W3C RDFCore WG 2001-05-18 Teleconference Minutes

W3C RDFCore WG 2001-05-18 Teleconference Minutes

Agenda:
- http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0084.html

Participants:
- Dave Beckett
- Art Barstow
- Brian McBride
- Dan Brickley (chair)
- Jos De Roo
- Aaron Swartz
- Frank Manola
- Frank Boumphrey
- Ron Daniel
- Bill dehOra
- Ghram Klyne
- Mike Dean
- Eric Miller (scribe)

Regrets:
- Ora Lassila
- R.V. Guha
- Jan Grant
- Rael Dornfest

----

Agenda Discussion

- http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0084.html

Review of Previous Action Items:

--
Item: Send feedback to Guha about use of RDF features
Owner: Guha
Discussion: Feedback by some, but not all...
Status: Continued, as owner is not on call

--
Item: Update the proposed changes for this issue and add test cases to
demonstrate how they worked.
Owner: Dave Beckett
Discussion:
(see Open Issue:
http://www.w3.org/2000/03/rdf-tracking/Overview.html#rdf-ns-prefix-confusion
discussion)
Status: Continued

--
Item: Send analysis (#rdf-container-syntax-ambiguity and
#rdf-containers-syntax-vs-schema) to rdfcore-wg list
Owner: Ora Lassilla, Dan Brickley
Status: Continued, as owner is not on call

--
Item: Reword #rdfms-empty-property-elements proposal and send a new message
with test cases, including one with just an ID attribute.
Owner: Jan Grant
Discussion:
Danbri notes Jan's post to the list on this item -
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0033.html
Status: Continue, as owner is not on call

--
Item: re #rdfms-not-id-and-resource-attr: Revise analysis and post to list
Owner: Dave Beckett
Discussion:
(see Open Issue:
http://www.w3.org/2000/03/rdf-tracking/Overview.html#rdfms-not-id-and-resource-attr
discussion)
Status: Continued

--
Item: re #rdfms-reification-required: Present analaysis to list for
discussion.
Owner: R.V. Guha
Status: Continued, as owner is not on call

--
Item: Link test cases, results etc. from issues list
Owner: Brian McBride
Discussion: Brian not feeling especially well, was unable to address
this issue... scribe hopes he feels better.
Status: Continued

--
Item: Do an analysis of the impact of XML Base and summarize to list.
Note: Owner not on call so item discussion deferred
Owner: Jan Grant
Status: Continue, as owner is not on call

--
Item: Contact Rael about hosting face to face at O'Reilly.
Owner: Brian McBride
Discussion:
Eric noted that previous communication with Rael indicated that
O'Reilly would indeed be willing to host this, and suggest we confirm
and move forward on this.
Status: Continued, as Brian will confirm.

----
Agenda Issues

--
Issue: Helping WG members new to RDF come up to speed.

Overview:
not much discussion... seems like good thing, but don't
want to slow working group down.

Resolution:
Danbri suggests contacting him off-list if help is needed

--
Issue: Ensuring DAML issues are captured in the issue list

Overview:
Brian wants to make sure that the DAML issues that are being
raised on other various lists are (if possible/appropriate) addressed
by RDFCore open issues list

Discussion:
Mike Dean: The various simple DAML issues by and large on list are
indeed addressed, however with there are still some issues that may
need to be taken into consideration.  Mike agree's to being contact
point for DAML/RDFCore and will solicit Pat Hayes (and others) to try
to get them to provide some concrete recommendations.

Danbri notes possible technical concerns with perhaps how DAML is
handling data-typing, with paticular concerns of how groups like Dublin
Core are handling data-types (e.g. value of dc:creator being an Agent
or a literal string)

Mike Dean responds to this indicating this isn't a problem, just in this 
case not
being able to take advantage of certain DAML capabilities.
Ghram Klyne also brings up the point about that reification is yet
another of these issues that DAML has raised as a potential problem

Resolution:
ACTION ITEM: Ghram Klyne and Frank Manola agree to summarize www-rdf-logic
perspective of reification as it applies to both logic and rdf and
report back to rdfcore wg

ACTION ITEM: Frank Manola agrees to summarize what the the logicians
(www-rdf-logic) like to see in RDF and provide references to issues
regarding clarification of RDF semantics, removing the extraneous
parts of RDF / modularizing RDF, etc. by early next week.

