W3C RDF Core Working Group IRC Chat Logs for 2001-10-12

13:01:03 Topic now RDF Core WG Meeting 14:00-15:00 UTC Friday

13:01:03 Users on #rdfcore: logger_1 @dajobe

13:24:28 * danbri is tied up in Euro-project stuff again today :(

13:39:56 * danbri will be late, distracted etc :(

13:42:08 <bwm> Not unexpected. You seem to have remembered the password :)

13:43:51 <danbri> Yes, that was really odd. They both (I forgot #ilrt too) came back to me in a flash. Some form of recovered memory syndrome...?

13:44:09 <dajobe> danbri: I set it up here so registered nick people can op themselves, set topicds

13:44:16 <dajobe> but auto-opping doesn't happen

13:44:33 <bwm> I just registered my nick

13:44:50 <bwm> quit

13:45:55 <danbri> I'm happy auto-opping RDFCore members in this channel; it was #rdfig that I wanted to minimise opping in, per http://www.openprojects.net/channel_guidelines.shtml

13:46:02 <dajobe> OK

13:46:10 <dajobe> Only aaron can fix that

13:54:55 <DanC> regrets for today's telcon... I'm gonna do some family stuff.

13:55:40 <DanC> reviewing the agenda...

13:55:55 <DanC> I'm pretty bummed the newsletter article was withdrawn. I think that's still important.

13:56:55 <bwm> Well it doesn't have to be withdrawn - just felt to me that it had got too late

13:57:08 <DanC> withdrawn is probably right; if he

13:57:23 <DanC> ... if he's gonna do a newsletter article, the ftf meeting is no longer really news.

13:57:44 <DanC> I still think he should do a newsletter article, but more along the lines of "progress in the SemWeb activity"

13:57:52 <DanC> but that's not really an RDF Core thing, per se.

13:58:05 <bwm> k

13:58:11 <DanC> before I run off, brian, about the list of active issues...

13:58:11 <dajobe> dialing

13:58:17 <DanC> are those all really active?

13:58:57 <DanC> I'd rather that the active list were pruned... I'd like to hear from the owner of every active issue every week.

13:59:26 <DanC> and I'd like to see older active issues get closed before new active issues are added.

14:00:05 <AaronSw> +Aaron

14:00:07 <DanC> (or get rid of the "active issues" section altogether, and just use the telcon agendas to see what's up)

14:00:26 <DanC> another time, I suppose.

14:00:30 <bwm> OK - I'll give it some thought - got a little tidying up done this week but not as much as I would have liked

14:00:38 * ArtB-logg promises not to cry :-)

14:00:48 <DanC> I'd really like to close about 5 issues/week.

14:00:55 <dajobe> agenda http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0173.html

14:01:12 <bwm> -hi

14:01:12 <Jema> Hi. I'm Jema, the Jena meeting assistant

14:01:54 <bwm> -open

14:01:55 <Jema> RDFCore WG Telecon 2001-10-12 is now open

14:01:55 <Jema> The agenda can be found at http://current/

14:01:55 <Jema> Agenda item 1: Allocate scribe

14:01:56 <jjc> jjc is now known as jjcscribe

14:02:12 <jjcscribe> -agenda next

14:02:12 <Jema> Agenda item 2: Roll Call

14:02:21 <jjcscribe> reg danb

14:02:25 <jjcscribe> brian here

14:02:29 <jjcscribe> abs ericm

14:02:38 <jjcscribe> art here

14:02:45 <jjcscribe> davebn here

14:02:48 <jjcscribe> frankb abs

14:02:51 <jjcscribe> jheremy c here

14:02:53 <jjcscribe> danc reg

14:02:57 <jjcscribe> ron here

14:03:04 <jjcscribe> bill dehora reg

14:03:09 <jjcscribe> jos abs

14:03:13 <jjcscribe> rael here

14:03:22 * danbri regrets; may be free to join call later (will try follow irc tho)

14:03:25 <jjcscribe> no rael abs

14:03:30 <jjcscribe> jang regrets

14:03:32 <jjcscribe> martin here

14:03:44 <jjcscribe> yoshi abs

14:03:49 <jjcscribe> grham here

14:03:51 <jjcscribe> michale abs

14:03:54 <jjcscribe> kwon abs

14:04:00 <jjcscribe> ora abs

14:04:04 <jjcscribe> frank m here

14:04:09 <jjcscribe> satoshi abs

14:04:21 <jjcscribe> steve gone?

