RDFCore WG minutes for the Telecon 2002-06-07

RDFCore WG minutes for the Telecon 2002-06-07

Transcript:
  Attached

Agenda:
  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0020.html

1: Volunteer scribe

  Dave Beckett

2: Roll Call

Participants:
    - Dan Brickley
    - Eric Miller (chair)
    - Dave Beckett
    - Dan Connolly
    - Ron Daniel
    - Bill dehOra
    - Jos De Roo
    - Jan Grant
    - Graham Klyne
    - Frank Manola
    - Stephen Petschulat
    - Patrick Stickler
    - Aaron Swartz
    - Mike Dean
    - R. V. Guha
    - Pat Hayes

Regrets:
    - Brian McBride
    - Jeremy Carroll

Absent:
    - Frank Boumphrey
    - Rael Dornfest
    - Arno Gourdol
    - Yoshiyuki Kitahara
    - Michael Kopchenov
    - KWON, Hyung-Jin
    - Ora Lassila
    - Satoshi Nakamura
    - Pierre G. Richard
    - Andew Salop
    - Sergey Melnik


3: Review Agenda

DaveB asked Item 9 to be skipped with Jeremy Carroll not present,
EricM supported this.

DaveB asked to report on status of faq-html-compliance

DanB noted some further notes on item 12 just emailed to the list
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0027.html
  

4: Next telecon?

10am Boston time, 2002-06-14

To be possibly chaired by Eric if Brian McBride not present.


5: Review Minutes of 2002-05-31

Minutes of 2002-05-31
  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002May/0159.html
with corrections:
  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002May/0160.html

APPROVED


6: Register for RDF Core F2F

EricM encouraged all to register at
  http://cgi.w3.org/Register/selectUser.pl?_w3c_meetingName=RDFCOREF2F3
even if not attending - the last two to register have to scribe :)


7: Status of previous Action Items

ACTION 2002-05-31#1, FrankM: Circulate list of issues needing
clarification
  -- CONTINUED

ACTION 2002-05-31#5, jos: Summarize and check decision with WebOnt

see:
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0001.html
  -- DONE

ACTION 2002-05-31#2, DaveB: Update syntax spec with above decisions
  -- CONTINUED

ACTION 2002-05-31#3, JanG: Update test case document with this, and
other, test cases
  -- CONTINUED

ACTION: 2002-05-24#4  danbri  update the isDefinedBy proposal (with 
help from Guha) in light of the discussion in the 2002-05-24 telecon
see: Issue 12
  -- DONE with update at
     http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0027.html


8: outstanding issues - 8 left

rdfs-xml-schema-datatypes
rdfms-contexts
rdfms-seq-representation
rdfms-assertion
rdfs-editorial
rdf-namespace-change
faq-html-compliance
rdfs-isDefinedBy-semantics


See:
  http://www.w3.org/2000/03/rdf-tracking


9: Definition of graph syntax

Discuss Jeremy's proposal.

See:
  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002May/0120.html
  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002May/0122.html

Not discussed since Jeremy gave regrets; EricM will put it on next
week's agenda if Jeremy can attend.


10: Issue: rdfms-assertion

2002-05-17#7  DaveB  Investigate recent TAG decision on registering
mime types for W3C specifications and make a recommendation to the WG

-- DONE after verbal report by DaveB

Discussion of the IETF mime type registration process and the
TAG finding http://www.w3.org/2001/tag/2002/0129-mime
and how the existing RDF I-D should be handled.  The finding
says that the registration form must be in the REC that handles
the syntax; for RDF Core, that is the RDF/XML Syntax WD.

Discussion of the assertion issue that this action addressed and
where to put the words; in the I-D, REC and/or both.  Some concerns
about this leading to long discussions in the type registration
process.

ACTION 2002-06-07#1, PatHayes: draft a resolution to rdfms-assertion
and send to the list

Issue CONTINUED


11: Approve test cases

Propose approve test cases reviewed by
  Aaron  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002May/0125.html
  Graham http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002May/0144.html

Aaron is happy with rdfms-xmllang test001-test005 but not test006
which was if xml:lang applies to attribute content.  Several people
on the telecon thought that it did or would be surprising if it didn't.

