Semantic Web Interest Group IRC Chat Logs for 2002-10-11

09:26:13 <ChanServ> [#rdfig] This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome

09:58:26 <dajobe> dajobe has changed the topic to: RDF/Semantic Web 24x7 chat - http://rdfig.xmlhack.com/ blog

11:48:46 <q|away> q|away is now known as qmacro

12:45:22 <lilo> [Global Notice] Hi all. Will be doing some major rehubbing. Please bear with us for a few moments.

12:46:41 <ChanServ> [#rdfig] This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome

12:54:16 <lilo> [Global Notice] Re-hi all. One more major leaf node and minor hub to go, and then we should be through with the rehubbing, assuming everything remains stable. Thanks for using freenode and have a great day!

14:59:20 <Code_Poet> Code_Poet is now known as Talliesin

15:31:01 * sandro_ reads the 112 rdfcore e-mail messages for october. wheeeee. :-)

15:31:57 <dajobe> a quiet month so far

15:32:31 <sandro_> Yep. It was readable. Took less than two hours.

15:32:51 <dajobe> we just voted on rdf datatyping

15:33:36 <sandro_> So yeah -- I'm up to date on the discussion except for the just-ended meeting. How did it go?

15:33:41 <dajobe> . http://ilrt.org/discovery/chatlogs/rdfcore/2002-10-11.html#T14-43-25

15:33:43 <dajobe> tidy

15:33:52 <danb_lap> tiiiidddyyyyy!

15:34:08 * sandro_ wanted to watch it on C-Span. (the american TV channel that shows the debates in congress (parlement), etc)

15:34:16 <dajobe> i.e http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Oct/0098.html was accepted, with the addtiions

15:34:23 <sandro_> Option C ?

15:34:27 <dajobe> above

15:34:37 <dajobe> the irc log is public, and live

15:36:17 <sandro_> So that vote is on option C, using rdf:datatype="foo" for each literal you want to datatype, approved 10-2.

15:36:29 <dajobe> yes

15:36:43 <sandro_> Is it over?

15:37:42 <sandro_> (Are there any parsers that do this yet?)

15:37:49 <dajobe> ahem, mine

15:38:06 <dajobe> jjc says next release of arp will have parseTYpe="collectioN" but not datatype

15:38:08 <sandro_> Int 37 minutes? Dave, you're a god.

15:38:09 <dajobe> next release=next week

15:38:18 <dajobe> lol

15:38:38 <sandro_> (funny typo of "int" for "in", in a datatyping discussion)

15:38:56 * JibberJim reckons he can get it into his parser quicker than "next week" :-)

15:39:04 <sandro_> Are you going with <a> <b> <datatypeUri>"10" in N-Triples?

15:39:34 <dajobe> yes

15:39:45 <sandro_> ... and is "XML" (as in XML literals) a datatype? I kinda seems like it.

15:39:51 <dajobe> not so far

15:40:27 <dajobe> that, and making [a b "foo"] == [a b <uriforxsd:string>"foo"] are questions to answer

15:40:43 <dajobe> however since rdf strings are really lang-strings, maybe not

15:40:55 <sandro_> Well, congratulations, y'all (WG members).

15:41:09 <sandro_> welcome home from the war. :-)

15:41:11 * dajobe sighs deeply, collapses

15:42:24 <danb_lap> do any parsers (except Eikeon's) support rdf:nodeID

15:42:25 <danb_lap> ?

15:42:35 <DanCon> well, a battle was decided; it remains to be seen (from review comments) whether the war is over.

15:42:45 <dajobe> or raptor, you mean also

15:43:06 <bijan> I think ours might.

15:43:07 <dajobe> oh, arp does it too now I expect

15:43:10 * DanCon wonders who will lead the interop-report charge...

15:43:14 <danb_lap> that was the question. i didn't remember Raptor had it yet... (though I now remember you saying it took n<60 mins)

15:43:15 <dajobe> since that's been around for a while

15:43:23 <dajobe> not even that

15:43:26 <danb_lap> nodeID is a big useful change...