14:04:30 <jjcscribe> piere abs

14:04:32 <jjcscribe> aaron here

14:04:39 <jjcscribe> mike dean abs

14:04:44 <jjcscribe> guha abs

14:04:45 <jjcscribe> pat here

14:04:57 <jjcscribe> sergey not yet up

14:05:20 <jjcscribe> -agenda next

14:05:21 <Jema> Agenda item 3: Review Agenda

14:05:21 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0173.html

14:05:26 <jjcscribe> jos arrived

14:05:27 <spetschu> steve here... sorry, computer problems

14:05:29 <jjcscribe> miked arrived

14:05:35 <jjcscribe> steve arrived

14:06:06 <jjcscribe> aaron adds singing to agenda

14:06:14 <jjcscribe> bwm: rednition by aaron

14:06:21 <jjcscribe> ron daniel now here

14:06:28 * ArtB-logg things we should sing Kumbaya :-)

14:06:48 <jjcscribe> ron d suggests parseType=Literal decision as topic

14:06:52 <jjcscribe> -agenda next

14:06:54 <Jema> Agenda item 4: Next telecon - 10am Boston time, 19th October 2001

14:07:01 <jjcscribe> -agenda next

14:07:02 <Jema> Agenda item 5: Review Minutes of previous meeting

14:07:02 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0005.html

14:07:06 <jjcscribe> -agenda next

14:07:06 <Jema> Agenda item 6: Confirm Status of Completed Actions

14:07:22 * em dialing

14:07:23 <jjcscribe> -agenda 4

14:07:23 <Jema> Agenda item 4: Next telecon - 10am Boston time, 19th October 2001

14:07:26 <em> +em

14:07:29 <jjcscribe> agreed

14:07:38 <jjcscribe> pat gives regrest for next week

14:07:40 <jjcscribe> -agenda 6

14:07:41 <Jema> Agenda item 6: Confirm Status of Completed Actions

14:07:48 <jjcscribe> ercim here

14:07:57 <jjcscribe> -agenda 7

14:07:58 <Jema> Agenda item 7: Confirm Folling actions are withdrawn

14:08:05 <jjcscribe> agreed bwm and ericm

14:08:07 <jjcscribe> -agenda 8

14:08:08 <Jema> Agenda item 8: Tidying up broken Test cases

14:08:08 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0322.html

14:08:22 <jjcscribe> 2201-09-28#5 and #6 done

14:08:44 <jjcscribe> 2001-09-28#7 continued

14:09:02 <jjcscribe> #8 Art has made proposal

14:09:10 <jjcscribe> #9

14:09:20 * AaronSw abstains from decision

14:09:21 <jjcscribe> -decisiion accept art's proposal

14:09:22 <Jema> command not recognised

14:09:32 <jjcscribe> -decision accept art's proposal

14:09:38 <dajobe>http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0322.html

14:10:03 <jjcscribe> -hi

14:10:04 <Jema> Hi. I'm Jema, the Jena meeting assistant

14:10:11 <jjcscribe> -agenda next

14:10:12 <Jema> Agenda item 9: Propose approve cycle in subClassOf test cases

14:10:12 <Jema> See: http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfs-no-cycles-in-subClassOf/

14:10:32 <jjcscribe> daveb refviewed

14:10:42 <jjcscribe> jos reviewed

14:10:46 <jjcscribe> art revire

14:10:59 <jjcscribe> pat reviewed

14:11:11 <jjcscribe> -decision agreed test cases subClassOf

14:11:16 <jjcscribe> -agenda next

14:11:17 <Jema> Agenda item 10: Propose approve cycle in subPropertyOf test cases

14:11:17 <Jema> See: http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfs-no-cycles-in-subPropertyOf/

14:11:25 <jjcscribe> daveb has reviewed

14:11:31 <jjcscribe> jos, art have reviewed

14:11:39 <jjcscribe> no disent

14:11:44 * AaronSw notes that we should be sure to be clear that these are not syntactical test cases... in the manifest or whatever

14:11:47 <jjcscribe> -decision approve test cases

14:11:57 <jjcscribe> -agenda next

14:11:58 <Jema> Agenda item 11: Review domain and range test cases

14:11:58 <Jema> See: http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfs-domain-and-range/

14:12:20 <jjcscribe> daveb has had quick look

14:12:23 <jjcscribe> art wants more time

14:12:38 <jjcscribe> -action art review domain and range test cases

14:12:39 <Jema> Jema could not parse the action command. Please separate the action holder from the action by a '/' character

