Semantic Web Interest Group IRC Chat Logs for 2003-07-16

00:11:17 <ChanServ> [#rdfig] This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome

05:52:25 <dmwaters> {global notice} Good morning everyone, we need to do some rehubbing in the us. I appologize for the inconvenience, any further messages will be given in wallops, 2500 users will be effected. Thank you for using freenode!

05:56:51 <ChanServ> [#rdfig] This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome

06:33:24 <anselm> anselm is now known as anselm_zzz

06:33:41 <nym> nym is now known as alsoalseep

06:34:05 <alsoalseep> alsoalseep is now known as sleepinym

07:43:51 <qmacro> morning danbri

11:09:25 <pxh> pxh is now known as pxhAway

13:11:49 <danbri> ndw, do you plan to add support for Adobe's XMP to your image metadata thing?

13:12:44 <libby> many ndw used the jigsaw code for exif? maybe can also use the xmp code?

13:13:07 <danbri> ah maybe. ndw, Bert was just trying to load up some data.

13:14:43 <ndw> I don't have a good source of XMP data, nor did I find the adobe spec illuminating on the subject.

13:15:04 <ndw> As libby suggests, I am using a (modified) version of the jigsaw code from rdfpic

13:15:54 <ndw> Nudged gently in the direction of an explanation of XMP and provided with a couple of sample images, I'll happily add support.

13:15:58 <danbri> thanks for the info!

13:16:14 <ndw> How's France? :-)

13:16:16 <danbri> support would be great... to show interop w/ rdfpic3 (looming)

13:16:26 <danbri> france is good, but i've got sick somehow :(

13:16:31 * danbri scribing too, betterg go

13:16:37 <ndw> Yuck. Sick on the road sucks.

13:17:05 <danbri> yeah... not so far from home tho

13:17:05 * ndw goes off to see if he can find the Adobe docs again. I expect to have a couple of hours today...

13:17:39 <libby> apparantly the xmp stuff is in rdfpic 3, not quite out yet

13:18:19 <ndw> Yes, I saw some code going in that direction in the classes I was hacking at.

13:18:50 <ndw> Hmmm, is Bert maintaining that code now? I sent some patches back (I'd like to stop relying on patched org.w3c.* classes). I wonder if my patches were accepted.

13:19:03 <libby> hm, lemmee ask him

13:22:55 <ndw> danbri: fyi: the images on http://www.w3.org/2001/sw/Europe/200206/imagemeta/extract/extract are 404

13:23:27 <danbri> i have just action'd myself on that.

13:24:03 <libby> ndw, bert says he will definitely get in touch with you

13:24:13 <ndw> Ok. Thanks!

13:24:14 <libby> - he hasn;t made the updates (yet)

13:35:46 <jordan> what's a good place to find not-too-large DAML data sets (by data set I mean an ontology and an ontology instance) -- other than the daml.org page?

13:47:26 <aharth> jordan: http://pulp.dyndns.org:8080/data.rdf some rss news, around 800k

13:49:08 <aharth> or just crawl RDF that's online

13:51:24 <danbri> anything in the wiki at http://esw.w3.org/topic/FrontPage somewhere?

13:57:49 * ndw wanders off for a day of xsl/query meetings

13:59:32 <bijan_> bijan_ is now known as bijan

14:06:25 <ericP> hi martin

14:06:33 <MJDuerst> hi ericP

14:06:54 <ericP> do you have a pointer to my scenarios?

14:06:57 <ericP> i'll look...

14:07:03 <ericP> that will frame our discussion

14:07:18 <MJDuerst> ok, so following the discussion we had yesterday,

14:07:55 * ericP looks for pointer...

14:08:17 <MJDuerst> it would be relatively easy for an application or a query language to treat <property>some text</property>

14:08:48 <ericP>http://www.w3.org/2003/07/15-XMLLiteral-CharData

14:08:48 <dc_rdfig> A: http://www.w3.org/2003/07/15-XMLLiteral-CharData from ericP

14:08:49 <MJDuerst> and <property parseType="Literal">some text</property> as the same, if it were just for that.

14:09:26 <MJDuerst> But the problem is that we may have <property xml:lang='en'>some text</property>

14:09:43 <MJDuerst> (a plain literal flagged as English)

14:09:50 <ericP> A:| Equivilence of Simple XMLLiterals to Simple RDF Literals

14:09:50 <dc_rdfig> Titled item A.

14:09:59 <MJDuerst> Due to the most recent changes, this does not correspond to

14:10:14 <MJDuerst> <property parseType="Literal" xml:lang='en'>some text</property>

14:10:36 <MJDuerst> (which is a literal of type xml Literal without any language info)

14:11:06 <MJDuerst> a query language should be able to deal with the textual equivalence as such

14:11:24 <MJDuerst> But the problem is that in order to say that we are dealing with 'en' in the second case,

14:11:37 <ericP> if we assume that the c14n of the XMLLiteral returned to including the xml:lang, do you see and addtional prob?

14:11:39 <MJDuerst> the RDF core WG has suggested that one use some markup like

14:12:10 <MJDuerst> <property parseType="Literal"><span xml:lang='en'>some text</span></property>

14:12:27 <MJDuerst> Now what the query language or application would have to do is to equate

14:12:38 <ericP> any XML Core folks around?

14:12:46 <MJDuerst> a) <property xml:lang='en'>some text</property>

14:13:03 <MJDuerst> and b) <property parseType="Literal"><span xml:lang='en'>some text</span></property>

14:13:30 <MJDuerst> with <span> being something potentially different every time, and no way to guess

14:13:37 <ericP> untenable since there are infinite number of ways to spell <span>

14:13:47 <ericP> (unless one gets recommended)

14:14:40 <ericP> does c14n provide a way to store the xml:lang?

14:14:51 <ericP> exclusive c14n, that is

14:15:34 <MJDuerst> no, exclusive canonicalization doesn't do that, and tells users that they should be careful

14:16:26 <MJDuerst> problem is that what gets canonicalized is not an element, but a mixed content fragment.

14:16:39 <MJDuerst> So there is no good place to put xml:lang in the fragment itself.

14:16:57 <MJDuerst> RDF manages to put xml:lang as a separate flag on plain literals.

14:17:34 <MJDuerst> They also did that for xml literals, up to last call, but then changed that to make their model theory simpler.

14:18:36 <MJDuerst> ericP, re http://www.w3.org/2003/07/15-XMLLiteral-CharData.html, use case 1 should clearly be NO.

14:18:39 <ericP> i wonder how much work is involved in getting the model theory up to snuff with it re-introduced

14:19:03 <MJDuerst> There is a very clear difference between markup and text that looks like markup. XML makes this distinction,

14:19:04 <ericP> and you say yes to two

14:19:12 <ericP> i should make it a votable form

14:19:25 <MJDuerst> and it would be a bad idea for RDF to remove that

14:20:00 <MJDuerst> Yes, I would clearly like 2 to be a YES.

