Semantic Web Interest Group IRC Chat Logs for 2003-06-11

00:01:58 <edd> P:[viewcvs access|http://svn.usefulinc.com/cgi-bin/viewcvs.cgi/trunk/foafbot/] is available too, for a slightly more sophisticated view on the source

00:01:58 <dc_rdfig> Added comment P6.

00:04:29 <DanC> <mid:000001c30039$9f76ba40$4ff0010a@matisse> a lca:OutstandingMessage;

00:04:29 <DanC> :subject "Remarks on OWL Guide and question about AS&S" .

00:23:56 <sbp`> sbp` is now known as blungenaut

00:24:12 <blungenaut> blungenaut is now known as sbp

00:39:23 <DanC> ugh; this is too hard.

00:50:46 <DanC> aha; here's what to do: reply to all outstanding threads, so that they all show up in the current month's archive.

00:51:39 <DanC> frap; evolution crashes when I move stuff between folders.

00:59:58 <deltab> H:[http://www.harmonise.org/aboutus_001.html|(skip Flash)]

00:59:58 <dc_rdfig> Added comment H3.

01:22:07 <earle> earle is now known as grault

01:45:13 <mea_cu|pa> mea_cu|pa is now known as mea_culpa

01:56:12 <deusx_away> deusx_away is now known as deusx

02:34:30 <jason> anyone know how to automaite the task of synchronizing source files from my linux box to my web server? So that clicking a button saves the newest versions of a set of files to htdocs?

02:34:57 <dnm> rsync

02:36:11 <jason> is this what you use?

02:50:46 <soccos> soccos is now known as soccos_away

02:54:27 <sbp> sbp is now known as sbp_

02:54:35 <sbp_> sbp_ is now known as sbp`

03:05:07 <AaronSw> dnm? congrats on cool new job

03:37:36 <golbeck_> golbeck_ is now known as golbeck

04:14:09 <bitsko>http://bitsko.slc.ut.us/blog/2003/06/10/rss-links-0dot2

04:14:09 <dc_rdfig> A: http://bitsko.slc.ut.us/blog/2003/06/10/rss-links-0dot2 from bitsko

04:14:18 <bitsko> A:|RSS Links 0.2

04:14:18 <dc_rdfig> Titled item A.

04:15:01 <bitsko> A:Adds another session's worth of links, plus many authors and subjects.

04:15:01 <dc_rdfig> Added comment A1.

05:12:30 <mea_cu|pa> mea_cu|pa is now known as mea_culpa

06:21:45 <mdupont> mdupont is now known as mdupont-away

08:37:53 <dajobe> chump bot ISP down, back later

11:27:33 <deusx> deusx is now known as deusx_away

11:55:49 <gromgull> Shameless self-plug:

11:55:49 <gromgull>http://www.csd.abdn.ac.uk/~ggrimnes/dev/n3php/

11:55:49 <dc_rdfig> B: http://www.csd.abdn.ac.uk/~ggrimnes/dev/n3php/ from gromgull

11:56:21 <gromgull> B:| PHP N3 Parser

11:56:21 <dc_rdfig> Titled item B.

11:56:40 <gromgull> B: Works with { RAP RDF API | http://www.wiwiss.fu-berlin.de/suhl/bizer/rdfapi/ }

11:56:41 <dc_rdfig> Added comment B1.

11:56:56 <gromgull> B: Shamelessly ripped from Spb's python parser

11:56:57 <dc_rdfig> Added comment B2.

11:57:07 <iand> Looks cool

11:57:32 <iand> Need to update my PHP/FOAF article now :)

11:57:37 <mattb> gromgull: chump needs square brackets not braces

11:57:58 <gromgull> right... I saw that in the manual a few seconds after I hit return :)

11:58:21 <mattb> you can edit lines too

11:58:25 <gromgull> How?

11:58:29 <mattb> just do B1:replacement text for line 1

11:58:57 <gromgull> B1: Works with [ RAP RDF API | http://www.wiwiss.fu-berlin.de/suhl/bizer/rdfapi/ ]

11:58:58 <dc_rdfig> Replaced comment B1.

11:59:08 <gromgull> magic

11:59:16 <mattb> heh, looks like it doesn't swallow whitespace either

11:59:25 <mattb> gotta drop the spaces around the [ and | characters

11:59:34 * mattb adds it to mental list of bugs to fix

11:59:35 <gromgull> B1: Works with [ RAP RDF API|http://www.wiwiss.fu-berlin.de/suhl/bizer/rdfapi/]

11:59:36 <dc_rdfig> Replaced comment B1.

11:59:45 <mattb> magic :)

11:59:48 <gromgull> even more magic : )

12:01:30 <iand> iand is now known as iand_away

12:08:30 <danbri> B:Cool! Is there a complete list of N3 parsers (and their status) anywhere?

12:08:30 <dc_rdfig> Added comment B3.

12:23:10 <jordan-> jordan- is now known as jordan

13:22:43 <deusx_away> deusx_away is now known as deusx

13:54:03 <edd> dc_rdfig:status

13:54:05 <dc_rdfig> I am the Daily Chump Bot, version 1.2. In public mode. Promiscuous. <http://usefulinc.com/chump/>

13:54:11 <edd> dc_rdfig:die

13:54:33 <edd> dc_rdfig:status

13:54:35 <dc_rdfig> I am the Daily Chump Bot, version 1.3. In public mode. Promiscuous. I am in UTF-8 mode. 日本語. See?<http://usefulinc.com/chump/>

13:55:37 <edd> BLURB:this chump is now UTF-8 aware

13:55:41 <dc_rdfig> C: this chump is now UTF-8 aware from edd

13:56:02 <edd> C:those with utf-8 enabled IRC clients will be able to take full advantage of that

13:56:03 <danbri> C:excellent :)

13:56:06 <dc_rdfig> Added comment C1.

13:56:06 <dc_rdfig> Added comment C2.

13:56:08 <danbri> C:Thanks Edd!

13:56:08 <dc_rdfig> Added comment C3.

13:56:13 <edd> C:there's fallback for latin 1 users, too, which should be automagic.

13:56:14 <dc_rdfig> Added comment C4.

13:58:04 <Morbus> is there a chat about chandler happening in here today?

13:58:40 <sbp`> hey Morb. kinda: one of the people working on Chandler should be along for an RDF Calendar meeting

13:59:07 <sbp`> from the agenda: [[[

13:59:08 <sbp`> > * Chandler and RDF

13:59:08 <sbp`> > http://osafoundation.org/for_developers_2003_05_06.htm

13:59:08 <sbp`> > Katie Capps Parlante

13:59:09 <sbp`> ]]]

13:59:34 <sbp`> - http://lists.w3.org/Archives/Public/www-rdf-calendar/2003Jun/0004

14:00:01 <sbp`> 1600UTC, so that's 1700BST, so that's in two hours

14:06:42 <libby> yep, 2 hours. Katie's going to talk to us about what they're doing, but not for thge full time - just for 10-15 mins

14:08:39 <Morbus> thank you kindly.

14:09:42 <libby> welcome. should be interesting

14:10:57 <sbp`> their current RDF API is interesting--I wonder if I can find a summary...

14:14:13 <sbp`> hmm. nope. seems subject to change, anyway

14:14:19 <sbp`>http://wiki.osafoundation.org/bin/view/Main/DataModelSchema

14:14:19 <dc_rdfig> D: http://wiki.osafoundation.org/bin/view/Main/DataModelSchema from sbp`

14:14:29 <sbp`> D:|Chandler PIM Schema

14:14:29 <dc_rdfig> Titled item D.

14:16:21 <sbp`> D:a lot of terms in there that could've (possibly, depending on what range/domain they use) been reused frm other schemata. cf. [http://esw.w3.org/topic/BuildOrBuyTerms|BuildOrBuyTerms]

14:16:21 <dc_rdfig> Added comment D1.

14:26:59 <bitsko> C:whoohoo!

14:26:59 <dc_rdfig> Added comment C5.

14:28:14 <bitsko> it is still accurate to say that Chandler is using RDF as an interchange format and not an internal representation, correct? in contrast to, say, Haystack?

