Semantic Web Interest Group IRC Chat Logs for 2003-04-30

00:27:35 <_joshua> may?

00:27:53 <may> hi there...just checking in

00:28:08 <_joshua> woo, right?

00:28:47 <may> yup :) i saw this channel in jo's email

00:29:06 <_joshua> ahh, yeah. i set up the geo list

00:29:40 <may> is everyone here on that list?

00:32:56 <_joshua> no, but a bunch

00:37:11 <may> okay, will check in tomorrow for the "official" mtg :)

00:49:40 <_joshua> heh

03:42:54 <DanC> "It\x{2019}s easy, it\x{2019}s fair and it\x{2019}s legal"

03:43:07 <DanC> -- http://www.apple.com/music/store/

03:43:27 <DanC> I just tried it out. bought (some rights to) a song for 99cents.

03:44:05 <DanC> I hope they make it easy for artists to market their wares direct thru them, without going thru record companies.

07:38:38 <mdupont_> mdupont_ is now known as mdupont-busy

08:51:08 <crysflame> i've been trying to figure out how to use ENT with MovableType, and not having much luck. Is there a plugin I need to get that associates my MT categories with multiple .opml topical schemas? It's not clear how to participate :)

08:53:50 * zool doesnt even know what ENT is ...

08:55:22 <crysflame> it's what drives http://k-collector.evectors.it/

08:55:25 <crysflame> i think.

09:00:19 <deltab> DanC: how did you get those apostrophes? I just get control character U+0092

09:01:52 * zool squints at http://www.purl.org/NET/ENT/1.0/ - a tangential world to this one

09:04:09 <crysflame> aha, thanks

09:04:11 <crysflame> that's why i brought it here

09:04:27 * zool smiles

09:12:52 <libby> morning mattb

09:13:33 <mattb> heya libby

09:56:32 <danbri> morning all

09:57:21 <zool> morning danbri

09:58:27 <crysflame> hi danbri

09:58:43 * crysflame holds up the sign labeled "i am soderberg", for clarification.

10:00:45 * danbri finds http://www.crystalflame.net/ plenty of interestingness

10:01:03 <mattb> heya danbri

10:01:10 <danbri> hi

10:01:20 <crysflame> heh

10:01:36 <crysflame> life is too much fun to not be interesting

10:28:52 <danbri>http://shelter.nu/x/

10:28:55 <dc_rdfig> A: http://shelter.nu/x/ from danbri

10:29:07 <danbri> A:|Topic Map graphical navigation demos

10:29:09 <dc_rdfig> Titled item A.

10:29:42 <danbri> A:(from topicmapmail@infoloom.com, in case you were curious)

10:29:44 <dc_rdfig> Added comment A1.

10:30:11 <danbri> A:Hmm a 'where I found this' convention in chump might be useful, so people could reveal their sources.

10:30:14 <dc_rdfig> Added comment A2.

10:30:30 <crysflame> oo, there's one here too

10:30:53 <danbri> a chump? yup. been here maybe couple years now...

11:12:38 <danbri> danbri has changed the topic to: RDF/SemWeb http://rdfig.xmlhack.com/ TODAY: GeoInfo chat 1400 UTC (3pm BST) http://esw.w3.org/topic/GeoInfo 60-90mins.

11:12:53 <zool> it's not that time already is it? :/

11:12:55 <danbri>http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0300.html

11:13:13 <dc_rdfig> B: http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0300.html from danbri

11:13:29 <danbri> No, I'm just doing a little prep and advanced warning. The agenda's pretty rough, and I'm not practiced at trying to chair IRC meetings

11:14:01 <danbri> B:|GeoInfo chat 1400 UTC (3pm BST) TODAY

11:14:35 <dc_rdfig> Titled item B.

11:14:47 <crysflame> B:that's 7am, for the pacific west coast people who can't do math that early

11:14:52 <danbri> one chump per agenda item...

11:14:56 <danbri> i guess

11:15:13 <crysflame> hm, this chump is differently tuned. okay.

11:15:15 <danbri> Yes, pretty early for them... I forget why we ended up earlyish.

11:15:20 <dc_rdfig> Added comment B1.

11:15:27 <danbri> which chump are you used to, crysflame?

11:15:35 <crysflame> ah, nevermind. the unlagged sort.

11:15:38 <danbri> Oh, I wasn't complaining re B:

11:15:42 <crysflame> i thought it rejected my comment

11:15:43 <danbri> ah right

11:16:07 <danbri> no, the internet's in slow mode. Someone must've knocked the turbo button by accident.

11:17:07 <danbri> BLURB:AGENDA - review [http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0300.html|actions]

11:17:07 <dc_rdfig> C: AGENDA - review [http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0300.html|actions] from danbri

11:17:39 <danbri> BLURB: AGENDA (D:) Libby's SVG demo

11:17:39 <dc_rdfig> D: AGENDA (D:) Libby's SVG demo from danbri

11:18:08 <danbri> OK, we're out of agenda material now. Suggestions welcomed... Oh no, one more.

11:19:27 <danbri> BLURB: AGENDA (E:) Relationship to GML, [http://www.mapbureau.com/notes/gml.html writeup from Chris Goad] ([http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0301.html|email msg])

11:19:27 <dc_rdfig> E: AGENDA (E:) Relationship to GML, [http://www.mapbureau.com/notes/gml.html writeup from Chris Goad] ([http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0301.html|email msg]) from danbri

11:23:18 <danbri> BLURB: AGENDA (F:) Geo/SVG/etc project idea brainstorm and wishlist (msg from [http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0306.html Charles] re student project ideas).

11:23:19 <dc_rdfig> F: AGENDA (F:) Geo/SVG/etc project idea brainstorm and wishlist (msg from [http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0306.html Charles] re student project ideas). from danbri

11:23:51 <danbri> Suggestions, anyone else?

11:24:21 <zool> i'd like to mention my noderunner stuff and talk about the nocat connection; i dont know if that merits a whole item

11:24:44 * zool sketching up a protoype but wont have anything adhoc / dynamic by 3pm

11:24:54 * danbri wonders about etcon debrief (at least sharing slides/links here again)

11:25:05 <crysflame> what's noderunner? i've been seeing it a lot

11:25:26 <danbri> you don't have to build it all by 3pm! I propose adding a 'heads up' agenda item on that...

11:26:02 <danbri> BLURB: AGENDA (G:) Heads up from zool re noderunner and nocat developments

11:26:03 <dc_rdfig> G: AGENDA (G:) Heads up from zool re noderunner and nocat developments from danbri

11:26:07 <zool> . http://www.noderunner.com - a wireless mapping game - i'm adjusting my rdf map model so that it can serve as a backend for noderunner annotation, and collect gps traces from a game i'm planning to run in london in 2 weeks

11:26:30 <zool> gps, spatial annotation, isomorphic mud world, and fun+games! ;)

11:26:41 <crysflame> mm, nice

11:26:46 <crysflame> you've seen the assassin people then

11:26:57 <crysflame> also, sounds like a good use for ntags.

11:26:59 <danbri> F:See Msg from [http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0306.html Charles] re student project ideas).

11:26:59 <dc_rdfig> Added comment F1.

11:27:31 <danbri> E:see [http://www.mapbureau.com/notes/gml.html writeup from Chris Goad] ([http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0301.html|email msg])

11:27:32 <dc_rdfig> Added comment E1.

11:27:47 <danbri> F1:See Msg from [http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0306.html|Charles] re student project ideas).

11:27:48 <dc_rdfig> Replaced comment F1.

11:28:10 <danbri> F:|Geo/SVG/etc project idea brainstorm and wishlist (msg from [http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0306.html Charles] re student project ideas).

11:28:10 <dc_rdfig> Titled item F.

11:28:19 <danbri> F:|Geo/SVG/etc project idea brainstorm and wishlist

11:28:20 <dc_rdfig> Titled item F.

11:28:26 * danbri getting there, sorry for the noise!

11:28:50 <danbri> E:|Relationship to GML

11:28:50 <dc_rdfig> Titled item E.

11:29:14 <danbri> E1:see [http://www.mapbureau.com/notes/gml.html|writeup from Chris Goad] ([http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0301.html|email msg])

11:29:15 <dc_rdfig> Replaced comment E1.

11:30:07 <danbri> C:|AGENDA (C:) - review actions

11:30:07 <dc_rdfig> Titled item C.

11:30:37 <danbri> F:| (F:) Geo/SVG/etc project idea brainstorm and wishlist

11:30:37 <dc_rdfig> Titled item F.

11:30:47 <danbri> E:| (E:) Relationship to GML

11:30:47 <dc_rdfig> Titled item E.

11:34:59 <danbri> ok, libby has an agenda item, swad-europe geo thingy for mapping SemWeb developers.

11:35:27 <danbri> libby, can you add a blurb to the agenda.

11:35:55 <libby> sure

11:36:39 * danbri does some dodgy mental arithmetic. Try for 10 minutes per item, which leaves 30 mins in reserve to spend on going overtime if needed.

11:37:05 <libby> BLURB: (H:) Semantic Web Developer Map: representing locations of people, research groups and projects

11:37:05 <dc_rdfig> H: (H:) Semantic Web Developer Map: representing locations of people, research groups and projects from libby

11:38:02 <danbri> can you add a link to it?

11:38:08 <libby> yep

11:38:43 <danbri> dc_rdfig:view

11:38:43 <dc_rdfig> D: AGENDA (D:) Libby's SVG demo (blurb)

11:38:44 <dc_rdfig> E: (E:) Relationship to GML (blurb)

11:38:45 <dc_rdfig> F: (F:) Geo/SVG/etc project idea brainstorm and wishlist (blurb)

11:38:46 <dc_rdfig> G: AGENDA (G:) Heads up from zool re noderunner and nocat developments (blurb)

11:38:47 <dc_rdfig> H: (H:) Semantic Web Developer Map: representing locations of people, research groups and projects (blurb)

11:38:49 <libby> H:a[http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0252.html|plea for data] and [http://www.w3.org/2001/sw/Europe/200303/geo/intro.html|notes on how we might do it]

11:38:49 <dc_rdfig> Added comment H1.

11:40:06 <libby> H:[svg demos and queries|http://www.w3.org/2001/sw/Europe/200303/geo/intro.html#demos], [http://esw.w3.org/topic/AnRdfHarvesterStartingPoint|where to put your file]

11:40:10 <dc_rdfig> Added comment H2.