14:20:46 <ericP> i vote 1:yes, 2:no

14:21:00 <ericP> we both cite consistency for our arguement

14:21:11 <MJDuerst> (a sequence of characters / plain text / whatever you call it) should not become something different just because its context is different

14:21:53 <ericP> (if you told RDF it's an XML subtree, it's an XML subtree)

14:22:17 <ericP> i'll include both of these arguments in the doc

14:22:34 <MJDuerst> For more details on 1), see http://www.tbray.org/ongoing/When/200x/2003/04/22/RSS-Problems

14:23:18 <sleepinym> sleepinym is now known as nym

14:23:23 <ericP> what's the forum for weighing in on this discussion?

14:48:48 <libby> erk

14:48:50 <libby>http://data.digicult.info/download/ti3_low.pdf

14:48:52 <dc_rdfig> B: http://data.digicult.info/download/ti3_low.pdf from libby

14:48:56 <bijan>http://lists.w3.org/Archives/Public/www-tag/2003Jul/0022.html

14:48:56 <dc_rdfig> C: http://lists.w3.org/Archives/Public/www-tag/2003Jul/0022.html from bijan

14:49:02 <bijan> C:|

14:49:02 <dc_rdfig> Titled item C.

14:49:19 <bijan> C:|"Social meaning" raises its ugly head, once more

14:49:19 <dc_rdfig> Titled item C.

14:49:27 <bijan> C: As a TAG issue.

14:49:27 <dc_rdfig> Added comment C1.

14:49:54 <libby> B:|Digicult: towards a semantic web for heritage resources

14:49:55 <dc_rdfig> Titled item B.

14:50:14 <libby> bijan: Paul Shabajee pointed me at it. interesting

14:50:17 <libby> sorry

14:50:25 <libby> B:Paul Shabajee pointed me at it. interesting

14:50:25 <dc_rdfig> Added comment B1.

14:50:43 <bijan> C: Said issue formulated and proposed by Tim Berners-Lee, without consulting any of the dissenters toward his view, and without including their views as anything other than, at best, "misunderstandings".

14:50:44 <dc_rdfig> Added comment C2.

14:51:41 <libby> B:caught me eye: paraphrasing, that since new web pages would mostly be database-driven, the SW would not work with them.

14:51:41 <dc_rdfig> Added comment B2.

14:51:53 <libby> B:|Digicult: towards a semantic web for heritage resources (pdf)

14:51:53 <dc_rdfig> Titled item B.

14:51:57 <bijan> C:The existence and substance of these views was made very clear to Tim on at least three occasions, once on this #channel (check the logs), once at the Tech Plenary, and once at the Social Meaning F2F at WWW2003.

14:51:57 <dc_rdfig> Added comment C3.

14:53:21 <bijan> C: At the tech plenary, many people, I believe, at least me, agreed to be part of some group, ad hoc or otherwise, that would try to say something sensible and useful about this issue. I believe that the consensus reached there was in part determined by the understanding that there would be such a group.

14:53:21 <DanC-AIM> .time

14:53:21 <dc_rdfig> Added comment C4.

14:53:21 <datum> Wed, 16 Jul 2003 14:53:21 GMT

14:53:33 <libby> B:this problem woudl be solved by OAI though

14:53:34 <dc_rdfig> Added comment B3.

14:53:49 <libby> B:(I've just been skimming, may have misrepresented them)

14:53:50 <dc_rdfig> Added comment B4.

14:54:52 <nym> hi lasse

14:54:59 <nym> er hi libby

14:55:04 <libby> heh

14:55:06 <bijan> C: The Semantic Web Coordination Group was tasked with figuring out the best form of this group and organizing it. There was some understanding, in me, at least, that the people who volunteered or indicated interest at the plenary would be given the opportunity to participate fully.

14:55:06 <dc_rdfig> Added comment C5.

14:55:09 <libby> hi nym

14:56:23 <DanC-AIM> Without consulting, bijan?

14:57:20 <bijan> He didn't consult, afaik, with me, or peter, or any similarly viewed person about his formulation of this issue

14:58:15 <DanC-AIM> He can't send mail without your permission?

14:58:59 <DanC-AIM> The issue hasn't even been accepted, let alone resolved.

14:59:26 <simonstl> has the tone of the conversation been set?

14:59:32 <bijan> Metacomment: I don't believe you and I can discuss this usefully in irc. This could be solely due to me.

15:00:41 <bijan> It's not about permission, anyway, it's about consensus building and fairness and consistency.

15:03:03 <DanC-AIM> If you don't want to talk about it here, please undo your chump comment

15:03:26 <bijan> No.

15:03:48 <DanC-AIM> Ok, forget permission. What's unfair about what he's done?

15:04:09 <bijan> Why should i have to discuss what I chump?

15:04:18 <bijan> When was that rule, even of ettiquete introduced?

15:05:05 <DanC-AIM> I gather it's law in europe. 1/2 :-)

15:05:09 <libby> libby is now known as danbri

15:05:11 <DanC-AIM> Right of reply

15:05:26 <danbri> logger_2, bookmark?

15:05:26 <danbri> See http://ilrt.org/discovery/chatlogs/rdfig/2003-07-16#T15-05-26

15:05:40 <bijan> Who said I had to exercise *my* right to reply in order foryou to exercise yours?

15:05:54 <DanC-AIM> I do

15:06:16 * Morbus scratches head.

15:06:26 <danbri> C:Bijan's commentary on this seems to be contested by danc, see [http://ilrt.org/discovery/chatlogs/rdfig/2003-07-16#T15-05-26|irc logs] for some more detail.

15:06:26 <dc_rdfig> Added comment C6.

15:06:34 <danbri> danbri is now known as libby

15:07:39 <DanC-AIM> B1:""

15:07:39 <dc_rdfig> Deleted comment B1.

15:08:51 <bijan> Well, that's interesting. It's acceptible to delete other people's chumpings because you disagree with them?

15:08:53 <bijan> B:

15:08:53 <dc_rdfig>http://data.digicult.info/download/ti3_low.pdf

15:08:54 <dc_rdfig> Digicult: towards a semantic web for heritage resources (pdf)

15:08:55 <dc_rdfig> (1:libby) caught me eye: paraphrasing, that since new web pages would mostly be database-driven, the SW would not work with them.

15:08:56 <dc_rdfig> (2:libby) this problem woudl be solved by OAI though

15:08:57 <dc_rdfig> (3:libby) (I've just been skimming, may have misrepresented them)

15:09:25 <bijan> Hmm. I don't see what you disagreed with, even:

15:09:27 <bijan> libbyB:Paul Shabajee pointed me at it. interesting

15:09:48 <bijan> dc_rdfigAdded comment B1.

