See also: IRC log
<DanC> agenda points to http://www.w3.org/2006/12/13-grddl-wg-minutes.html
<HarryH> RESOLVED to approve http://www.w3.org/2006/12/13-grddl-wg-minutes.html as a true record
<HarryH> ACTION: Fabien to post to sawsdl list relevant questions about RDF mapping and relationship to GRDDL [CONTINUES] [recorded in http://www.w3.org/2006/12/20-grddl-wg-minutes.html#action01]
Chime: xml primer example use-case, lightweight ontology for medical records
<HarryH> This would fulfill Dave Booth's request in our comments list: http://lists.w3.org/Archives/Public/public-grddl-comments/2006OctDec/0000.html
Chime: GRDDL transform, HL7 document -> RDF
Dan: how to fit GRDDL transformation into the document without breaking the schema?
<HarryH> namespace document is a URN, which is a slight problem.
Chime: schema might be closed.
Chime: But we could use schema annotations, don't think schema is closed to import infoitems from other namespaces.
Harry: plain XML would be a nice
addition to the primer
... ian is ok with additions to the primer document
Dan: check it into the test suite
<HarryH> ACTION: Chime's add HL7 plain XML health care use-case and check it into test suite. [recorded in http://www.w3.org/2006/12/20-grddl-wg-minutes.html#action02]
<HarryH> ACTION: DanC to add N3/turtle mime type to Atom/turtle test case. noting the unregistered status [CONTINUES] [recorded in http://www.w3.org/2006/12/20-grddl-wg-minutes.html#action03]
<HarryH> ACTION:DanC to write rules about XSLT 1.0 processing [CONTINUES] [recorded in http://www.w3.org/2006/12/20-grddl-wg-minutes.html#action04]
Ben: in RDFa html metadata by providing transforms, microformats -> RDFa
keeps the RDFa in the DOM, so you can grab individual items
use GRDDL like mechanism to make RDFA
can this be done with the current GRDDL spec?
<chimezie> we could test this with the current test suite impl..
Ben: output-issue might solve
... requires clients to request RDFa output of given profile
<HarryH> http://www.w3.org/2004/01/rdxh/spec#issue-output-formats is technically still open.
Dan: GRDDL transformation pointer, source document => HTML/RDFa
Ben: author must use specific profile to make a transformation to RDFa
<chimezie> the agent/client needs some way to know how to parse as RDFa
Ben: how would the client know what transformation to use? RDF, turtle or HTML?
Harry: browser would decipher the transformation and look for RDFa results
murray: GRDDL tells us how to find 1 or more transformations
agent is RDFa aware and GRDDL aware
agent is able to look into class attributes, link element has special @class
chime: RDFa, does it have a specific mimeType?
Ben: no special mimeType - schema
or DTD instead
... DTD mostly - murray's direction is a good one
murray: to identify a grddl transformation can be in a LINK element
Dan: hCard page, profile="hcard uri here", this works for GRDDL transform
Murray: namespace transformation keyword, profile transformation keywork... HTML profile document could have link element with special class name for RDFa URI
RDFa aware user agant can follow its nose and find the correct LINK/REL uri and do correct transformation
<chimezie> that's a very specific follow your nose scenario
Dan: worries about choosing a good class attribute
<DanC> (I'm worried about squatting on the author's choice of class attribute values)
Harrry: run all transformation, user agent chooses the right output it wants
murray: try not to complicate GRDDL by using existing attributes in HTML namespace
Harry: mimeType for HTML RDFa not part of GRDDL
Ben: differenciate between transforms and/or type the transforms
Dan: run the transform and see what you get instead of typing
Ben: annotate transformation or not?
<HarryH> Annotate the transformation pre-running it
murray: leave it up to the agents to discover the annotation
Dan: GRDDL spec says all are relevant, but if you only want to run certain ones, go ahead
<DanC> (ben, as fodder for hGRDDL testing, there's an existing hCard test: http://www.w3.org/2001/sw/grddl-wg/td/testlist2#hcard1 )
<HarryH> ACTION: BenA to write a sample hGRDDL transformation [recorded in http://www.w3.org/2006/12/20-grddl-wg-minutes.html#action05]
<DanC> (yes, output formats is closed: http://lists.w3.org/Archives/Public/public-grddl-wg/2006Dec/att-0011/06-grddl-wg-minutes.html__charset_us-ascii )
<DanC> (though there are some pending actions)
murray: RDFa, transformation that converts to 'augments HTML RDFa with notations'
wouldn't you want ALL the RDF to annotate ALL of it back into HTML?
Ben: this could be done by keeping other GRDDL pointers
Dan: not all the transformation tell you where an RDF triple belongs back into the HTML
harry: pending edit, to remove the words "scrapping"
<HarryH> ACTION: HarryH to make sure edit "scraping" edit is. [recorded in http://www.w3.org/2006/12/20-grddl-wg-minutes.html#action06]
Harry: what do do about this without Ian?