ACTION 2002-06-07#2, AaronS: Check rdfms-xmllang/test006 and XML
specification.

Graham gave detailed comments, approval on many test cases.

JanG reported on comments from JeremyC and DaveB on other test cases
that needed fixing.

ACTION 2002-06-07#3, JanG: Fold in the consensus test cases and bring
back the rest to the group

ACTION 2002-06-07#4, JanG: Update the test case problems reported by
JeremyC and DaveB


12 Issue: rdfs-isDefinedBy-semantics

2002-05-24#4  danbri  update the isDefinedBy proposal (with help from
Guha) in light of the discussion in the 2002-05-24 telecon

DanB sent http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0027.html
just before the meeting and noted that the issue discussion had
exploded across multiple lists and yet to be summarised.

Discussion of namespaces, namespace URIs and RDF schemas, how to
relate these things with RDF properties, use of rdfs:isDefinedBy and
rdfs:seeAlso for this purpose and defining a way to say when or how
to retrieve rdf/xml at a URI about schema terms.

CONTINUED


13: F2F Agenda

Suggestions for agenda items were received including
- write press release for proposed recommendation (chairs to draft)
- write draft request for proposed recommendation status (chairs to draft)
- datatypes (see item 14)
- clarification of use of n-triples in documents
- layering
- primarily consider "what's left?" in order to complete


14 : Datatypes

Discussion after Pat's latest revision of the DataType WD based on
Patrick's draft as announced in
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0000.html
and the corresponding document
http://www.coginst.uwf.edu/~phayes/RDF_Datatyping060102_draft.html

PatH noted that the document needed to pass on to SergeyM (not
present) who will make a revision, completing a revision by each of
the editors.  The current plan is to carve it into bits and then fit
them together.  PatH will try to contact SergeyM.  Discussion
of GrahamK's comments on the draft
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0017.html


15: faq-html-compliance

DaveB reported on an analysis on approaches of RDF in HTML made by
Sean B. Palmer independently and encouraged by DaveB as input to
this issue. Announced in
http://lists.w3.org/Archives/Public/www-rdf-comments/2002AprJun/0170.html
and archived at
http://lists.w3.org/Archives/Public/www-archive/2002Jun/att-0002/01-index
Discussion of some of the conclusions, and DaveB prefered linking
approach <link> rather than embedding.  Some agreement with this.

ACTION 2002-06-07#5, DaveB: Forward Sean B. Palmer's analysis to the WG
list and propose some recommendations from it to close the issue.

-- end --

(Note times in UTC)