15:11:59 <DanC-AIM> Deleting comments is part of the protocol. If you won't discuss your comment, I don't see why it should be there. I'm off to the dentist now, so I won't be able to prevent you from putting it back if you choose.

15:12:16 <simonstl> I think you deleted the wrong entry

15:12:20 <bijan> You deleted Libby's comment

15:12:22 <bijan> As I showed

15:12:36 <libby> delete away

15:12:37 <bijan> And I said I wouldn't discuss it with you, in irc

15:12:44 <libby> grumpies

15:12:47 <bijan> Given your tactics thus far, I think that's reasonable

15:13:42 <bijan> After all, I never said, nor do I need to, that you needed my permission to post to my chump entry

15:13:46 <bijan> As danbri did for you

15:16:51 <bijan> Hmm. I was leaving it implicit, but clearly you owe libby an apology

15:17:58 <bijan> C:

15:17:59 <dc_rdfig>http://lists.w3.org/Archives/Public/www-tag/2003Jul/0022.html

15:18:00 <dc_rdfig> "Social meaning" raises its ugly head, once more

15:18:00 <dc_rdfig> (1:bijan) As a TAG issue.

15:18:01 <dc_rdfig> (2:bijan) Said issue formulated and proposed by Tim Berners-Lee, without consulting any of the dissenters toward his view, and without including their views as anything other than, at best, "misunderstandings".

15:18:03 <dc_rdfig> (3:bijan) The existence and substance of these views was made very clear to Tim on at least three occasions, once on this #channel (check the logs), once at the Tech Plenary, and once at the Social Meaning F2F at WWW2003.

15:18:04 <dc_rdfig> (4:bijan) At the tech plenary, many people, I believe, at least me, agreed to be part of some group, ad hoc or otherwise, that would try to say something sensible and useful about this issue. I believe that the consensus reached there was in part determined by the understanding that there would be such a group.

15:18:05 <dc_rdfig> (5:bijan) The Semantic Web Coordination Group was tasked with figuring out the best form of this group and organizing it. There was some understanding, in me, at least, that the people who volunteered or indicated interest at the plenary would be given the opportunity to participate fully.

15:18:09 <dc_rdfig> (6:danbri) Bijan's commentary on this seems to be contested by danc, see [http://ilrt.org/discovery/chatlogs/rdfig/2003-07-16#T15-05-26|irc logs] for some more detail.

15:18:27 <bijan> heh. If he'd delted C1, he would have deleted something uncontestably true :)

15:21:57 <bijan> C:I don't consider moving it to the TAG, raised and championed by people advocating a tendentious view, without even a cc or other notice, to be giving the opportunity to participate fully.

15:21:58 <dc_rdfig> Added comment C7.

15:23:49 <kota_> kota_ is now known as xkota

15:24:02 <bijan> C:Some meta-commentary: I'm aware of the provocative content of my comments here. I've tried to make them reasonably factual, i.e., reporting on *my sense* of what was expected and how what's happing conflicts with those expectations.

15:24:02 <dc_rdfig> Added comment C8.

15:26:32 <bijan> C8:Some meta-commentary: I'm aware of the provocative content of my comments here. I've tried to make them reasonably factual, i.e., reporting on *my sense* of what was expected and how what's happening conflicts with those expectations.

15:26:32 <dc_rdfig> Replaced comment C8.

15:35:25 <bijan> C:More meta-commentary: Given that I was an active, ongoing player in these discussions, my reactions seem at least somewhat noteworthy. Hence, I so note them. I'll probably do so in other fora, as well.

15:35:25 <dc_rdfig> Added comment C9.

15:38:36 <sandro> C: Bijan, I can't tell how much you and Jim Hendler act in concert here or not. He tells me you and he consult on these issues constantly. And he was a party to the SWCG decision to move this issue to the TAG, wasn't he? (It's a matter of public record that he's on the CG, but the records of the meeting in question are not public.)

15:38:36 <dc_rdfig> Added comment C10.

15:39:39 <bijan> C: There's two issues, I take it: 1) moving the issue to the TAG, and 2) the manner in which it was done.

15:39:39 <dc_rdfig> Added comment C11.

15:40:19 <bijan> C: Well, there's a third, about dealing with the old expectation.

15:40:19 <dc_rdfig> Added comment C12.

15:41:32 <sandro> C: And a forth -- the fact that the decision in this area was relegated to a body which deliberates in private.

15:41:34 <dc_rdfig> Added comment C13.

15:41:35 <bijan> C: I would have expected that *before* resolving the issue *this way*, the SWCG would make some effort to canvass the people who volunteered and reached consensus at the Tech Plenary session. If they have been so consulted, i don't have strong evidence thereof.

15:41:37 <dc_rdfig> Added comment C14.

15:42:21 <sandro> C: in what sense is this a resolution? This is just a bit of routing.

15:42:22 <dc_rdfig> Added comment C15.

15:42:24 <xkota> xkota is now known as kota

15:42:27 <DanC-AIM> Ahh... Clean teeth

15:43:24 <DanC-AIM> "resolving the issue"??? I don't exprect the issue to be resolved until, maybe next spring, if we're lucky.

15:43:30 <bijan> C:The resolution of what body would handle the social meaning issue.

15:43:31 <dc_rdfig> Added comment C16.

15:44:38 <DanC-AIM> I hope you'll work with a few others to figure out what the tag should do with this issue, bijan

15:45:17 <sandro> C: I don't expect the TAG to figure this out by itself. It should and I hope will ask for help.

15:45:18 <dc_rdfig> Added comment C17.

15:45:49 <DanC-AIM> I also hope we have more running code and experiments to inform the discussion. Experience with owl:imports, for example

15:49:38 <sandro> The Joint Committee yesterday discussed a parallel to owl:imports in the rules language. (sigh) Got into another fairly-brief round of Social Meaning with Ian and Pat and me and Mike.

15:50:01 <DanC-AIM> "made very clear to tim"??? Wow.. You think tim is very clear on your position? I don't understand anybody's position on this issue very clearly... Not even my own.

15:50:16 <_joshua> i feel all tingly

15:50:21 <sandro> lol

15:50:30 <sandro> why, _j ?

15:51:05 <_joshua> Just a non sequitur thrown in to defuse the situation

15:51:06 <_joshua> Proceed

15:52:58 <DanC-AIM> This gizmo works ok, but it's no match for a desktop... I thought social meaning was B:, but I see that it's C:

15:53:12 * sandro has vague images of a Pennsic battlefield when you say that. http://www.ogrecave.com/features/pennsic/2_pennsic.battle.jpg

15:54:46 <DanC-AIM> C:tim's request to the tag is part of setting up "some group, ad hoc or otherwise, that would try to say something sensible and useful on the issue"

15:54:46 <dc_rdfig> Added comment C20.

