Session Start: Fri Jan 18 15:01:48 2002 [15:01] *** Now talking in #rdfcore [15:01] *** Topic is 'RDF Core WG meeting Friday 15:00-16:00 UTC Agenda: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0125.html' [15:01] *** Set by dajobe on Fri Jan 18 14:05:07 [15:03] *** gk is now known as gk-scribe [15:03] Role call [15:03] DanBri regrets, Eric regrets [15:03] Dave here [15:03] FrankB abs, Jeremy here [15:03] * DanC dials in... [15:03] DanC nearly here? [15:03] Rion jhere [15:03] Bill Deh here, Jos regrets [15:04] \\\ Jos abs [15:04] Maretyn here [15:04] Yoshi abs [15:04] Graham here (almost) [15:04] *** jjc has joined #rdfcore [15:04] Michael K [15:04] ... abs [15:04] Kwon, Ora abs [15:05] Frank M regrets [15:05] Satoshi N abs [15:05] Steve P abs [15:05] Pierre R abs [15:05] Patrick S here [15:05] * DanC looks around for EricM; suspects he's in Hawaii for WWW2002 PC meeting still [15:05] Aaron here [15:05] +DanC [15:05] Mike D regrets [15:05] Guha abs [15:05] PatH here [15:06] DanC here [15:06] Jos here [15:06] Agenda comments, AOB [15:06] re webont ftf, minutes are in progress: http://www.w3.org/2001/sw/WebOnt/ftf1.html [15:07] Report from WebOnt F2F??? DanC will report, interest is here [15:07] Register for F2F [15:07] * DanC looks for ftf registration info on the WG home page; doesn't see it [15:08] *** dajobe-jang is now known as dajobe-scribe [15:08] item 5 [15:08] please register [15:08] item 6 - minutes of last meeting [15:08] waiting for echo to go... [15:09] * DanC tries to grab an operator... [15:09] *** dajobe-scribe is now known as dajobe-jang [15:09] * DanC reaches somebody at MIT, where the telcon hardware is [15:10] APPROVED [15:10] Completed actions... [15:10] APPROVED [15:10] item 8 reification [15:10] MIT folk say the echo comes when GK dials in. They'd like to know where GK is calling from. [15:10] * gk-scribe calling from UK [15:10] GK is in UK [15:11] Frank will gather his thoughts [15:12] gk: use 60# to speak again [15:12] ralph visits... [15:13] ... and report back to the group [15:13] ... to kick off the discussion (in email?) [15:14] DanC wants to say... [15:15] Example like "Mary hits ball" ... reify this some parts of spec suggest this like a word beginning with "M", others suggest (???missed this???) [15:15] * gk-scribe DanC: can you be clear about what you think *is* userful? [15:16] works like quoting in other systems (DanC) [15:16] Only useful if quoting (subject is word beginning with "M", rather than subject is a person) [15:17] So DanC wants reification to show subject as a string, not a URIref [15:17] * gk-scribe I find this counter to my intuitions [15:17] I don't feel confortable with Dan's proposal [15:18] What about reifying a reification: quoting a literal is identity, so there's a point of idempotence there [15:20] Can Dan offer a brief example/summary explaining why one way works and the other doesn't [15:21] >>> DanBri joins [15:21] item 9 [15:21] Item 9, status of test cases WD [15:21] Actions; Brian hasn't done his [15:21] jjc, I wouldn't expect you to be comfortable with it. I'm not comfortable with it, after all. ;-) [15:22] jang said he hadn't time to work on his issues, continued [15:23] All acrions continue [15:23] Item 10 -- primer [15:23] Pass over, Frank is not here [15:23] Disappointment is expressed [15:23] There is concern for the rate of progress [15:23] frown... primer is in the someday pile. [15:23] No current estimate of progress [15:24] Next item, model theory draft [15:24] item 11 [15:24] Pat is working through the various comments recived [15:25] * DanC saw a bunch of MT comments in a message from PatrickS [15:25] jos's comment today: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0127.html [15:26] Jos and ??? actions have been completed [15:26] jeremy [15:27] questions about substantialness of commetns [15:27] jjc: some, interested in pat's replies [15:27] Estimate will be public WD ... after substantive discussion next week [15:27] sounds to me like 1 to 2 weeks till next MT WD [15:28] item 12 [15:28] Unrecognized attributes... [15:29] Test cases? Needed. [15:29] ACTION: Dave B will do some [15:29] item 13 [15:29] Item 13, xml:base [15:30] Options: (1) honour xml:base of enclosing document [15:30] I'm uncomfortable with "ignored"; I'd much prefer "not allowed" for xml:base within rdf:RDF [15:31] (1) xml:base not allowed in stand-alone document, honoured from enclosing document [15:32] Question: how will RDF parser find xml:base? A: not our problem [15:32] APPROVED [15:33] A2: ARP finds xml:base [15:33] proposal: is that we allow xm:base ONLY on the rdf:RDF eleemtn [15:33] gah, "xml:base" [15:33] rosalto allow xml:base on rdf root elemeent only [15:33] to be clear: this is a change to the spec. [15:34] Note: never ignore xml:base - either honour it or disallow it [15:34] * gk-scribe I agree with that last point [15:35] Proposal: (1) allow xml:base on outer element, disallow it inside, never ignore it. [15:36] *** danb-lap has joined #rdfcore [15:36] What about invisible inclusion? Embedded RDF literals? [15:36] Proposal (2): supprt xml:base anywhere [15:37] DaveBN would like an analysis, JJC offers to provide it. [15:37] ACTION, JJC, (above) [15:37] danbri asks for a test case shipping xhtml [15:37] ... parsetype literal with xhtml [15:38] Request for test case dealing with embedded RDF , and stuff embedded in RDF (?) [15:39] JJC's analysis to include test cases to illustrate the core issues [15:39] * DanC wonders why we can't decide this today. oh well. [15:39] * gk-scribe Hmmm... can we decide now, and allow retraction if there;s a proiblem??? [15:41] APPROVED xml:base to be allowed anywhere [15:41] ACTION (supersedes above) JJC, to provide analysis of the implications of this decision, with test cases to illustrate the issue [15:41] item 14 [15:42] Next item -- bagIDs [15:42] DaveB has madea proposal, with test cases [15:43] DanC -- will this have the "wrong" form of reification? [15:43] martynh has reviewd them [15:43] jjc verifies with test cases, and will check changes [15:43] indeed, I'm uncomfortable with this test; it says [15:43] *** sergey has joined #rdfcore [15:43] . [15:43] but it should say "http://example.org/b" . [15:44] * gk-scribe the line ech is back with a vengeance, and I'm muted [15:44] * gk-scribe s/ech/echo/ [15:44] Not approved yet -- too many abstentions [15:45] It seems we need some better understanding of reification??? [15:46] JJC: Test cases are about *which* statements get reified, not *how* they get reified... [15:46] ... can we agree on one pending the other [15:46] Probably not. [15:47] Issue continues, after reification is clearer [15:48] I'd be surprised and dissapointed if the syntax doc said anything about reification at this point. [15:48] Next item, MIME types [15:48] Candidate Rec? did I miss something? we're doing a clarification of a REC, no? [15:49] Propose that MIME type registration will be handled when document goes to candidate rec [15:49] * gk-scribe DanC, I'd say the changes are big enough to need recycling [15:49] * gk-scribe ... at CR [15:51] * gk-scribe will the final REC have the same URI as M&S??? [15:51] make the wording about registration date vague [15:51] * gk-scribe (I meant *current* M&S) [15:52] * gk-scribe How about "whe we have a stable document to reference"?? [15:52] bwm: when we are read [15:52] y [15:52] Next item ... Datatypes [15:52] (item 16 skipped) [15:52] item 17 [15:53] gk: submitted some minor changes to desdiratta [15:53] .. hopeing this is a good basis for discussion [15:53] "The datatyping proposal should not preclude the use of non-XML-schema datatypes" <- hmm... seems like a high bar to set. [15:54] I guess I can buy the above as "should" and "desiderata"/ [15:54] . [15:54] +sergey [15:54] (earlier) [15:55] * gk-scribe Sergey, I'm happy if the desiderate goes back into your doc, or not, as we see fit down the line. [15:56] Patrick has offered 1st draft [Insert URI] .. to be finalized next week?? .. of description of P/D proposal [15:58] * DanC hunts for PatrickS's draft [15:58] Jeremy has yet to provide the math for P/D -- current target is next Friday [15:59] ACTION: Sergey, to analyze both proposals against the desiderata [16:00] ACTION: ALL, propose new idioms that should be in the desiderate document, ready for next week. [16:01] gk's latest doc: v3 http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/att-0137/01-RDF-Datatyping-Desiderata.html [16:02] discussion of desdirata and idioms [16:03] bwm: to gk, can you recast as use cases? [16:03] patrick: use cases, not answers [16:03] would expect not to see ntriples or XML (jjc) [16:04] gk v2 http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/att-0109/01-RDF-Datatyping-Desiderata.html [16:04] gk: more like legacy data, data uses/idioms seen in the wild [16:05] patrick: more focus on apps using particular idioms, cart before horse [16:06] ACTION: gk, clarify that these are idioms already in use (with citations if possible) [16:06] ACTION: all, use cases of idioms, send to gk [16:06] ACTION: GK, review "idioms" section to clarify that these are claimed examples of existing use, provide specific references where possible. Accept and incorporate new use-cases as provided. [16:07] END OF MEETING