14:00:05 <logger> logger has joined #rdfcore
14:00:05 <irc.w3.org> Users on #rdfcore: @logger 
14:56:58 <cmjg> cmjg has joined #rdfcore
14:57:20 <Zakim> Zakim has joined #rdfcore
14:57:25 <cmjg> zakim, this is rdfcore
14:57:27 <Zakim> sorry, cmjg, I do not see a conference named 'rdfcore'
14:57:47 <gk> gk has joined #rdfcore
14:58:17 <bwm> bwm has joined #rdfcore
14:58:53 <cmjg> zakim, this is rdfcore
14:58:54 <Zakim> ok, cmjg
14:58:56 <Zakim> +??P5
14:59:03 <cmjg> zakim, who is on the phone?
14:59:04 <bwm> Zakim, ??p5 is HP
14:59:04 <Zakim> On the phone I see ??P2, ??P5
14:59:05 <Zakim> +HP; got it
14:59:16 <cmjg> zakim, ??p2 is jang
14:59:17 <Zakim> +Jang; got it
14:59:19 <Zakim> +FrankM
14:59:49 <Zakim> +GrahamKlyne
15:00:29 <Zakim> +??P11
15:00:41 <Zakim> +PatH
15:00:43 <bwm> Zakim, ??P11 is SteveP
15:00:46 <Zakim> +SteveP; got it
15:00:47 <em> em has joined #rdfcore
15:02:09 <bwm> Zakim, HP has bwm
15:02:10 <Zakim> +Bwm; got it
15:02:20 <cmjg> bwm: zakim, hp has xxx; zakim, hp also has yyy
15:02:28 <cmjg> ^^^ like that
15:02:31 <em> em has changed the topic to: rdfcore 2003-02-07
15:02:36 <cmjg> with perfect syntax, obviously :-)
15:02:43 <em> em has changed the topic to: rdfcore 2003-02-07 teleconference
15:02:50 <em> ack... logger again has ops
15:02:52 <cmjg> cmjg is now known as jang
15:02:56 <bwm> Zakim, who is on the phone?
15:02:58 <Zakim> On the phone I see Jang, HP, FrankM, GrahamKlyne, SteveP, PatH
15:02:58 <Zakim> HP has Bwm
15:02:59 <jang> jang is now known as jang_scri
15:03:11 <jang_scri> we start...
15:03:14 <Zakim> +EMiller
15:03:28 <jang_scri> http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2003Feb/0073.html
15:03:31 <bwm> Zakim, who is on the phone?
15:03:31 <jang_scri> agenda
15:03:32 <Zakim> On the phone I see Jang, HP, FrankM, GrahamKlyne, SteveP, PatH, EMiller
15:03:34 <Zakim> HP has Bwm
15:03:36 <jang_scri> roll call...
15:03:56 <jang_scri> ...regrets: dave beckett
15:04:12 <jang_scri> (bwm checks for others)
15:04:22 <jang_scri> regrets patrick s
15:04:59 <jang_scri> agenda comments/ aob?
15:05:02 <jang_scri> nope
15:05:10 <jang_scri> next telecon 14 feb 2003
15:05:21 <jang_scri> scribe next week..?
15:05:33 <jang_scri> jan does it again
15:05:47 <jang_scri> minutes of last meeting:
15:05:48 <jang_scri> approved
15:06:03 <jang_scri> http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2003Jan/0241.html
15:06:16 <jang_scri> item 6: last call comment solicitation
15:06:18 <jang_scri> all done
15:06:31 <jang_scri> item 7: last call comments handling
15:06:41 <jang_scri> bwm's algorithm:
15:06:51 <jang_scri> http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2003Jan/0198.html
15:06:59 <jang_scri> change IDs?
15:07:17 <jang_scri> path: what level of change needs a change id?
15:07:53 <Zakim> +Mike_Dean
15:07:56 <jang_scri> http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2003Feb/0048.html
15:07:58 <jjc> jjc has joined #rdfcore
15:08:01 <jang_scri> ^^ summary of process
15:08:16 <jjc> Jeremy is at HP on the telecon.