15:54:50 <jimh-scribe> hmm, JH wanted to refer to the CG minutes from July 07 with respect to soc mng discussion - is surprised to find them ACLed Team - can someone chacl them

15:54:59 * bijan points out the existence of these views was, at least, made very clear :)

15:55:49 <jimh-scribe> +1 to what DanC said - I think the CG spirit was that TAG would be a starting place for producing a document that would be more widely discussed

15:56:20 <DanC-AIM> Well, is there something about tim's request that says there are no other views? Is not requesting a tag issue a claim that there are various issues to reconcile?

15:56:47 <bijan> C:"The proposal is that a draft finding be written which pulls this together, with elaborations pointing into the various specs. Members of the SWCG have volunteered and some members of members of the SWCG have been volunteered to read early versions." (last paragraph of the message)

15:56:48 <dc_rdfig> Added comment C21.

15:57:25 <bijan> C: I read "pulls this" as, roughly, take the text tim sent, fill in the -etc. etc., add some links, etc.

15:57:26 <dc_rdfig> Added comment C22.

15:57:41 <DanC-AIM> More that the tag would be a place to finish a discussion started in a task force/writing group

15:57:44 <bijan> C:Also, I don't know if I'm a member of a member of SWCG. I don't think I am.

15:57:44 <dc_rdfig> Added comment C23.

15:57:46 <jimh-scribe> whoops - July 07 cg was cancelled - never mind

15:58:14 <jimh-scribe> I think you are Bijan - MIND is represented on the CG by its AC member

15:58:27 <bijan> C: And, finally, those volunteered might like to have known what they were being volunteered for.

15:58:28 <dc_rdfig> Added comment C24.

15:58:38 <DanC-AIM> Jim, you're not quoting from cg records without permission, are you?

15:59:23 <DanC-AIM> You're not a cg member, bijan

15:59:29 <jimh-scribe> no - I'm consulting CG records so I can recall what we resolved - these are just my own opinions

15:59:42 <bijan> Nor a member of a member?

15:59:45 <DanC-AIM> No, jim, mind is not on the cg.

15:59:50 <DanC-AIM> The webont chair is

15:59:59 <bijan> That's what I thought.

16:00:03 <jimh-scribe> well, then what does TimBL mean by "A member of a member"?

16:00:11 <bijan> So some of jim's parts have been volunteered to read it? :)

16:00:13 <DanC-AIM> ?

16:00:27 <sandro> I think it's another typo of Tim's. :-)

16:00:48 <sandro> (that is, he didn't really only mean members of the CG.)

16:00:58 <jimh-scribe> I don't think so -- look at what he said -- Members of the SWCG have volunteered and some members of members of the SWCG have been volunteered to read early versions

16:01:19 <bijan> So, I stand by my reading that almost all of the people who reached consensus in Boston, or who participated in the F2F in hungary are not (yet) cut in.

16:01:44 <kendallclark> i read that as some swcg members have volunteered, others have been compelled, to read early versions -- fwiw

16:01:46 <jimh-scribe> I'll beleive he didn't mean this literally, but that he was referring to some of the people Bijan is asking about

16:01:51 <DanC-AIM> Re lack of cc... The plan was that once the tag had accepted the issue, we'd notify www-rdf-interest and the bof participants in particular

16:03:05 <jimh-scribe> DanC - it would be nice if the CG would decide that our specific resolution on this could be shared with the public - not sure how we achieve that (next CG meeting)?

16:03:10 <bijan> Hmm. well, evidently, that wasn't tactically the *best of all possible* plans :)

16:03:29 <DanC-AIM> Ok, stipulated

16:03:57 <DanC-AIM> But can you give us a break? We're doing out level best to be fair.

16:04:01 <DanC-AIM> Our

16:04:12 <sandro> +1 jimh. I don't remember what the resolution was, but I think the important bit would be saying something about WHY the CG decided to do this.

16:04:56 <sandro> (which I thought you would have told Bijan, at least....)

16:05:47 <DanC-AIM> Would that the CG had time to get clear written justifications of its decisions. Let's get real, shall we? That would take a lot more time, and would rarely be worthwhile.

16:05:51 <jimh-scribe> Bijan and I have a high bandwidth back channel going on as we speak -- as a member I did also point him tothe resolution

16:06:39 <DanC-AIM> Written as in: written for public consumption

16:06:45 <jimh-scribe> I think it fair to point out, hwoever, that he's not objecting primarily to process (although this has gone there), his original complaint was how timbl had writtem the issue

16:07:00 <DanC-AIM> Heading home now...

16:07:25 * bijan pondering the "give you a break" issue

16:07:33 <jimh-scribe> s/writtem/written - i.e it had Tim's spin on it, as opposed to, say, the more balanced (and too long) document that sandro had produced (members can see it in the CG archive)

16:08:11 <jimh-scribe> +1 to Dan's attempting to be fair - adding "within the rules of W3C"

16:08:13 <DanC-AIM> What's wrong with how tim wrote the issue? How could he be expected to incorporate other views before the discussion begins?

16:08:39 <jimh-scribe> go read Bijan's comments C 1- 8ish

16:08:46 <jimh-scribe> :->

16:08:56 <DanC-AIM> I did

16:08:59 <jimh-scribe> jimh-scribe is now known as Jimh_

16:09:11 * Jimh_ realises he is not scribing :->

16:09:38 * Jimh_ goes back to figuring out how to get OWL LC comments into better OWL/RDF

16:09:54 * sandro goes back to implementing an OWL reasoner

16:09:56 <bijan> """3) There may be some need to clarify frequent misunderstandings by making some things clear.

16:10:06 <DanC-AIM> Heading back to the office

16:10:10 <bijan> ...- that use of a URI in RDF implies a commitment to its ontology """

16:10:28 <Morbus> can't we all just ... ..... get alooOng?!

16:10:31 <Jimh_> Bijan - that could be you making misunderstandings clear to Tim as much as it could be him making them clear to you...

16:10:46 <Jimh_> Morbus - we haven't had a good fight in weeks - don't spoil it :->

16:10:52 <Morbus> heh, heh.

16:11:07 <bijan> Given the text of the message, especially the last passage, I don't see how, consistent with that, one could end up with text that didn't endorse that claim.

16:11:49 <bijan> I don't agree with that claim, on several of the readings I give to "use" "in rdfs" and "commintmetn" not to mention "its ontology"

16:11:57 <bijan> er "in RDF"

16:12:15 <Jimh_> Bijan - be fair - the proposal was not for the TAG to accept Tim's text, but to start from it - that is far from saying anyone other than Tim has endorsed this view

16:12:36 <Jimh_> this was the raising of the issue, not the closing of it

16:12:40 <bijan> I think I am being far.