15:43:41 <dajobe> I did it 2002-08-31 says the changelog

15:43:46 <danb_lap> wonder how localised Mozilla's datatyping stuff is

15:43:54 <DanCon> yes, nodeID means the log:forSome hack in cwm's RDF serializer can go away.

15:44:15 <DanCon> (the log:forAll bug/kludge remains, meanwhile :-{ )

15:44:45 <sandro_> nodeID does not occur in http://www.w3.org/2000/03/rdf-tracking/, but it is in the minutes somewhere, I imagine? (no need to find it.)

15:45:11 <DanCon> nodeID not in the "attention developers" stuff? bummer, that...

15:45:44 <DanCon> should be under http://www.w3.org/2000/03/rdf-tracking/#rdfms-syntax-incomplete

15:46:14 <dajobe> it's in there somewhere

15:46:24 <sandro_> The text "nodeid" is not. :-)

15:46:27 <dajobe> I add all that to http://ilrt.org/discovery/2001/07/rdf-syntax-grammar/

15:46:51 <danb_lap> people are going to start seeing it in a lot of rdf files, since creationwise we're in early adopted stage still and early adopters use new stuff

15:47:06 <danb_lap> but not if it barfs peoples parsers

15:47:25 <DanCon> "agreement. congrats, jeremy." -- http://ilrt.org/discovery/chatlogs/rdfcore/2002-07-26.html

15:47:31 <DanCon> seems nodeID was decided 27 July

15:48:09 <dajobe> yes, that's what I have below http://ilrt.org/discovery/2001/07/rdf-syntax-grammar/#rdfms-syntax-incomplete

15:48:12 <DanCon> danbri, wanna update rdf-tracking real quick? or should I? or should I mail BrianM?

15:48:36 <dajobe> [[On 26th July 2002, the WG decided to re-open this issue and accept the proposal (as amended) to add an rdf:nodeID to the syntax for specifying blank nodes in triple subject and object positions.]] -- i got froms omewhere

15:48:40 * sandro_ is very happy he's going to have to change his RDF-Core-Update talk slides for next week's DAML meeting. (They currently have a summary of the one-remaining open issue. Now they can say there are none, and summarize option C.) And I'll mention NodeIDs.

15:49:20 <DanCon> might be interesting to have a slide on DAML's impact on RDF: parseType="Collection", model theory, [others?]

15:49:44 <DanCon> i.e. the impact on the RDF specifications of the DAML program.

15:51:21 * DanCon starts assembling WebOnt meeting record bits...

15:51:26 <sandro_> You mean that DAML got RDF to be more theoretically-well-specified? (I have "Collection" in there already).

15:51:33 <sandro_> That's a great point.

15:52:10 <DanCon> yes, I think credit is due to the DAML program for focussing the attention of folks with formal logic experience on RDF.

15:52:19 * sandro_ does wonder a little how much XSD code will have to go into RDF code now......

15:52:39 <sandro_> sandro_ is now known as sandro

15:52:43 <DanCon> how much XSD code: good question. That was my reservation about C.

15:53:35 <danb_lap> <DanCon> danbri, wanna update rdf-tracking real quick? or should I? or should I mail BrianM?

15:54:02 <danb_lap> hmm, it's really brian's doc nowadays.

15:54:11 <sandro> But maybe alot of software can just pass around pair(datatypeURI, string) without caring about what's inside.

15:54:13 <danb_lap> I'd suggest mailing him (and perhaps ask re edit policy)

15:54:41 <DanCon> if I were brian, I'd rather get mail ala "I just updated the issues lists" than "would you please update the issues list?"

15:55:16 <danb_lap> I suspect so too. Go for it; we can always back out the change.

15:55:35 * sandro heads off to MIT to hear about some real theorem proving (from one of Boyer & Moore; can't remember which).

15:55:58 <DanCon> Moore

15:56:14 * DanCon hopes for sandro-powered notes

15:56:28 * danb_lap too

15:56:48 <danb_lap> <DanCon> yes, I think credit is due to the DAML program for focussing the attention of folks with formal logic experience on RDF.

15:56:53 <sandro> I'll try.

15:57:00 <danb_lap> Yup. I remember early days of www-rdf-logic...