--

Issue: Acquiring control of the RDF errata document

Overview:
Danbri requests write access to to the RDF M&S/Schema errata documents

Discussion:
Aaron notes that Ralph Swick has provided edits of the errata several
times based on feedback from public list

Eric suggests that if we can additional provide errata write access to
those maintaining RDF Core lists, this would be easier for
maintanence.

Resolution:
ACTION ITEM: Eric Miller agrees to work with chairs and provide a solution

--

Issue: Past Action Item / RDF Open Issue Discussions

Open Issue: 
http://www.w3.org/2000/03/rdf-tracking/Overview.html#rdf-ns-prefix-confusion

Discussion:
DaveB reflects Dan C.'s comments about using the term "must" and
suggests changes to "may" in describing solutions to this problem.
General discussion forms on how the working group anticipates future
decisions and the working groups attitude toward suggested decisions
and effects on backward compatibility.

The notion of 'deprecation' of previous work was discussed and Frank
M. and Martin H. remind the group this the term 'deprecation' has a
Art suggest that these changes (what ever they may be ) be reflected
in a separate page so developers can easily find these
issues/resolutions.  General agreement on this, however, a more
detailed write up on this would be needed.  Ron D. notes, however, to
make clear that all of decisions are subject to change until final
spec is out.

Resolution:
General consensus was that Dave B. would re-word this to reflect this
notion of deprecation in this position

ACTION ITEM: Art B. to formalize his suggestion rdfcore changes be
reflected in a separate page so developers can easily find these
issues/resolutions

--

Open Issue: 
http://www.w3.org/2000/03/rdf-tracking/Overview.html#rdfms-not-id-and-resource-attr

Discussion:
Brian notes, that Dave B's analysis of this issue went to list late
yesterday and suggest more time is needed for discussion. Further, he
suggests this is not a closed issue, and we postpone discussion here and
take this off-line for continue discussion.

Resolution:
Agreed to discuss this on the list

-- 

Open Issue: 
http://www.w3.org/2000/03/rdf-tracking/Overview.html#rdfms-resource-semantics

Discussion:
Martin H. notes that a lot of time was spent on this issue. Two
different questions:
  1) is a uri a pack of atoms (suggests answer is no)
  2) can resources point to the same object? (suggests answer yes)
Martin notes that the conversation has been dangerously close to more
philosophical in nature.  Frank M. suggests that what is needed is
more test cases, and suggestions that whatever conclusions we come to
be illustrated with test cases to ensure the distinctions we want to
convey are clear.

Resolution:
ACTION ITEM: Martin H. agrees to provides an initial set of test cases

--

Open Issue: 
http://www.w3.org/2000/03/rdf-tracking/Overview.html#rdf-terminologicus

Discussion:
Not really enough time to effectively discussion this. Ghram doe note,
however, that
http://public.research.mimesweeper.com/RDF/RDFTerminologyConcepts.html
might be some help to terminology

----

Meeting Adjourn

----

Raw IRC log - (to help clarify any potential confusion of the scribe's minutes)