14:02:13 <DaveB-scr> DaveB-scr is now known as db-scribe
14:02:34 <em> zakim, who is here?
14:02:35 <Zakim> I see ILRT, AaronSw, PatrickS, EricM, Manola, SteveP, DanBri, JosD, Pat.Hayes
14:02:37 <Zakim> ILRT has JanG, DaveB
14:02:38 <Zakim> -PatrickS
14:03:10 <db-scribe> regrets brian
14:03:10 <Zakim> +PatrickS
14:03:14 <db-scribe> regrets jeremyc
14:03:20 <db-scribe> item 4
14:03:38 <db-scribe> next telcon 14th
14:03:39 <DanCon> DanCon has joined #rdfcore
14:03:41 <db-scribe> item 5
14:03:47 <db-scribe> minutes review
14:03:54 <db-scribe> approved
14:04:01 <DanCon> * DanCon Zakim, what's the passcode?
14:04:03 <Zakim> * Zakim saw 7332 given for the conference code, DanCon
14:04:13 <db-scribe> back to item 3
14:04:23 <em> current agenda - http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0020.html
14:04:39 <db-scribe> * db-scribe raises item 9, like to have jjc present
14:04:45 <db-scribe> em: propose postponing
14:04:51 <db-scribe> item 9 removed
14:05:03 <db-scribe> em: askes dave to update on issue-faq-compliance
14:05:09 <db-scribe> dave: ok
14:05:14 <Zakim> +DanC
14:05:17 <db-scribe> danbri: issue 12, just sednt to list
14:05:29 <db-scribe> em: ok, see if we have time
14:05:35 <db-scribe> item 6
14:05:47 <db-scribe> em: some have registered, some not
14:05:57 <db-scribe> ... last two to register, have to scrib e:)
14:06:00 <db-scribe> so do it now!
14:06:05 <db-scribe> item 7
14:06:28 <db-scribe> action 2002-05-31#5 done, rest continued
14:06:39 <Zakim> +MDean
14:06:41 <db-scribe> danbri: update on the isdefined by issue, not closure
14:06:59 <db-scribe> item 8
14:07:04 <db-scribe> outstanding issues
14:07:05 <db-scribe> 8 left
14:07:10 <danbri> my action closed with update at http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0027.html
14:07:12 <db-scribe> item 9
14:07:17 <db-scribe> postponed, no jjc here
14:07:23 <db-scribe> em: like to target for next week
14:07:35 <db-scribe> item 10
14:08:40 <Zakim> +??P20
14:08:51 <AaronSw> Zakim, ??P20 is RonD
14:08:52 <Zakim> +RonD; got it
14:09:31 <db-scribe> * db-scribe discuss this, not scribing...
14:09:46 <db-scribe> danc: can't see conflict ietf registring process, w3c rec
14:09:56 <db-scribe> aaron and danc discuss approach
14:10:13 <Zakim> +??P21
14:10:53 <AaronSw> zakim, ??P21 is GrahamK
14:10:54 <Zakim> +GrahamK; got it
14:11:50 <db-scribe> re assertion issue, where to say statements are asserted
14:13:32 <db-scribe> more discussion of whether to put assertion in the type rfc
14:17:32 <db-scribe> proposal to put assertion words in syntax doc
14:17:59 <db-scribe> and in type registry too?
14:18:35 <Guha_> Guha_ has joined #rdfcore
14:19:05 <db-scribe> gk suggests (concerns on ratholing) get our own agreement right and then strip to a sentence in the mime spec, pointing to more detailed words
14:19:08 <Zakim> + +1.650.965.aaaa
14:19:18 <AaronSw> zakim, where is 1.650?
14:19:19 <Zakim> sorry, AaronSw, I do not understand your question
14:19:24 <Zakim> -MDean
14:19:25 <DanCon> the 1st RDF WG got away with sorta ducking rdfsms-assertion; I'm not sure we're gonna get away with that this time. But really handling it looks tricky, as PatH points out in ... oop; in private mail to me.
14:19:44 <DanCon> Zakim, where is 650?
14:19:46 <Zakim> North American dialing code 1.650 is California
14:20:18 <db-scribe> action pathayes: draft words on rdfms-assertion
14:20:21 <AaronSw> zakim, +1.650 is Guha
14:20:22 <Zakim> +Guha; got it
14:20:52 <Zakim> +MDean
14:20:55 <DanCon> em, you don't have to do actions twice. As soon as somebody says "I'll do XYZ", you can record it as an action.
14:20:57 <db-scribe> 2002-05-17#7 done
14:21:03 <db-scribe> item 11
14:21:10 <DanCon> resolutions, however, have to be carefully put.
14:21:29 <db-scribe> jang: short, but had feedback from jjc, daveb on test case fixes...
14:21:35 <db-scribe> ...no problem
14:21:48 <gk> gk has joined #rdfcore
14:22:27 <db-scribe> * db-scribe and aaron discuss approvement