11:40:17 <libby> that ok?

11:40:41 <crysflame> MORE LINKAGE

11:40:54 <crysflame> that's a lot of linkage. i'm happy, as an aside

11:42:59 <libby> H:see also [http://jibbering.com/foaf/foaf-people-map.svg|Jim ley's SVG map of foaf nearest airport data]

11:43:02 <dc_rdfig> Added comment H3.

11:43:22 <libby> H won't take long danbri, I'm just begging for data :)

11:43:30 <crysflame> mm, that's all foafy goodness

11:43:50 <crysflame> i suppose it's relevant to point out that i just integrated the xmlns:geo namespace into my foaf file

11:44:08 <crysflame> so now i'm ICBM-targetable; a lot more accurate than the nearest airport.

11:44:24 <crysflame> i don't know if i did it right, but parsers seem to be eating it well.

11:44:38 <libby> yeah....nearestairport is a sort of hack, but a useful one

11:44:39 <danbri> crysflame, cool... got a pointer?

11:44:42 <crysflame> oh, hah

11:44:43 <crysflame> The eventual aim is to represent the location of people and groups more accurately using lat/long coordinates

11:44:46 <crysflame> quote unquote

11:44:52 <crysflame> danbri: www.crystalflame.net/foaf.rdf

11:47:13 <crysflame> oh, libby, you're the author. right. so no need to go find your email to tell you about that url. hee.

11:47:25 <crysflame> nice paper, by the way. reads well.

11:50:26 <libby> thanks :)

11:52:08 <libby> your file won;t show up on my visualization yet, but I'll hack it later in the week (at the moment only looks for nearestairports with a lat/long, but lots of foafers just have nearestaiport data, or like you, lat/long data for themselves)

11:52:23 <crysflame> i can add lat/long to my nearestairportdata?

11:52:25 <crysflame> how so?

11:52:33 * crysflame skips backwards

11:53:30 <crysflame> ah, hm

11:53:59 <libby> like this: http://jibbering.com/rdf/airports.1?EUG

11:54:04 * crysflame tries to find the lat

11:54:18 <libby> lat/long using http://jibbering.com/rdf/airports.1?EUG

11:54:38 <libby> as I say, I can hack it anyway, it's just esier for me like this

11:54:38 <crysflame> aaaugh

11:54:48 <crysflame> that is SO USEFUL

11:55:02 <crysflame> uh, how do i integrate this into my foaf

11:55:09 * crysflame looks flustered

11:55:19 <crysflame> do you require xmlns:pos vs xmlns:geo?

11:55:19 <libby> e.g <contact:nearestAirport>

11:55:19 <libby> <airport:Airport rdf:about='http://www.daml.org/cgi-bin/airport?BRS'>

11:55:19 <libby> <airport:iataCode>BRS</airport:iataCode>

11:55:19 <libby> <airport:icaoCode>EGGD</airport:icaoCode>

11:55:19 <libby> <pos:lat>51.383333333333</pos:lat>

11:55:20 <libby> <pos:long>-2.7166666666667</pos:long>

11:55:22 <libby> </airport:Airport>

11:55:24 <libby> </contact:nearestAirport>

11:55:27 * danbri thinks about the lat/long in http://www.crystalflame.net/foaf.rdf ...

11:55:50 <libby> as long as it points to http://www.w3.org/2003/01/geo/wgs84_pos#, doesn;t matter whether call it geo or what

11:55:54 <crysflame> i'd eventually like that data to be live-updated, but it's static (and fuzzed, precision-wise) for now.

11:55:56 <crysflame> okay, cool

11:56:01 <danbri> crysflame, the way you've used foaf:topic with value of the lat/lon'd point makes sense when applied to documents, but not so much with people...

11:56:03 <crysflame> this newfangled xmlns thing

11:56:11 <danbri> (I'm not complaining, its a reasonably subtle point...)

11:56:25 <danbri> a foaf:topic relates a document to a thing that the document is about.

11:56:34 <danbri> So you implicitly assert that you are a Document.

11:56:39 <crysflame> danbri: i dislike the topic involvement here entirely. can i just move the lat/long out of the container?

11:56:47 <crysflame> couldn't figure out what topic was, but it seemed necessary

11:56:51 <danbri> And the FOAF schema says that Document and Person are disjoint classes, ie. there are no things that are both.

11:57:03 <crysflame> heh

11:57:04 <danbri> So... smart Semweb tools should be able to detect the contradiction.

11:57:16 * crysflame raises the bar

11:57:17 <danbri> I'm not sure what there is online currently which could've spotted this though...

11:57:20 <danbri> :)

11:57:23 <crysflame> hee

11:57:35 <danbri> Libby, does your doc propose markup to attach geo lat/long to people?

11:57:39 <zool> can't i be a document in some derridean para-world?

11:57:40 <crysflame> it'd be invalid foaf in its current form, right?

11:57:44 <libby> hm

11:57:47 <danbri> Zool, No.

11:57:48 <danbri> ;)

11:57:54 <danbri> well, maybe...

11:57:58 <libby> only nearestairport

11:58:03 * danbri waves bye; -> lunch

11:58:22 <crysflame> you say right at the beginning that you'd like to see it documented closer, which is horribly serendipituous.

11:58:29 * crysflame waves hi

11:59:03 <danbri> re taking it out of the container, basically need to pick some relations, like 'livesNear' to link the place and the person.

11:59:13 <danbri> RDF insists on being explicit about relations...

11:59:15 <crysflame> hrm. okay.

11:59:27 <danbri> the geoonion thing might help. bbiab.

11:59:29 <libby> I did it with a group - CCLRC because they gave me coords: - <pos:within100mof rdf:parseType="Resource">

11:59:40 <crysflame> oo, i like that

11:59:41 <crysflame> perfect

11:59:48 <libby> erk - oh well see http://www.w3.org/2001/sw/Europe/200303/geo/CCLRCSWGroup.rdf

12:00:02 <libby> but I wouldn;t necessarily recommend it - spur of the moment thing - geoonion

12:00:11 <libby> oop, lunch, back in a bit

12:00:12 <danbri> cf http://esw.w3.org/topic/GeoOnion

12:00:22 <libby> cheers crysflame :)

12:00:45 <crysflame> ah, draft xml schema

12:01:05 <crysflame> somewhere

12:02:19 <danbri> f***, lightening! hailstones!

12:02:28 <crysflame> doom!

12:02:35 <crysflame> lunch will be denied.

12:02:41 <danbri> worse, lightening and hailstones between me and food...

12:05:12 <crysflame> <pos:within10mof rdf:parseType="Resource">

12:05:12 <crysflame> <pos:lat>44.0728</pos:lat>

12:05:12 <crysflame> <pos:long>-123.0207</pos:long>

12:05:13 <crysflame> <pos:alt>150</pos:alt>

12:05:13 <crysflame> </pos:within100mof>

12:06:48 <crysflame> golly, that's a lot of namespace includes.

12:06:52 <crysflame> this is like writing perl, only uglier.

12:09:55 <crysflame> WOOHOO

12:10:07 <crysflame> i now have two GeoURL tags showing up in the foaf explorer

12:10:08 <crysflame> *awesome*

12:11:31 <zool> the pos:* is GeoOnion ? nice

12:11:43 <crysflame>http://xml.mfd-consult.dk/foaf/explorer/?foaf=http://www.crystalflame.net/foaf.rdf

12:11:44 <dc_rdfig> I: http://xml.mfd-consult.dk/foaf/explorer/?foaf=http://www.crystalflame.net/foaf.rdf from crysflame

12:11:45 <crysflame> oh hell

12:11:49 * crysflame bats forehead

12:11:59 <crysflame> I:|Richard Soderberg's FOAF, explored

12:12:00 <dc_rdfig> Titled item I.

12:12:14 <crysflame> I:The GeoURL links are interesting :)

12:12:16 <dc_rdfig> Added comment I1.

12:12:38 <crysflame> libby: thanks for the example.

12:18:29 <zool> i can never remember what namespace is canonically used for ical/cal stuff

12:19:18 <zool> ah, http://www.w3.org/2002/12/cal/ical

12:20:08 * crysflame is apparently using that namespace

12:34:59 <mortenf> E:seeAlso [http://esw.w3.org/topic/InterpretationProperties|InterpretationProperties] regarding conversion issues.

12:34:59 <dc_rdfig> Added comment E2.

13:09:02 <zool> G:[http://makeashorterlink.com/?D2DA14464|nocat maps], example output from [http://mapserver.gis.umn.edu/|mapserver]; nice things that can be done with open-source GIS; outputs PNG, but SVG output is near-future plausible

13:09:02 <dc_rdfig> Added comment G1.

13:11:03 <crysflame> fwiw, masl.to is a cname for makeashorterlink.com

13:11:30 * zool smiles knows this yes

13:14:34 <zool> G:[http://www.noderunner.com/|noderunner], the wireless node annotation adventure game, of which planning a [http://twenteenthcentury.com/uo/index.php/CCNoderunnerLondon|london version]

13:14:35 <dc_rdfig> Added comment G2.

13:25:16 <_joshua> hi

13:25:59 <crysflame> morning!

13:26:15 * crysflame grins

13:26:34 <crysflame> got the foaf geourl thing worked out. now the foaf browsers see the pos: tags and give me green clicky links.

13:26:46 * crysflame does the small happy dance.

13:26:56 <crysflame> yay to #rdfig #foaf for being helpful, tolerant

13:28:25 <zool> morning _joshua

13:29:10 <_joshua> heh

13:29:14 <_joshua> hey jo

13:29:20 <zool> G:sketch of [http://space.frot.org/rdf/noderunner_sketch.rdf|noderunner rdf] to be built into [http://space.frot.org/mudlondon.html|mudlondon's] API

13:29:20 <dc_rdfig> Added comment G3.

13:30:01 <zool> G:[http://space.frot.org/rdf/noderunner_sketch.png|graphviz-ualisation] of a noderunner route - comments apprec.

13:30:01 <dc_rdfig> Added comment G4.

