W3C RDF Core Working Group IRC Chat Logs for 2002-01-11

14:00:02 Users on #rdfcore: @logger

14:00:03 <ChanServ> You do not have AutoOp access to [#rdfcore]

14:00:03 <ChanServ> You do not have AutoOp access to [#rdfcore]

14:53:01 <dajobe> dajobe has changed the topic to: RDF Core WG Meeting 15:00-16:00 UTC Friday

14:56:06 * DanC wonders where EricM is...

14:56:25 <dajobe> he was on #rdfig

14:56:28 <DanC> I'd like to attend, for the datatyping stuff, but I haven't been to my other Friday meeting for some time, and I think I need to.

14:57:04 <dajobe> dajobe has changed the topic to: RDF Core WG Meeting 15:00-16:00 UTC Friday - agenda http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0048.html

14:58:50 * dajobe dials in

14:58:53 <AaronSw>http://www.w3.org/1998/12/bridge/Zakim.html

14:58:57 <AaronSw> +Aaron

15:00:50 * em-mit appolgies for being about 10 min late

15:02:09 <jjc> dialing

15:03:50 <dajobe> dajobe is now known as jang_scribe

15:03:53 <jang_scribe> roll call

15:04:09 <jang_scribe> danbri abs, bwm here, em reg (10 mins late?)

15:04:31 <jang_scribe> dave b here, frank b abs, jeremy here, danc reg

15:04:46 <jang_scribe> ron d abs, bill deh abs, jos de roo here, rael abs

15:05:09 <jang_scribe> jang here, martin h here, yoshi abs, graham here, michael kop abs

15:05:25 <jang_scribe> kwon abs, ora abs, frank m here, satoshi abs

15:05:43 <jang_scribe> steve pet abs, pierre abs, patrick here, aaron here, mike dean here

15:05:59 <jang_scribe> guha abs, pat h reg, sergei probably en route

15:06:26 <jang_scribe> review agenda

15:06:28 <jang_scribe> aob?

15:06:42 <jang_scribe> next telecon next week, same time

15:06:47 <jang_scribe> minutees of last meeting...

15:06:50 <jang_scribe> approved

15:06:56 <jang_scribe> completed actions: any problems?

15:06:58 <jang_scribe> all approved

15:07:08 <jang_scribe> item 7: syntax wd

15:07:15 <jang_scribe> congratulations!

15:07:28 <jang_scribe> any more to add dave?

15:07:44 <jang_scribe> daveb: plan to do another one soon, to try to close as many (all) issues by the f2f

15:08:11 <jang_scribe> jeremy: we need to reopen i18n issues

15:08:28 <jang_scribe> ...but I've not got around to it yet

15:08:45 <jang_scribe> I won't be able to do this until return from webont

15:08:53 <jang_scribe> item 8

15:09:13 <jang_scribe> (ron d turns up)

15:09:28 <jang_scribe> jang: need better access to repository. can someone from w3c please expedite?

15:09:56 <jang_scribe> action: bwm to persue with em

15:10:02 <jang_scribe> continued

15:10:29 <jang_scribe> action: jang to post current changes to mailing list

15:10:49 <jang_scribe> jos: just sent mail re progress on entailment tests

15:10:58 <jang_scribe> item 9: primer

15:11:04 <jang_scribe> frank m:

15:11:17 <jang_scribe> there's a new draft of the intro, xml intro

15:11:35 <jang_scribe> we've got suggested material from gk

15:11:40 <jang_scribe> not yet been incorporated

15:11:40 <AaronSw> all should be linked from http://www.w3.org/2001/09/rdfprimer/todo eventually

15:11:46 <jang_scribe> cheers aaron

15:12:31 <jang_scribe> aaron, i didn't catch that...

15:12:47 <jang_scribe> item 10: model theory wd

15:12:49 <jjc> Aaron said: We weren't slaving dspite erics request

15:12:53 <jang_scribe> heh

15:13:14 <jang_scribe> reminder on review actions prior to path's return next week... is that ok?