14:28:38 <Morbus> bitsko: have you fiddled with Haystack? sadly, I don't believe it'll work on OS X.

14:28:51 <Morbus> unless I install a jvm for linux, as opposed to apple's native.

14:29:26 <danbri> they talk about mapping their python-based schema to rdf triples for OODB mapping etc

14:29:28 <bitsko> I stopped as soon as I saw it was in Java. I've always had problems getting "large" java apps running, much less even smaller ones

14:29:44 <Morbus> bitsko: the best java app I've ever used (and continue to use, daily) is jedit for win32.

14:33:16 <bitsko> . http://staticred.net/archives/2003/06/04/11_46_51.html

14:33:35 <bitsko> "Update: I tried haystack last night, and the authors weren't lying about how slow it was... it took 20 minutes to load everything, and was glacially slow to use. I think it'll have to wait until I get a new processor and more RAM before playing with it in any detail."

14:34:22 <shellac> I was going to have a look at it on OS X - see if I could get it going

14:34:35 <shellac> but download is huge - sorry HUGE

14:34:56 <shellac> chandler works well - and is small

14:35:02 <bitsko> that's really odd, and unfortunate

14:35:37 <Morbus> shellac: have you *used* chandler in your day to day? i've played with it, but not actually set out to integrate.

14:36:24 <shellac> no - it isn't really doing that much AFAICT. a proof of concept

14:36:44 <shellac> reminds me, in some ways, of mozilla and XUL

14:37:14 <Morbus> shellac: heh!

14:37:22 <Morbus> i just used that example to describe chandler to my boss the other day.

14:37:56 <shellac> I mean tech wise, rather than in a pejorative way, I hasten to add

14:38:37 <Morbus> same here :)

14:39:44 <shellac> the impression I got is that is presents (partial) views of a database

14:40:10 <shellac> and that database is semi-structured

14:40:44 <danbri> yes, it is in a v similar space to moz and xul

14:40:55 <danbri> i wonder what abstraction they'll use for creating views of the data.

14:40:59 <shellac> I'd like to know how to make views

14:41:08 <shellac> jinx!

14:41:10 <shellac> :-)

14:41:13 <danbri> moz approach is basically rdfq + a templating language

14:41:16 <Morbus> anyone ever use Entourage on OS X?

14:41:18 <arnarl> hi

14:41:25 <Morbus> that has some very very useful features.

14:41:34 <bitsko> one thing that will be very interesting is how either or both Chandler and Haystack map existing formats to RDF (and if we can all come to agreement on the mappings)

14:41:34 <Morbus> if it didn't use a single point of failure, I'd probably be using it.

14:42:47 <bitsko> as I was recently gathering links from a dozen different forums, I came face-to-face with how incredibly difficult it is to get useful information from them. I did it all by hand.

14:43:21 <sbp`> would it have been much help if the fora had formalized indexes?

14:43:30 <shellac> bitsko: acutally I'd like them not to agree

14:43:44 <bitsko> eek. why?

14:43:53 <shellac> then try inferencing / rules to convert

14:44:00 <bitsko> gotcha

14:44:27 <bitsko> I'm cool with that. even using a common framework is half the battle.

14:44:56 <danbri> nice clean inferences shade off into perl-esque hackery very easily... lots of string munging etc often needed

14:46:31 <botlars> botlars is now known as larsbot

14:46:36 <bitsko> sbp`: yes, even that would have been useful. xml-dev was by far the easiest, because I could download the mbox tar files!

14:46:54 <shellac> hmm - client died before I fully backed up statement

14:47:13 <bitsko> the only missing piece of info I had to supply by hand was the archive's generate URL for the message

14:47:51 <bitsko> s/tar/gzip/

14:48:44 <shellac> my point was if everyone agrees on formats then we aren't showing of the strength of RDF

14:49:04 <bitsko> by comparison, yahoogroups makes it intentionally difficult to fetch messages. one cannot actually fetch a message using just wget/curl.

14:49:50 <Morbus> bitsko: funny you should mention that.

14:49:58 <Morbus> my new book has some two hacks on that one ;)

14:50:51 <shellac> do we have to buy the book to get the info? ;-)

14:50:53 <bitsko> shellac: I'll wait for some practical examples. I'm supportive of it, but I need to see how it works in practice.

14:53:11 <shellac> bitsko: that's true. There are quite a few tools with inferencing support, now, but not too many with rules (which I suspect will be essential)

14:53:52 <sethl> shellac: how do you define rules?

14:54:25 <shellac> I point at cwm :-)

14:55:00 <sethl> another way to phrase it, what's the difference between inferencing and rules (which, given cwm, can provide inferencing)?

14:55:03 <shellac> rules are essentially modus ponens + axiom schemas

14:55:15 * sethl runs off to google

14:56:11 <shellac> essentially rules are statements like foo -> bar

14:56:32 <shellac> eg AxFx -> AxGx

14:56:37 <sethl> so the act of rules is itself inferencing?

14:56:45 <sethl> I found: http://www.wikipedia.org/wiki/Modus_ponens

14:56:57 <mattb> oh, sethl, did you get anywhere with your redfoot/rdflib query engine workings?

14:57:06 * mattb remembers such things being mentioned during irc at www

14:57:12 <sethl> yup, sent it off to eikeon

14:57:15 * danbri didn't get shellacs distinction

14:57:20 <shellac> modus ponens is just a fancy way of saying A, A then B therefore B

14:57:24 * mattb wonders if source code preview is possible

14:57:25 <sethl> never wired it into test cases, though

14:57:33 <shellac> one inference

14:57:41 <shellac> modus ponens

14:57:47 <danbri> you mean some RDF tools hardcode a set of built in inferences (eg. RDFS + OWL stuff), future ones will allow more inferences via rules as a data format.

14:58:01 <danbri> ?

14:58:07 <mattb> sethl: i've been doing a bunch of hacking on redland's python wrapper, curious to try adapting your code whatever state it's in :)

14:58:09 <sethl> ah... a rule specifies an inference

14:58:27 <sethl> mattb: I've been trying to get ahold of eikeon to see what the state of that is

14:58:41 <sethl> I can send you what I have, and you can have at it

14:58:55 <mattb> that'd be great

14:59:17 <shellac> danbri: sort of, yes

15:00:10 <shellac> rules are like what you enter in prolog

15:01:01 <shellac> the inferencing is what prolog does using what you entered

15:01:24 <sethl> morbus: what book are you refering to?

15:01:35 * danbri hmms re noun / verb

15:01:57 <Morbus> seth: my second oreilly book. can't say much more than that. not publicly announced yet.

15:02:19 <sethl> morbus: not even the subject matter? :)

15:02:26 <Morbus> nope :)

15:03:01 <shellac> screen scraping for fun and profit?

15:05:15 <darobin> ora has secrecy policies on its books?!?

15:15:57 <danbri>http://wiki.osafoundation.org/bin/view/Main/DemoTranscript2003-04-23

15:15:58 <dc_rdfig> E: http://wiki.osafoundation.org/bin/view/Main/DemoTranscript2003-04-23 from danbri

15:16:07 <danbri> E:|Transcript of a Chandler presentation

15:16:07 <dc_rdfig> Titled item E.

15:17:17 <danbri> E:Mentions restaurant guide scenario; fodder for compare/contrast w/ approaches in [http://esw.w3.org/topic/RestaurantRecommendation|esw:RestaurantRecommendation].

15:17:18 <dc_rdfig> Added comment E1.

15:19:04 <libby>http://lists.w3.org/Archives/Public/www-rdf-calendar/2003Jun/0004.html

15:19:04 <dc_rdfig> F: http://lists.w3.org/Archives/Public/www-rdf-calendar/2003Jun/0004.html from libby

15:19:22 <libby> F:|Agenda for RDFcalendar meet this afternoon

15:19:22 <dc_rdfig> Titled item F.

15:19:49 <libby> A:[2003-06-11, at 1600UTC|http://www.timeanddate.com/worldclock/fixedtime.html?day=11&month=6&year=2003&hour=16&min=0&sec=0&p1=0]

15:19:49 <dc_rdfig> Added comment A2.