16:12:50 <bijan> "The proposal is that a draft finding be written which pulls this together..."

16:12:56 <bijan> First sentence of last paragraph.

16:13:26 <bijan> Item 2 is framed in a questiony way

16:13:35 <bijan> but 1 and 3 seem substantive

16:14:13 <bijan> that I object to 1 is certainly well know to tim :)

16:15:22 <Jimh_> yeah, so he knows you'll be free to send mail to the public TAG mailing list about it

16:16:38 <sandro> Then maybe you should explain that to the TAG, Bijan. I think it would be more productive to focus on showing test cases about where you think Tim is wrong, though, than on process issues. Arguably, you should wait until the TAG accepts this and gives it an issue number, so you can tag your comment.

16:17:17 <bijan> Well, look first, nothing I do here precludes, or even vaguely suggests that I would preclude, sending email to the tag

16:18:08 <bijan> I felt like discussing it here first, and chumping about it

16:18:22 <bijan> I'm also carrying out email discussions with other people.

16:19:27 <sandro> true

16:19:30 <bijan> Second, it's more than formal process issues, per se, it's about the spirit of consensus and good acting and the *perception* of good acting within the process.

16:19:40 <sandro> (I can see you composing those messages in my xTIA window)

16:20:03 <bijan> Also, from what you just said, it entails that I should shut up until the TAG accepts the issue, as framed.

16:20:07 <bijan> Boo to that.

16:20:45 <anselm_zzz> anselm_zzz is now known as anselm

16:21:02 <bijan> I'll also note that you and I spent a *lot* of time discussing this since Hungary.

16:21:03 <sandro> In the sense that you "should" stay inside the station until the train arrives, rather that waiting in the rain, yeah.

16:21:27 <bijan> And that I worked on framing the issue in a position neutral way with you

16:21:34 <bijan> I'll walk, thanks.

16:21:38 <bijan> Which is what I've done thus far

16:23:58 <sandro> Have you read my member-confidential proposal to the SWCG? Does it hit your betrayal button that way? I tried to be fair and neutral, but in all honesty it's really hard to do that (at least for me). Anyway, I'm trying to get myself (and hopefully you) focused on running code, etc, now. (and also trying to read some philosophy and educate myself, too.)

16:24:37 <bijan> I've not read anything you've not directly pointed me to

16:24:54 <bijan> But I'd be surprised if there were substantial overlap between anythign you wrote or we discussed and tim's note

16:24:56 <bijan> Which was my point

16:25:14 <bijan> it's not like work hadn't been done, and presented to tim, on position neutral ways of doing it

16:28:45 <sandro> Are TAG issues supposed to be raise in a neutral way? I don't know..... It does bother me when the issues page links to only some sides of the issue, but I don't much care about the original message.

16:29:01 <sandro> ??? dc_rdfig has quit (Excess Flood)

16:30:00 <bijan> Sandro: please read the final paragraph again

16:30:01 <jordan> sandro: I think it's my fault, I asked dc_rdfig what C: was in /msg.

16:30:15 <bijan> And note the difference in item 2 and 3 of the message

16:32:41 <bijan> And, generically, I have no idea if issues are *supposed* to be raised in such and such a way. Does that matter to the messages communicated by raising *this* issue *this* way?

16:33:47 <bijan> Thanks to mattb for restoring the chump

16:34:18 <mattb> its server is having issues today, apologies

16:35:11 <mattb> oops, does look like a bug

16:35:14 * mattb will look into it

17:18:12 <DanC>http://www.w3.org/People/maxf/textorizer/

17:18:12 <dc_rdfig> D: http://www.w3.org/People/maxf/textorizer/ from DanC

17:18:23 <DanC> D:|SVG Textorizer Tool

17:18:23 <dc_rdfig> Titled item D.

17:18:30 <DanC> D:Max's latest hack. way cool.

17:18:30 <dc_rdfig> Added comment D1.

17:28:56 <Wack> wow, nice

17:29:30 <sandro> referring to D?

17:29:36 <Wack> yeah

17:30:06 <ericP> i like the "how does it work?" not-link

17:31:27 <Wack> have to make mental note about it when I have to design a flyer for a party, can imagine some very nice layouts/designs with it, especially considering the fact it uses a given list of words (great eye catcher)

17:31:48 <Wack> kinda like those images created by large amounts of movie stills or photographs

17:32:15 <sandro> I wonder if you can tune how close a fit it finds. I wouldn't recognize danbri from that.

17:32:18 <Wack> photograp of darth vader created by a lot of smaller star wars frames etc

17:32:46 <Wack> I guess it heavily depends on the input impage

17:32:48 <Wack> er image

17:33:17 <Wack> if you look at the png it used of danbri, it doesnt have much contrast in the face (nose, mouth etc)

17:34:01 <ericP> Wack, so "impage" was not a typo for "mpeg"?

17:34:10 <ericP> that would probably require some heavy CPU...

17:35:06 <Wack> well, considering the ascii mpeg players, it could be doable

17:35:19 <DanC>http://www.w3.org/2002/ws/chor/

17:35:19 <dc_rdfig> E: http://www.w3.org/2002/ws/chor/ from DanC

17:35:22 <Wack> the problem with svg is not creating the abstract image but rendering the glyphs

17:35:30 <DanC> E:|Web Services Choreography Working Group

17:35:30 <dc_rdfig> Titled item E.

17:36:07 <Wack> something like aqua or that new gui engine of the next ms windows might be able to speed things up enough (especially if it prerenders the words as bitmaps)

17:36:09 <DanC> E:"45 members representing 29 organisations"

17:36:10 <dc_rdfig> Added comment E1.

17:38:27 <DanC> E:they seem to be discussion requirements... can somebody help me find the motivating use cases?

17:38:27 <dc_rdfig> Added comment E2.

17:38:40 <DanC> E2:they seem to be discussing requirements... can somebody help me find the motivating use cases?

17:38:40 <dc_rdfig> Replaced comment E2.

17:45:27 <DanC> E:hmm... no WDs yet... 1st meeting was in March... they're stretching the [http://www.w3.org/2003/06/Process-20030618/groups.html#three-month-rule|3 month heartbeat requirement]

17:45:27 <dc_rdfig> Added comment E3.

17:47:26 <DanC> E:[http://www.w3.org/2003/01/wscwg-charter#schedule|WG schedule] calls for 1st WD of their spec this month. I wonder what's a more realistic expectation at this point.

17:47:26 <dc_rdfig> Added comment E4.