15:13:33 <jang_scribe> jeremy: those of us going to webont have to review this twice!

15:13:56 <jang_scribe> thanks to gk for his review

15:14:15 <jang_scribe> daveb: is pat going to tidy this up for publishing when he gets back?

15:14:18 <jang_scribe> bwm: that's the idea

15:14:36 <jang_scribe> item 11: test cases for para 196...

15:14:54 <jang_scribe> proposal to accept these test cases...

15:15:21 <jang_scribe> jeremy: another test case would help

15:15:27 <jang_scribe> ...with a fragment id, say

15:15:37 <jang_scribe> ...if dave thinks this is enough, I'm happy

15:15:50 <jang_scribe> proposal: daveb to add test case for same thing but with frag id,

15:15:54 <jang_scribe> then to approve and close

15:16:02 <jang_scribe> "so ordered"

15:16:11 <jang_scribe> item 12: qname-uri-mapping

15:16:25 <jang_scribe> daveb's proposal to close the issue, not making any change

15:17:01 <jang_scribe> objections?

15:17:06 <jang_scribe> APPROVED

15:17:18 <jang_scribe> issue 13: rdfms-reification-required

15:17:41 <jang_scribe> proposal: parsers aren't required to reify all rdf:Description elements

15:17:47 <jang_scribe> except in the explicit case

15:18:00 <jang_scribe> comments?

15:18:28 <jang_scribe> jos: this is part of the larger reification issue, which we keep putting off

15:18:42 <jang_scribe> so I'm going to abstain until we get to the larger issue

15:19:05 <jang_scribe> bwm: to clarify: there's no implication for other aspects of reification here

15:19:16 * DanC has input to reification too; i.e. don't think that just because I haven't said anything about reification that I don't have anything to say about it.

15:19:39 <jang_scribe> jeremy: I'd like to have the reification discussionsooner rather than later

15:19:57 <AaronSw> jjc: if you want reification dropped, you need to raise it.

15:20:13 <jang_scribe> pat st: can you clarify what you mean by dropping reification?

15:20:18 <jang_scribe> bwm: sense a rathole

15:20:37 <jang_scribe> if anyone wishes to propose to just drop this, start a discussion after the meeting or on the mailing list

15:20:40 <jang_scribe> I'd rather move on

15:21:05 <jang_scribe> jos abstains, on the grounds that reification is so ambiguous at the moment

15:21:15 <jang_scribe> frank m abstains

15:21:28 <jang_scribe> daveb: closing this doesn't mean we can't modify this later

15:21:36 <jang_scribe> no dissent, so ordered

15:21:56 <jang_scribe> frank m: would it be in order to suggest that the WG takie an action item to really discuss reification

15:22:00 <jang_scribe> this keeps on coming up

15:22:19 <jang_scribe> a suggestion has been made previously to drop it; sooner or later we've got to deal with it

15:22:27 <jang_scribe> aaron: www-rdf-reification@ ..?

15:22:32 <jang_scribe> :-)

15:22:41 <AaronSw> subscription is optional. :)

15:22:57 <jang_scribe> bwm: we've an outstanding issue on MT to deal with reification

15:23:06 <jang_scribe> frank: ok, as long as it comes up!

15:23:19 <jang_scribe> it's not the MT per se that makes the decision; the MT reflects what we decide to you

15:23:24 <jang_scribe> bwm: I absolutely agree

15:23:39 <jang_scribe> bwm: I'll check for an issue on reification...