13:30:10 * zool just pre-meeting braindumping

13:31:55 <_joshua> the noderunner stuff was amusing

13:39:48 * zool nods, a london one should be fun, even though we don't have anything like the wireless density of NYC... then can do wireless reach maps, gps trace visualisations of streets, rdf connection model enhancement, all in one go! (but hide all that complexity in the game)

13:40:15 <mattb> games are good

13:40:22 <mattb> learning is fun! :)

13:40:22 <mortenf> comments re the viz: runner's don't connect to persons?

13:40:35 <zool> don't they? oops

13:40:35 <mortenf> brb

13:41:24 <zool> ah, doh; cheers, fixing

13:53:36 <zool> was it something we said?

13:54:51 <chrisgoad> e:updated the [notes|http://www.mapbureau.com/notes/gml.html] about my obvious proposal; 2nd paragraph now states it concisely

13:54:51 * danbri back

13:55:16 <libby> chisgoad, you need to use capital E I think

13:55:23 <libby> chrisgoad, you need to use capital E I think

13:55:51 <chrisgoad> Thanks, libby. I'll try again.

13:56:21 <libby> .time UTC

13:56:22 <datum> 2003-04-30T13:56:21

13:56:40 <chrisgoad> E:updated the [notes|http://www.mapbureau.com/notes/gml.html] about my obvious proposal; 2nd paragraph now states it concisely

13:56:40 <dc_rdfig> Added comment E3.

13:56:41 * mortenf is back, though still at work, only partially present...

13:59:05 <danbri> Folks, I didn't finish my action on chefmoz data example... but need help anyways: can someone London-y recommend a restaurant/pub that is in both Chefmoz and Grubstreet?

13:59:16 <danbri> .time UTC

13:59:17 <datum> 2003-04-30T13:59:17

13:59:48 <danbri> Non-Geo folks, we're about to take over the channel for RDF/Geo chat for 90 mins. Thanks for your patience and passing interest!

13:59:57 <danbri> see http://rdfig.xmlhack.com/

14:01:21 <libby> C:[actions email|http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0300.html]

14:01:21 <dc_rdfig> Added comment C1.

14:01:31 <danbri> .time UTC

14:01:32 <datum> 2003-04-30T14:01:31

14:01:37 <danbri> OKay, getting started.

14:01:40 <danbri> =======================================================================

14:01:58 <danbri> ============= RDF/GeoInfo chat, agenda: http://rdfig.xmlhack.com/

14:02:14 <danbri> Hi everyone. thanks for coming...

14:02:29 <danbri> Let's have folk sign up using the IRC/weblog bot, so we know who's here...

14:02:38 <libby> too B:?

14:02:41 <danbri> B:Attending...

14:02:42 <dc_rdfig> Added comment B2.

14:02:53 <danbri> B:[http://www.w3.org/People/DanBri/|Dan Brickley]

14:02:55 <dc_rdfig> Added comment B3.

14:02:58 <danbri> (like that...:)

14:03:12 <zool> B:jo walsh

14:03:13 <libby> B:[Libby Miller|http://ilrt.org/people/libby]

14:03:13 <dc_rdfig> Added comment B4.

14:03:14 <dc_rdfig> Added comment B5.

14:03:19 <mortenf> B:Morten Frederiksen

14:03:19 <dc_rdfig> Added comment B6.

14:03:23 <danbri> Just write B:Your Name, if you don't have a homepage. And do sign up even if lurking, it helps us know if folks are interested!

14:03:26 <crysflame> B:[http://www.crystalflame.net/|Richard Soderberg]

14:03:26 <dc_rdfig> Added comment B7.

14:04:07 <chrisgoad> B:[Chris Goad|http://www.mapbureau.com/cgfoaf.rdf]

14:04:07 <dc_rdfig> Added comment B8.

14:04:29 <mattb> B:[http://www.hackdiary.com|Matt Biddulph]

14:04:29 <dc_rdfig> Added comment B9.

14:04:40 <hiramatu> B:Kaoru Hiramatsu

14:04:40 <dc_rdfig> Added comment B10.

14:05:02 <danbri> Brief comment on what/why/etc, and scope. This is one of several topical chats (see http://esw.w3.org/topic/ScheduledTopicChat) we've been having in the RDF Interest Group, picking topics that have grabbed the attention of a lot of RDF, XML, Semantic Web developers.

14:05:48 <danbri> It's a 'show and tell' thing, get to find what others have been doing, and collaborate on things that help our systems/data/etc work together.

14:06:27 <danbri> Any comments on the agenda?

14:06:40 <chaasBRS> B:[http://www.w3.org/People/Charles|Charles McCathieNevile]

14:06:40 <dc_rdfig> Added comment B11.

14:06:46 <Cedric_> Cédric Kiss (Chaals' intern)

14:06:53 <libby> does everyone know about the geowankers list?

14:06:54 <crysflame> -- RDF/GeoInfo chat, agenda: http://rdfig.xmlhack.com/

14:07:07 <danbri> the listing at http://rdfig.xmlhack.com/ is in reverse order. You can add comments, links under each item if you know it's name. Eg. the first item is called C:

14:07:13 <chaasBRS> B: Cédric Kiss

14:07:14 <dc_rdfig> Added comment B12.

14:07:28 <grault> grault is now known as earle

14:07:30 <danbri> Is DanC here?

14:07:43 <crysflame> agenda item to discuss geowankers?

14:07:50 <earle> B:[Earle Martin|http://c2.com/cgi/wiki?EarleMartin]

14:07:50 <dc_rdfig> Added comment B13.

14:07:52 <crysflame> or just link it and describe

14:07:55 <libby> or maybe just to say it's there....

14:07:57 <libby> yeah

14:08:03 <crysflame> let me get the link, archived

14:08:16 <danbri> OK, let's cover that under actions, as DanC had an action re mailing lists.

14:08:20 <danbri> ACTIONS

14:08:23 <danbri> C:

14:08:24 <dc_rdfig> blurb

14:08:25 <dc_rdfig> AGENDA (C:) - review actions

14:08:26 <dc_rdfig> (1:libby) [actions email|http://lists.w3.org/Archives/Public/www-rdf-interest/2003Apr/0300.html]

14:09:02 <crysflame> hm.

14:09:12 <danbri> C:Continuing [http://rdfig.xmlhack.com/2003/04/16/2003-04-16.html#1050505090.657704|action DanC discussion of time/place/people/documents/money].

14:09:12 <dc_rdfig> Added comment C2.

14:09:27 <crysflame>http://lists.burri.to/mailman/listinfo/geowanking

14:09:27 <dc_rdfig> J: http://lists.burri.to/mailman/listinfo/geowanking from crysflame

14:09:34 <crysflame> J:|Geowankers list

14:09:34 <dc_rdfig> Titled item J.

14:10:28 <danbri> C:DanC isn't here, but I can say a few words. Last time we discussed setting up a general forum for RDF/SW vocab discussion rather than a new w3c-hosted geo specific list. After discussion I believe it was decided not to do this while the new Semantic Web Activity (phase 2) is being designed...

14:10:29 <dc_rdfig> Added comment C3.

14:11:07 <danbri> C:In particular, [http://www.w3.org/2001/sw/interest/charter.html|a new Semantic Web Interest Group charter] is being drafted for discussion.

14:11:07 <dc_rdfig> Added comment C4.

14:11:31 <libby> C:[perhaps use this Geo list instaed?|http://lists.burri.to/mailman/listinfo/geowanking]

14:11:31 <dc_rdfig> Added comment C5.

14:11:32 <danbri> C:In meantime, we have (see J:) Joshua's new list, and possibility of hosting a SW/RDF geo vocab list at W3C.

14:11:32 <dc_rdfig> Added comment C6.

14:12:16 <danbri> I don't really mind where I discuss things so long as there archives are persistent and URL addressable...

14:12:33 <danbri> (I do hate the 'geowankers' name though, I think it'll scare off some folks... sorry joshua!)

14:12:36 <crysflame> i don't want to see traffic split between two lists

14:12:54 * crysflame remembers the debate attached to smokers@perl.org

14:13:06 <danbri> Joshua, any comments on geowankers?

14:13:21 * danbri should probably have made this a separate item rather than getting into detail during actions. ho hum.

14:13:28 <zool> i like the idea of their being a dedicated semweb list for discussing converging vocab; it's not so much the geo layer and what can be built when it's there...

14:13:36 <zool> s/and/but/

14:13:43 * chaasBRS wonders if we have a zakim bot here for doing agendas

14:13:45 * crysflame nods.

14:13:48 <libby> next action item?

14:13:48 <libby> [[ACTION: danbri produce some sample chef-moz based data using

14:13:48 <libby> onion + wgs_84 vocab.]]

14:13:52 <crysflame> that's a nice seperation.

14:13:58 <danbri> Let's add this to end of the agenda...? There does seem a lot of energy and interest around geo discussions.

14:14:37 <danbri> C:Continuing ACTION danbri produce some sample chef-moz based data using onion + wgs_84 vocab

14:14:38 <dc_rdfig> Added comment C7.

14:15:08 <danbri> C:I didn't get much done on this, got sick last week. I did get my harvester to slurp up all the Grubstreet data (incl geo stuff) though.

14:15:09 <dc_rdfig> Added comment C8.

14:15:26 <danbri> (Libby reports problems with IRC laggyness btw.)

14:15:41 <danbri> Can someone (or several) confirm they can read >this< ?

14:15:42 * chaasBRS not seeing them (sitting next to lib)

14:15:46 <mortenf> this.

14:15:48 * chaasBRS can

14:15:53 <danbri> ok thanks

14:15:57 * shellac can

14:16:35 * dajobe checks logs

14:16:37 <danbri> [

14:16:38 <danbri> 3. http://rdfig.xmlhack.com/2003/04/16/2003-04-16.html#1050501564.795588

14:16:38 <danbri> DanC: ACTIOn libby write a converter from chefmoz opening housr format

14:16:38 <danbri> to rdf calendar format (or icalendar format maybe?) [continues]

14:16:38 <danbri> geodanbri: Continued, ACTION danbri to have a draft of the 'concentric

14:16:40 <danbri> circles'

14:16:42 <danbri> ]

14:16:51 <danbri> Libby, any progress with that?

14:16:53 <dajobe> the logs http://ilrt.org/discovery/chatlogs/rdfig/2003-04-30.html are live (per minute)

14:17:09 <libby> nope, sorry

14:17:17 <libby> it's on two action lists now :(

14:17:28 <danbri> C:No progress from libby on chefmoz opening hours format. Continued.