**** BEGIN LOGGING AT Fri May 18 10:01:10 2001

10:01:10 -->	em (em@dhcp065-024-049-027.columbus.rr.com) has joined #rdfcore
10:02:11 -->	mdean (mdean@63.219.164.68) has joined #rdfcore
10:02:26 ---	mdean is now known as _mdean
10:02:32 <bwm-dan-dave>	+eric
10:02:35 <bwm-dan-dave>	+aaron
10:02:42 <bwm-dan-dave>	+klyne
10:03:08 <ArtB>	+Jos
10:03:10 <bwm-dan-dave>	+jos
10:03:20 <bwm-dan-dave>	+mike
10:04:52 -->	AaronSw (aswartz@216.146.78.254) has joined #rdfcore
10:05:45 -->	horner (chatzilla@ANice-101-2-1-161.abo.wanadoo.fr) has joined 
#rdfcore
10:05:48 *	em agrees to scribe
10:05:48 <bwm-dan-dave>	regarding minutes from last week, approved except 
amend re frank manola ref
10:06:46 <ArtB>	Agenda: 
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0084.html
10:07:41 <em>	Issue: A1: ALL           Send feedback to Guha about use of 
RDF features.
10:07:49 <bwm-dan-dave>	regrets from ora, guha, jan grant, rael
10:08:07 <em>	A2: Dave Beckett  Update the proposed changes for this issue 
and add test
10:08:07 <em>	                  cases to demonstrate how they worked.
10:08:24 <em>	issue: 
http://www.w3.org/2000/03/rdf-tracking/Overview.html#rdf-ns-prefix-confusion
10:08:44 <em>	daveb: action addressed, open for discussion
10:09:09 *	em scribe raises hand
10:09:13 <ArtB>	Containers: 
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0113.html
10:09:47 <em>	A2: discussion to be postponed to end
10:09:53 <em>	A4: Jan Grant     reword #rdfms-empty-property-elements 
proposaland send
10:09:53 <em>	                  a new message with test cases, including 
one with just an
10:09:53 <em>	                  ID attribute.
10:10:07 <em>	danbri: jan sumbitted to list
10:10:18 <em>	daveb: missing still last point
10:10:26 <em>	daveb: of action
10:10:28 AaronSw ArtB 10:10:35 <em>	A5: Dave Beckett  re 
#rdfms-not-id-and-resource-attr: Revise analysis and
10:10:35 <em>	                  post to list
10:10:50 <em>	daveb: done, discussion postponed
10:10:55 <em>	A6: Guha:         re #rdfms-reification-required: Present 
analaysis to list
10:10:55 <em>	                  for discussion.
10:11:11 <em>	danbri: not completed.. danbri to poke guha
10:11:14 <em>	A7: Brian McBride Link test cases, results etc. from issues list
10:11:27 <em>	brian-ill: still open as not feeling well
10:11:48 <em>	group wishes brian well
10:11:53 <em>	A8: Jan Grant     Do an analysis of the impact of XML Base 
and summarise to
10:11:53 <em>	                  list.
10:11:56 <em>	danbri: still open
10:12:06 <em>	danbri: anyone want to help  with this?
10:12:11 <em>	A9: Brian McBride Contact Rael about hosting face to face at 
O'Reilly.
10:14:12 <em>	brian: agress to continue this
10:14:30 <em>	danbri: point raised... helping RDF people comming up to speed
10:15:01 *	ArtB wants DanBri to justify reification :-)
10:15:04 <em>	danbri: suggests contacting danbri offlist if help is needed
10:15:16 <em>	20  Ensuring DAML issues are captured in the issue list
10:15:45 <ArtB>	Brian: by DAML issues, do you mean issues raised on 
www-rdf-logic?
10:15:55 <em>	brian: wants to make sure that the DAML issues that are being 
raised on other lists are fully addressed by RDFCore open issues list
10:16:16 <em>	mikedean: current status... simple issues by and large on list
10:16:25 <em>	mikedean: lots of discussion on rdf-logic,
10:16:45 <em>	mikedean: solicit pat hayes and others to take a look at this
10:17:02 <em>	danbri: can you say at this stage if there are glaring 
inconsistenciese?
10:17:10 <em>	mikedean: not really...
10:17:41 <em>	danbri: technical concerns with perhaps how DAML is handling 
datatyping
10:17:59 <em>	danbri: seems to be potential problem with groups like DublinCOre
10:18:20 <em>	danbri: eg. literal as the value of dc:creator
10:18:41 <em>	mikedean: (missing something here... somethink about changing 
damml-)
10:18:46 *	AaronSw thinks that it's preferable to be clear -- authorName 
vs. author
10:19:35 <em>	gk: question about reification
10:19:44 <em>	group seems to agree
10:19:47 <AaronSw>	many laughs
10:20:00 <em>	danbri: reification... big issue, suggests to defer