15:24:07 <jang_scribe> action: bwm to add an issue on reification: "fix/drop reification"

15:24:20 <jang_scribe> (em turns up)

15:24:52 <jang_scribe> item 14: rdfms-replace-value

15:24:58 <jang_scribe> proposal 1 from dave (drop it)

15:25:11 <jang_scribe> prop 2 aaron : open new issue

15:25:11 <em-mit> em: regrets from DanC

15:25:22 <jang_scribe> prop 3: bwm (a synthesis of the two)

15:25:40 <jang_scribe> gk: this depends on the outcome of another discussion, but the DT/literal discussion...

15:25:57 <jang_scribe> ...at least one interp of rdf:value is to regard it as an equivalence relation

15:26:11 <jang_scribe> daveb: we're not discussing its meaning, just "do we rename it"?

15:26:23 <jang_scribe> aaron: I think brian's proposal is the way forward

15:26:30 <jang_scribe> (text in agenda)

15:26:40 <jang_scribe> we need to reclarify rdf:value

15:26:54 <jang_scribe> proposal: to go with brian's option

15:27:02 <jang_scribe> comments? dissent? abstentions?

15:27:04 <jang_scribe> so ordered

15:27:06 <jang_scribe> ace.

15:27:30 <jang_scribe> item 15: datatypes

15:27:53 <jang_scribe> bwm thanks mike dean for joint committee input

15:28:01 <jang_scribe> bwm: sergei here yet? apparently not

15:28:16 <jang_scribe> mike dean: the dt was a big issue for JC when we addressed it

15:28:27 <jang_scribe> there were two camps: implicit vs explicit

15:28:57 <jang_scribe> we ended up with a compromise which allows content producer to decide which mechanism they use

15:29:08 <jang_scribe> you may indeed end up with models that use both approaches

15:29:19 <jang_scribe> I suspect there'll be two camps here too

15:29:35 <jang_scribe> we'll (i suspect) satisfy a bit of all camps

15:29:54 <jang_scribe> jeremy: clarify; implicit means associated with property

15:30:00 <jang_scribe> (ie, range or somesuch)

15:30:10 <jang_scribe> gk: can you point to some examples of how this works with daml?

15:30:22 <jang_scribe> mike: not on the spur of the moment, ...

15:30:42 <jang_scribe> (but he sketches example based on geographical location)

15:30:49 <jang_scribe> gk: can you drop an example to the mailing list?

15:30:58 * danb_awol multitasks, badly

15:30:59 <jang_scribe> ACTION: miked to drop example of both approaches to the mailing list

15:31:23 <jang_scribe> gk: I'd like to see the examples before I comment on this

15:31:48 <jang_scribe> bwm: makes remarks from the chair...

15:32:04 <jang_scribe> I've had a break from DT over the holiday period, and I'm getting back up to speed...

15:32:13 <jang_scribe> there's nowhere I can go for a summary of all the proposals

15:32:26 <jang_scribe> I really think this is a problem, that we don't have that easily ot hand

15:32:56 <jang_scribe> also, my observation is that we're moving away from the generation of proposals and discussion of their characteristics toward people stating their opinion of what the right answer is

15:33:22 <jang_scribe> it's now time to capture all this discussion in a succinct form and move on to being clear about our requirements, our options, our criteria

15:33:27 <jang_scribe> and move towards making a decision

15:33:38 <jang_scribe> gk: there are placeholders in sergei's document for these...

15:33:59 <jang_scribe> jeremy: pats and jjc were discussing a PD document on the list today

15:34:04 <jang_scribe> gk: great!

15:34:13 <jang_scribe> daveb: so _how_ many proposals do we have?

15:34:18 <jang_scribe> bwm: that's a good question

15:34:43 <jang_scribe> we've had input from the JC, the basic thrust of which is we should allow implicit & explicit / global/local typing

15:34:55 <jang_scribe> do we accept that this should be one of our requirements?

15:35:15 <jang_scribe> frankm: the claim is that daml+OIL supports that requirement

15:35:29 <jang_scribe> if that claim is true, there's at least an existence proof that this criterion can be met

15:35:47 <jang_scribe> bwm: daml+oil want to build on our work, they're strongly urging us to meet that constraint

15:36:07 <jang_scribe> daveb: I'm not quite clear on this requirement; for those who do understand it, is it a legit requirement that we should meet?