14:17:28 <dc_rdfig> Added comment C9.

14:17:47 <danbri> C:DanBri [http://esw.w3.org/topic/GeoOnion|GeoOnion] draft up for comments.

14:17:48 <dc_rdfig> Added comment C10.

14:18:02 <danbri> [

14:18:02 <danbri> 4. http://rdfig.xmlhack.com/2003/04/16/2003-04-16.html#1050501597.408405

14:18:02 <danbri> libby: ACTION libby translate a sample bus timetable to zool/jim's

14:18:02 <danbri> formats, feedback to some list (www-rdf-calendar maybe?)

14:18:03 <danbri> ]

14:18:08 <danbri> Libby?

14:18:28 <libby> yep, sorta, it's down as an agenda item...(D)

14:18:46 <danbri> So shall we mark this one DONE and move on to the meeting meeting?

14:18:52 <libby> yeah

14:19:04 <danbri> I've seen it and can vouch for its done-ness.

14:19:27 <danbri> C:COMPLETED ACTION: libby translate a sample bus timetable... (see D: item)

14:19:28 <dc_rdfig> Added comment C11.

14:19:30 <danbri> ===

14:19:37 <danbri> D: Libby's SVG demo.

14:19:38 <dc_rdfig> Added comment D1.

14:19:57 <danbri> Libby, wanna take the floor? approx 10 mins, 15 if you get the crowds fired up... ;)

14:19:59 <libby> D:[SVG+js (based on Jim Ley's code Bus and train route visualization|http://sw1.ilrt.org/discovery/2003/04/svggeo/jj/travels/bristol-bus.svg?url=http%3A%2F%2Fswordfish.rdfweb.org%2F%7Elibby%2Fbusmap%2Falltimetables.rdf]

14:19:59 <dc_rdfig> Added comment D2.

14:20:05 <libby> ok

14:20:10 <libby> not sure I need so long

14:20:26 <danbri> Can you give some context for folk who weren't here before...

14:20:34 <libby> basiccally I want to create RDF data describing bus and train routes and fastest paths betweejn them.

14:20:38 <libby> sure

14:21:22 <libby> D:[what I want to do with bus/trai data|http://www.ilrt.bris.ac.uk/discovery/2003/01/public-transport/]

14:21:23 <dc_rdfig> Added comment D3.

14:21:35 * danbri nods

14:21:36 <crysflame> !

14:22:05 <libby> so the idea is that with timetables and geo points in RDF, we gcen get the data in an accessible way to people who don;t know their way around a place (bristol in this case)

14:22:13 <danbri> D:Background on [http://www.w3.org/Graphics/SVG/Overview.htm8|W3C Scalable Vector Graphics (SVG)]: it's W3C's XML based language for 2-d graphics in XML.

14:22:14 <dc_rdfig> Added comment D4.

14:22:31 <libby> with the added advantage that bus companies etc can see what routes people want to uuse and espond accordingly

14:22:38 <danbri> Can anyone confirm they can see the demo working?

14:22:44 <danbri> (will need Adobe SVG plugin)

14:23:01 <libby> it's probbaly a bit duff, as my javascript and SVG as dodgy

14:23:20 <mortenf> looks fine

14:23:25 <danbri> D:Ted (tguild) pointed me at this [http://www.theregister.co.uk/content/59/30454.html|piece on taxis and gps] earlier.

14:23:25 <dc_rdfig> Added comment D5.

14:23:35 <libby> anyway, so I haven;t written it yet, but I'm going to do a shortest path demo with it

14:23:38 <danbri> Libby, what're plans for it next?

14:23:41 <danbri> shortest in what sense?

14:23:45 * crysflame tries to find a shortest path link..

14:23:58 <libby> shortest in time

14:24:02 <danbri> quickest strategy for getting someplace versus ...

14:24:28 <danbri> it'd be nice to see some more of the rdf-calendar work mixed in, for reasoning about things like making connections (bus/train) and suchlike.

14:24:28 <libby> i.e. tere are 3 ways of getting from A to B, what's the best given next bus/train, time of journey in each case

14:24:40 * chaasBRS just getting a page of text :(

14:24:43 <libby> yep, there would be ifmy perl didn;t suck so much

14:24:46 <zool> libby, that is looking lush

14:25:08 <libby> thanks zool :)

14:25:21 <libby> my real probalem will be finding the GPS points for busstops!

14:25:23 <danbri> its very purty, indeed!

14:25:29 * chaasBRS having brain problems.

14:25:33 <libby> any ideas on how to dop that (or map coords) would be neat

14:25:46 <danbri> Has anyone studied the bit of the SVG spec relating to GIS?

14:25:50 * danbri digs for reference

14:26:05 * chaasBRS saw it before and it looked good.

14:26:15 <danbri> can't find link :(

14:26:16 <crysflame> put people on the bus, mapping points

14:26:20 <libby> zool, this also sounds like noderunner maybe?

14:26:26 <chaasBRS> This is related to the projects item of the agenda.

14:26:27 <crysflame> also, this is data that bus organizations may keep around

14:26:29 <zool> where can i see the underlyng rdf model?

14:26:46 <danbri> Libby, can you say a bit about how you go from having the lat/long/alt of things to getting them to show up like this? what do you need to know about a map to plug it into the code?

14:26:53 <_joshua> Is this the meeting already?

14:27:00 <crysflame> yes

14:27:00 <danbri> yes!

14:27:04 <chaasBRS> (Particularly to what Cedric is going to try and work on)

14:27:09 * earle can't see the svg demo, Adobe SVG dies (old story) and sodipodi doesn't like it either...

14:27:14 <libby> D:[RDf example|http://www.ilrt.bris.ac.uk/discovery/2003/01/public-transport/]

14:27:14 <dc_rdfig> Added comment D6.

14:27:21 <_joshua> oops

14:27:29 <danbri> I have trouble w/ latest mozilla and svg on my desktop. looks fab on laptop though.

14:27:47 <libby> hm, danbri, I very cruftily did the scaling by hand by trial and error

14:27:48 <danbri> shame bristol isn't in london or it'd fit well w/ zool's tubemap model

14:28:08 <libby> it's all based on jim ley's code

14:28:09 <_joshua> is there any agenda?

14:28:13 <zool> my model can theoretically extend england-wards now i have rewritten the interface

14:28:14 * danbri invites geo-ish people to sign up under B:

14:28:22 <chaasBRS> map: need to believe in the projection (using mercator projection is nice because you get straight scales to work with graphically (but they don't match distance scales)

14:28:25 <danbri> joshua, see http://rdfig.xmlhack.com/

14:28:25 <zool> just be able to state /a_space/Bristol/Foo

14:28:27 <libby> _joshua, see http://rdfig.xmlhack.com

14:28:42 <zool> except i still havent resolved naming clashes for streets within one town :/

14:28:50 <danbri> (libby, that was 9 minutes...)

14:28:50 <_joshua> zool: i am having that problem

14:28:53 <libby> it's tricky

14:28:59 <mortenf> yep.

14:29:19 <_joshua> I noticed in California, El Camino Real reuses street addresses in neighboring towns

14:29:19 <libby> ok, well, I just wanted to show it people; I want to extend it to different places, maybe get some funding for it.

14:29:21 <danbri> can you get away without using unique ids for streets? and just have their descriptions?

14:29:28 <nwalsh> Is there a "standard" vocabulary for geographic place names?

14:29:30 <danbri> yes, funding's always nice...

14:29:35 <zool> apart from arbitrary numeric diffrentiators (which i think danbri's wordnet namespace does for senses?)

14:29:39 <danbri> hi norm...

14:29:49 <nwalsh> hi danbri et. al. :-)

14:29:49 <_joshua> think about how you would explain it to another person

14:29:55 <danbri> (yes, zool... I use wordnet's sense numbers plus the term to disambiguate).

14:29:58 <_joshua> "60th street, between columbus and amsterdam"

14:30:30 <danbri> (can someone dig out the SVG spec that talks about GIS? I just can't find it).

14:30:49 <_joshua> I wasn't aware one existed

14:31:07 <zool> in some cases i have used postcode as a suffix

14:31:08 <danbri> OK, we should move on, unless folks have more questions for Libby now?

14:31:18 <danbri> joshua, i should've said bit of spec...

14:31:19 <libby> thanks all, advice welcomed

14:31:19 <zool> but that gives arbitrary priority to the first street to go it

14:31:31 <_joshua> on svg and gis: http://www.directionsmag.com/article.php?article_id=198

14:31:52 <danbri> libby, a server-side version would rock. Too few people have SVG plugin currently, and many developery people use Moz, which has issues sometimes.

14:32:00 <danbri> moving on...

14:32:17 <danbri> ==== E: Relationship to GML

14:32:20 <danbri> - http://rdfig.xmlhack.com/2003/04/30/2003-04-30.html#1051701567.272332

14:32:37 <danbri> Chris raised this, and it was discussed briefly last time.

14:33:06 <_joshua> hmm, http://www.svgmapper.com/

14:33:08 <danbri> GML is a big, professionally put together and very rich XML representation of Geographic/mapping/etc info.

14:33:14 <mortenf> chris, did you see the conversion/InterpretationProperties link?

14:33:38 <danbri> It was originally done as an RDF app, and retains a very RDFish feel at the instance data level. But is specified in XML and doesn't currently parse to RDF.

14:33:48 <chrisgoad> Yes, I looked at that, and commented to Jim Ley about it. I'll add a comment

14:34:16 <danbri> Meanwhile, in the RDF IG and nearby we've been playing with relatively lightweight (even scruffy) RDF vocabularies for geo-related info, with an emphasis on mixing geo with other data (people, events, social, document etc).

14:34:34 <danbri> hi Chris... want to say a bit about your doc you circulated?

14:34:48 <chrisgoad> E:[about interpretatedValues vs interpretationproperies | http://www.mapbureau.com/notes/interpretedvalues.html]

14:34:49 <dc_rdfig> Added comment E4.

14:35:11 <chrisgoad> Well, I just wanted to state an obvious idea explicitly. Geometry should

14:35:27 <chrisgoad> not be a part of a GEO Rdf spec, but should be it's own RDF language

14:35:32 <danbri> Chris reminds us that there's potential for richer integration. I wouldn't want us to replicate the hard work of the GML folk, but rather, help figure out how data in their format could be represented within RDF, for mixing with other SemWeb data streams.

