifying
ideas that we may not be the direction we want to go.
... The whole @profile issue is still there.
... Maybe we can do content negotiation?
... Maybe the thing at the end of the URI could be anything?
like have an RDFa document or JSON?
... We may want to see if we're happy with the general idea of
prefix-less CURIE
... Then I think the second step is how to give somebody a list
of how to define these tokens externally.
Ivan: So, just to clarify -
next/prev/stylesheet would all be considered "tokens"
... So, what we are trying to find is how to define those
reserved words.
... XHTML is the only one that has these reserved words, SVG
doesn't?
Manu: Not sure that's true
Steven: Yes, not clear in the SVG Tiny 1.2 spec.
Manu: We might want to just do some spec text, then?
Mark: Might be good to discuss
some other points first?
... So, we got to a certain point with the discussion with Ben
- whether you regard a reserved word as a relative URI or a
token
... Both will work, if there is no token defined - it's a
relative URI, if a token is defined, we use that.
<Steven> I think the fact that we have relative URIs and keywords in the same position makes it harder to find a solution
Mark: With this recent change to allow URIs everywhere, we've got a few more options.
Manu: I'm concerned about
intermediate formats.
... What happens when you have vocab="http://example.org/terms"
property="foo" and you can't resolve "http://example.org/terms" ?
Mark: Well, you generate a triple
with "http://example.org/terms#foo"
and perhaps resolve that in the future.
... Well, Ben mentioned that "foo" would be placed in the
default namespace... which we didn't really want.
Ivan: my problem is a bit more
fundamental than that.
... I'm not fully convinced that we need to go down that
route...
... none of these solutions seem to be simple (for an
implementer) - still not convinced that we need to do
that.
... When we go to other dialects of SVG - I don't see what the
problem we're trying to solve for those other languages
are...
Mark: The problem is about trying to get Microformats-style simplicity for markup.
Manu: Yes
Mark: It could potentially give
RDFa the ability to do everything good that Microformats does
in a way that is rigorous.
... The good things about Microformats is that it's easy to
understand. Quick to get something up and going.
... The second thing that is related is that you get people
like Google that have one uber-vocabulary.
... It's the ability to just stick a @profile at the top of the
document rather than 12 namespaces, is useful for authors.
Steven: i am definitely a big fan
of simplicity of authoring.
... There are good proposals for automatic namespaces by Liam
et. al
Ivan: I agree with Steven on the
point that @tokens become less important when we have a
mechanism for defining prefixes/tokens.
... If we have that, then I think we can forget about this
@token issue, except for the XHTML tokens, which are
included.
Mark: If there is a URI mapping defined, then using it in an attribute that can take a CURIE, even without a colon, is sufficient.
Ivan: if we have a token/keyword
without a colon, what do we do with it.
Mark: So, let me give an
example...
<markbirbeck> @xmlns:dc="http://xyz"
<markbirbeck> dc:creator
<markbirbeck> dc:
<markbirbeck> @xmlns:creator="http://xyz#creator"
<markbirbeck> creator:
<markbirbeck> property="creator"
Mark: We can still argue that you don't need @tokens or you do - but this is a very tiny change.
Ivan: So, I could create a ton of
xmlns:xyz for all DC terms.
... This make sense only if there is a way to specify those
xmlns elsewhere.
<markbirbeck> <html
<markbirbeck> xmlns:Agent="http://purl.org/dc/terms/Agent"
<markbirbeck> xmlns:Person="http://xmlns.com/foaf/0.1/Person"
<markbirbeck> xmlns:title="http://xmlns.com/foaf/0.1/title"
<markbirbeck> xmlns:fn="http://xmlns.com/foaf/0.1/name"
<markbirbeck> >
<markbirbeck> <div
<markbirbeck> about="http://www.ivan-herman.net/me"
<markbirbeck> typeof="Person Agent"
<markbirbeck> >
<markbirbeck> <h1>
<markbirbeck> <span property="title">Dr</span>
<markbirbeck> <span property="fn">Ivan Herman</span>
<markbirbeck> </h1>
<markbirbeck> </div>
<markbirbeck> </html>
Mark: if we go for this small change, it still doesn't matter how we get those URI mappings.
<Zakim> ShaneM, you wanted to speak for being able to define new reserved words
Shane: I don't have a problem
with this @token proposal.
... Without having an external mechanism, it's not terribly
useful.
... In todays spec, we have these things called "reserved
words"
... Those only work for @rel/@rev
... That's fine.
... So, we're extending these to all CURIE-aware
attributes?
Mark: Not necessarily - we have this new mechanism for @property for resolving reserved words/CURIEs.
Shane: So, putting on my DAISY
hat, they want to be able to define reserved words.
... We still need to be able to have a way to change the
default collection of reserved words.
Manu: We still don't have a solution for defining extended reserved words.
<ShaneM> aside: what if I have multiple profile references and there are collisions after taking the union of the prefix mappings? Last takes precedence?
Ivan: Let's say we have a @profile - we could pull @tokens in from an external document.
Mark: yes, agree with Ivan
... With this minor change, it helps us solve the other
Microformats-like-markup problem.
<ShaneM> I want a "prefix import mechanism" not a "namespace import mechanism"
<Zakim> ShaneM, you wanted to say that I dont want to rely on these external namespace proposals. we are too tied to namespaces already!
<scribe> ACTION: Mark to generate spec text for @token and @prefix [recorded in http://www.w3.org/2009/12/03-rdfa-minutes.html#action06]
<ShaneM> I suddenly grok why @prefix is a bad name
<scribe> ACTION: Mark to generate spec text for pulling in external vocabulary documents [recorded in http://www.w3.org/2009/12/03-rdfa-minutes.html#action07]
<scribe> ACTION: Manu to generate spec text for pulling in external vocabulary documents. [recorded in http://www.w3.org/2009/12/03-rdfa-minutes.html#action08]
<ShaneM> @cmap="token URI"
<ShaneM> where cmap == curie mapping
http://rdfa.digitalbazaar.com/test-suite/
<p xmlns:xml="http://www.w3.org/XML/1998/namespace" property="xml:test">Test</p>
Manu: Thoughts?
Shane: keep it on hold.
Ivan: Maybe we could disallow it in the spec?
Mark: It's an edge case that we
don't refer to.
Manu: So, we'll keep it on hold and I'll contact LibXML
developers to fix the library, once it is fixed, we will approve the
test case.
<scribe> ACTION: Manu to get in touch with LibXML developers about TC 142. [recorded in http://www.w3.org/2009/12/03-rdfa-minutes.html#action09]
<span xmlns:test="http://example.org/error/" property="testŀ:test">Test</span>
Ivan: So I can only have ASCII characters there?
Shane: Yes.
Manu: Ok, so this is an invalid XHTML test, but we will be revisiting this issue for HTML4 and HTML5.