15:19:59 <libby> A2:""

15:19:59 <dc_rdfig> Deleted comment A2.

15:20:05 <libby> F:[2003-06-11, at 1600UTC|http://www.timeanddate.com/worldclock/fixedtime.html?day=11&month=6&year=2003&hour=16&min=0&sec=0&p1=0]

15:20:05 <dc_rdfig> Added comment F1.

15:20:37 <libby> BLURB: RDF calendar agenda item (G): outstanding actions

15:20:37 <dc_rdfig> G: RDF calendar agenda item (G): outstanding actions from libby

15:21:22 <libby> BLURB: RDF calendar agenda item (H) InterpretationProperties, esp as applied to timezones

15:21:22 <dc_rdfig> H: RDF calendar agenda item (H) InterpretationProperties, esp as applied to timezones from libby

15:21:47 <libby> BLURB: RDF calendar agenda item (I): Chandler and RDF - Katie Capps Parlante

15:21:48 <dc_rdfig> I: RDF calendar agenda item (I): Chandler and RDF - Katie Capps Parlante from libby

15:22:07 <libby> I:[more on Chandler|http://osafoundation.org/for_developers_2003_05_06.htm]

15:22:08 <dc_rdfig> Added comment I1.

15:22:36 <libby> G:see [last meeting record|http://rdfig.xmlhack.com/2003/05/14/2003-05-14.html]

15:22:36 <dc_rdfig> Added comment G1.

15:22:57 <libby> H:[InterpretationProperties wiki page|http://esw.w3.org/topic/InterpretationProperties]

15:22:57 <dc_rdfig> Added comment H1.

15:30:50 <bitsko> D:WARNING: this page is huge and full of tables.

15:30:50 <dc_rdfig> Added comment D2.

15:31:03 <libby> H:[mortenf explaining teh tz problem last time|http://ilrt.org/discovery/chatlogs/rdfig/2003-05-14.html#T16-15-19]

15:31:04 <dc_rdfig> Added comment H2.

15:31:09 <libby> .time

15:31:09 <datum> Wed, 11 Jun 2003 15:31:09 GMT

15:35:47 * danbri wonders if Chandler attributes come with anything like domain/range metadata, to say which types they make sense to use with.

15:35:56 <danbri> can't tell from http://wiki.osafoundation.org/bin/view/Main/ItemsAndAttributesDataModelIssues

15:41:52 <shellac> I'm curious about the versioning - any other links?

15:42:15 <shellac> (i.e. beyond http://wiki.osafoundation.org/bin/view/Main/VersioningIssues)

15:43:45 <danbri> hi bijan

15:43:56 <bijan> Heya

15:53:06 <libby_> libby_ is now known as libby

15:53:55 <danbri> hi katie

15:53:59 <katie_> hello

15:54:03 <katie_> katie_ is now known as katie

15:58:43 * DanC is deep in WebOnt WG business... tries to switch context

15:59:50 <libby> DanC, it's not on the agenda, but at some stage you were going to say somethign about roundtripping from rdf->ical and back....feel like talking about that maybe?

16:00:04 <DanC> willing to, sure.

16:00:12 * DanC finds record of decision to meet now ... http://rdfig.xmlhack.com/2003/05/14/2003-05-14.html#1052929573.964914

16:00:35 <libby> .time

16:00:37 <DanC> odd... the RdfCalendar wiki page doesn't seem to have been cited from the 14May record

16:00:38 <datum> Wed, 11 Jun 2003 16:00:35 GMT

16:00:49 <libby> -----rdf calendar meet------

16:00:53 <libby> no?

16:01:08 * DanC finds the agenda all set up at http://rdfig.xmlhack.com/ . nifty.

16:01:13 <libby> I've been updating but maybe not everything

16:01:57 <libby> F:[http://ilrt.org/people/libby|libby miller] attending

16:01:58 <dc_rdfig> Added comment F2.

16:02:00 <DanC> F:last time: [http://rdfig.xmlhack.com/2003/05/14/2003-05-14.html|14 May]

16:02:00 <dc_rdfig> Added comment F3.

16:02:17 <DanC> oops; redundant. oh well.

16:02:22 <libby> hi everyone who is ehre for the calendar meet - could you add yourself to the weblong as I just did?

16:02:28 * mortenf is sorry he's late.

16:02:43 <libby> everyone else - we'll be chatting about rdf and calendaring for about 90 minutes...

16:02:48 <DanC> F:nearby [http://esw.w3.org/topic/RdfCalendar|RdfCalendar wiki topic]

16:02:48 <dc_rdfig> Added comment F4.

16:02:58 <libby> hey mortenf, not really started yet

16:03:05 <mortenf> great.

16:03:10 <mortenf> i think. :)

16:03:12 <DanC> DanC has changed the topic to: RDF calendar chat http://rdfig.xmlhack.com/ 1600Z for ~90min

16:03:21 <danbri> F:[http://www.w3.org/People/DanBri/|Dan Brickley] attending

16:03:21 <dc_rdfig> Added comment F5.

16:03:42 <libby> ok we have three scheuled agenda item, plus danc has offered to say soemthign about roundtripping

16:03:42 <DanC> F:[http://www.w3.org/People/Connolly/|Dan Connolly] attending

16:03:42 <dc_rdfig> Added comment F6.

16:03:56 <libby> RDF calendar agenda item (G): outstanding actions

16:04:00 <RSwick> F:[http://www.w3.org/People/all#swick|Ralph Swick] attending

16:04:01 <dc_rdfig> Added comment F7.

16:04:11 <mortenf> F:[http://purl.org/net/morten/|Morten Frederiksen] attending

16:04:11 <dc_rdfig> Added comment F8.

16:04:22 <libby> I think outstanding actions are mostly me, and I havn;t dine anythign since last time except update the pages a bit :(

16:04:23 <timbl> F:[http://www.w3.org/People/Berners-Lee|Tim Berners-Lee] attending

16:04:23 <dc_rdfig> Added comment F9.

16:04:26 <libby> apoologies

16:04:50 <mortenf> yeah, poo ;)

16:04:57 <libby> heh

16:05:08 <libby> so I guess

16:05:09 * DanC thinks he owes an action on {} vs. subAbstrac; hunts...

16:05:13 <katie> I'm here, from OSAF

16:05:14 * danbri reminds #rdfig occasionals that the dialog with the 'dc_rdfig' bot results in real-time weblog at http://rdfig.xmlhack.com/ organised by A:-to-Z: headings

16:05:48 <DanC> katie, if you put F: in front of that, it'll be easier to find later.

16:06:03 <katie> F: Katie Parlante is here from OSAF

16:06:04 <dc_rdfig> Added comment F10.

16:06:12 <katie> thanks DanC

16:06:14 <DanC> well played :)

16:06:32 <libby> G:continuing actions: ACTION libby look at skical optimeset and report back; ACTION libby send a mail to rdf-interest on interprettatyon properties and datatypes...also the thing about a registry of xproperties in icalendar

16:06:32 <dc_rdfig> Added comment G2.

16:07:09 <libby> anhyone else think of any?

16:07:26 <DanC> libby, katie, folks, would it be useful if I sorta re-hashed the history of RdfCalendar in a few minutes?

16:07:30 <danbri> (continuing them sorta assumes we'll meet again... which i for 1 am willing to do...)

16:07:39 * danbri would be up for that, yes

16:07:46 <libby> yes, go for it DanC

16:07:56 * katie would also appreciate it

16:07:58 <DanC> (whether to meet again is a non-trivial issue for me...)

16:08:07 <DanC> ok, some history...

16:08:33 * danbri rummages around the ill-documented RDFIG f2f notes at http://www.w3.org/2001/02/rdfig-f2f/#outcomes

16:09:11 <DanC> ok, do you see the "RdfCalendar wiki topic" link under "Agenda for RDFcalendar meet this afternoon" in http://rdfig.xmlhack.com/ , katie (and everybody)?

16:09:40 * danbri nods

16:10:13 <danbri> (someone else nod!)