15:36:17 <jang_scribe> I think we should strongly listen to what daml+oil says?

15:36:22 <AaronSw> proposals: G, P, P++, PD, PL, S, X

15:36:35 <jang_scribe> action: jeremy to write a proposal that we accept this as one of our requirements

15:37:08 <jang_scribe> bwm: pl fails the daml requirement

15:37:14 <AaronSw> aaron: it does?

15:37:28 <jang_scribe> gk: we can only consider alternatives for which we have comparable level of writeups?

15:37:29 <jjc> PL is doesn't fail it

15:37:53 <jang_scribe> pats: do we want a deadline?

15:37:58 <jang_scribe> bwm: what's a good date?

15:38:12 <jang_scribe> I want this decision to be taken by the end of the f2f

15:38:26 <jang_scribe> preferably before that, but we need to get the doc out for feedback by the f2f

15:38:35 <jang_scribe> I'd push for next week, were it not for webont

15:38:41 <jang_scribe> two weeks from today.

15:38:50 <jang_scribe> who proposed G?

15:39:06 <jang_scribe> gk: sounds like me, it's been folded into S.

15:39:11 <jang_scribe> G is no longer a distinct proposal

15:39:24 <jang_scribe> P/P++ are natural for path, is there anyone else to pick this up?

15:39:39 <jang_scribe> pats: p++ is out by reason of charter, those are all folded into PD

15:39:52 <jang_scribe> jeremy: pd is the natural place to put all those proposals

15:40:09 <jang_scribe> who is going to write up PD?

15:40:15 <AaronSw> sergey's writeup: http://www-db.stanford.edu/~melnik/rdf/datatyping/

15:40:28 <jang_scribe> action: pats, jeremy to do the prose/math (resp.) for PD

15:40:36 <jang_scribe> PL writeup in place from danc

15:41:04 <jang_scribe> gk: the problem is that PL writeup doesn't address specific points like sergei's doc does

15:41:12 <jang_scribe> jos: I think pl is s(b)

15:41:17 <jang_scribe> gk: that's not clear to me...

15:41:32 <jang_scribe> can we ask danc to complete or drop the pl writeup?

15:41:35 <jang_scribe> ^^^daveb

15:41:54 <jang_scribe> jeremy: the document standard bar shold not be set too high

15:42:25 <jang_scribe> pats: we need the math for pd so that we can compare it to s

15:42:37 <jang_scribe> gk: that's what jeremy is actioned to do

15:43:00 <DanC> yes, PL is S(b)

15:43:18 <DanC> I'm happy to drop PL in favor of S(b)

15:43:28 <jang_scribe> right, cheers danc

15:43:37 <jang_scribe> s(b) is written up in sergei's doc

15:43:39 <jang_scribe> what about X?

15:43:50 <jang_scribe> pats: X dropped due to charter considerations

15:44:06 <jang_scribe> pats: U can be left out as an application on top of the PD approach, it can be left out for now

15:44:14 <jang_scribe> we're downto three then, are there any others?

15:44:28 <jang_scribe> namely, PD (subsuming P/P++)

15:44:46 <jang_scribe> bwm: those are around until pat agrees they're subsumed

15:44:59 <jang_scribe> we've got PD, S

15:45:06 <jang_scribe> do we need more on S?

15:45:18 <jang_scribe> jeremy: can we do this writeup independently of sergei's document?