14:35:42 <chrisgoad> to be used for many purposes. One idea for this is to "port" the

14:35:50 <chrisgoad> geometry part of SVG (eg) to RDF.

14:35:53 <danbri> Polygons and suchlike...

14:36:23 <chaasBRS> similar to the way that Jim Ley used SVG stuff for identifying bits of pictures?

14:36:33 <danbri> So this is interesting to me, in that we've in a different context been using fragments of SVG inside RDF graphs to represent overlays on photos.

14:36:34 <chrisgoad> Luckily, GML is "striped" already, which may help integrate with RDF

14:36:51 <chaasBRS> slurped into RDF namespaces, but effectively able to be dropped straight back into SVG

14:36:51 <danbri> Yes, chaals, exactly. Your hassling me into learning Amaya's SVG being an ancestor of that...

14:37:50 <danbri> so http://www.mapbureau.com/notes/interpretedvalues.html looks useful, i haven't had a good look/think yet...

14:38:04 <mortenf> chris, it seems there are still problems.

14:38:27 <_joshua> me careful with that mapbureau stuff, it's more about maps than territories

14:38:48 * danbri remembers mortenf was going to look at representing shapes, but didn't get ACTION'd on it explicitly...

14:38:55 <danbri> did you get a chance to look further, morten?

14:39:03 <chrisgoad> Problems with interpreted values, do you mean Morten?

14:39:08 <_joshua> and via a computer you can always project a representation of a territory into some sort of map

14:39:31 <mortenf> ?

14:39:39 <mortenf> danbri, i don't remember that?

14:39:44 <danbri> no worries!

14:40:09 <chrisgoad> What do you mean by territories?

14:40:15 <zool> _joshua, i'm not sure how you differentiate. by maps you are saying geometries? pure, with or without semantics?

14:40:16 <mortenf> chris, yeah, you're right about the n*m properties, but there's no way around it in the rdf model (if you want to be able to do conversions).

14:40:55 <tguild> B:Ted Guild

14:40:55 <dc_rdfig> Added comment B14.

14:41:09 <mortenf> but danbri, i may have said that i'm looking into regions in spacenamespace:dk - in the not-short term.

14:41:11 <chrisgoad> Morten, yes, I thought my m*n argument worked.

14:41:17 <danbri> yep, rdf forces some funny modeling conventions sometimes, due to the binary relation requirement.

14:41:18 <chrisgoad> Glad you agree.

14:41:20 <_joshua> Basically. information about a map is just a vector of data. likewise information about a territory is also a vector of data.

14:41:44 <mortenf> i agree that it's a lot of properties, but it's still the way to do it... :(

14:41:44 <tguild> sorry for not logging in earlier, mostly lurking right now but very interested in rdf geo

14:41:46 <_joshua> you apply a transformation and a filtering on the territory data to get the map.

14:42:01 <danbri> hi ted :)

14:42:12 <chrisgoad> Ok, I see, the territory is the description before projection and clipping, right?

14:42:24 <_joshua> I think that the territory data + transformations + filterings are more valuable than the map itself. Which can be regenerated, redesigned, etc

14:42:28 <danbri> OK we've done 10 minutes on GML-related... Anyone want to spend another 5 minutes on this?

14:42:46 <chrisgoad> Yes, absolutely. The map is derivative data.

14:42:59 <libby> do we have any actions? a way forward?

14:43:01 * chaasBRS thinks he should have read up on GML more before talking about this.

14:43:02 <mortenf> however, if one defined generic properties (xmin, xmax, etc.) the only difference would be the actual namespace, that could be semi-automatigically generated?

14:43:06 <danbri> I'd like to see some next steps on representing geometry more explictly in RDF.

14:43:20 <_joshua> Right. It's important to not confuse these things. We are marking up territories, and generating maps from that.

14:43:27 <_joshua> Traditional GIS is about making maps, in many senses.

14:43:32 <chrisgoad> Danbri, me too - next steps would be good.

14:43:40 <danbri> The representation of points is useful, and things like GeoOnion are handy hacks, but there's so much richness in SVG that I think we could link somehow...

14:43:43 <mortenf> (when's the next meet?)

14:44:00 <libby> motenf, not specified (yet)

14:44:02 <danbri> (next meet: none arranged yet... get your diaries out...)

14:44:14 <_joshua> that's representation. also different

14:44:20 * danbri looks around the see-thru room in search of volunteers

14:44:41 * mortenf wants to chat with chris (some other time) about this (gml linkage vocabs)

14:44:49 <danbri> SVG is representation... in UI sense, but it also has a lot of primitive data structures we could shadow into RDF rather than re-invent slowly and painfully.

14:45:07 <chrisgoad> The maps are part of a linked structure that includes territories and the physical

14:45:14 <danbri> Geo vocab discussions are always welcome here btw! Doesn't need to be in a scheduled meet

14:45:20 <chrisgoad> contents thereof; great RDF app.

14:45:27 <_joshua> Although SVG does let you use raw coordinates and then layer the transformation, so it's closer to my conceptual ideal

14:45:36 <zool> i'd like to get more involvd in the SVg output side of things, help nocat sort out that aspect of exporting from their maspserver/GRASS server

14:45:46 <zool> postGIS and GML... i dont understand the connection fully enough

14:45:55 <danbri> I'm going to dig up that SVG spec reference. Won't be a minute...

14:46:06 <chaasBRS> coordinates plus transformation - this is really important.

14:46:15 <_joshua> look at, for example, the SVG on http://nyc.burri.to/W_72nd_St/Between_Columbus_Ave_And_Amsterdam_Ave

14:46:30 <danbri> morten and chris, any chance we could persuade one of you (with others help) to take an action to progress the mapping to GML thing?

14:46:36 <_joshua> I use the raw measurement dataset (latlon * 1e6) and transform it to a map

14:47:21 <chrisgoad> Sure. I'm trying to work on these issues in any case.

14:47:36 <mortenf> sure, we'll talk later (gotta go in 10).

14:48:01 <danbri> timecheck, we have 4 issues left to cover in ~45 mins.

14:48:04 <mortenf> joint action...

14:48:31 <libby> stick it under E?

14:48:33 <chrisgoad> Yes, good.

14:48:35 * danbri picks one person to own it and one to collaborate, so every action has an owner. That OK?

14:48:43 * danbri flips a coin, nominates Chris as owner...

14:48:58 <chrisgoad> Fine, happy to "own" this one.

14:49:08 <danbri> Just gotta figure out what it is.

14:49:12 <chrisgoad> rights

14:49:17 <chrisgoad> right, I mean.

14:49:38 <mortenf> +1

14:49:40 <danbri> E:ACTION ChrisGoad (with MortenF) to propose next steps re showing interop between GML and RDF/SVG-based systems.

14:49:41 <dc_rdfig> Added comment E5.

14:49:45 <danbri> that cover it?

14:49:51 <chrisgoad> Yes.

14:49:59 <mortenf> oops, how did that SVG enter there? :)

14:50:02 <danbri> E:Extra points for sample data / examples...

14:50:02 <dc_rdfig> Added comment E6.

14:50:09 <mortenf> np.

14:50:23 <danbri> E:Mentioned SVG 'cos its from the W3C family of happily-interoperating specs :)

14:50:24 <dc_rdfig> Added comment E7.

14:50:36 <danbri> E:Hopefully relevant re polygons and suchlike

14:50:37 <dc_rdfig> Added comment E8.

14:50:48 <chrisgoad> Meanwhile, as a last note, any thoughts about porting SVG geometric formalism

14:50:48 <mortenf> easy now...

14:50:50 <chrisgoad> into RDF?

14:51:06 <crysflame> geo wha huh?

14:51:07 <chaasBRS> E: For porting SVG into RDF see the work of Jim Ley on image annotation

14:51:08 <dc_rdfig> Added comment E9.

14:51:09 * crysflame lurks more.

14:51:26 <danbri> Chris, I've _thought_ of it. For now all I've done (and Jim Ley took it further) is to represent SVG paths (which have a flat textual representation anyway) as textual values of RDF properties.

14:51:31 <chrisgoad> Ok, I'll look at that.

14:52:01 <danbri> E:see [http://esw.w3.org/topic/ImageDescription] notes for more SVG/RDF experiments (SVG overlays on photos rather than maps)

14:52:02 <dc_rdfig> Added comment E10.

14:52:08 <danbri> Let's move on...

14:52:16 <danbri> ====== (F:) Geo/SVG/etc project idea brainstorm and wishlist

14:52:25 <danbri> see http://rdfig.xmlhack.com/2003/04/30/2003-04-30.html#1051701799.099352

14:52:39 <chaasBRS> Cedric (who is here) is about to have 4 months to work on a project...

14:53:03 <chaasBRS> on "SVG interfaces for working with / creating RDF Geo data...."

14:53:05 <danbri> Semweb/Geo hacking is one of those areas where everyone wishes there were more hours in the day, so many potentially cool things to build.

14:53:08 <chaasBRS> or something of the sort.

14:53:37 <danbri> So specifically re Cedric, and more generally, I thought having an agenda slot for wishlisting, project brainstorming etc.

14:53:48 <zool> i'm hoping to do an SVG version of the buckminster fuller dymaxion map, based on work that metamute.com have half-done and have full permission for...

14:54:00 <danbri> logger_1, pointer?

14:54:00 <danbri> See http://ilrt.org/discovery/chatlogs/rdfig/2003-04-30#T14-54-00

14:54:19 <danbri> E:see [http://ilrt.org/discovery/chatlogs/rdfig/2003-04-30#T14-54-00|chat logs]

14:54:20 <dc_rdfig> Added comment E11.

14:54:26 <DanC> .time

14:54:26 <datum> Wed, 30 Apr 2003 14:54:26 GMT

14:54:30 * DanC oops

14:54:34 <danbri> hi DanC

14:54:47 <mortenf> i'd like to see an example of a zoomable map that changes the underlying image when getting "closer", i.e. (jim's world map)++

14:54:57 <danbri> (sign-in is at B:)

14:55:07 <danbri> Zool, what's a "dymaxion map"?

14:55:17 * DanC wasn't here, figures the record shouldn't show I was

14:55:20 * danbri intrigued and mystified in equal measure :)

14:55:35 <danbri> (we're running until half-past; you could note partial attendence perhaps)

14:55:49 * DanC boots up...