16:10:15 <DanC> esp the para "The www-rdf-calendar announcement of April 2001 followed an RDF Interest Group meeting at which several of us realised we shared an interest in this area."

16:10:51 * DanC realies the path to http://www.w3.org/2002/12/cal/ is too long/obscure

16:10:53 * katie finally sees it

16:11:03 <mortenf> got it.

16:11:35 * danbri finds a passable, partial history is provided by the wayback machine, http://web.archive.org/web/*/http://ilrt.org/discovery/2001/04/calendar/

16:11:44 <DanC> ok, so we got together in Cambridge forever ago, started a www-rdf-calendar list, and eventually had an RDF calendar workshop in Bristol. (for which thanks again, libby and company)

16:12:36 <DanC> I think timbl wrote a review of the icalendar RFC with RDF glasses forever ago too... (LinkMe)... libby picked up on that and put together an RDF schema...

16:12:45 * danbri enjoyed the meeting; good mix of people, backgrounds, interests yet didn't have that 'we're all speaking different languages' feeling that mixed backgrounds sometimes bring.

16:13:13 <danbri> cf http://www.w3.org/2000/01/foo 'A quick look at iCalendar' (timbl)

16:13:37 <DanC> ... various hacks followed libby's work, including a conference organizer challengning the world to rdf-ize the program, which challenge was met some 48 hours later (by Terry payne and company?)

16:14:35 <DanC> ... by the Oct 2002 workshop in bristol, I was trying to go from icalendar tools to RDF and back by machine...

16:15:03 <DanC> ... and I wasn't sure how to relate my work to the various bits of libby's schema. And libby wasn't sure how to compare her schema to mine either.

16:15:07 <danbri> cf http://lists.w3.org/Archives/Public/www-rdf-calendar/2001Oct/0003.html "We've finally released the Alpha version of the Retsina Calendar

16:15:07 <danbri> Agent for the Semantic Web. This is the followup from Stefan's Semantic

16:15:07 <danbri> Web Challenge at the Semantic Web Working Symposium in Stanford this

16:15:07 <danbri> summer. It runs on Windows machines, and allows users to browse events

16:15:07 <danbri> marked up in RDF, and import them into MS Outlook 2000."

16:15:17 <danbri> (--Terry Payne)

16:15:49 <DanC> so both libby and I obsoleted our work in favor of a new http://www.w3.org/2002/12/cal/ical schema

16:16:20 <DanC> where the rule is... [[ At the Bristol workshop, we agreed, roughly...

16:16:20 <DanC> * we announce all changes to the schema www-rdf-calendar

16:16:20 <DanC> * if anyone screams, within a week or so, we'll back out the changes (for further discussion)

16:16:20 <DanC> ]]

16:16:40 <danbri> Michael Arick's contribution (via a UML representation) worth noting; some more details in Libby's status report from last year, http://lists.w3.org/Archives/Public/www-rdf-calendar/2002May/0000.html

16:16:40 <DanC> and we've been having occasional IRC chats like this one ever since.

16:17:13 <DanC> that's a short-ish version. make sense? any questions?

16:17:44 * danbri has a question: do you believe the schema at http://www.w3.org/2002/12/cal/ical is reasonably in sync with tools/code you use?

16:18:01 * katie ran across a lot of these materials several months ago, has found it useful

16:18:15 <DanC> yes, danbri, the consistency between the schema and the tools is enforced by machine.

16:18:19 <danbri> cool

16:18:23 <DanC> enforcement could be tigher... esp with test data.

16:18:38 <danbri> logger_1, bookmark?

16:18:38 <danbri> See http://ilrt.org/discovery/chatlogs/rdfig/2003-06-11#T16-18-38

16:19:37 <DanC> so we could transtion to either the chandler calendar agenda item (I) or an item on round-trip testing.

16:19:50 * danbri would like to hear about Chandler

16:19:55 <libby> I'd be einterestedin the chandler discussion, if that's ok with all

16:20:08 <katie> perhaps I could give a short introduction

16:20:10 * timbl would like to hear about Chandler

16:20:30 * danbri adds link rdfcal history chat to http://esw.w3.org/topic/RdfCalendar

16:21:00 <DanC> ---- agenda item (I): Chandler and RDF - Katie Capps Parlante

16:21:09 * DanC yields the floor to katie

16:21:15 <katie> Mitch Kapor worked on a product years ago (so long ago it was a DOS application) called Agenda

16:21:39 <katie> Ecco is a more recent PIM, inspired by Agenda

16:21:58 <katie> Both had the flavor of being able to manage data flexibly

16:22:46 <katie> Mitch has been thinking about related problems for years, and founded OSAF to build an open source PIM that captured this spirit

16:23:13 <DanC> flexibly... in the psion agenda, you can move an item from the calendar to the todo list and back easily. I miss that on the palm and danger sidekick platforms.

16:23:33 <katie> yes, you might create a simple 'note' first

16:23:41 <katie> and then later turn it into a calendar item

16:24:25 <katie> early on, before I got there, folks ran across RDF and the stuff about the semantic web, and realized it had a lot in common with what Mitch wanted to do

16:25:00 <katie> Also, Mitch wanted a "serverless calendar"

16:25:20 <katie> by "serverless", we don't literally mean no-servers-absolutely

16:25:35 <katie> but a group calendaring application where the group doesn't have to admin a server

16:26:00 <katie> so we started out with this idea of a p2p PIM, where you could share your flexible data, including calendar data

16:26:23 <DanC> server-optional ++

16:26:33 <timbl> So the system is quite p2p with no central node?

16:26:38 <katie> server-optional => exactly

16:27:02 <katie> we've realized that there will in some sense often be a central node

16:27:13 <katie> for discovery if nothing else

16:27:52 <katie> we've received a lot of interest in the calendar from universities

16:27:52 <DanC> sounds like: if the organization has a natural center, you might as well put a server there. but if it doesn't, chandler is still happy.

16:28:28 <katie> in particular, for calendar information, you might want some of the information to be widely available

16:28:36 <katie> not gone when someone turns off their laptop

16:29:04 <katie> so a small workgroup might run a version of chandler on a machine under the desk

16:29:22 <katie> and people in the group might replicate part or all of their repositories to that machine

16:29:49 <katie> but two people might browse each other's contacts, or calendar information

16:29:57 <katie> even if there is no "server" in the picture

16:30:17 * DanC is torn between blue-sky discussion of the whole p2p partial-understanding-sync space and detailed "how does the chandler release do X?" questions

16:30:31 <katie> yes, best to move back from blue sky :)

16:31:13 <katie> So there's the possible set of questions: what kind of functionality will the calendar have?

16:31:14 <timbl> Can my chandler can use data from many different groups?

16:31:27 <DanC> ok, how does chandler do authorization? i.e. how does it decide whether requests to look at my calendar are authorized?

16:31:29 <katie> And the set of questions: how do you manage data? what's your relationship to RDF?

16:31:32 <danbri> the blue sky sounds good... how's it going so far?

16:31:58 <katie> timbl: yes, as I understand your question, chandler can use data from many different groups

16:32:16 <katie> a given chandler client can browse any chandler repository, if one has permission

16:32:17 <danbri> I'm interested in the Chandler data model, how you name and document item types, attributes etc... (maybe w/ calendar as specific example to stay on topic :)

16:32:36 * timbl feels that Katie should feel free to answer questions -- ours or hers -- in any order she sees fit or not at all. :)

16:32:54 <katie> just remind me if I forget something :)

16:33:13 <katie> A lot of answers are still up in the air

16:33:19 * DanC looks for docs on how "if one has permission" is implemented

16:33:24 <DanC> e.g. does it support email callback auth?

16:34:01 <katie> authorization: we don't have that worked out yet, ongoing discussions

16:34:24 * danbri notices that the no1 google hit for 'chandler rdf' is the invite to this discussion, so stops googling and listens...