14:12:55 <jjcscribe> -action pat/review d&r test cases

14:12:56 <Jema> Jema notes action 2001-10-12#1

14:13:12 <jjcscribe> -action art/review d&r test cases

14:13:13 <Jema> Jema notes action 2001-10-12#2

14:13:24 <jjcscribe> art suggests uses of example.org as default hostname

14:14:00 <jjcscribe> -action jos/change test cases to use example.org

14:14:01 <Jema> Jema notes action 2001-10-12#3

14:15:24 <jjcscribe> jos clarifies that entailment is excluded from current discussion

14:15:44 <sergey> +sergey

14:17:06 <jjcscribe> sergey here

14:17:33 <jjcscribe> -agenda next

14:17:34 <Jema> Agenda item 12: Issue: Identity of anonymous resources

14:17:34 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0147.html

14:17:41 <jjcscribe> -agenda next

14:17:42 <Jema> Agenda item 13: Syntax Sub-group Report

14:17:42 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0119.html

14:17:50 <AaronSw> -agenda prev

14:17:51 <Jema> Agenda item 12: Issue: Identity of anonymous resources

14:17:51 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0147.html

14:18:28 <jjcscribe> bwm anyone wnat to discuss?

14:19:16 <jjcscribe> ron: questions no URI on bNode

14:20:43 <gk> Ron is saying that he didn't notice the decision that bNodes are existentially quantified, as opposed to just having unknown identifier

14:20:50 <jjcscribe> pat: must be able to distingush bNode from other nodes

14:21:38 <gk> Ron is suggesting special URI scheme; Frank disagrees

14:22:25 <jjcscribe> jos notes bNodes are local not global scope

14:22:53 <jjcscribe> chair asks ron if he is satisified

14:23:01 <gk> I would like to observe that bNode *names* exist *only* in N-triples; not in RDF/XML or model

14:23:36 <jjcscribe> ron wonders if we are making life harder

14:24:14 <jjcscribe> frank notes uris are permanent

14:24:53 <jjcscribe> chair intervenes

14:25:07 <jjcscribe> general surpirse

14:25:20 <jjcscribe> frank adds ...

14:25:31 <jjcscribe> .... qu about specific text

14:26:00 <jjcscribe> ... sympathy with DanC's comment: where does this text go?

14:26:59 <jjcscribe> bwm says text is for issue list

14:27:34 <jjcscribe> frank says also stuff in draft primer

14:27:46 <jjcscribe> pat answer special case issue

14:28:00 <jjcscribe> pat claims this distinctionis fundamental and is needed

14:28:09 <jjcscribe> pat sees this as one off

14:28:24 <jjcscribe> not the beginnging of the hordes of special cases

14:28:26 <danbri> sorry to barge in. A minor nit pick:

14:28:27 <danbri> [[

14:28:28 <danbri> To directly address the question of the issue: a so-called anonymous

14:28:28 <danbri> resource has no URI.

14:28:28 <danbri> ]]

14:29:00 <gk> to DanBri: I think that's what the text says, with explanation

14:29:01 <danbri> ...should be qualified by noting that the original question can mislead; the resource itself may have a URI; we are concerned here however with some specific description of that resource.

14:29:02 <jjcscribe> chair claims we reahed consensus at f2f

14:29:08 <jjcscribe> chair unhappy with late objections

14:29:12 <danbri> I was quoting from http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0147.html

14:29:38 <danbri> the rest of the text is great; that little bit (taken without carefully considering rest) can propagate the confusion.

14:29:44 <danbri> </delurk>

14:29:58 <jjcscribe> chair wants substantial cases against the consensus that has been forming

14:30:00 <gk> to DanBri: existential quanitification doesn't necessarily mean only one such resource

14:30:01 * danbri not objecting; just a textual nitpic (pls ignore)

14:30:27 <danbri> I agree. I'm on IRC only, responding to text. I should go away, not being helpful doing this.

14:31:02 <jjcscribe> ron suggests we have a vote

14:31:15 * gk to DanBri: no, it's OK ... it provides some additional viewpoint

14:31:56 <jjcscribe> who supports model theory on bnodes

14:32:38 <jjcscribe> vote aborted

14:32:56 <jjcscribe> chair postpones to next week - wants more discussion

14:33:07 <jjcscribe> -action gk/to bring this back to table

14:33:08 <Jema> Jema notes action 2001-10-12#4

14:33:48 <jjcscribe> sergey makes procedural suggestion

14:34:09 <jjcscribe> sergey is unbconfortable without case

14:34:32 <gk> GK note to self: formulate a simple question that is votable

14:34:50 <jjcscribe> chair puts onus on obj

14:35:17 <jjcscribe> -action frankm/respond to gk text

14:35:18 <Jema> Jema notes action 2001-10-12#5

14:35:42 <gk> GK note to self: separate existential interpretation from explanatory text

14:35:43 <jjcscribe> -agenda next

14:35:43 <Jema> Agenda item 13: Syntax Sub-group Report

14:35:43 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0119.html

14:35:56 <jjcscribe> dave quick or long?

