See also: IRC log
<jjc> PROPOSED: to approve Minutes of GRDDL WG meeting, 23rd May 2007 as a true record
<jjc> Chime seconds
<jjc> so RESOLVED
<jjc> (decision on next meeting to be taken at end)
jjc: decide later if next conf next week is needed
<jjc> three ACTIONS done
JJC: Embeded RDF 4 is embeded is implemented and implementations pass the tests
<jjc> PROPOSED to approve #embedded-rdf4 http://www.w3.org/2001/sw/grddl-wg/td/grddl-tests#embedded-rdf4
<jjc> RESOLVED to approve #embedded-rdf4 http://www.w3.org/2001/sw/grddl-wg/td/grddl-tests#embedded-rdf4
JJC: Test case for error in a transform http://www.w3.org/2001/sw/grddl-wg/td/pendinglist#error1
<jjc> http://www.w3.org/2001/sw/grddl-wg/td/pendinglist#error1 input file is a test case for errors
<jjc> http://www.w3.org/2001/sw/grddl-wg/td/broken4.xsl is the transform that produces some triples
JJC: this test over specifies the behaviour ; this test starts to produce triples and then fails.
<john-l> GRDDL.py passes the single test in <http://www.w3.org/2001/sw/grddl-wg/td/pendinglist>.
JJC: I will abstain rather than vote against but I don't like it because it over specifies the behaviour
<chimezie> so we need an explicit (yes, this resolves our issues) to close the comment? i.e., it's mostly a process issue?
chimezie: we should discuss more on the list.
JJC: we have adequate answers from the group.
<chimezie> from XSLT specification:
<chimezie> If the terminate attribute has the value yes, then the XSLT processor should terminate processing after sending the message.
JJC: if a transform fails the produced triples should be discarded
<chimezie> PROPOSED to approve test (with URI) http://www.w3.org/2001/sw/grddl-wg/td/grddl-tests#error1
<chimezie> input: http://www.w3.org/2001/sw/grddl-wg/td/withErrors
<chimezie> output: http://www.w3.org/2001/sw/grddl-wg/td/one.rdf
<rreck> i second
<jjc> jeremy abstains
<jjc> APPROVED test (with URI) http://www.w3.org/2001/sw/grddl-wg/td/grddl-tests#error1
<jjc> ACTION: chime to update test editors draft with #error1 [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action01]
<jjc> ACTION: john to produce new EARL output including #error1 [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action02]
<jjc> ACTION: jjc to produce new EARL output including #error1 [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action03]
<scribe> ACTION: john to update result page [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action04]
<scribe> ACTION: jeremy to email back to the working group [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action05]
<chimezie> ACTION: chimezie update test spec to indicate approval for #embedded-rdf4 [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action06]
<chimezie> i assumed that was included with my action on #error1
Re: base-uri and XSLT 1.0 Dave Beckett (Saturday, 5 May), and informal reply, 2007AprJun/0071. http://lists.w3.org/Archives/Public/public-grddl-comments/2007AprJun/0051.html http://lists.w3.org/Archives/Public/public-grddl-comments/2007AprJun/0071.html
<jjc> [[Maybe it's a bug in the above sheets.]] from Dave Beckett
Fabien: see my message http://lists.w3.org/Archives/Public/public-grddl-wg/2007May/0069.html
I the RDFa2RDFXML is producing triples with relative URIs when there is no parameter specified
<jjc> ACTION: jjc to e-mail Dave Beckett with not a bug in the spec, but maybe in some stylesheets [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action07]
<john-l> ACTION: DanC to remove base param from online GRDDL service and home2rss.xsl [CONTINUED] [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action08]
<john-l> ACTION: HH to remove or clarify base param in eRDF transformation [CONTINUED] [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action09]
<chimezie> i recall a continued action (which blocks the test case doc from going PR) to cross reference tests to GRDDL mechanisms by section
<john-l> ACTION: jjc to write text about base-uri, which may become normative. [CONTINUED] [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action10]
<scribe> ACTION: Fabien to clarify purpose of uri param in RDFa2RDFXML.xsl [DONE] [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action11]
issue-dbooth-9a: GRDDL should be usable in a messaging pipeline http://lists.w3.org/Archives/Public/public-grddl-comments/2007AprJun/0069
<jjc> Hence, the spec
<jjc> should be clear that the lack of a base URI is only a problem if it is
<jjc> needed (to resolve a relative URI) but not present.
<jjc> David Booth
JJC: David says the lack of a base URI should be a problem only for relative URI.