14:56:08 <zool> danbri, cf http://www.spaceshipearth.org/EARTHSCOPE/earthscope.html

14:56:09 <danbri> (I made some comments re mailing list(s), you could check for accuracy. My last week was a bit blurry)

14:56:29 <zool> the classic bucky fuller squashed globe... an open version (standards, data) of an application that looks like that one

14:56:39 <zool> also cf http://blogosphere.headmap.org

14:56:53 <earle> also http://www.westnet.com/~crywalt/unfold.html (with animations)

14:57:17 <danbri> E:Wishilist: I started hacking a PalmOS app for creating Geo-ready records in Calendar, MemoPad etc. for my Visor+GPS. I even wrote C! But I never figured out how to create a new Memo doc (with some geo-related data in it as primer).

14:57:17 <dc_rdfig> Added comment E12.

14:57:19 <earle> dymaxion map = global surface mapped onto a geodesic object

14:57:35 <zool> . http://www.metamute.com/look/metamap.pdf

14:57:56 <danbri> E:Idea being that someone with Palm+GPS could easily write restaurant reviews or whatever in memos or calendar, and scoop it off the device later. Possible student project in doing that properly. But maybe gadgets have moved on.

14:57:56 <dc_rdfig> Added comment E13.

14:58:13 <chaasBRS> dynamic mapping relies on being able to replace bits of info (or having a very big file to start with)

14:58:44 <danbri> not fair! I'm chairing, can't look at all these interesting links. Later I guess!

14:58:58 * DanC catches up...

14:59:12 <_joshua> Projections are for people without computers

14:59:18 <_joshua> er, static projections, rather

14:59:31 <DanC> hmm... the mapbureau stuff still doesn't use InterpretationProperties. wasn't it clear why that style is wrong?

14:59:52 <mortenf> DanC, chris and I will get back to that.

14:59:53 <chaasBRS> What I had hoped Cedric would do is something that lets you zoom into a map to mark something at a relevant resolution

14:59:54 <mortenf> (action)

15:00:17 <danbri> E:I do think [http://esw.w3.org/topic/RestaurantRecommendation|restaurant recommendations] (in text, audio, photos etc too) is a good Geo use case. Also links to people data (FOAF etc), eg find me recommendations from (other) lactose intollerants / vegans / meat eaters

15:00:17 <dc_rdfig> Added comment E14.

15:00:26 <chaasBRS> and select between naming schemes (eg lat/long of a given resolution, or postcodes)

15:00:35 <mortenf> gotta go, will read logs and and look for more DanC comments...

15:00:40 <mortenf> see y'all

15:00:44 <chrisgoad> E4: about [http://www.mapbureau.com/notes/interpretedvalues.html| interpreted values] vs interpretation properties

15:00:44 <dc_rdfig> Replaced comment E4.

15:00:55 <danbri> Chaals, yes an SVG-based zoomer that could be plugged into websites of the sort that folk here are building...

15:01:18 <danbri> that'd be cool. If you're building a site now and you want someone to supply geo-data, its very non-trivial to get it from them.

15:01:39 <danbri> Maybe mapquest or whoever could offer a callback/shopping cart interface that could be plugged in, but i've not seen such.

15:01:45 <danbri> Or streetmap/upmystreet/etc...

15:02:10 <danbri> (we've had 9 mins on brainstormery / project ideas)

15:02:21 <danbri> Any more suggestions, wishlists?

15:02:36 <_joshua> Those guys are paranoid with their data. Good luck

15:02:39 <danbri> Things you expect to build when you (yeah right!) get a free weekend or two and it's rainy...

15:03:21 <danbri> yeah... but it might get them more banner adds. Idea is that you'd step into their site, passing them a return-to-here URL, they'd keep that and give you a (banner ad strewn) UI for finding a place, then chuck you out the other end...

15:04:20 <_joshua> I guess.

15:04:42 <_joshua> The problem with maps is that anyone can make them; the barrier to entry is just effort and not much else. So they are hyper paranoid about ownership and access to data

15:04:57 <zool> that describes many thing

15:05:00 <danbri> OK, let's move on. #rdfig is 24x7 brainstorming anyways...

15:05:10 <danbri> Joshua, it's trickier in UK, less free data.

15:05:12 <DanC> B:regrets from DanC; neglected to integrate this into my schedule

15:05:12 <dc_rdfig> Added comment B15.

15:05:18 <_joshua> danbri: get a GPS, and fix it.

15:05:19 <chaasBRS> Vincent Hardy from W3C is interested in maps and SVG in particular - I will try to talk to him too next week.

15:05:26 <_joshua> Metadata will not fix the lack of data.

15:05:46 <danbri> ====== AGENDA (G:) Heads up from zool re noderunner and nocat developments

15:05:54 <zool> yes, all this G: business

15:06:02 <danbri> Zool, sounds like you've been having fun. What've you bin up to?

15:06:03 <zool> well this is an etcon feedback loop

15:06:08 <_joshua> oh - hey, in real life I know the people who organized the noderunner stuff.

15:06:13 <_joshua> In case you want an intro.

15:06:18 <zool> talking to nocat, who doing wireless mapping demos using mapserver GRASS postgis

15:06:23 <zool> url already chumped

15:06:27 <danbri> Yeah, I heard good things about your talk at etcon, and etcon in general.

15:06:43 <zool> reckon they can do SVG output; whoever groks that best maybe talk to them

15:06:47 <danbri> I take it this is the same nocat that give code/tools for opening up one's own wireless for sharing?

15:06:55 <_joshua> yes

15:06:57 <zool> and slurp in RDF in whatever point/route format we come up with

15:07:03 <zool> yes, maps.nocat.net

15:07:11 * danbri tried it, pretty slick install; i got stuck w/ lack of two working eth cards thats all

15:07:13 <zool> Rich_G and Schuyler and collord

15:07:19 <zool> so

15:07:34 <zool> noderunner is the wireless node hunting/ photography game

15:07:48 <danbri> chumped already? links/urls welcome...

15:07:59 <zool> i'm going to run one in london in a couple weeks, and extend my RESTful rdf interface to express it

15:08:02 <crysflame> that talk was great

15:08:07 <zool> yes all this is already on the chump sorry

15:08:19 <libby> can re chump under (G:)

15:08:19 <jben> ..get socialfiction people involved

15:08:31 <zool> . http://space.frot.org/rdf/noderunner_sketch.rdf

15:08:32 <zool> yes

15:08:45 <zool> this will all be part of the Cartographic Congress

15:09:09 <zool> psygeocon, socialfiction, IAA, everyone already involved with that wanting to play the bigger game

15:09:18 <jben> hehe

15:09:23 <zool> that sketch includes a namespace i made up to express community wireless networks

15:09:25 <danbri> 'socialfiction' people?

15:09:42 <jben> ..psychogeography crowd

15:09:54 <zool> which i think is needed to persuade consume.net to release everything in RDF - they're already doing some RSS feed of new nodes

15:09:57 <danbri> 'psychogeography crowd'?

15:09:58 <zool> G:http://consume.net

15:09:58 <dc_rdfig> Added comment G5.

15:10:08 <danbri> (sorry, don't know those labels!)

15:10:20 <zool> G:[http://socialfiction.org/psychogeography|socialfiction]

15:10:21 <dc_rdfig> Added comment G6.

15:10:42 <danbri> ta for urls!

15:10:45 <zool> G:[http://psychogeography.co.uk|psychogeographic association]

15:10:45 <dc_rdfig> Added comment G7.

15:11:03 <jben>http://socialfiction.blogspot.com/

15:11:03 <dc_rdfig> K: http://socialfiction.blogspot.com/ from jben

15:11:07 <zool> G:[http://www.glowlab.com/psygeocon/pgc_index.html|psygeocon]

15:11:08 <dc_rdfig> Added comment G8.

15:11:22 <zool> G:[http://blogosphere.headmap.org|blogosphere]

15:11:22 <dc_rdfig> Added comment G9.

15:11:29 <zool> okay

15:12:06 <zool> so, i think that was the story; any advice on the suitability or not of the rdf model i sketched out at http://space.frot.org/rdf/noderunner_sketch.png would be appreicatred

15:12:25 <zool> that's not implemented in my RESTful interface yet

15:12:37 <danbri> G:aside re weblogs, I made a quick [http://rdfweb.org/mt/foaflog/archives/000014.html|note on foaf and weblogs] recently..

15:12:38 <dc_rdfig> Added comment G10.

15:12:53 <zool> but it is based on the stuff i talked about at the last geo meet, the 'en route' mode following thing, e.g. http://space.frot.org/show_route/test

15:13:33 * danbri remembers, forgot quite how it differed from Jims

15:13:49 <danbri> ta for http://space.frot.org/rdf/noderunner_sketch.png thats useful for digesting the structure

15:13:52 <zool> i think that's all i have to say right now, apart from lots of semi-irrelevant stuff about proxying bots which i should be saying elsewheretime

15:14:02 * danbri grins

15:14:12 <zool> i could talk through that if nesc... the thing about groups of foafs following a route

15:14:24 <danbri> any more re ETCON? what was it _like_? geekfest? academicy conference? businessy? cliquey? fun?

15:14:32 <danbri> bit of all the above?

15:14:35 * crysflame stifles himself

15:14:46 <zool> the model kind of assumes that the people who are on a route all start it and end it and follow it all the way; i don't want that, want to be able to annotate arrival and departure times

15:14:47 <crysflame> it was like living in irc for four days

15:14:54 <zool> it was spiffing

15:15:07 <crysflame> even the conversations had that wavering never-quite-on-focus quality

15:15:07 <danbri> lol

15:15:08 <danbri> re irc

15:15:18 <danbri> someday everything will be like this ;)

15:15:22 <zool> very collaborative feeling, breaking down the geek/blogger dichotomy a bit, i hope to think

15:15:26 <crysflame> and the small groups of five people would reform into split groups temporarily

15:15:35 <danbri> OK we have 15 mins left on the clock...

15:15:46 <danbri> re breaking down the geek/blogger dichotomy a bit, i hope to think

15:16:01 <danbri> that's good. sometimes a bit of tension there, coders versus commentators...

15:16:19 <crysflame> didn't see any

15:16:22 <danbri> Shall we move on?