17:50:06 <DanC> E:hmm... [http://lists.w3.org/Archives/Public/public-ws-chor/2003Jul/0024.html|semantics task-force kick-off msg]

17:50:07 <dc_rdfig> Added comment E5.

17:50:11 * sandro wonders at the penalty provided by law for violating the three-month rule.

17:50:32 * sandro notes last TR from RDF Core was 23January.

17:51:44 <DanC> penalty isn't specified. One can draw a conclusion that there is no WS-choreography WG, though.

17:52:01 * sandro notes last public version of most OWL docs (except TEST) is 2+ weeks over 3 months.

17:52:19 <DanC> but the TEST publication counts

17:53:10 <DanC> RDF Core really, really needs to publish something soon.

17:53:38 <DanC> the fact that they haven't published is a reflection of a lot of angst, I think.

17:53:46 <bijan> How does one reference an XML Schema document, with datatypes, in an RDF or OWL document?

17:53:47 <sandro> by the letter of the law yes, but not the stated intent, at least: "People cannot be expected to read several months of a group's mailing list archive to understand where the group stands". Of course there public editors drafts, which makes it tricky.

17:54:12 * bijan notes problems with dereferencing the schema namespace is raising this issue

17:54:49 <DanC> the only thing the OWL spec says is how to use the dozen or so built-in datatypes.

17:55:01 <Jimh_> DanC - I'm on the Chor WG -- things got derailed for a while due to the political stuff

17:55:02 <DanC> for user-defined datatypes, we punted.

17:55:12 <bijan> Ok, that's what I figured, thanks.

17:55:19 <Jimh_> they're working on a requirments and use case document now.

17:55:54 <Jimh_> I'm pretty sure they stil lexist - either that or I spent a couple of hours last night scribing a non-existing telecon...

17:56:00 * DanC wonders if the Web Services CG discusses schedule changes ala "things got derailed for a while due to the political stuff"

17:56:15 <DanC> my dream is that WGs learn how to notify the AC etc. of schedule slips *before* their charter expires

17:56:46 <DanC> I'm just saying there's conflicting evidence about whether the choreography WG exists.

17:57:07 <DanC> (likewise RDFCore)

17:58:27 <Jimh_> yes, that is true. Question - when a group (like Core or WOWG) goes to LC, what are they expected to publish (i.e. if it takes more than the 3 months to get to CR/pR)?

17:59:29 <DanC> if it takes more than 3 months to clean up your LC comments, that suggests your design doesn't have consensus.

17:59:51 <DanC> you can always do another LC WD.

18:00:11 <sandro> What if there can never be consensus on RDF?

18:00:29 <sandro> or XHTML2

18:01:40 <DanC> well, you can ask to advance without consensus; you can argue that you've got something that's an advance in the state of the art, regardless of outstanding dissent.

18:01:55 <DanC> And The Director can either grant you proposed rec status or not, and the AC can either agree or not.

18:02:29 <DanC> (formally, the director can either determine a critical mass of support from the AC or not)

18:03:10 <sandro> IE the W3C *could* make a REC with nothing like consensus.

18:03:14 <sandro> ?

18:03:23 <DanC> yup

18:03:33 <Jimh_> Dan - more than 3mo on LC - don't forget that includes 4 weeks of LC (from when docs come out) plus all the time spent in paperwork tracking comments, etc.

18:03:37 <sandro> I didnt even know that. :-)

18:03:52 <Jimh_> not really complaining, just saying that the 3 months becomes a lot more of a heuristic than some other times ...

18:03:58 * sandro supposes he should actually READ the process document.

18:04:27 <DanC> er.. jim, how could I *possibly* forget?

18:04:39 <sandro> lol

18:04:44 <DanC> I'm knee deep in comments paperwork

18:05:06 * sandro appreciates how deep in it both DanC and JimH are.

18:05:37 * sandro also appreciates the iron will it takes programmers to do paperwork.

18:05:59 * Jimh_ wonders if it would be taking me and Dan so long if we weren't trying to do so much using RDF

18:07:08 <DanC> I think eating our dogfood is worthwhile. yes, it takes a little longer. but at the end, I can say with a straight face: this stuff works.

18:08:31 <Jimh_> ditto

18:09:53 <Jimh_> in fact, I did 50+ comments from your LC-status to my spreadsheet in about 1.5 hours, but the other 15-20 which I did after took almost twcie as long

18:10:18 <DanC> ooh! a product endorsement!

18:11:36 <DanC> sandro, jim, I'd like to talk tactics... I don't want another 2 week slip for no reason... I'm off to vancouver this Sat, and I want to be sure we're set to decide to go to CR next Thu, 24 July

18:12:34 <sandro> Is PR a possibility?

18:13:12 <Jimh_> Dan - I'd like to do the following - if we could get a draft of the CR request (with maybe a couple of things hanging) to the WG before you go, then we should be able to close the last loose ends on 24th and

18:13:22 <Jimh_> either take a CR vote.

18:13:39 <DanC> right...

18:13:43 <Jimh_> we did set tyhe expectation that that would happen at the last telecon (with the WBS straw poll)

18:13:44 <DanC> I have a draft of the CR request now...

18:13:59 <DanC> what I'd like to do is go over each of the @@s in that draft and make sure we've got a plan for getting rid of ti

18:14:26 <Jimh_> sandro - group straw poll pref was a short CR, which also seems to work in terms of scheduling w/RDF

18:14:34 <DanC> I have one more appointment today (now, in fact) but I'd like to spend a couple hours on this later today and/or tomorrow

18:15:40 <sandro> My sense from the comments in the meeting was that the straw poll was pretty ... straw, with people not understanding all the issues.

18:16:17 <Jimh_> I wasn't on the call - so all I know is what I see in the irc log...

18:16:39 <sandro> yeah. but probably RDF Core *is* enough of a reason to wait.

18:17:19 <sandro> (famous last words, as RDF Core takes 7 more years)

18:17:23 <Jimh_> DanC - I'm free tomorrow afternoon (from about 2:30P on) - today is pretty tough

18:18:47 <DanC> that's 2:30pET/3:30pCT?

18:18:52 <DanC> phpht.

18:18:59 <DanC> no, 2:30pET/1:30pCT

18:19:12 <Jimh_> yes, ET

18:20:48 <DanC> ok

18:21:59 <DanC-AIM> byebye

18:54:32 * sandro gets a Canon G3 for his birthday!!!! My first digital camera!!!!!

18:55:30 <Morbus> whoa, rock!

18:55:33 <Morbus> i want firstpic!

18:55:48 <sandro> what woudl that mean? you coming over?

18:55:57 <Morbus> no, the first picture you ever take with it. i want it.

18:56:04 <Morbus> i don't travel. i phear.

18:56:14 <sandro> phear what?

18:56:33 <sandro> I guess I need some RDF photo archive software now.