16:34:26 <katie> data model: the basics of the data model are 'items', a collection of attributes

16:34:26 * DanC is also curious about how the chandler project is organized; who can commit to the CVS repository? but I'll probably follow that up outside this meeting

16:35:20 <katie> you can think of an 'item' as a series of triples where the subject is the same

16:35:30 <katie> the repository is 'item' based, not triple based

16:35:49 <DanC> "Take a look at the bug-filing guidelines and give it your best shot." a link to the bug-filing guidelines there would help. (random comment re http://osafoundation.org/for_developers_2003_05_06.htm)

16:35:55 <katie> popping up to the calendar schema, I hope it has a lot in common with ical/your rdf schema

16:36:32 <DanC> item based... that makes a lot of sense w.r.t. sync protocols

16:36:33 <katie> right now, only staff can commit to the CVS repository

16:36:34 * timbl wonders if I have a location which is my office and it has an address, whether the nested stuff about the address witll ber serialized with info about me and how that is deternmined.

16:36:44 <katie> but we hope to gradually become more open

16:36:49 <danbri> item-based in terms of indexing; given some item you can find values of it's attributes more cheaply than finding which other items it is an attribute of?

16:37:14 <timbl> Suppose I use two address books, and they have various bits of information about the same person -- is that one item or two?

16:37:36 <katie> we need to get our architecture in place

16:37:40 <timbl> (ie is the item a subject within a contect, or a subject in all contexts)

16:37:55 <katie> we need to get to the point where I can answer your questions more specifically :)

16:38:02 <katie> we're still in a pretty early phase

16:38:11 <DanC> in our RDF model of icalendar, many of the properties of an icalendar component have the same subject, but not all; there is some nesting. esp. with recurrence rules, timezones, and a few other things

16:38:29 <katie> item-based in terms of indexing: yes

16:38:39 <katie> item-based in terms of results you might get back from a query

16:39:19 <katie> two address books, one item or two: that depends

16:39:27 <DanC> I vaguely recall reading that you're using ZODB in chandler. is that (still) the case?

16:39:55 <katie> two address books: you might have two repositories, say, one on your laptop and one on your desktop. In this case, I think the items are the same

16:40:30 <danbri> Have you had difficulty reconciling the extensible schema approach (items can always have unexpected attributes attached) with goal of synching? (at least re synching with less flexible representations, like icalendar)

16:40:42 <mortenf> it seems the "Chandler UID" assigned to items are repository dependant?

16:41:16 <katie> timbl: could you clarify for me: subject within a context, or a subject in all contexts

16:41:25 <katie> I think the answer is that the subject is in all contexts

16:41:36 <katie> A location is an item

16:41:40 <katie> A calendar event is an item

16:41:46 <katie> A person is an item

16:42:02 <timbl> Clarification - it is a question of keepingtrack of where you get data from.

16:42:27 <katie> each attribute may have a link to another item

16:42:30 <timbl> In practice a triple comes from some place, so you can think o fit mroe as a quad of (context, subject, pred, object)

16:42:34 <katie> so, yes, subject in all contexts

16:42:45 <danbri> item is just the most general class of things, like 'Thing', 'Resource', 'Object' etc?

16:43:00 <timbl> It seems there are three competeing architectures. 1) Use OO design, and serialize in RDF. 2) Use RDF design, and build objects (items) at runtime; 3) Use RDF design, and use RDF API at runtime instead of OO store. Any feeling which one Channdler is looking like?

16:43:02 * DanC surfs for docs on item architecture

16:43:09 <katie> item is like resource, I think

16:44:14 <danbri> danc, http://wiki.osafoundation.org/bin/view/Main/DemoTranscript2003-04-23 has some pointers. ah yup http://wiki.osafoundation.org/bin/view/Main/DataModel -> http://wiki.osafoundation.org/bin/view/Main/DataModelIntro

16:44:22 <katie> there's definitely a tension between whether items are essentially oop 'objects' or more rdf like 'items'

16:44:26 <RSwick> earlier, "the basics of the data model are 'items', a collection of attributes" -- I interpret that as a bag of statements

16:44:43 <katie> we're struggling with how to make the items useful to python programers, as objects

16:44:52 <DanC> I:nearby: [http://osafoundation.org/architecture.htm|osaf architecture]. RDF, ZODB, python, ...

16:45:00 <dc_rdfig> Added comment I2.

16:45:00 <katie> and still retain the data flexibility/mutability that we desire

16:45:14 * DanC doesn't see the connection with bags that ralph sees

16:45:44 <timbl> struggling: me too.

16:45:45 <katie> timbl: thinking about your question about competing architectures

16:46:31 <danbri> you can play tricks with Python's object system (and Perl's, Ruby's, ...), catching method calls at runtime and suchlike... but that doesn't feel quite right as a general solution

16:46:46 <timbl> struggling: There are a number of eamples of making python objects which actually arebacked by an RDF store not python storage.

16:47:00 <katie> you can play a lot of tricks with python, very flexible

16:47:16 <katie> so, we're not implementing a general RDF store

16:47:16 <danbri> Is the plan to stick with Python for finished product, or is it yr prototyping language with C/++/etc for the final thing?

16:47:27 <katie> we're implementing an 'item' store, which we are defining

16:47:34 <katie> we are defining the features items can have

16:47:48 <katie> an item can have a 'kind' (like a class)

16:48:06 <katie> a 'kind' can have a list of attribute-definitions

16:48:36 <katie> I was just looking on our twiki for the notes on the latest meeting, but looks like someone has just rearranged everything

16:48:48 <katie> doh! twikis

16:48:53 <danbri> can an item have multiple, independently defined 'kinds'? eg. item_232 might be both a chandler:Person and a mozilla:Recipient...

16:48:58 * DanC brb

16:49:20 <katie> danbri: yes, the latest thinking is that an item can have multiple kinds

16:49:30 <katie> although one 'kind' might be primary

16:49:42 <katie> the kind used for the python mapping

16:49:42 <danbri> good, that helps with the loosly-coupled thing... you don't have to contend to be 'the' type of a thing

16:50:13 <katie> we realized we have two clients in mind when designing the data model

16:50:15 <danbri> when you say a kind has a list of attribute-definitions, could you give an example...? like 'expected slots'?

16:50:29 <katie> one client wants an oop object: someone writing python code to implement a calendar

16:50:42 <katie> the other client is the end user, who wants to be able to restructure data

16:50:49 * danbri nods

16:50:56 <katie> example of attribute definitions:

16:51:08 * timbl suggets that the item store be defiend to be equivalent to an RDF store, and yo test it my making RDF API to it too -- and also see whether you can implement an item store on top of a general RDF store. That would surface the differences.

16:51:12 <katie> I might have an 'Event' kind, for calendar events

16:51:24 <katie> timbl: yes, that's a good idea

16:51:43 <katie> timbl: one of the tasks on our plate was to map our data model to rdf

16:51:56 <danbri> I would love to try that... if I could get a dump of sample items/attributes, loading into an SQL-backed RDF store and running some queries...

16:52:24 <katie> back to example: I might have a start-time attribute-definition

16:52:39 <katie> and I might have some 'restrictions' on the attribute definition

16:52:45 <katie> such as type/domain: datetime

16:52:54 <katie> perhaps cardinality

16:53:26 <katie> I might err on the side of loose restrictions

16:53:41 <mortenf> have you tried making a (partial) OWL schema?

16:53:53 <katie> yes, I did for an earlier incarnation

16:54:10 * DanC is afraid ZODB has 'the type' embedded pretty deeply

16:54:10 <mortenf> so, it seems possible?

16:54:28 <katie> we're not married to ZODB

16:54:31 <danbri> In the RDF/OWL design -- because we don't allow a class to dictate an exhaustive list of the properties/attributes its members can have -- it is a bit different. It's more attribute centric...

16:54:39 <katie> its going to be useful in the short run, certainly

16:54:48 <DanC> there are a lot of good things about ZODB: transactions, replication, python integration.

16:55:02 <katie> yes, that's why we're interested

16:55:17 <danbri> You might say of the chandler:start-time property/attribute that it applies to the class chandler:event and has values of kind chandler:datetime.