15:16:26 <crysflame> .

15:16:30 <danbri> ==== (H:) Semantic Web Developer Map: representing locations of people, research groups and project

15:16:41 <danbri> Libby?

15:16:49 <libby> hello...me again....

15:16:58 <zool> re libby

15:17:05 <libby> this is something we've been working on as part of our EC project

15:17:08 <danbri> what do you want this time? our data no doubt?

15:17:22 <libby> but somethign that's already been done in quite a cool way with foaf data

15:17:28 <libby> give me your data

15:17:37 <libby> I don;t really need the time, just give me your data!

15:17:45 <danbri> Our beloved EC-funded project, SWAD-Europe. Which incidentally helps pay for me, libby, chaals, dajobe and others to hack on RDF/SW tools, collaboration etc.

15:17:52 <libby> I appreciate it's a nuisance, esp if you've done it all before for foaf

15:18:07 <danbri> H:see also [http://www.w3.org/2001/sw/Europe/|SWAD-Europe] site.

15:18:07 <dc_rdfig> Added comment H4.

15:18:15 <libby> but I think it will be quite cool to see everyone workign on the SW, in a group or not

15:18:34 <danbri> So can folk just tweak their FOAF files (if they have 'em) to add the new stuff...?

15:18:37 <libby> ...and if you're based in Europe, you get to be on the SWAD-E postcard (as a black dot) :)

15:18:53 <libby> sure, that would be cool

15:19:08 <libby> this is how i suggest could do it: http://www.w3.org/2001/sw/Europe/200303/geo/intro.html

15:19:13 <danbri> We want to know where people, groups etc are... and merge that with info about weblogs, RSS feeds etc... picked up by RDF harvesters...

15:19:21 <chaasBRS> E: [http://ckiss.lanetcie.com/divers/carte_monde/|A work in progress on mapping the globe] (not RDF based, but a fun thing nevertheless. Even without Greece in it :)

15:19:21 <dc_rdfig> Added comment E15.

15:19:45 <libby> I'm inclined to think that I may just adapt the foaf data anyway, do another query...

15:20:24 <libby> several people have kindly suppl;ied me with some data: http://esw.w3.org/topic/AnRdfHarvesterStartingPoint thanks!

15:20:58 <danbri> OK, does that cover H:? Any questions? More pleaing from libby?

15:21:01 <libby> I've harvested it, and queried it using an RDF Query language squish, and displayed the result as SVG, as jim did for foaf data

15:21:04 <libby> please

15:21:06 <libby> please

15:21:14 <libby> go on go on go on go on

15:21:17 <libby> that's it

15:21:19 <libby> thanks

15:21:29 * danbri wonders how to make it so these things are rewarding enough as to not need pleaing

15:21:31 <DanC> A:nifty... hmm...

15:21:31 <dc_rdfig> Added comment A3.

15:21:34 <_joshua> danbri: re etcon, bitter about not being invited :/

15:21:35 <chrisgoad> Libby, at what URL will you put the results of the harvest? I might try mapping it myself.

15:21:36 <danbri> geourl sorta hit that spot

15:21:42 <jben> ..a post ecton summary of geolinks http://www.headmap.org/wiki/index.php?ExternalLinks

15:21:50 <danbri> was it invitational? i didn't notice :)

15:21:55 <zool> cheers jben

15:21:58 <danbri> ta

15:22:05 <libby> chrisgoad: people are putting files here: http://esw.w3.org/topic/AnRdfHarvesterStartingPoint

15:22:20 <libby> I don;t really have an output in RDF of queries of the harvester

15:22:39 <chrisgoad> Ok; I guess I knew about the starting point, since I put myself there.

15:22:47 <libby> see here for a list of foaf data: http://rdfweb.org/topic/FOAFBulletinBoard

15:22:52 <libby> oh yes :) thanks chris

15:22:54 <_joshua>

15:23:22 <libby> chrisgoad, can you handle ntriples?

15:23:47 <chrisgoad> I can if you give me couple of hours to write the parser.

15:24:05 <chrisgoad> Seriously, I will support triples, and even N3, soon.

15:24:24 <danbri> OK, next agenda item?

15:24:25 <libby> hang on a sec

15:24:33 <danbri> ok

15:24:34 <libby> chris i mean, go ahead with agenda

15:24:41 <libby> sorry, danbri, move on

15:25:03 <danbri> righto... well we're off the agenda'd items except to revisit Mailing list discussion, GeoWankers, etc.

15:25:08 <danbri> DanC is here now, too.

15:25:16 <danbri> === AOB: mailing lists, GeoWankers etc.

15:25:34 * DanC is mostly elsewhere

15:25:57 <danbri> We discussed within W3C setting up of either a general RDF/SemWeb vocab collaboration list, or a geo-specific one, and it isn't something we can rush into, since we ought to first get the new Interest Group charter out for comments.

15:26:10 <DanC> oh... I owe the group on this mailing list stuff...

15:26:12 <danbri> But that's no excuse for not continuing discussions

15:26:41 <chaasBRS> Like danbri I can see people being uptight about joining a list that suggests they are wankers

15:26:53 <crysflame> yeah

15:27:07 <zool> otoh, it provides a nice alternative of levity to the often serious and demanding GIS world

15:27:13 <zool> reminds us we're playing

15:27:27 <DanC> I got as far as drafting http://esw.w3.org/topic/SeedApplications , but I didn't call for discussion on semantic-web@w3.org, because I decided (after consulting with folks, as danbri says) not to preempt discussion of RDFIG-NG charter.

15:27:42 <_joshua> sorry. I got distracted. What was the question?

15:27:45 <crysflame> geohippies!

15:28:01 <libby> chrisgoad, can get ntriples on nearest airport from top query here: http://swordfish.rdfweb.org/discovery/2001/08/codepict/ntriplequery.html

15:28:03 <_joshua> (I'm at work)

15:28:22 <chrisgoad> Libby, excellent - thanks.

15:28:26 <danbri> Yes, I like the levity angle. (hence various FOAF follies, and introducing technical terms like 'smush' and 'scutter')...

15:28:29 <libby> no prob

15:28:37 <danbri> Sorry joshua, forgot about dayjob!

15:29:28 <DanC> http_s_ for geowankers archive?

15:29:38 <_joshua> You can refer people to geowankers, but I reserve the right to keep people on topic, moderated, etc

15:30:11 <danbri> Ok, we have 1 minute left. My chairing-bad... I propose extending the meeting by 15 minutes to talk more about mail list and scheduling a followup chat.

15:30:11 * DanC is glad there are no popup ads etc

15:30:16 <crysflame> wankers forces informality, interestingly enough

15:30:23 * DanC can hang for :15

15:30:45 <_joshua> DanC: Yes

15:30:46 <danbri> Yes, mailman and informality is good. I'm not sure re moderation, depends how hands-on. But chairing/nudging is good!

15:31:21 * libby seconds extending meeting

15:31:36 <danbri> any objections re extending for 14 mins, till 45 past?

15:31:38 <_joshua> DanC: the reason it's https is spiders tend to ignore https

15:32:12 <danbri> Joshua, is geowankers somewhere we can say 'if you're interested in rdf/geo etc., its a list you probably want to join?'

15:32:17 <danbri> or would you rather keep it cosier?

15:32:21 <DanC> er... yeah... but... don't we *like* spiders? e.g. google?

15:32:33 <danbri> google++ #beats remembering

15:32:38 <crysflame> danbri: hahaha

15:32:56 <danbri> ok, no objections. meeting extended till .45 past.

15:33:04 <_joshua> it's not about rdf per se. It's about discussing projects, not just sortof generic "i wish this was in rdf" stuff

15:33:32 <danbri> i find http://esw.w3.org/topic/SeedApplications a good start towards SW phase two scoping...

15:33:43 * DanC thought "I wish this was in RDF" stuff *was* wankering. 1/2; -)

15:33:53 <_joshua> Heh heh

15:34:21 <danbri> Okay, I guess I see 'lets get this into RDF' projects as being good focal point for collaboration, versus 'and heres another geo project that overlaps with all these others'...

15:34:39 <danbri> ...ie a focus on data exchange / interop amongst them is healthy (and i see that through RDF-tinted glasses ;)

15:34:39 <zool> well, i am keen to get feedback on the viability of the rdf geo descriptive application i do have

15:34:39 <_joshua> Non-rdf stuff is interesting to me and thus the list

15:34:51 <earle> _joshua: I was wondering, will geourl ever support lat/long embedded in RDF?

15:35:07 * earle happy to use ICBM, but curious

15:35:08 <zool> i'm rethinking / rewriting a bit post-etcon though so this is not the alpha-best-time for it

15:35:10 <_joshua> I want a good place to discuss lessons learned, implementation considerations, etc

15:35:23 <_joshua> earle: it's mostly about markedup HTML documents. RDF is not viewable to most humans now

15:35:32 <earle> _joshua: right.

15:35:36 <zool> sets of points connected to each other is all i fundamentally feel i need to express for e.g. throwing rdf at a hackedup mapserver

15:35:37 <_joshua> so is unfair to declare it "out of scope" for now?

15:35:40 <danbri> Sure, same as with www-rdf-calendar, we have plenty of non-RDF things there, eg. interop amongst IETF ical implementations (and associated pain)

15:36:02 <_joshua> Ok.

15:36:35 * DanC wonders where to follow up about InterpretationProperties and mapbureau stuff

15:36:46 <_joshua> my inclination is to say

15:36:58 <_joshua> implement something, even if it's wrong

15:37:02 <danbri> Good question. I'd default to www-rdf-interest but I'm waiting to see what Joshua reckons...

15:37:03 <_joshua> then you'll get a feel for the issues

15:37:16 * danbri thinks folk _have_ been implementing...

15:37:22 <chrisgoad> I'll hang around after the meet if you like to discuss interpretatioProps

15:37:34 <DanC> we've institutionalized the "implement something, even if it's wrong" in http://esw.w3.org/topic/DontWorryBeCrappy

15:37:54 <DanC> chrisgoad, do you follow the geowanker list?

15:38:09 <jben> G.O.W's

15:38:13 <chrisgoad> Hadn't heard of it until now, but I'll check it out.