15:08:31 <jang_scri> bwm: we resolved one too quickly, the software can't do that yet :-)
15:09:01 <mdean> mdean has joined #rdfcore
15:09:08 <jang_scri> general lauds for using rdf to manage our issues
15:09:27 <jang_scri> path: how large a change requires a change id?
15:09:31 <jang_scri> bwm: suggest...
15:09:34 <jang_scri> typo, no change ID
15:09:41 <gk> Note process URI in minutes is wrong one ... see later URI entered by Jan above
15:09:45 <jang_scri> anything in response to a comment should have a change ID
15:09:55 <jang_scri> jjc: what about URi/URI?
15:11:20 <jang_scri> frank: change ids apply to post-lc and stuff that's a holdover from pre-lc, right?
15:11:31 <danbri> danbri has joined #rdfcore
15:11:32 <jang_scri> bwm: yes, the director needs a list of changes
15:11:43 <danbri> * danbri waves, had network trouble, dialing...
15:11:56 <jang_scri> frank: eg, I've had all these prior-lc comments that we agreed to postpone until after LC
15:11:58 <Zakim> +DanBri
15:12:13 <jang_scri> bwm: if you make a change as a result of one of those, then it SHOULD be recorded (unless completely trivial)
15:12:32 <jang_scri> path: in the ed's draft I sent you...
15:12:42 <jang_scri> ... I've got a section "changes since LC"
15:12:51 <jang_scri> with links into the document
15:13:04 <jang_scri> bwm: that sounds very diligent...
15:13:26 <jang_scri> path: these are anchors in a document that I'm not supposed to admit exists (?)
15:13:30 <danbri> * danbri catches up
15:13:50 <jang_scri> bwm: just hand out the id, 
15:14:00 <em> q+
15:14:01 <Zakim> * Zakim sees Em on the speaker queue
15:14:04 <jang_scri> bwm: want to come onto giving people the URL of the updated document
15:14:08 <em> ack em
15:14:09 <Zakim> * Zakim sees no one on the speaker queue
15:14:19 <jang_scri> comment last week was to keep people reviewing the LC document, not a moving target
15:14:44 <jang_scri> em: what bwm said. Get reviews on LC
15:14:52 <jang_scri> summarise all changes in the end
15:15:01 <jang_scri> that's the easiest way to manage all of this
15:15:16 <jang_scri> reviewing interim versions is a recipie for disaster
15:15:37 <jang_scri> bwm: I'm now thinking putting off all changes until the end of the LC period...
15:15:45 <jang_scri> first reason: don't want a moving target
15:15:59 <jang_scri> second reason: don't want to change, then (say) change back due to second comment, then again, etc.
15:16:27 <jang_scri> path: another consideration, the way I'm doing this:
15:16:36 <jang_scri> as far as semantics is concerned wrt webont...
15:16:42 <jang_scri> people on the webont wg are doing semantics there
15:16:51 <jang_scri> and the relationships between the details are critical
15:17:16 <jang_scri> if we have to go through a process in which a wg acts in secret then hands over a fait accomplit, we won't finish for ages
15:17:21 <jang_scri> webont's under intense pressure to finish
15:17:28 <em> q+
15:17:30 <Zakim> * Zakim sees Em on the speaker queue
15:17:31 <jang_scri> pfps and path &co MUST keep talking for this to work
15:17:42 <jang_scri> em: I'm part of the reason why this pressure is there
15:17:53 <jang_scri> I'd absolutely encourage you to continue this dialogue
15:18:23 <jang_scri> we're after the best engineering approach for making a foundation to owl
15:18:31 <jang_scri> don't consider yourself muzzled...
15:18:55 <jang_scri> so, is the big issue, that in order to have that dialogue, must an entire document be discussed?
15:19:05 <jang_scri> path: YES! in fact, three or four documents are involved
15:19:19 <gk> q+ to note that I think capture of comments should be separate from recording changes;  also that the process should allow editors some discretion in how to handle comments provided that the relationship to LC baseline is clear
15:19:21 <Zakim> * Zakim sees Em, Gk on the speaker queue
15:19:38 <jang_scri> bwm: I've got a suggestion to cope with this:
15:19:51 <jang_scri> we collect comments on -comments until the end of LC period
15:19:59 <jang_scri> then and only then we have a document update period
15:20:05 <jang_scri> * jang_scri lags: damn anacron, sorry
15:20:19 <jang_scri> path: webont won't like that
15:20:32 <jang_scri> to be fair to pfps, he's got a larger task because OWL's semantics are more complex
15:20:47 <jang_scri> each time I change a little bit, he's got a lot of work to do
15:21:13 <jang_scri> he's also highly critical and unlikely to take for granted that small changes won't introduce bugs...