<chimezie> it seems to me that rfc39XX covers the case where there is no URI associated with the input document
<chimezie> either the containing context provides it, or the application provides it (the outermost ring)
JJC: the source is not part of RDF
<jjc> RFC 3986, section 5.1.4 application provides default base URI
Murray: I have a resource and I extract statements A B and C from it these statements should be "somewhere"
<jjc> http://www.w3.org/2000/03/rdf-tracking/#rdfms-contexts is a postponed issue about contexts
Chime: base is used only to resolve the relative URIs not to locate the resource.
<jjc> Subject to security considerations below and local policy as expressed in its configuration, given a URI I of an information resource IR, and an XPath node N for a representation of IR, a GRDDL-aware agent should:
<jjc> ]] grddl spec
<chimezie> the dereference operation happens here: ?IR log:uri [ fn:doc ?R ].
JJC: the context of a triple is a postpone issue of the RDF group and outside the scope of this group
john-l: providing a default base URI shouldn't be a big problem
<scribe> ACTION: Chime to respond to David on issue-dbooth-9a: GRDDL should be usable in a messaging pipeline http://lists.w3.org/Archives/Public/public-grddl-comments/2007AprJun/0069 [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action12]
<jjc> (and also 9c Base URI should only be required if needed)
issue-dbooth-9c: Base URI should only be required if needed http://lists.w3.org/Archives/Public/public-grddl-comments/2007AprJun/0074
JJC: in the pipeline example if you have to give a URI it would have to be a default one.
<john-l> [[ the problem is that it *fails* to define results in terms of representations ]] -- #issue-dbooth-9a
Murray: after three steps in the pipeline you don't have the URI anymore
<jjc> If F and G are GRDDL results of IR, then the merge [RDF-MT] of F and G is also a GRDDL result of IR.]
Murray: all information resources have a URI no URI means it is not an information resource
<chimezie> "" rdf:type foaf:Person
<chimezie> "" :isTheContextOf <..rdfStatement URI..>
Fabien: wouldn't reification be another mechanism too?
Fabien: Although the triples and their reification are not really linked...
<rreck> i agree with that statement
Murray: it is important to have that URI available all along the process
JJC: the base URI can come from a lot of sources / mechanisms
JJC: too big change of the spec to be made at this point.
<chimezie> practically speaking the URI is only valuable for resolving relative URIs, (empty URI references can be used to make statements about the containing document, but this is different from making statements about provenance)
<rreck> chimezie: you mean the baseURI, correct?
<rreck> otherwise provenance would be inferred
<chimezie> rreck: i mean the URI from where the XML concrete syntax was dereferenced (fetched) from - which sometimes is the same as the BaseURI if no BaseURI is specified within
<scribe> ACTION: chimezie to start an index of tests by normative assertion (proxy for feature) [CONTINUE] [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action13]
<scribe> ACTION: HarryH to start an index of tests by issue (not urgent; due 30 May) [CONTINUE] [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action14]
<scribe> ACTION: HH to ask Dom to check his XSLT against current GRDDL tests [CONTINUE] [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action15]
<scribe> ACTION: john-l add link from report to spec explaining multiple results and not applicable relationship [DONE] [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action16]
<scribe> ACTION: jjc to review primer [CONTINUE] [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action18]
<scribe> ACTION: john-l to draft an Excel spreadsheet primer section [CONTINUE] [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action19]
<scribe> ACTION: HH to e-mail GRDDL Logo contest announcement - semantic web at w3.org. (pending SWEO coordination) [CONTINUE] [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action20]
<jjc> ACTION: jjc to clarify to discussion participants on RDFa that it is not formal comments [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action21]
JJC: Dave Booth comment issue-dbooth-10: Does an XML namespace necessarily imply semantics? http://lists.w3.org/Archives/Public/public-grddl-comments/2007AprJun/0072
<jjc> [[ I am satisfied with the WG's response.
Murray: you cannot impose on someone to ingnore an XInclude in a document. Noone has the authority to do it
<john-l> [[ That begs the question. GRDDL doesn't have the authority because we chose not to give it that authority. There are other choices we could have made. ]] - Dan in <http://lists.w3.org/Archives/Public/public-grddl-comments/2007AprJun/0086.html>
Murray: the decision was that we work on XPath node set ; we should not reopen the issue
JJC: I agree we should not reopen it.
<chimezie> it seems that he is motivated both by having a 'complete' rendition (which we don't compute) and in giving some precedence in a 'default' XML processing pipeline (which it seems - to me - GRDDL has no authority to mandate)
Murray: If you want to guarantee you have the right answer you can run everything yourself (e.g. xinclude). I can talk to him on the phone. I prefer somebody else to send the formal response.
<jjc> suggested proposal (not from me) "We resolve not to make changes in response to DBooth-3, and not to reopen the faithful infoset issue"
<scribe> ACTION: Murray to phone/msg Dave Booth [recorded in http://www.w3.org/2007/05/30-grddl-wg-minutes.html#action22]
<jjc> Adjourned by lack of numbers