16:55:33 <katie> yes, you might say that

16:56:09 <katie> you might define the attribute-definition for start-time globally

16:56:10 * DanC recommends the cyc ontology re TemporalThings and SpatialThings. i.e. people have start-times too, i.e. birthdays.

16:56:34 <katie> yes, you might want to add start-time to a variety of items

16:56:45 <katie> and then, you might want to view them in your calendar viewer

16:57:04 <katie> or you might want to view your calendar-event-items in a more general viewer, an outline

16:57:12 * mortenf wonders if there's some "linkage" between the cal schema and cyc?

16:57:52 * DanC downloads chandler source... holy cow! 38MB!

16:57:57 <danbri> If folks here wanted to play around with Chandler data in an RDF context, could you advise us on how to get started? eg. dumping a "live" calendar to RDFesque triples somehow?

16:58:01 <katie> yes, its a problem!

16:58:17 <katie> you know, its going to get more interesting in a few months

16:58:24 <katie> I know I keep saying that...

16:58:26 <mortenf> have you given any thought to chandler:uid vs uris?

16:58:52 <DanC> does the 38MB include copies of libraries from elsewhere? wxWindows and such? or is that all OSAF-written code?

16:58:58 <katie> but we've just hired our repository person, and we've been in heavy design mode

16:59:29 <katie> and we hope to have an API, a strawman schema, and some minimal repository implementation by September

16:59:36 <katie> an earlier version even in the next month

16:59:49 <katie> it would be great to get feedback from you folks as we go, though

17:00:04 <katie> 38MB includes wxWindows, wxPython and Python

17:00:10 <mortenf> sounds great!

17:00:12 <katie> the actual chandler source is very small

17:00:48 <katie> there is a repository, but its mostly scaffolding

17:00:56 <katie> partly to start playing with python mappings

17:00:59 <katie> for items

17:01:11 <danbri> katie, you were talking about how Chandler data is viewed by application code (eg. OO-style calendar app code). Is there any plan to have some abstraction roughly akim to Mozilla's XUL, for encoding data views without doing it in Python?

17:01:36 <danbri> (XUL being javascript+css+xml + rdf query templates for data binding, roughly)

17:01:41 <katie> there's a plan in the works for a 'document architecture'

17:01:56 <katie> where a 'document' is a description of a view

17:02:11 <katie> a hierarchical set of queries (defining sets of items)

17:02:22 <katie> but I'm handwaving, because I haven't been in those meetings

17:02:46 <katie> yes, played with XUL some, we considered using it

17:03:04 <danbri> Who is working on the query language aspects? is that Rys McCusker?

17:03:19 <katie> Rys McCusker has left OSAF

17:03:37 <katie> Andi Vadja is the new repository person

17:03:59 <danbri> The RDF query language implicit in XUL templates, fwiw, is pretty similiar to most of the other RDF query systems out there...

17:04:04 <katie> Brian Skinner is working on the data model, John Anderson is the system architect

17:04:09 * danbri nods

17:04:31 <katie> John, Brian, Andi and I have been working out the API, some initial discussions about the query language

17:04:38 <katie> well, query mechanism

17:05:02 <katie> How happy are people with the existing RDF query languages?

17:05:13 <katie> in practice?

17:05:20 <danbri> straw poll?

17:05:51 <libby> I love them :)

17:05:57 * zool holds up a short straw

17:06:02 <danbri> for me, I think the basic graph match stuff we've all implemented is nearly good enough, but for practical apps it needs extending to allow for retrieval of optional bits of the graph. Mozilla needed (and has) that.

17:06:06 <libby> if I could add one feature it would be optional variables, like XUL

17:06:08 * zool -> dorkbot

17:06:15 * sandro gets nosey and reads Andi's resume. He looks very well qualified.

17:06:18 <danbri> Probably also need support for xml datatypes

17:06:24 <katie> yes Andi was a find!

17:06:43 <katie> there was an earlier question about uris and uids

17:07:00 <mortenf> yep, just curious.

17:07:03 <katie> the current thinking was that each item in the repository has a uid

17:07:22 <katie> the uid may be part of a uri for someone remotely referring to the item

17:07:25 * danbri asking too many questions, gives others a chance

17:07:30 <katie> where the repository itself is part of the uri

17:08:02 <mortenf> would (a representation of) the item be retrievable from its uri?

17:08:17 <katie> I may have missed other questions (not totally keeping up!)

17:08:35 <katie> yes, I think that should be possible

17:09:00 <katie> each item needs to have a path in the repository, for implementation reasons

17:09:11 <katie> so this path can also constitute a uri

17:09:23 <danbri> I did ask one about how RDF IG folk might be able to help/experiment etc with RDF views of Chandler data... maybe we could take that to email if short of time here... would be good to stay in touch after this chat.

17:09:29 <katie> starting to handwave :) I hope this will be more clear in a few weeks

17:09:44 <mortenf> np

17:09:51 <katie> yes, would be great to stay in touch

17:10:02 <katie> I can tell that you all 'get it' :)

17:10:03 * DanC looks for http-browsable cvs repository

17:10:08 <libby> we are somewhat short on time, but if people want to continue for 10-15 mins that's fine. the only thing we really need to talk about is date of next meet (if any)

17:10:19 <mortenf> a calendar question: how does the chandler model of an event relate to the cal schema?

17:10:29 <mortenf> i.e. do they "match"?

17:10:32 <DanC> which cal schema?

17:10:49 <mortenf> the revised one at http://www.w3.org/2002/12/cal/ical

17:10:57 <katie> we're committed to working with icalendar as an open standard, itip/imip as well

17:11:07 <katie> which implies that we need to understand ical data

17:11:13 <timbl> These URIs of items, which can be retrived and return info about item, they will be HTTP URIs, I hope.

17:11:14 <DanC> I doubt they match. unless chandler uses exactly icalendar's data model.

17:11:28 <danbri> there seems to be a shortage in the ical (.ics) test suite space

17:11:38 <katie> timbl: yes, http uris

17:11:48 <katie> I wouldn't say 'match'

17:11:48 * sandro woohoo!

17:12:02 <katie> but there needs to be some reasonable mapping

17:12:05 <mortenf> would it be possible/desirable to try to merge them to some extent?

17:12:11 <DanC> what other itip/imip client/servers have you seen?

17:12:27 <katie> entourage does itip/imip

17:12:32 <mortenf> yeah, mapping is probably better (easier).

17:12:33 * timbl katie: HTTP URIs - Good!

17:13:03 <katie> we're talking with folks from universities about CAP, to open up another can of worms

17:13:17 * DanC finds it frustrating when folks make new protocols to go with every new kind of data

17:13:30 <katie> yes, have heard that frustration before

17:13:54 <katie> the universities are really motivated to find a calendaring standard

17:14:01 <katie> they like IMAP

17:14:09 <katie> want the equivalent for calendars

17:14:12 * danbri believes that's true of Uni Bristol too

17:14:28 <katie> they don't want to be beholden to any one vendor

17:14:42 <timbl> HTTP is pretty vendor-neutral.

17:14:45 <katie> I guess that's true of most folks :)

17:14:53 <katie> yes, they like http too :)

17:15:00 <timbl> But it needs on top of it a for RDF a way to post diffs to an RDF-mapped store.

17:15:41 * mortenf would like to talk a bit about InterpretationProperties before end-of-meet.

17:15:51 <timbl> You can do a huge amount by reading and writing RDF using HTTP. But the whole p2p multi party sync needs a layer on top.

17:16:29 <katie> well, I should probably wrap up, but am happy to converse over email or in another chat sometime

17:16:36 <timbl> Ok, may be too lateto get into p2p synch

17:16:49 <katie> yes, but interesting topic for later

17:16:55 * timbl another chat ... another chat ... !

17:16:59 <mortenf> +1

17:17:13 <danbri> +1

17:17:19 <katie> what if I schedule a chat with more osaf folks, not taking over rdf-calendar :)

17:17:24 * DanC is having trouble finding time for rdf-calendar while WebOnt is in end-game mode

17:17:24 <katie> interest?

17:17:26 <mortenf> DanC: you had objections re future chats?

17:17:39 <mortenf> ah.