18:57:24 <Morbus> phear traveling.

18:57:56 <sandro> Huh. Me too, but I do it anyway.......

18:58:15 <Morbus> well, look at mr. hoity toity! <G>

18:58:41 <sandro> :-( I didnt mean it like that

18:58:48 <sandro> sorry

18:59:36 <sandro> I meant it like a sympathy thing. I understand, I think, and can imagine if it were worse.

19:01:01 <_joshua> I don't mind being someplace else, but I hate being In Transit

19:03:02 <sandro> I have trouble until I've been in the new place 2-3 days. Then I'd particularly want to travel *home*.

19:06:03 <Morbus> sandro: no, no, i wasn't being serious. sorry if it came out like that ;)

19:07:24 * DanC considers chumping RDDL to discuss it in preparation for next week's TAG ftf

19:10:09 * sandro considers chumping Pat's reply to www-tag, so we can appreciate the blow-by-blow.

19:12:30 <DanC> pls do

19:13:15 <DanC> sandro, I'd like to make sure you'll be attending the WebOnt telcon next thu (24July)

19:15:26 <sandro> Yes, I plan to.

19:15:29 <sandro>http://lists.w3.org/Archives/Public/www-tag/2003Jul/0147.html

19:15:29 <dc_rdfig> F: http://lists.w3.org/Archives/Public/www-tag/2003Jul/0147.html from sandro

19:15:56 <w1k1_> w1k1_ is now known as w1k1

19:16:24 <sandro> F|"Social Meaning" rebuttal from Pay Hayes

19:16:43 <sandro> F:|"Social Meaning" rebuttal from Pay Hayes

19:16:43 <dc_rdfig> Titled item F.

19:17:37 <gromgull> Hi all, does anyone know if there is a way of refering back to an anonymous node in N3? Like rdf:nodeID?

19:17:56 <sandro> F: replying along the lines of Bijan [earlier today|http://rdfig.xmlhack.com/2003/07/16/2003-07-16.html#1058366936.561735]

19:17:56 <gromgull> i.e. [ a :Book ; :title "My Book" ].

19:17:56 <dc_rdfig> Added comment F1.

19:18:15 <gromgull> then later doing : :Gunnar a :Person ; :hasWritten ???

19:18:31 <gromgull> where ??? should be a reference to afore mentioned book?

19:22:08 <gromgull> If not, there would be (as far I can see) things you could represnet in RDF/XML but not in N3.

19:47:56 <DanC> gromgull, write _:b1 a :Book; :title "My Book".

19:48:08 <DanC> then: :Gunnar a :Person; :hasWritten _:b1.

19:48:25 * gromgull was just typing : To answer my own question, since no-one else did: Yes, you can use _: .... :)

19:48:37 <gromgull> after looking at some Jena2 test-cases :)

19:49:09 <DanC> it's even documented: http://www.w3.org/2003/Talks/0520-www-tf1-b3-rules/slide23-0.html

19:49:11 <gromgull> Thanks

19:49:54 <DanC> aka Variables with ? and _: in http://www.w3.org/2000/10/swap/doc/Rules

19:50:37 <DanC> gromgull, how did you learn N3?

19:50:54 <DanC> I wonder if there's documentation that discusses [] but not _:foo... I guess the N3 primer.

19:51:03 <gromgull> Looking at different sources, mainly the primer

19:51:22 <gromgull> And looking by hacking other n3 files...

20:01:19 <DanC> ok. cool.

20:43:31 <mdupont> hey

22:14:15 <mdupont> anyone awake

22:14:17 <mdupont> hey mortenf

22:14:25 <mdupont> hi DanC

22:14:30 <mortenf> hi mike

22:14:52 * DanC waves, goes back to TAG ftf prep...

22:15:15 <mdupont> re

22:15:23 <mdupont> mortenf: what you up to?

22:15:31 <mortenf> exif :)

22:19:43 <mdupont> what is that, me googles

22:20:17 <mdupont>http://www.exif.org/

22:20:18 <dc_rdfig> G: http://www.exif.org/ from mdupont

22:20:28 <_joshua> you know what'd be amusing? using a google I Feel Lucky uri to indicate someting

22:20:37 <mdupont> G:EXIF.org is a site dedicated to EXIF and related resources. EXIF stands for Exchangeable Image File Format, and is a standard for storing interchange information in image files, especially those using JPEG compression. Most digital cameras now use the EXIF format. The format is part of the DCF standard created by JEIDA to encourage interoperability between imaging devices.

22:20:37 <_joshua> that way the most popular schema or whatever is chosen

22:20:37 <dc_rdfig> Added comment G1.

22:20:57 <mdupont> cool

22:21:01 <mdupont> google rocks

22:21:16 <mortenf> i'm not sure i get it, but it sounds interesting

22:22:05 <mdupont> that what you are working on?

22:22:41 <mortenf> well, just some exif2rdf for my photos (got me a canon 10d)

22:23:58 <_joshua> i wish i had more time to toy with that

22:24:08 <mortenf> me too...

22:29:05 <mdupont> cool

22:44:08 <wkearney99> wkearney99 is now known as wkearney|away

23:20:04 <ericP> mdupont, i've got a bit of coding challenge

23:20:12 <mdupont> cool

23:20:21 * mdupont is going to fall over

23:20:46 <ericP> what does op_or on two vectors do?

23:21:33 <mdupont> wow

23:21:38 <ericP> (?a color "blue" || ?b color "brown")

23:21:48 <mdupont> it should apply to the containers, no?

23:21:50 <ericP> where i have some bindings for ?a and ?b

23:22:01 <ericP> containers?

23:22:20 <ericP> if you're about to fall over, i can bug you with this later

23:22:38 <mdupont> applies the iterator the set?

23:23:05 <ericP> yeah, that's how it's implemented now

23:23:18 <mdupont> is that what you want?

23:23:34 <ericP> i was trying to make it so my semantic actions could do the calculation for me

23:23:51 <ericP> ask (

23:24:27 <ericP> ?a shape "rectangle". # yields ?a == (o1, o2, o3)

23:25:49 <ericP> ?a inverstion ?b. # yields ((o1, b1), (o3, b2), (o3, b5))

23:26:14 <ericP> ie, o2 didn't have any inversions and o3 had 2

23:27:07 <ericP> so it's possible to imagine that every time i get a sentence, i take the working set and manipulate it according to the product of the queries

23:27:50 <mdupont> woah

23:27:52 <ericP> the first would have called triplesMatching(_, shape "rectangle")

23:27:59 <mdupont> looks crazy!

23:28:06 <mdupont> wow

23:28:06 <ericP> tx!

23:28:09 <mdupont> as telmplates?

23:28:24 <mdupont> triplesmatching as a template class?