10:20:53 <em>	gk: notes simple notes that reification is yet another of 
these issues that DAML has raised
10:21:38 <em>	danbri: gk? interested in summerizing rdf-logic?
10:21:44 <_mdean>	i'm hard to hear from a hotel room in Staunton Virginia
10:21:49 <AaronSw>	mike agrees to being contact point for daml/rdfcore and 
will try to get them to provide some concrete recommendations
10:21:59 <_mdean>	here's a summary of some of the DAML/RDF issue discussion
10:22:17 <em>	gk: agrees to summerize rdf_logic list
10:22:26 <em>	frankm: agrees to help with this
10:22:56 <em>	ACTION ITEM: gk, and frankm: agreed to summerize rdf-logic 
perspective and report back to rdfcore wg
10:23:32 <em>	wrt reification dsicussion... summerise rdf-logic discussion
10:23:42 <_mdean>	i haven't yet finished comparing the DAML concerns to the 
RDF issues list, but i think they capture the simple issues -- there has 
also been a lot of discussion (particularly this week) on www-rdf-logic 
dealing with higher-level issues (reification, minimalist RDF, etc.)
10:24:09 <em>	as it applies to both logic and rdf
10:24:29 <em>	frankb: agrees to have something (at least issues and 
distinctions) by nbext weel
10:24:49 *	ArtB pats FrankM on the back ...
10:25:07 <em>	frankb: agrees to have something by early next week
10:25:47 <_mdean>	for next week's Joint Committee telecon, I'll plan to 
open a discussion of providing specific issues/recommendations to RDF Core
10:25:57 <_mdean>	i'm willing to serve as a weekly coordination point
10:25:58 <em>	danbri: "what do the ontologists want" thread
10:26:19 <em>	danbri: asks jos to summarise the fact that the issue hasn;t 
been addressed
10:26:30 <em>	frankb: ehat would the logicians like to see in rdf
10:26:57 <em>	eg clarification of semantics, removing the cruft, etc.
10:27:24 <em>	ACTION ITEM: frank to agree to provide references to this work
10:27:38 <em>	danbri: 25  Acquiring control of the RDF errata document
10:28:15 <_mdean>	also, danbri asked about the object/datatype separation 
for DAML properties, particularly related to Dublin Core -- if you use 
daml:ObjectProperty or daml:DataTypeProperty, you're limited -- if you 
continue to use rdf:Property [names/namespaces from memory] the value can 
still be either a data type (literal) or object, but you may lose some of 
the power of DAML
10:28:39 *	AaronSw raises hand
10:29:10 <bwm-dan-dave>	action on emiller: find out how to get rdfcore 
control of rdf errata doc
10:29:16 <bwm-dan-dave>	aaron: ack'd
10:30:00 <ArtB>	Errata for M&S: 
http://www.w3.org/TR/1999/REC-rdf-syntax-19990222/errata
10:30:23 <em>	issue status: 
http://www.w3.org/2000/03/rdf-tracking/Overview.html#rdf-ns-prefix-confusion
10:30:33 <em>	discussion:...
10:30:46 *	AaronSw comments that Ralph Swick has edited errata several 
times based on feedback from public list
10:31:31 *	em is unable to scribe for sec.
10:31:42 *	em asks group to fill in here
10:32:11 *	em is back
10:33:09 <em>	daveb: refelcting danc point... instead of issue *nust* to 
this, change this to *may*
10:33:24 <em>	which reflects our atticute toward backward compatability
10:33:34 <em>	ArtB: are we guided by w3c process
10:33:35 <em>	?
10:33:50 <em>	danbri: in charter, but this is loose provacative interpretation
10:34:10 <em>	daveb: suggests keeping this as is
10:34:45 *	horner raises hand
10:34:53 <em>	danbri: are best judgement of the group is that we deprecate 
this at this time
10:35:13 <em>	martinh: deprecate is a hard line, one level removed...
10:35:38 <em>	martinh: advocating deprecation... just remembering this is 
an evolutionary concept
10:35:50 <em>	frankb: deprecation means... next version is gone
10:35:50 -->	GK (GK@host213-123-53-236.dialup.lineone.co.uk) has joined 
#rdfcore
10:36:22 <em>	frankb: be very clear... if we say deprecation, its gone next 
version.
10:37:02 *	em raises hand
10:37:30 <ArtB>	Dave's last comments re ns-prefix-confusion: 
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0110.html
10:37:56 <AaronSw>	danbri: is it gone in the rec we publish or the rec 
after that?
10:39:10 *	ArtB raises hand
10:40:17 <bwm-dan-dave>	danbri: this wg anticipates making a hardline spec 