15:21:28 <jang_scri> ... which I can't blame him for!
15:21:45 <jang_scri> path: pfps is also under a lot of time pressure
15:21:57 <jang_scri> em: if you can't have dialogue without sharing collections of documents
15:21:59 <jang_scri> ...
15:22:08 <jang_scri> path: we're talking "offline" at the moment
15:22:31 <jang_scri> em: bwm's strategy, em doesn't like, but it seems sensible to take
15:22:57 <jang_scri> bwm: this is the sort of thing that's going to happen, it can't be helped I think...
15:23:13 <jang_scri> ... obviously with multiple versions of documents around, you want to see the latest one
15:23:29 <jang_scri> em: that doesn't preclude having a private dialogue, or we can sort you out telecon bridges, etc.
15:23:57 <jang_scri> path: when I send pfps a list of detailed changes, his (natural) reaction is, wow, that's a lot, let me see them integrated into a document
15:24:11 <jang_scri> frankm: I can see that point of view, it's completely reasonable
15:24:33 <bwm> q?
15:24:34 <Zakim> * Zakim sees Em, Gk on the speaker queue
15:24:36 <gk> q+ doesn't this (multiple changes to semantics) mean that changes being contemplated will in any case need another last call?
15:24:37 <Zakim> * Zakim gk, you typed too many words without commas; I suspect you forgot to start with 'to ...'
15:24:52 <gk> q+ to say doesn't this (multiple changes to semantics) mean that changes being contemplated will in any case need another last call?
15:24:54 <Zakim> * Zakim sees Em, Gk on the speaker queue
15:24:59 <jang_scri> q+
15:25:00 <Zakim> * Zakim sees Em, Gk, Jang on the speaker queue
15:25:14 <jang_scri> frankm: the issue isn't with private changes, it's to do with making public changes, right?
15:25:26 <jang_scri> em: this is like software, we've "shipped a beta"
15:26:01 <jang_scri> bwm: path, you and the rest of the webont people can talk, of course. Exchange documents by email,
15:26:11 <jang_scri> just don't let the URLs leak out onto the web
15:26:58 <jang_scri> em: again, yes, we want ONE LC document up for review
15:27:06 <jang_scri> path: the problem is preventing all leaks
15:28:21 <jang_scri> bwm: in fact, please DON't put these on the web
15:28:26 <jang_scri> send these documents via email
15:28:58 <jang_scri> frankm: what about updating editor's copy in CVS space?
15:29:08 <jang_scri> bwm: this is me saying, "I changed my mind about that"
15:29:24 <jang_scri> bwm: in a nutshell, change, don't put back until end of LC
15:30:09 <jang_scri> bwm: other document editors: you happy to hold off for a couple of weeks?
15:30:12 <jang_scri> everyone sounds happy
15:30:24 <jang_scri> bwm: is everyone going to be in a position to do those updates at the end of that time?
15:31:02 <gk> * gk I think concentrated time is easier than drip-feed
15:31:07 <jang_scri> bwm is talking about the last week of february, basically
15:32:03 <jang_scri> ACTION: bwm put a schedule together for dealing with LC comments
15:32:08 <em> q?
15:32:09 <Zakim> * Zakim sees Em, Gk, Jang on the speaker queue
15:32:11 <bwm> q?
15:32:12 <Zakim> * Zakim sees Em, Gk, Jang on the speaker queue
15:32:12 <em> ack em
15:32:13 <Zakim> * Zakim sees Gk, Jang on the speaker queue
15:32:19 <em> ack Gk
15:32:20 <Zakim> Gk, you wanted to note that I think capture of comments should be separate from recording changes;  also that the process should allow editors some discretion in how to handle
15:32:23 <Zakim> ... comments provided that the relationship to LC baseline is clear and to say doesn't this (multiple changes to semantics) mean that changes being contemplated will in any case
15:32:25 <Zakim> ... need another last call?