14:36:05 <jjcscribe> dave - quick version

14:36:11 <jjcscribe> we like relax ng

14:36:36 <jjcscribe> jeremy is worjking on xform engine

14:36:51 <jjcscribe> we have a list of syntactic issues from issue list

14:36:55 <jjcscribe> -agenda next

14:36:55 <Jema> Agenda item 14: Issue rdfs-xml-schema-datatypes: Review Progress

14:36:55 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0444.html

14:36:55 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0003.html

14:36:55 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0002.html

14:37:05 <jjcscribe> -agenda 13

14:37:05 <Jema> Agenda item 13: Syntax Sub-group Report

14:37:05 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0119.html

14:37:26 <jjcscribe> jos asks whetehr we will use XSLT as technology

14:37:54 <jjcscribe> daveb ducks

14:38:04 <jjcscribe> -agenda 14

14:38:04 <Jema> Agenda item 14: Issue rdfs-xml-schema-datatypes: Review Progress

14:38:04 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0444.html

14:38:04 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0003.html

14:38:04 <Jema> See: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Oct/0002.html

14:38:33 <jjcscribe> sergey notes e-mail discussion is ongoing

14:39:10 <jjcscribe> pat and peter have been progressing

14:40:01 <jjcscribe> -action jos/send interpretaion properties URL to list

14:40:02 <Jema> Jema notes action 2001-10-12#6

14:40:13 <jjcscribe> -action sergey/sumamrise this issue fo rnext weeks telecon

14:40:13 <Jema> Jema notes action 2001-10-12#7

14:40:14 <ArtB-logg> Interpretation properties: http://www.w3.org/DesignIssues/InterpretationProperties.html

14:40:24 * ArtB-logg just did 2001-10-12#7 for Jos!

14:41:08 <jjcscribe> AOB - ron parseType=Literal

14:42:06 <jjcscribe> DAnc suggested parseTyper =Lieeral is dropped

14:42:11 <jjcscribe> rond and daveb are opposed

14:42:22 <jjcscribe> rond use case 1 prism working group

14:42:38 <jjcscribe> editiorial style for magazine article einterview

14:42:55 <jjcscribe> abbreviated name in italic at ebginning using html markup

14:43:16 <jjcscribe> <I>intervioew: kkk </I>

14:43:23 <jjcscribe> use case 2:

14:43:36 <AaronSw> Actually, I think it was <title><i>CRN</i> Interviews: Tom Cruise</title>

14:43:41 <jjcscribe> titles of things need embededd markup

14:43:43 <spetschu> <i>mag:</i>interview_name

14:44:01 <jjcscribe> again html markup for paragaph break

14:44:10 <jjcscribe> use case 3

14:44:51 <jjcscribe> i18n wg late in process for M&S required markup for bidi and ruby

14:45:27 <jjcscribe> i18n may require arbitrary XML

14:45:27 <AaronSw> (arbitrary xml)

14:46:19 <gk> I note that Ron's cases 1 and 2 refer to markup tags ... if the literal is a unicode sequence then the tags could be &lt;...&gt; encoded

14:47:50 <jjcscribe> rond sees it as a parser escape

14:47:59 <jjcscribe> pat wonders whether it involves typing

14:49:25 <AaronSw> Parser escape for RDF, but leaves it as XML for any other engine that wants to look at it. PatH understands now.

14:49:44 <jjcscribe> jeremy suggests further discussion on e-mail

14:49:46 <AaronSw> Aaron explains example doing minutes in RDF

14:50:08 <jjcscribe> jeremy suggests there is little support for dorpping it

14:52:09 <gk> Aha.. N-triples has *only* unicode strings, no?

14:52:18 <dajobe> gk: yes

14:52:19 <jjcscribe> jeremy asks for opinion on string versus well-formed XML

14:52:27 <jjcscribe> jos likes that

14:52:47 <jjcscribe> rond wants the perl hacker to be able to leave string unchanged

14:53:16 <jjcscribe> rond wants a flag on string saying it is parsetype literal

14:53:36 <jjcscribe> namespaces could be listed explciitly (rond)

14:54:28 <jjcscribe> we note that was no support in telecon for delting parseType = lityeral (abandoning?)