14:22:34 <AaronSw> * AaronSw endorses rdfms-xmllang/test001-test005
14:22:39 <db-scribe> endorse 001-005 of aaron's msg 0125.html
14:22:57 <db-scribe> gk reviewed all other tests
14:23:00 <db-scribe> and had some comments
14:23:29 <db-scribe> jang: action to fold in consensus test cases and bring back the rest
14:24:09 <db-scribe> aaron asked if xml:lang applies to attribute content
14:24:15 <db-scribe> gk: says yes, as he reads XML spec
14:24:26 <danbri> seems reasonable, unsuprising
14:25:02 <db-scribe> jang & dave discuss more test case things need to amend
14:25:09 <db-scribe> jang: will fix
14:25:14 <DanCon> good to hear that folks are going over the tests in bulk.
14:25:34 <db-scribe> item 12
14:25:40 <db-scribe> danbri has just sent 
14:25:43 <danbri> isDefinedBy: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0027.html
14:26:01 <db-scribe> isdefinedby issue exploded over multiple lists...
14:26:19 <db-scribe> (rewinding) action jang: to fix things after jjc and dave's comments
14:26:27 <gk> My review of test cases is at: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002May/0144.html
14:26:42 <db-scribe> naemspace URI and rdf schecma, isdefinedBy between them
14:26:50 <db-scribe> some impls used that property to point to anything
14:27:10 <db-scribe> danbri propose that we tighten this and make isdefinedby point to just one t hing
14:27:22 <db-scribe> danc: why constrain more?
14:27:31 <db-scribe> danbri: since they've used for wider things
14:29:25 <DanCon> re-reading the CR-rdf-schema text, it says isDefinedBy MAY be used in this fashion; danbri's argument why constrain doesn't hold.
14:29:52 <danbri> read it alongside M+S chunks
14:30:23 <DanCon> "The most common anticipated usage" could be translated SHOULD, but not MUST. how do you derive MUST?
14:31:06 <db-scribe> discussion of defining an rdfs:Class for saying "retrieve this" to get rdf/xml for more info
14:31:25 <Zakim> -SteveP
14:32:08 <db-scribe> discussion of uri "ownership"
14:32:24 <danbri> aside: the RDFS CR gave uuid: URIs as a use case
14:32:32 <danbri> +
14:32:35 <danbri> q+
14:32:37 <Zakim> * Zakim sees Danbri on the speaker queue
14:33:28 <danbri> q-
14:33:30 <Zakim> * Zakim sees no one on the speaker queue
14:34:01 <gk> So I say;  http://www.w3.org/GKs-property is defined by http://www.nminebynine.org/GKs-schema?  Is this contradictory?
14:34:08 <db-scribe> patrick doesn't like namespaces :)
14:34:22 <DanCon> PatH asks whether { <http://dans-stuff.example/stuff#term> rdfs:isDefinedBy <http://pats-stuff.example/doc> } is legitimate.
14:34:34 <db-scribe> frankm askss if you can get from property to namesapce?  No, many people
14:35:23 <db-scribe> danbri: would like to use some rdf:ns thing, but name might be suggestive of wrong thing
14:35:32 <db-scribe> ... also datatypes will be defined by iothers
14:35:49 <db-scribe> namespace / schema
14:35:53 <DanCon> yes, a namespace (document) is a schema document.
14:36:38 <DanCon> Aaron asks whether { <uud:23lkj23> rdfs:isDefinedBy <http://example/doc> } is legitimate.
14:36:51 <DanCon> * DanCon likes to get the example/test cases in the record
14:36:55 <db-scribe> patricks says has a namespace defined by many different schemas
14:37:12 <DanCon> I disagree with PatrickS about what a namespace is.
14:37:45 <danbri> * danbri thinks ericm was right to try moving on in the agenda; this is a rathole
14:37:45 <db-scribe> hasn't this got a TAG finding/draft? this=what is a namespace
14:38:39 <danbri> q+
14:38:40 <Zakim> * Zakim sees Danbri on the speaker queue
14:38:45 <db-scribe> rathole++
14:39:07 <gk> Isn't that just provenance?  That seems to me like a good way to describe it.
14:39:59 <em> ack danbri
14:40:00 <Zakim> * Zakim sees no one on the speaker queue
14:40:06 <DanCon> well, it's related to provenance, but I was reporting on real implementation experience. picking a new term undercuts this implementation experience, btw.
14:40:44 <em> q+