17:17:50 <mortenf> timeframe for WebOnt?

17:18:02 <danbri> katie, that'd be great. We tend to have chats on various RDF-related topics here on wednesdays; eg. rdf cal, rdf query, rdf geo/mapping, ...

17:18:11 <libby> katie, interest in rdf and calendar chats has been a bit low until you turned up :) feel free to use the list/and this slot

17:18:19 <timbl> Suppose that the chat wasn't just "Chandler" but sepcific topics which arose.

17:18:19 <katie> sure, happy to

17:18:31 <katie> a list of topics would be good

17:18:34 <DanC> WebOnt last call comments were due 9May. We had hoped to respond to all of them and go to the next step in June. It's looking like about Jun 45th now.

17:18:42 <mortenf> :)

17:18:50 <danbri> yep, having some interop-testing with real data would be good topic focus...

17:19:01 <katie> danbri: good goal

17:19:03 <libby> real data++

17:19:19 <danbri> ...like, can we take data from Chandler and get it into Cwm/Squish/Redland/Jena/RDFLib/... and do anything useful with it...

17:19:22 * DanC has lots of real data, but little of it that isn't sensitive

17:19:31 <danbri> (and then get it back into Chandler again)

17:19:50 * libby using apple ical more - not all sensitive - can become rdf....

17:19:57 <DanC> agenda request: icalendar schema vs. general-purpose calendar ontology

17:20:11 <libby> shall we schedule another chat now? or wait?

17:20:12 <DanC> agenda request (old): round-trip testing of icalendar

17:20:20 * mortenf is slowly building a list of events for a local group.

17:20:37 * libby interested i n that one DanC - it was one of our 'when are we done' conditions...

17:20:41 <mortenf> also: InterpretationProperties

17:20:44 <libby> sigh

17:20:47 <katie> I'm sure we're going to have to create some data in the next few months

17:21:11 <mortenf> sigh?

17:21:14 <libby> PROPOSED: another chat on rdf and calendars in 2 weeks

17:21:14 <dc_rdfig> Label PROPOSED not found.

17:21:27 <danbri> katie, any rough snapshot of a 'live' chandler item/attribute db would be useful... even if the schema changed

17:21:29 <libby> (sigh - remembers all my actions :)

17:21:40 <libby> any seconders for another chat in 2 weeks?

17:21:40 <DanC> libby, I can't answer "shall we schedule another chat?" in general. I can evaluate proposals to meet at some specific time

17:21:44 <mortenf> :)

17:21:55 <mortenf> sure, i'll be here.

17:21:58 * DanC checks calendar...

17:22:27 <DanC> I have no conflicts for 25June.

17:22:40 <libby> actually I dont know if I can make that

17:22:50 <libby> but I'm happy to create an agneda etc

17:22:51 <DanC> but I feel kinda silly coming to these chats without having spent 2 minutes preparing.

17:23:14 <mortenf> well, if DanC would be happier in start-july, perhaps 2/7?

17:23:26 <libby> the propsal is for 2003-06-25, 1600UTC: http://www.timeanddate.com/worldclock/fixedtime.html?day=25&month=6&year=2003&hour=16&min=0&sec=0&p1=0

17:23:26 * danbri can make 25th June; can't make 2nd july.

17:23:30 * DanC wonders what 2/7 is... is that the 2nd day of July?

17:23:42 <mortenf> july 9th? :)

17:23:46 <mortenf> DanC: yep.

17:23:53 * mortenf is European...

17:24:15 <libby> I'm at risk for then....might be able to make it

17:24:30 <mortenf> hmm, a non-wednesday?

17:24:37 <libby> can;t make 2nd july...

17:24:42 * timbl notes international stds are godo an 2003-07-02 is proposed date

17:24:53 <danbri> regrets

17:25:03 <libby> me too

17:25:18 <libby> again, I can create agenda etc ok

17:25:18 * DanC doesn't see critical mass around any particular date

17:25:35 <mortenf> calendarbot: find timeslot for libby, danbri, katie, timbl, danc, mortenf

17:25:39 <katie> I can make any of these times, btw

17:25:56 <DanC> DanC is now known as calendarbot

17:26:01 <libby> :)

17:26:02 <calendarbot> sorry, problem overconstrained. you lose

17:26:06 <calendarbot> calendarbot is now known as DanC

17:26:08 <mortenf> d'oh.

17:26:16 <libby> sigh

17:26:20 <timbl> calendarbot, what is the least damage solution?

17:26:38 <mortenf> seems June 25th is best?

17:27:03 <libby> that's my inclination, though perhaps a bit soon

17:27:27 * libby might particpate by P800

17:27:28 <mortenf> well, there are enough carried actions and items...

17:28:14 * timbl calendarbot, Could we meet any time for less than 3 bedtime stories missed, 4 annual reviews potsponed and 6 meals converted to pizza?

17:28:26 <danbri> p800 I tried that recently from Paris; it's tough... (and i nearly got run over ;)

17:28:31 <timbl> I can make 2003-06-25

17:28:34 <libby> let's go with 25th June, see how we get on

17:28:38 <mortenf> k.

17:28:41 <libby> I can make it too actually (duh)

17:28:47 <libby> hurray :)

17:28:59 <DanC> BLURB: next meeting (agendum H)

17:29:00 <timbl> Great.

17:29:02 <mortenf> mortenf is now known as calendarbot

17:29:03 <dc_rdfig> J: next meeting (agendum H) from DanC

17:29:09 <DanC> oops. I can't count.

17:29:15 <timbl> I will expect an iCalendar inviation by email

17:29:22 <calendarbot> 2003-06-25Z18:00 for 2.5 minutes.

17:29:24 <DanC> J: RESOLVED: to meet next 25 Jun 2003, 1600Z.

17:29:24 <dc_rdfig> Added comment J1.

17:29:25 <timbl> if not in RDF

17:29:26 <calendarbot> calendarbot is now known as morten

17:29:49 <libby> J:[http://www.timeanddate.com/worldclock/fixedtime.html?day=25&month=6&year=2003&hour=16&min=0&sec=0&p1=0|2003-06-25, 1600UTC]

17:29:49 <dc_rdfig> Added comment J2.

17:29:50 <timbl> Bye everyone ....

17:29:58 <timbl> timbl is now known as tim-away

17:29:59 <libby> ok, we';re out of time....

17:30:09 <tim-away> tx to libby

17:30:12 <AaronSw> AaronSw has changed the topic to: RDF hack and chat http://rdfig.xmlhack.com/

17:30:20 * RSwick thanks Katie, for whetting appetite to learn more

17:30:20 <morten> morten is now known as mortenf

17:30:29 <katie> thanks all, good questions!

17:30:30 <libby> katie, if there's somethign you'd like to talk about wrt chandler in particular can get you a slot for necxt time

17:30:37 <libby> thanks katie, very interesting stuff

17:30:42 <danbri> yeah, thanks Katie, much appreciated

17:30:44 <mortenf> thanks all.

17:31:17 <katie> libby: I'll let you know when I see what progress we've made in the next week or so

17:31:33 <katie> perhaps in a month we'll have more concrete stuff

17:31:43 <DanC> J:|next meeting (agendum J)

17:31:45 <dc_rdfig> Titled item J.

17:32:23 * DanC wonders what mortenf wanted to discuss re interpretation properties

17:32:25 <libby> cool katie, just let me know...

17:33:20 <mortenf> DanC, just the basics, i've been coming across quite a few vocabs with the same problem, e.g. currencies (and the holdover from geo coodinates).

17:34:07 <DanC> hmm... if you're able to spot the problem, you grok the basics.

17:34:16 <mortenf> i'm however about to give up on the the geo problem...

17:34:30 * DanC wonders if mortenf finds http://esw.w3.org/topic/InterpretationProperties usefule

17:34:41 <mortenf> but how to convince others that it's really a problem?

17:34:45 <mortenf> oh yes.

17:35:42 <AaronSw> AD: I'm looking for someone to develop an XML Schema for Creative Commons.

17:35:42 <dc_rdfig> Label AD not found.

17:35:45 <mortenf> and the other question: what to about timezones?