for eventual w3c recommendation; therefore we as a wg currently discourage 
the use of x, y, z constructs as our best judgement is that these things 
will go...
10:40:31 <em>	ArtB: suggests that these changes be reflected in a seperate 
page so developers can easiliy find these uissues/resolutions
10:40:52 <em>	ArtB: a page for resolution, a page for issues
10:41:06 *	AaronSw thinks that that would be the errata page
10:41:12 *	em agrees
10:41:42 *	GK (testing IRC feature)
10:41:48 *	em waves to gk
10:41:56 *	GK waves to em
10:41:57 <em>	jos: agrees to arts suggestion
10:42:25 <em>	daveb: am I done?
10:42:44 <em>	other than some additional prose (danbri's suggestion)
10:43:25 <AaronSw>	+Ron Daniels on phone?
10:43:34 <em>	ron: agrees to artb's suggestion
10:43:48 <em>	ron: notes all of this is subject to change until final sepc 
is out
10:43:52 *	AaronSw raises hand
10:44:29 <em>	ACTION ITEM: artb to formalize his suggestion rdfcore changes 
be reflected in a seperate page so developers can easiliy find these 
uissues/resolutions
10:44:50 <em>	Issue: 
http://www.w3.org/2000/03/rdf-tracking/Overview.html#rdfms-not-id-and-resource-attr
10:44:55 *	AaronSw lowers hand but wonders whether danbri's text lets us 
decide if it goes in the rec or not
10:46:19 <em>	AaronSw: notes there does seem to be an issues of empty prop 
elements
10:46:35 <em>	brian: dave's ananlysis went to list late yesterday
10:46:56 <em>	brian: suggest more time is needed for disussion... suggests 
this is not a closed issue
10:47:24 *	em requests dave to submit the url of the issue response
10:47:59 <em>	danbri: suggests this be taken offline and continue discussion
10:48:14 <em>	end discussion of issue...
10:49:01 <em>	issue: 
http://www.w3.org/2000/03/rdf-tracking/Overview.html#rdfms-resource-semantics
10:49:09 <em>	martin - owner
10:49:25 <em>	martinh: spent a lot of time on this... more than expected
10:50:10 <em>	martin: 2 questions - is a uri a pack of atoms (suggest 
answer is no), can 2 resources point to the same object (suggests answer yes)
10:50:22 <em>	amrtin: suggests cutting this loose
10:50:56 *	ArtB wonders if that is Bill speaking ?
10:50:58 <bwm-dan-dave>	+bill (not in roll call?)
10:51:00 <em>	+bill
10:51:12 <em>	Bill deOra (sp?
10:51:53 <em>	martin: very close to philosophical discussions...
10:52:00 <em>	danbri: rathole (em agrees)
10:52:22 <em>	frankb: notes that this perhaps is being too quickly 
suggested this is philosphy
10:52:31 <em>	frankb: suggest perhaps more test cases
10:53:01 <em>	frankb: suggests conclusiosn we come to illustrated with test 
cases to ensure distinctions we want to convey are done by terms we use
10:53:32 <em>	frankb: e.g. the thing on the web that is obtained with 
derefencing uri amd the thing the uri denotes
10:53:49 <GK>	Web-copy of the old terminologicus document is at 
http://public.research.mimesweeper.com/RDF/RDFTerminologyConcepts.html
10:53:56 <AaronSw>	em, was that not frankM?
10:53:56 <em>	danbri: as fars as issue... technically critcal, but clear 
rathole
10:54:05 <em>	s/frankb/frankm
10:54:08 *	em thanks
10:55:17 <em>	general: group agres to talk about test cases rather than 
philosophy about this
10:55:30 <em>	end discussion
10:55:40 <em>	issue: 
http://www.w3.org/2000/03/rdf-tracking/Overview.html#rdf-terminologicus
10:55:44 <em>	owner martin
10:56:46 <em>	ACTION ITEM: martin agrees to provide test cases
10:56:51 <em>	wrappup...
10:56:57 <em>	danbri: other business?
10:58:26 <em>	GK: notes 
http://public.research.mimesweeper.com/RDF/RDFTerminologyConcepts.html as 
perhaps help to terminoligy
10:59:05 <AaronSw>	all leave
10:59:08 <--	AaronSw has quit (Gotta Run)
11:00:17 <--	horner (chatzilla@ANice-101-2-1-161.abo.wanadoo.fr) has left 
#rdfcore
11:07:02 <--	bwm-dan-dave (brian_mcbr@phobos.hpl.hp.com) has left #rdfcore
11:07:33 <--	GK has quit (Leaving)

--
eric miller                              http://www.w3.org/people/em/
semantic web activity lead               mailto:em@w3.org
w3c world wide web consortium            tel:1.614.763.1100

Received on Friday, 18 May 2001 15:08:52 UTC