Dan: spec pretty much remains the same, otherwise the primer doesn't get the introduction
Murray: happy to withdraw the cross-document intro
<HarryH> ACTION: Ian to reconsider comments on cross-document introduction [WITHDRAWN] [recorded in http://www.w3.org/2006/12/20-grddl-wg-minutes.html#action07]
murray: given a doc with a DTD or schema or xinclude....
<DanC> (the HTML DTD has a default 'rect' attribute)
murray: xinclude agumnets document if it is realized
XSLT may or maynot be all of the values if it were xinclude processed
might return two different RDF graphs given if or if not the includes are included
people have to deal with it themselves, or
MUST do things so you get what the author intended
with XI namespace
transformation in xi namespace, so that xinclude were incountered - discover it, and add something into result
<HarryH> I think there's a problem the xi namespace is not usually the root element.
<DanC> Subject: XInclude or Not, that is the question
<DanC> Date: Thu, 23 Nov 2006 12:24:30 -0500 (11:24 CST)
murray: authors intention to get with xinclude expanded
xinclude namespace is in the document
not in the same namespace, follow you nose for the subtree?
Dan: different GRDDL design
murray: not xi namespace, but explain
simple transform which you can include and make the result graph aware that you missed an xinclude
Dan: if you want you transformation to do xinclude, then make your transformations do xinclude
harry: not mandate any further xml processing, but instead make recommendations
murray: if we add a paragraph, about issues and problems
in the spec
Chime: we currently recomment xproc
<DanC> +1 The GRDDL WG does not mandate XInclude and further XML processing on the source document is wanted to be run before the transformation to RDF, and write a caveat in the specification recommending XPROC or using (or something like Murray's xinclude transformation) when this is the case.
<DanC> hmm... faithful-infoset
<HarryH> PROPOSAL: The GRDDL WG does not mandate XInclude and further XML processing on the source document before the transformation to RDF, and write a caveat in the specification recommending XPROC when this is the case.
harry: what exactly is the mimetype issue now?
<DanC> PROPOSED: to add issue faithful-infoset (split it out of issue-mt-ns) and close it a la: The GRDDL WG does not mandate XInclude and further XML processing on the source document before the transformation to RDF, and write a caveat in the specification recommending XPROC when this is the case.
<benadida> not me, but I'm comfortable abstaining
<DanC> brian, you'd rather think more? you'd rather abstain?
<DanC> I don't know if this has been mentioned before, but that rather than us
<DanC> trying to give a definitive answer on xinclude processing, we could
<DanC> leave it to the publisher. " -- McBride
<DanC> Wed, 13 Dec 2006 09:16:51 -0000
<HarryH> RESOLVED: to add issue faithful-infoset (split it out of issue-mt-ns) and close it a la: The GRDDL WG does not mandate XInclude and further XML processing on the source document before the transformation to RDF, and write a caveat in the specification recommending XPROC when this is the case. (benA and BrianS abstaining)
<HarryH> ACTION: Murry to draft paragraph giving us caveat for faithful infoset issue closure. [recorded in http://www.w3.org/2006/12/20-grddl-wg-minutes.html#action08]
<HarryH> PROPOSAL: Given that a base URI parameter is a parameter whose value is the base URI of the source document, the WG RESOLVES not to define a base URI parameter for transforms.
Dan: is ok, and results in no edits to spec
eRDF uses base-construct
harry: can't change
... defining default para
<HarryH> RESOLVED: Given that a base URI parameter is a parameter whose value is the base URI of the source document, the WG RESOLVES not to define a base URI parameter for transforms.
<HarryH> ACTION: Harry to check in Bwm's test case: http://lists.w3.org/Archives/Public/public-grddl-wg/2006Dec/0028.html [recorded in http://www.w3.org/2006/12/20-grddl-wg-minutes.html#action09]
<HarryH> BrianS - are you available in 2 hours or so?
<HarryH> OK, let's try to finish some of this off then :)
<HarryH> ACTION: Harry and Brian to rewrite second part of primer to use Brian and Dan's instance data. See message from Brian and message from Harry. [CONTINUES] [recorded in http://www.w3.org/2006/12/20-grddl-wg-minutes.html#action10]
<DanC> (hmm... when to do another release of the GRDDL primer WD?)
harry: next meeting will be...
<DanC> RESOLVED: to cancel 27 Dec
<DanC> anybody _want_ to meet on 3 Jan?
<HarryH> Does anyone want to meet on 3rd?
<DanC> RESOLVED: to cancel 3 Jan
<HarryH> Next meeting: Janurary 10th
<DanC> RESOLVED: to meet next 10 Jan, benadida to scribe