23:28:25 <mdupont> wow

23:28:27 <mdupont> good idea

23:28:31 <mdupont> fast

23:28:49 <ericP> hmm, i wonder who invented that idea

23:28:59 <ericP> now that you think i have, could you explain it to me?

23:29:12 <ericP> perhaps that will clarify at least the inventor

23:29:39 <ericP> i was just using triplesMatching as a primitive query into the DB

23:29:44 <ericP> most DB's seem to support it

23:30:15 <ericP> but i'm interested in what appears to be your invention

23:30:51 <ericP> but at any rate, the second time the semantic action is called, the working set has (?a) = (o1, o2, o3)

23:31:44 <ericP> so it calls triplesMatching(o1, inversion, _) and triplesMatching(o2, inversion, _) and triplesMatching(o3, inversion, _)

23:32:08 <ericP> it gets back 1 result for the first, replaces that row in the working set with (o1, b1)

23:32:27 <ericP> it gets back none for the second and replaces that row with nothing

23:33:05 <ericP> get gets back three for o3 and replaces that row with two rows: ((o3, b2), (o3, b5))

23:33:24 <ericP> so this mechanism works fine for conjunction

23:33:47 <mdupont> ok

23:33:58 <ericP> all the queries can happen when yacc driver calls the semantic action at the end of the statement production

23:34:03 <ericP> now a disjuction:

23:34:14 <ericP> (?a color "blue" || ?b color "brown")

23:34:25 <mdupont> template class<class Subject, class Object, class Predicate> triplesMatching { }

23:34:38 <mdupont> basically it would take three template parameters

23:34:42 <mdupont> defaulting to all objects

23:34:54 <ericP> ooo

23:34:57 <mdupont> then you build expressions as types in c++

23:35:02 * ericP starts to get it

23:35:06 <mdupont> and it results in a nice c++ query language

23:36:14 <mdupont> the idea is that you would override functions

23:36:23 <ericP> would that do more than allow me great flexibilty on the types of the objects i passed in as nodes?

23:36:30 <mdupont> that take functors with default constructors as parameters

23:36:45 <mdupont> you have constructors with various combinations of parameters

23:37:03 <mdupont> it would allow a user to specify specfic callback functions for some types

23:37:10 <ericP> so far, i get the nodes from an atom dictionary when the go into the db or into the query object

23:37:26 <mdupont> and allow template libs per uri predicate

23:38:02 <ericP> ahh, that would have some interesting implicatins in terms of first order/higher order logic

23:38:18 <ericP> let me get back to my conundrum, if i may

23:38:37 <ericP> you followed the logic that i descript for the conjucntion?

23:39:09 <mdupont> no

23:39:14 <mdupont> it is way over my head

23:39:18 <mdupont> your notation is killing me

23:39:36 <mdupont> please tell me, how can i quote a name in a uri

23:39:37 <mdupont> ?

23:39:44 <mdupont> in n3?

23:39:56 <ericP> i'm not so good on the established conventions

23:40:04 <mdupont> ericP: i am sorry to be so thick, let me see if i can cecypher

23:40:09 <mdupont> decypher

23:40:22 <ericP> i'll spew some ground facts in the mean time...

23:40:31 <mdupont> ?a shape "rectangle". # yields ?a == (o1, o2, o3) ?a inverstion ?b. # yields ((o1, b1), (o3, b2), (o3, b5))

23:40:37 <mdupont> as (?a) = (o1, o2, o3)

23:40:42 <mdupont> so it calls triplesMatching(o1, inversion, _) and triplesMatching(o2, inversion, _) and triplesMatching(o3, inversion, _)

23:41:25 <mdupont> you want to distribute the variables

23:41:43 <mdupont> the inversion function introduces funky results

23:41:55 <mdupont> what does it do,please explain to a bird brain?

23:44:05 <ericP> inversion should have been spelled "foo"

23:44:10 <ericP> it's just a property

23:44:15 <ericP> o1 shape "rectangle";

23:44:16 <ericP> inversion b1.

23:44:16 <ericP> color "red", "blue".

23:44:16 <ericP> o2 shape "rectangle";

23:44:16 <ericP> color "grey".

23:44:17 <ericP> o3 shape "rectangle";

23:44:20 <ericP> inversion b3.

23:44:22 <ericP> inversion b5.

23:44:25 <ericP> color "red", "green".

23:44:27 <ericP> b1 color "grey".

23:44:30 <ericP> b3 color "grey".

23:44:33 <ericP> b5 color "brown", "grey".

23:44:48 <ericP> (some typo's probable)

23:45:05 <mdupont> node_void_type String AddProperty(String, String)intrspctr-gcc-node:type.

23:45:11 <mdupont> is that valid n3?

23:45:23 <mdupont> ahh

23:45:24 <mdupont> fooo

23:46:45 <ericP> should be valid n3

23:47:06 <ericP> ';' means expect a new property,value. ',' means expect a new value.

23:47:41 <ericP> so first set means that o1 color "red". o1 color "blue".

23:47:57 <ericP> plus a shape and an inversion, whatever that is

23:48:35 <ericP> now the query, annotated by what i expect the semantic action to return for each statement:

23:48:38 <ericP> ask (?a shape "rectangle". # ((o1, o2, o3))

23:48:41 <ericP> ?a inverstion ?b. # ((o1, b1), (o3, b2), (o3, b5))

23:48:43 <ericP> (?a color "blue" ||

23:48:46 <ericP> ?b color "brown"))

23:49:18 <deltab> what's "inverstion"?

23:49:50 <deltab> oh, typo?

23:49:57 <ericP> 2nd line means o1 inversion b1. [o2 has no inversion] o3 inversion b2. o3 inversion b5.

23:50:24 <ericP> inversion (and all nearby variants) are a made up property

23:51:15 <ericP> mdupont, do you follow hwo the semantic actions would behave for the first two statements in the ask?

23:51:46 <mdupont> i dont get your syntax

23:51:57 <ericP> the # (())s stuff?

23:51:58 <mdupont> ask ( boo far )

23:52:36 <ericP> ask (statement1. statementn) is the algae syntax

23:52:53 <ericP> the comment to the right is the expected results

23:53:40 <ericP> the variable ?a will be bindable to three values after the first statement is queried

23:53:51 <ericP> o1, o2, and o3

23:54:31 <_joshua> how do I say "and" in an XPath? /foo/bar[@x=y and @z=q]

23:54:45 <ericP> i think so

23:54:57 <ericP> that is, i think you got it right

23:55:29 <deltab> [@x=y][@z=q]

23:55:35 <_joshua> sweet

23:55:52 <ericP> i'm glad i didn't swear to it

23:56:28 <ericP> come to think if it, i should know, i think i implemented a bit of it at some point.

23:56:31 <deltab> or there's the and operator