15:45:28 <jang_scribe> gk: it would be good if it did fit in, but the timeframe is limiting

15:45:52 <jang_scribe> pats: I'd like to steal most of sergei's prelim material :-)

15:46:01 <jang_scribe> but there's too much overhead to try to synch on a single document

15:46:20 <jang_scribe> unles sergei is willing to do the editorial work to make neutral some of his document

15:46:35 <jang_scribe> bwm: we're going to be sending something out to the wider community seeking feedback

15:46:53 <jang_scribe> ...possibly it's got _too_ much detail, we might want to send out something more discursive

15:47:41 <jang_scribe> gk: I'm worried that we're going to hobble ourselves without other considerations coming into play: presentation issues obfuscate the detail needed for making the decision

15:47:59 <jang_scribe> pats: if we need this done quickly, we can't have too much editorial overhead

15:48:13 <jang_scribe> jeremy: in two weeks, we go to rdf-interest. we've been skirting this for ages

15:48:29 <jang_scribe> if sergei wishes to reapparaise his doc for that audience, so be it; if he doesn't want to, fine

15:48:46 <jang_scribe> so the decision is that we have as many separate docs (put together as best we can) by then

15:48:51 <jang_scribe> that sounded like a DECISION

15:49:09 <jang_scribe> bwm: something that could be done in parallel is collecting and writing up the criteria/requirements

15:49:14 <jang_scribe> anyone ready to take this up?

15:49:22 <jang_scribe> gk: sergei's done this, section 1.2 of his doc

15:49:37 <jang_scribe> bwm: if that's uptodat and complete, fine, I'd like to ensure that's the case

15:49:50 <jang_scribe> gk: I couldn't think of anything else when I reviewed his document

15:49:56 <jang_scribe> daveb: is the daml+oil requirement in there?

15:50:03 <jang_scribe> pats: I think they are

15:50:25 <jang_scribe> daveb: we should put it out as a separate document

15:50:32 <jang_scribe> ...but I'm not volunteering(!)

15:50:50 <jang_scribe> daveb tries to volunteer graham

15:52:38 * jang_scribe turns up his trenchcoat collar and adjusts dark galsses

15:54:24 <jang_scribe> action: pats: excise intro material from sergei's document which is common and represents current agreement from the document

15:54:57 <jang_scribe> (this is a repeat of the previous pats/jjc action)

15:55:31 <jang_scribe> action: gk to produce desiderata / requirements / criteria document

15:55:55 <jang_scribe> action also all WG members: review sergei's document, anything missing, send it to graham

15:56:07 <jang_scribe> (that is, the requirements in the .5 page of the document)

15:56:42 <jang_scribe> gk: I'd like to avoid "requirements" incase they turn into a straighajcket

15:56:45 <jang_scribe> 4 minutes remain...

15:56:54 <jang_scribe> bwm: quick AOB.

15:57:04 <jang_scribe> face to face is coming, can people who are coming register please?

15:57:10 <jang_scribe> aaron: don't think I'll be able to make it

15:57:14 <DanC> I take it rdfms-assertion isn't happening today, right?

15:57:14 <jang_scribe> frank m neither.

15:57:26 <jjc> prob not

15:57:32 <jang_scribe> danc: can't see it on the agenda

15:57:42 <jang_scribe> ah, probably not

15:57:45 <jjc> item 17, bit we hjaven't dfoned e 16

15:57:47 <jang_scribe> item 17, right?

15:58:10 <jang_scribe> action: bwm to repost pointer to f2f details, registration. also in agenda for next week

15:58:24 <jang_scribe> bwm: we're about done on evrything today

15:58:46 <AaronSw> gk, http://cgi.w3.org/Register/selectUser.pl

15:58:46 <jang_scribe> bwm: are you going to have more syntax issues for next week dave?

15:58:54 <jang_scribe> daveb: yes, expect some more stuff monday/tuesday

15:58:59 <jang_scribe> any other AOB?

15:59:01 <jang_scribe> meeting closes

15:59:19 <jang_scribe> ...but there's another informal 30 minutes if you want it.

15:59:25 <jang_scribe> thanks everyone

16:44:00 <danb_awol> danb_awol is now known as danb_desk