17:35:51 <DanC> convince others... I dunno if I can help you there.

17:36:15 * mortenf is just a little frustrated perhaps

17:36:32 <DanC> what to do about timezones is something I hope to study... but I need a good 2hr slot to get all the context swapped in.

17:36:56 <DanC> and I can't justify using 2hr slots of rdf-calendar until webont is out the door.

17:36:57 <mortenf> fine, looking forward to end of June :)

17:37:08 <DanC> s/slots of/slots on/

17:37:19 <mortenf> np

17:43:14 <mortenf> well, EC soccer coming up, back later...

17:43:47 <DanC> G:continues: ACTION DanC: follow up in email about how folks should relate themselves to events they plan to attend. ical:attendee? cyc:socialParticipants? from [http://rdfig.xmlhack.com/2003/04/23/2003-04-23.html#1051113769.199079|23 Apr]

17:43:47 <dc_rdfig> Added comment G3.

17:44:17 <DanC> hmm... "RESOLVED: that folks should be encouraged to link their RDF calendars from their foaf home thingies via rdfs:seeAlso (@@e.g. danc's, libby's)"

17:44:29 <DanC> does anybody do that besides me and libby?

17:45:41 <libby> mines out of date

17:46:03 <libby> apple ical stores it's info in icalendar so easy to automate though

17:47:02 <bijan> Is there a list of w3c notes, specifically?

17:47:32 <bijan> Aha, foudn ti

18:01:40 <danbri> I would do it, but I don't have an rdf calendar.

18:01:56 <libby> eh?

18:01:59 <danbri> When Apple isync starts to work with the P800, I'm all for it...

18:02:10 <libby> you can use a converter for icalendar to rdf calendar

18:02:13 <danbri> it: <DanC> does anybody do that besides me and libby?

18:02:44 <danbri> not enough of my calendar is written down in a proper machine calendar, embarrasingly

18:05:16 <DanC> hmm... if you can survive without telling the machine your plans, more power to you.

18:06:03 * em pops by, willing to do this if I can find a set of rules that I can filter personal & private bits into different rdf stores

18:06:35 * em curses, puts blinders back on

18:07:44 <danbri> Yeah, my mailbox and todo files play criticial role... haven't got hooked enough on a desktop/laptop cal package since bailing on my Palm

18:10:24 <Morbus> danbri: same here. email sent to myself, and omnioutliner, are my biggest PIM utils.

18:12:04 <DanC> I'm pretty much completely dependent on a machine to beep at me 10 or 15 minutes before each appointment.

18:12:27 <Morbus> i could never find a good beeper, so i just don't go outside anymore.

18:12:31 <bijan> I'm pretty much similarly dependent on it not to

18:12:40 <bijan> If it did, I'd miss fewer meetings

18:12:45 <bijan> Be late for fewer meetings

18:12:53 <bijan> And thus NEVER get any work done :)

18:13:23 * DanC considers meetings a way of getting work done; doesn't go to meetings that aren't.

18:13:48 <bijan> Oh come come.

18:13:56 <bijan> That wasn't even a LITTLE funny?

18:14:03 <Morbus> ROFFLE.

18:14:04 <Morbus> :)

18:14:19 <bijan> If you can't get a cheap laugh out of a meeting bash, what *can* you get a cheap laugh out of? ;)

18:45:15 <bitsko>http://bitsko.slc.ut.us/blog/2003/06/11/rss-links-0dot3

18:45:19 <dc_rdfig> K: http://bitsko.slc.ut.us/blog/2003/06/11/rss-links-0dot3 from bitsko

18:45:29 <bitsko> K:|RSS Links 0.3

18:45:31 <dc_rdfig> Titled item K.

18:46:31 <bitsko> K:Collected links from the announcement of RSS in March 1999 up thru the release of the RSS 1.0 Proposal, August 2000.

18:46:33 <dc_rdfig> Added comment K1.

18:48:02 <bitsko> K:This release includes date, author, and subject for all the more substantive links in the discussion.

18:48:03 <dc_rdfig> Added comment K2.

18:51:43 <bitsko> K2:This release includes date, author, and subject for all the more [http://bitsko.slc.ut.us/cgi-bin/rss-links|substantive links] in the discussion.

18:51:44 <dc_rdfig> Replaced comment K2.

19:12:24 <mea_culpa> mea_culpa is now known as EdTivrsky

19:41:05 <tim-away> Ah. I see why phpicalendar gets the dalight savings time wrong. it checks DST on the local machine, instead of working it out from the timezone in question. It needs DanC's timezone data I guess.

19:41:09 <tim-away> tim-away is now known as timbl

19:47:17 <DanC> ouch!

19:48:49 <timbl> Is the algorithm fairly simple, given the data?

19:49:27 <deusx> deusx is now known as deusx_away

19:49:45 <timbl> (and do we have RDF parser in PHP?)

19:52:09 <DanC> I've seen announcements of RDF parsers in PHP

19:52:20 <DanC> and I've implemented (part of) the algorithm in N3

19:52:31 <DanC> there are closed-world assumptions in the data, unfortunately.

19:52:57 <timbl> You have to find the latest change before the date in question.

19:53:58 <DanC> cf http://www.w3.org/2002/12/cal/tzrules.n3

19:55:26 <DanC> btw... phpical is just a viewer, right?

19:55:40 <DanC> i.e. it doesn't use any of php's special database smarts, does it?

19:55:42 * timbl googles for php rdf and gets lots of in-your-face urius iwth .rdf and .php :-(

19:56:03 <timbl> phpicalendar is just a viewer, yes.

19:56:22 <timbl> It parses icalendra files, generates HTML and RSS

19:56:23 <DanC> bert's ical2html is awfully nice.

19:56:35 <timbl> what is it written in?

19:56:46 <timbl> It is as nice as phpicalendar?

19:56:51 <DanC> written in C

19:56:54 <DanC> nicer, to my eye

19:57:29 <timbl> pointer to an example?

19:57:48 * DanC is hunting; wonders why he didn't make a better link last time he found it...

19:58:09 <timbl> It might get the daylight saving right

19:59:49 <timbl> Hmmm... http://phpxmlclasses.sourceforge.net/rdf_parser.html is a parser

20:01:10 <DanC>http://www.w3.org/Tools/Ical2html/

20:01:11 <dc_rdfig> L: http://www.w3.org/Tools/Ical2html/ from DanC

20:01:20 <DanC> L:|Tools/Ical2html

20:01:20 <dc_rdfig> Titled item L.

20:01:26 <DanC> L:by Bert Bos Jan 2003

20:01:26 <dc_rdfig> Added comment L1.

20:01:36 <DanC> L:ical2html-0.1.tar.gz

20:01:36 <dc_rdfig> Added comment L2.

20:01:47 <DanC> L:oh yeah; it coredumped when I tried to use it the other day.

20:01:47 <dc_rdfig> Added comment L3.

20:28:20 * timbl submits phpicalendar bug report

21:09:07 <lilo> [Global Notice] Hi all. News up on the website (http://freenode.net/news.shtml): expired nicks and channels being dropped; new release of dancer-ircd on the testnet; RSS feed available for Computerworld's SCO coverage page. Please take a look. Thank you for using freenode, and have a great evening!

21:13:19 <lilo> [Global Notice] (RSS url corrected, thanks to several people for pointing out the typo :)

21:46:49 <deusx_away> deusx_away is now known as deusx

21:52:44 <lilo> [Global Notice] Hi all. We've lost some data from services and will need to restore a half-hour-old backup. Please bear with us. Any further information will be on wallops (/mode yournick +w). Thanks.

22:10:50 <DanC> F:[http://ilrt.org/discovery/chatlogs/rdfig/2003-06-11.html#T16-00-49|irc log]

22:10:51 <dc_rdfig> Added comment F11.

22:23:45 <deusx> deusx is now known as deusx_away

22:24:35 <mdupont> mdupont is now known as mdupont-zzz

22:30:06 <mdupont-zzz> mdupont-zzz is now known as mdupont

22:34:27 <mdupont> mdupont is now known as mdupont-zzz