14:55:00 <dajobe> proposals to change ntriples were also discussed a bit

14:55:14 <jjcscribe> gk concerned about complexity

14:55:52 <jjcscribe> gk wonders how parseType="Literal" and non-XML syntax e.g. n-triple

14:56:20 <jjcscribe> bwm asks ron for original motivations ...

14:56:30 <jjcscribe> bwm: people wanted to embed html /xml

14:56:43 <jjcscribe> bwm people could use clunky escaping

14:56:52 <gk> By way of explanation: the use of parsetype=literal seems to be very bound up with the behaviour of XML

14:56:54 <dajobe> so "literal"(XML)

14:56:56 <dajobe> >?

14:57:04 <jjcscribe> bwm it was put in to make it esay to write in markup

14:57:35 <jjcscribe> no discussion of namespace impact

14:58:24 <jjcscribe> art ask rond

14:58:31 <jjcscribe> about prism expectations on namespace

14:58:55 <jjcscribe> rond says <I> and <P> are understood as in html

14:59:04 <gk> parsetype=literal is beginning to sound like a [CDATA[ equiv for RDF???

14:59:10 <dajobe> bleugh

14:59:22 <jjcscribe> rond origianl use case was for moving metadata and meta data is hwon to people

14:59:23 <gk> yeah

15:00:27 <DanC> nobody else sees parseType="Literal" like a literal expression in a programming language, ala python's dictionary {} and list [] and tuple () syntax? i.e. something you could build up with API calls, but there's a syntactic shortcut for it.

15:00:31 <gk> Pat: best RDF can do is preserve markup without damage; not reasonable to interpret it.

15:03:10 <gk> See also DanC's piece on shoe sizes...

15:03:10 <jjcscribe> -close

15:03:12 <Jema> The meeting is closed

15:03:14 <Jema> the action/decision list has been mailed

15:03:38 <gk> (Jos just referred to TimBL's interpretation properties)

15:05:53 <gk> Got it: http://www.w3.org/2001/01/ct24 (DanC's note)

15:07:29 * DanC is back

15:07:31 * DanC is back

15:07:48 * DanC is away: back later today

15:09:58 <dajobe> discussion of parseType literal things...

15:12:16 <dajobe> "literal"(xmlns:ns1="URI1"; xml:lang="...") ?

15:13:59 <dajobe> "<i>ABC</i> blah"(xmlns="HTML4-ns-URI")

15:15:01 <gk> [ rdf:value "<i>ABC</i> ; ex:xmlns="HTML4-ns-URI" ] ?

15:15:08 <dajobe> that is N3

15:15:15 <gk> Yup. That was the point.

15:15:48 <gk> I.e. the extra info could be encoded in the RDF graph, and the literal left as just a unicode string.

15:15:48 <dajobe> one thing at a time, not adding [], i and n3 abbreviations all at once

15:16:03 <gk> OK ... just a thought experiment.

15:16:07 <dajobe> me too

15:16:55 <dajobe> I also wondered about using xmlns as a property like

15:17:10 <dajobe> <foo> <xml:xmlns> "ns:NSURI"

15:17:31 <dajobe> so for any <foo> a query can get all namespacea about it

15:17:35 <dajobe> just an impl. thought...

15:21:19 <gk> Yes... seems reasonable (in the spirit of coding the NS info in RDF, rathjer than in the literal?)

15:21:48 <gk> ... and if we could have properties of literals???

15:25:05 <gk> Key point: we need to be able to treat different instances of the same string differently.

15:28:26 <gk> If literals as subjects OK, do we need to relax the MT tidiness constraint for literal nodes?

15:29:03 <dajobe> pathayes outlining literals as subjects needs

15:29:26 <dajobe> _:node1"string" versus _:node2"string"

15:29:44 <dajobe> to distinguish different occurences of string

15:31:48 <dajobe> ron talking about <Description><rdf:type rdf:resource="an-xml-literal"><rdf:value>the string</rdf:value></rdf:Description> in future

15:32:21 <dajobe> roundtripping, yes but limitatiions with rdf/xml 1.0 - parsertype literal and xml:lang yes

15:32:31 <dajobe> but other things with string as subject - no

15:41:19 <dajobe> discussion of literal subjects ...

15:41:26 <dajobe> OK

15:41:35 <dajobe> but literal properties being rather barmy (my words)

15:41:41 <dajobe> in MT

15:41:49 <dajobe> and confusing with respect to datatype issues