15:32:26 <Zakim> * Zakim sees Jang on the speaker queue
15:32:41 <jang_scri> gk: what I heard Pat say about the MT and the semantics work
15:32:49 <jang_scri> sounds to me like we'll need another LC
15:32:52 <jang_scri> is that the case?
15:33:02 <jang_scri> does it help if we recognise that's coming?
15:33:08 <jang_scri> (just to mull over)
15:33:12 <jang_scri> also, wanted to comment:
15:33:21 <jang_scri> with all this talk about process to do things,
15:33:35 <jang_scri> is there sufficient discretion for editors to do the right thing?
15:33:49 <jang_scri> bwm: what do you need discretion for?
15:34:08 <jang_scri> gk: I'm not saying that _I_'m feeling the lack of this, just worried that there's a lot of discussion about process
15:34:44 <jang_scri> frank: bwm, it seems to me that part of the problem is that we're reading too much into the terminology, "last call"
15:34:50 <jang_scri> if this is any kind of a sensible process,
15:35:06 <jang_scri> you stick out LC, you know perfectly well that things are likely to change, people are going to have to read the revisions
15:35:15 <jang_scri> bwm: eric..?
15:35:17 <em> http://www.w3.org/Consortium/Process-20010719/tr#last-call
15:35:24 <jang_scri> em: if people haven't, have a look at that ^^
15:35:32 <jang_scri> it sums up what frank just said.
15:35:49 <jang_scri> in short, as a WG we believe that all of the work we were chartered for have been addressed to the best of our ability
15:35:59 <jang_scri> this is a stake in the ground, it's now open for review by the community
15:36:12 <jang_scri> there will be minor modifications
15:36:28 <jang_scri> that's what we're grappling with
15:37:03 <jang_scri> bwm: danc said, we go to the director, he looks at the changes, and he has to ask himself if any of the reviewers would change their review on the basis of any of those changes
15:37:31 <jang_scri> jjc: I think specifically in the case of semantics, asking pfps to review the document would be reasonable
15:38:04 <jang_scri> path: a lot of changes are essentially changes to the mathematical style
15:38:19 <jang_scri> so they don't affect anyone except those who are checking the proofs
15:38:31 <jang_scri> bwm: I've got pfps saying that these changes are substantive?
15:38:42 <jang_scri> path: for other spec writers they are,
15:38:59 <jang_scri> by and large these changes are to do with, "x isn't defined right" and so on
15:39:09 <jang_scri> it's getting the math right, not the language
15:39:50 <jang_scri> bwm: eric, any more comments?
15:39:59 <jang_scri> what might the director view as substantial?
15:40:03 <jang_scri> em: only He knows.
15:40:22 <gk> * gk the best we can do is what *we* think is reasonable
15:40:24 <jang_scri> before we let him take a crack at it, we have to ask ourselves that too
15:40:38 <jang_scri> bwm: i don't think that a second LC is unavoidable at this stage
15:41:18 <jang_scri> eg, "here are the substantive changes that nonmathematicians will observe; here are mathematically _stylistic_ changes (which are ok by webont); etc."
15:42:01 <jang_scri> jjc: current substantive changes you're aware of: xml literal entailment changes, that it?
15:42:12 <jang_scri> path: a new inference rule that lets you infer that any literal is a real thing
15:42:28 <jang_scri> em: the _intent_ was always there, though?
15:42:31 <jang_scri> path: oh, yeah, sure.
15:44:14 <jang_scri> (that discussion goes around once more, everyone sounds like they more or less agree with this)
15:45:20 <jang_scri> em: it makes a stronger case for future semantic web activities if our current work is at a particular stage
15:45:36 <jang_scri> but people aren't going to give up and go home if we're not quite there...
15:45:54 <jang_scri> ITEM 8
15:45:55 <jang_scri> q-
15:45:57 <Zakim> * Zakim sees no one on the speaker queue
15:46:17 <bwm> q?