15:57:18 <danb_lap> Stefan Decker too did a bunch of early work hooking that community into RDF

15:57:26 <DanCon> yup

16:04:29 <DanCon> logger, pointer?

16:04:29 <DanCon> See http://ilrt.org/discovery/chatlogs/rdfig/2002-10-11#T16-04-29

16:09:27 * DanCon updated issues list, mailed Brian and rdf-comments

16:27:25 <danbri> ta

17:00:30 * DanCon working on http://www.w3.org/2001/sw/WebOnt/ftf4

17:18:27 * DanCon suffers amaya crash. frap

17:19:15 * DanCon recovers. whew!

17:23:45 <danb_lap>http://www.fawcette.com/xmlmag/2002_10/online/xml_kcagle_10_09_02/default_pf.asp

17:23:46 <dc_rdfig> A: http://www.fawcette.com/xmlmag/2002_10/online/xml_kcagle_10_09_02/default_pf.asp from danb_lap

17:24:01 <danb_lap> A:|

17:24:02 <dc_rdfig> Titled item A.

17:24:02 <danb_lap> Print Article

17:24:02 <danb_lap> Keep XML Naked

17:24:23 <danb_lap> A:|Keep XML Naked, by Kurt Cagle (XML and Web Services magazine)

17:24:24 <dc_rdfig> Titled item A.

17:24:59 <danb_lap> A:Interesting idea but somewhat innacurately executed (eg. re Xlink and RDF; XTM and OWL).

17:25:00 <dc_rdfig> Added comment A1.

18:09:05 <sandro> Sorry, DanC, no notes from J. Moore talk. It was a fast guided tour & demo through some lisp code. I didnt have time to write anything down.

18:09:19 <DanCon> oh.

18:09:29 <DanCon> ACL2 stuff?

18:09:42 <sandro> But I did get a feel for using ACL2, yeah.

18:09:47 <DanCon> ok.

18:09:53 <DanCon> I think I'm pretty familiar with that work anyway.

18:10:17 * DanCon continues putting the pieces of the WebOnt record together... http://www.w3.org/2001/sw/WebOnt/ftf4

18:10:20 <sandro> And they're proving things about medium-simple program (via their java byte codes).

18:10:39 <sandro> Did you have Moore as a prof at UT?

18:10:51 <DanCon> no, never met Moore. had Boyer as a prof, though.

18:11:23 <DanCon> they've been proving stuff about code with ACL2 by looking at machinecode for a while... found a bug in the quicksort() in the BSD libc, as I recall!

18:12:13 <sandro> yep, he mentioned some work on libc. he's hoping JVM's being higher level will help them do more, I gather.

18:13:02 <sandro> It was nice seeing a talk presented mostly inside emacs (and some ghostview) on a KDE desktop.

18:13:20 <DanCon> nice in a sorta retro way, I guess. ;-)

18:13:48 <sandro> compared to PowerPoint.....

20:57:36 <AaronSw> AaronSw is now known as zooko

21:00:11 <zooko> zooko is now known as AaronSw

21:13:48 <qmacro> qmacro is now known as q|away

22:20:03 <DanCon> hoot! "that's what God put them there for when he designed the Internet; you know, during the prequal.

22:20:10 <DanCon> " -- mnot http://www.mnot.net/blog/archives/000012.html

22:21:37 <AaronSw> does he mean prequel?

22:21:48 <DanCon> dunno.

22:22:03 <DanCon> I think I've heard 'prequal' in testing/QA discussions.

22:22:15 <DanCon> meaning pre-qualifiication, before it goes golden.

22:24:14 <DanCon> hoot! "Baby's head which look like it has just been squeezed through a tube, because, well, it had just been squeezed out through a tube.

22:24:19 <DanCon> " -- http://www.ironycentral.com/archives/babyvol/babyvol1.html

22:26:58 <sbp> wow, that Baby Photo Album B idea is quite cruel

22:27:10 <sbp> but I have to admit that it's pretty neat

23:21:26 * dmiles wonders who here is going the the Cyc-101 classes

23:21:59 <dmiles> at http://www.opencyc.org/announce/cyc101