14:40:45 <Zakim> * Zakim sees Em on the speaker queue
14:40:51 <db-scribe> * db-scribe feels adding rdfs:ns would be more confusing than rdfs:isDefinedBy
14:41:13 <db-scribe> ack em
14:41:15 <Zakim> * Zakim sees no one on the speaker queue
14:41:46 <db-scribe> * db-scribe wonders if patricks can say how he's using namespaces, schema docs, properties, ..
14:42:38 <AaronSw> another testcase: { <http://foo/#bar> rdfs:isDefinedBy <http://danbri/>, <http://tap> .
14:42:40 <AaronSw> }
14:42:41 <db-scribe> frankm: if term changes to rdfs:namespace, it must point to a namespace
14:42:59 <DanCon> where's this mail proposing rdfs:namespace?
14:43:15 <db-scribe> DanCon: danbri's last mail to the list, see above?
14:43:27 <db-scribe> June 0027
14:43:48 <DanCon> I don't see a proposal for rdfs:namespace in there; where is it?
14:44:09 <Zakim> -Guha
14:44:41 <danbri> there isn't a full proposal as such, just http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0027.html
14:44:58 <db-scribe> item 13
14:45:04 <danbri> (no actions...)
14:45:38 <db-scribe> em: suggestions for best agenda itetems?
14:45:46 <db-scribe> danc: write press release for proposed rec?
14:45:50 <db-scribe> suggest chairs write it
14:46:07 <db-scribe> danc: want to finish rdf; last f2f before rec
14:46:22 <db-scribe> and write request for proposed rec
14:46:34 <db-scribe> and by end of meeting, should be closer
14:46:51 <db-scribe> em: useful proces
14:46:53 <db-scribe> s
14:46:58 <db-scribe> DataTypes
14:47:09 <db-scribe> (item 14)
14:47:21 <db-scribe> phayes: clarify use of n-triples
14:47:36 <db-scribe> ... in any of our documents
14:48:26 <danbri> * danbri fwd's WebOnt review of RDFS WD 
14:50:18 <db-scribe> other issues...
14:50:55 <db-scribe> danbri: 2 issues - datatyping and layering
14:51:06 <db-scribe> should we risk f2f time on potential rathole topics
14:51:09 <DanCon> I'd really rather there were no consideration of low-hanging fruit, people not being there, whatever. The only question is: what's left? all of it should be done at the ftf.
14:51:20 <db-scribe> patricks: more attn on datatypes
14:51:29 <db-scribe> since path and patricks there
14:51:33 <gk> I agree with DanC -- we're trying to get this as nearly wrapped as possible
14:51:59 <db-scribe> item 14
14:52:05 <db-scribe> dataypes
14:52:15 <db-scribe> pat revised draft based on patricks
14:52:23 <db-scribe> em: what is left?
14:52:27 <DanCon> I gather a pass by Sergey is next.
14:53:28 <db-scribe> phayes: emailing with sergey, rather slowly
14:53:55 <db-scribe> ... current plan is to carve in bits and then fit together
14:54:13 <db-scribe> em: telcon time?
14:54:21 <db-scribe> phayes: are converging
14:54:38 <db-scribe> still working on this,g etting it acceptable to all
14:54:47 <db-scribe> larger q is, is this proposal still what we want to do?
14:55:16 <danbri> * danbri revisits http://www.coginst.uwf.edu/~phayes/RDF_Datatyping060102_draft.html "rdfd:Datatype properties have exact domains and ranges. (This is an 
14:55:16 <danbri>   exception to the normal rule for rdfs:Property.) "
14:55:25 <danbri> * danbri should send mail on that
14:55:32 <em> yes please
14:55:53 <db-scribe> phayes: gk mail of thursday caused some pushback
14:56:00 <db-scribe> gk: didn't intend that
14:57:14 <db-scribe> http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jun/0017.html
14:57:17 <db-scribe> discussion of above
14:57:40 <db-scribe> timecheck
14:57:47 <db-scribe> em: agenda
14:58:36 <DanCon> * DanCon q+
14:58:38 <Zakim> * Zakim sees DanCon on the speaker queue
14:58:49 <db-scribe> action: daveb forward sbp analysis on faq-html- issue
14:59:03 <AaronSw> em: did pros cons but no conclusion
14:59:06 <db-scribe> em: can you make some recs? and give context
14:59:35 <AaronSw> DanC: linking works, rest is iffy. stick it in the primer and be done
14:59:47 <db-scribe> also action includes proposing a recommendation; linking (as I discussed)
15:00:00 <Zakim> -RonD
15:00:04 <db-scribe> END OF MEETING