15:46:19 <Zakim> * Zakim sees no one on the speaker queue
15:46:22 <jang_scri> massive punning (unsubscribed)
15:46:35 <jjc> (correction my comment earlier was on rdfs:Literal not xml:literal)
15:46:47 <jang_scri> item 8: context uses subject predicate object of a triple
15:46:51 <jang_scri> ... are uris
15:47:05 <jang_scri> the subject of a triple is a uriref, not a resource...
15:47:11 <jang_scri> ... or a bnode...
15:47:27 <jang_scri> the point here is a terminological one
15:48:02 <jang_scri> subject predicate object of triples are uriref
15:48:13 <jang_scri> the subject predicate object of statements are resources
15:48:27 <jang_scri> ie, one syntactic (I think)
15:49:08 <jang_scri> frank: often the connection between a triple and a statement isn't explicitly specified
15:49:18 <jang_scri> it's not just reification vocab
15:50:12 <jang_scri> jjc: is this in response to LC comment?
15:50:21 <jang_scri> bwm: it's in response to a comment I made on primer
15:50:38 <jang_scri> frank has pointed out that we ought to check all the document to ensure consistent usage
15:50:46 <jang_scri> I'd like to see coming out of this:
15:51:09 <jang_scri> action on each editor to check their usage of subject, predicate, object, triple, etc against proposed use here
15:51:51 <jang_scri> ^^^ this isn't an action yet(!)
15:52:11 <jang_scri> [scribed missed the latter half of the proposed action, however]
15:52:16 <jang_scri> jjc: this sounds like a lot of work
15:52:32 <jang_scri> ... I'm not sure if the benefit justifies it?
15:53:00 <jang_scri> frankm: the basic comment is that primer starts talking about english statements, goes on to talk about rdf statements,
15:53:11 <jang_scri> and talks about rdf statements being represented by triples or a graph
15:53:31 <jang_scri> path: it seems to me harmless to use language like this in the _primer_
15:54:04 <jang_scri> danbri: I have a question about "statement"
15:54:25 <jang_scri> is a statement the kind of thing that's either true or false, or is it the kind of thing that's true or false under some interpretation?
15:55:03 <gk> I(triple)~=statement, I(subject-of-triple)=subject-of-statement, etc ... is this it?
15:55:27 <jang_scri> path: even most logic texts are careless abobut statement/stating distinction
15:55:56 <em> q+ to ask for 2 min at end of the meeting to again remind wg members to (if they havent already) contact their AC rep and make sure they respond to the rdf core charter extension request
15:55:57 <Zakim> * Zakim sees Em on the speaker queue
15:56:27 <jang_scri> action: jjc ensure concepts is accurate on this
15:56:32 <jang_scri> we're more relaxed about primer
15:57:36 <jang_scri> action: jang check syntax, report back to the list if there's a big issue about "statement" terminology there
15:57:47 <jang_scri> action: danbri: similar for schema
15:58:17 <danbri> ACTION: danbri to reference the appropriate definition of 'triple' 
15:58:23 <danbri> er, in rdfs
15:58:41 <jang_scri> em: if wg members haven't already contacted their AC reps
15:58:43 <jang_scri> DO SO TODAY
15:58:58 <jang_scri> make sure they respond today!
16:01:18 <Zakim> -PatH
16:01:22 <Zakim> -SteveP
16:01:24 <Zakim> -FrankM
16:01:42 <Zakim> -Mike_Dean
16:01:52 <Zakim> -GrahamKlyne
16:02:03 <danbri> jang, http://lists.w3.org/Archives/Member/w3c-ac-members/2003JanMar/0003.html
16:02:34 <Zakim> -HP
16:03:26 <Zakim> -Jang
16:03:38 <Zakim> -DanBri
16:03:45 <Zakim> -EMiller
16:03:46 <Zakim> SW_RDFCore()10:00AM has ended
16:06:10 <jjc> jjc has left #rdfcore
16:08:21 <bwm> bwm has quit
16:08:56 <danbri> danbri is now known as Danrol
16:13:06 <em> em has left #rdfcore
16:19:55 <gk> gk has quit
16:28:47 <jang_scri> jang_scri has quit
16:46:59 <Danrol> Danrol is now known as danbri