15:38:14 <_joshua> How about Geowankers is a good place to discuss Geo stuff and how it relates to real world applications, not ivory-tower GIS stuff. Thus including interop with RDF, XML, perl, whatever etc mumble frotz

15:38:18 <danbri> GOW?

15:38:39 <jben> prisoners of geography = geowankers

15:38:42 <DanC> if you're not already following geowankers, chrisgoad, I don't either, yet.

15:38:57 <earle> who is this G. O'Wanker, anyway?

15:38:58 <danbri> (it's a v new list, DanC)

15:39:23 * DanC finds it pretty hard to reconcile the name "wanker" with "not ivory-tower stuff"

15:39:52 <danbri> We have 7 minutes left. We have some continued ACTIONs and a new one for Chris w/ MortenF. Wanna get thinking about a followup meeting...

15:40:02 <chrisgoad> DanC, I'm anxious to understand the objections to interpretedvalues; I'll respond on rdf-interest, if you like.

15:40:06 <_joshua> It's just my sense of humor. Don't take it so literally.

15:40:16 <jben> ..the name should keep the list relatively small

15:40:19 <danbri> Joshua, too late to change the name? geo-hackers?

15:40:21 <_joshua> "maillist" is a synonym for "wanking"

15:40:22 <danbri> heheh

15:40:23 <crysflame> jben++

15:40:53 <DanC> chrisgoad, I'll take you up on your offer to discuss it here after the meeting.

15:41:01 <chrisgoad> Ok.

15:41:06 <_joshua> the other lists there are ____wankers

15:41:17 <danbri> I'm happy with geowankers as a list I'll hang out on, like rss-dev but without the banner ads. I aspire to having a general RDF app/vocab forum at W3C and that'll take a while.

15:42:13 <danbri> OK, 4 mins. Anybody up for meeting again in IRC?

15:42:24 <libby> yep!

15:42:26 <danbri> danbri@fireball:~$ cal 06 2003

15:42:27 <danbri> June 2003

15:42:27 <danbri> Su Mo Tu We Th Fr Sa

15:42:27 <danbri> 1 2 3 4 5 6 7

15:42:27 <danbri> 8 9 10 11 12 13 14

15:42:28 <danbri> 15 16 17 18 19 20 21

15:42:28 <_joshua> It'd be easier if I werent' at the office

15:42:30 <danbri> 22 23 24 25 26 27 28

15:42:32 <danbri> 29 30

15:42:46 <danbri> default would be in 2 weeks time, perhaps later in the day for US west coasters?

15:43:16 <libby> there's a cal meet 14th

15:43:17 <chrisgoad> I don't mind early.

15:43:20 * danbri happy with IRC meetings on saturdays too, fwiw.

15:43:20 <libby> may

15:43:21 * chaasBRS wonders how to ask the semantic web about calendar meetings

15:43:35 <chrisgoad> I don't want to encourage earlier, though.

15:43:44 <danbri> libby, a cal meeting on sat 14 july?

15:43:45 <_joshua> Clearly we need threaded IRC

15:43:52 <libby> oops, no, may :)

15:44:04 <libby> i.e. in 14 days

15:44:07 <danbri> libby, why are we in this earlier slot? artifact of shift to summertime?

15:44:19 <libby> because of jim ley I think

15:44:33 * zool notes his conspic absence

15:44:44 <danbri> PROPOSED: to meet again 2003-06-14 1500 UTC

15:44:45 <dc_rdfig> Label PROPOSED not found.

15:44:46 <libby> yeah shame

15:44:50 <crysflame> heh

15:44:53 <crysflame> proposal?

15:44:56 * danbri missed Jim... guess he's off having fun on travels

15:45:09 * danbri looks for a bit of 2nding re next meeting

15:45:13 <libby> I can make that I think

15:45:18 <danbri> anyone else?

15:45:18 * DanC has a hard time seeing that far (14Jun) into the future

15:45:18 <libby> seconded

15:45:19 <_joshua> If you want me to change the list name, you're only going to get something more perverse

15:45:21 <chrisgoad> 2nded

15:45:33 <_joshua> Geohackers would have been a better name, perhaps

15:45:37 <zool> that is a long time away

15:45:39 <danbri> geotasm

15:45:44 <zool> thtpht

15:45:49 <_joshua> Perhaps I should have named it somethign that wouldn't get put in print

15:45:50 <crysflame> geogasm

15:45:50 <danbri> oh crap, i got the wrong month. sorry guys.

15:45:59 <_joshua> Geofuckers? etc

15:46:02 <danbri> danbri@fireball:~$ cal 05 2003

15:46:02 <danbri> May 2003

15:46:02 <danbri> Su Mo Tu We Th Fr Sa

15:46:02 <danbri> 1 2 3

15:46:02 <danbri> 4 5 6 7 8 9 10

15:46:03 <crysflame> hee

15:46:03 <danbri> 11 12 13 14 15 16 17

15:46:05 <danbri> 18 19 20 21 22 23 24

15:46:07 <danbri> 25 26 27 28 29 30 31

15:46:11 <crysflame> i see.

15:46:19 <danbri> someone else suggest a slot, i'm crap at this!

15:46:20 <_joshua> georgasm pretty funny too

15:46:22 <libby> 4th, 15th there are irc meets already

15:46:28 <libby> sorry 14th, 15th

15:46:43 <danbri> then we're in www2003 travel territory...

15:46:43 <_joshua> what about 20030528

15:46:59 <libby> I could make that

15:47:07 <libby> some people maybe stlill at www2003?

15:47:09 <danbri> I might be on the road then, but i'm expendable... libby, could you chair?

15:47:23 <libby> yep could do...

15:47:38 <danbri> what time?

15:48:00 <libby> 1500?

15:48:18 * DanC is probably not going to sign up for any more ScheduledTopicChats until stepping back and looking around in the WWW2003 context

15:48:53 * DanC hunts for chrisgoad's writing on interpretationproperties

15:49:00 <danbri> I could make that date. I'll just be back in bristol that day.

15:49:28 <danbri> Re DanC's comments, I propose we do it in www2003-debrief mode... Or at least I'd like some agenda time for that.

15:49:53 <danbri> BLURB:Next meeting

15:49:54 <dc_rdfig> L: Next meeting from danbri

15:49:55 <libby> so 1500 UTC would be http://www.timeanddate.com/worldclock/fixedtime.html?day=28&month=5&year=2003&hour=15&min=0&sec=0&p1=0

15:50:11 <chrisgoad> DanC, my interpretedvalues note is linked from chump; http://www.mapbureau.com/notes/interpretedvalues.html

15:50:12 <danbri> any objections to 1500 UTC 20030528 ?

15:50:25 * danbri looks at his watch, oopsie we're well over.

15:50:47 * _joshua blinks

15:50:55 <_joshua> Actually that's bad for me

15:51:02 <_joshua> well, maybe. Hm

15:51:06 <libby> time or date _joshua?

15:51:13 <_joshua> No, should be ok.

15:51:29 <DanC> ah; thx, chrisgoad

15:51:29 <_joshua> Well it's 7 days before my wedding so I may or may not be indisposed

15:51:41 <libby> let's go for it, see how the agenda shapes up?

15:51:45 <danbri> L:RESOLVED, meet again [http://www.timeanddate.com/worldclock/fixedtime.html?day=28&month=5&year=2003&hour=15&min=0&sec=0&p1=0|2003-05-28 1500 UTC], #rdfig. Agenda to include WWW2003 debrief and ACTION status from todday's actions.

15:51:46 <dc_rdfig> Added comment L1.

15:51:49 <libby> heh, congrats _joshua

15:51:53 <_joshua> thanks

15:52:04 * danbri hopes _joshua can make it, but understands if attention is elsewhere!

15:52:09 <danbri> congrats!

15:52:40 <DanC> L:regrets DanC

15:52:40 <dc_rdfig> Added comment L2.

15:52:49 <danbri> OK cool. Thanks everyone for coming. Normal #rdfig service is now resumed.

15:52:52 <danbri> ADJOURNED.

15:53:03 * zool claps

15:53:14 <libby> thanks danbri

15:53:24 <chrisgoad> Yes, thnx.

15:53:29 <DanC> ok, chrisgoad let's look at your example in http://www.mapbureau.com/notes/interpretedvalues.html that uses owl:sameAs.

15:53:37 <chrisgoad> Ok .

15:53:39 <danbri> danbri has changed the topic to: RDF/SemWeb hack'n'chat http://rdfig.xmlhack.com/ - now with residual Geo flavour, http://esw.w3.org/topic/GeoInfo

15:54:25 <DanC> for the purpose of conversation, we need names for the points... let's call them pt84 and pt72, ok?

15:54:31 <chrisgoad> Sure.

15:54:41 <danbri> logger_1, pointer?

15:54:41 <danbri> See http://ilrt.org/discovery/chatlogs/rdfig/2003-04-30#T15-54-41

15:55:01 <DanC> now the owl:sameAs triple says that <#pt84> owl:sameAs <#pt72>, right?

15:55:08 <chrisgoad> Exactly

15:55:25 <chrisgoad> I think I can guess where you're going.

15:55:45 <DanC> so anything true of pt84 is also true of pt72. for example, <#pt84> map:srs <http://nurl.org/0/geography/SRSCatalog/wgs72>.

15:55:45 <danbri> B:See [http://ilrt.org/discovery/chatlogs/rdfig/2003-04-30#T14-01-40|full chat logs

15:55:46 <dc_rdfig> Added comment B16.

15:55:54 <danbri> B16:See also [http://ilrt.org/discovery/chatlogs/rdfig/2003-04-30#T14-01-40|full chat logs]

15:55:55 <dc_rdfig> Replaced comment B16.

15:56:16 <chrisgoad> Yes, you're right. owl:sameAs is not the correct property. We need a geo:sameas

15:56:39 <danbri> E:Conversation continued post-meeting, see [http://ilrt.org/discovery/chatlogs/rdfig/2003-04-30#T15-54-41|irc logs].

15:56:39 <dc_rdfig> Added comment E16.

15:57:09 <DanC> well, geo:sameAs is one work-around (cyc calls it cospatial, btw). but I think you're going to regret that. InterpretationProperties is a much better fit.

15:57:39 <chrisgoad> What do you think of my n*m argument at the bottom of the note, though?

15:58:00 <DanC> good idea to point to the logs from the meeting chump item, danbri. I missed that in the case of earlier meetings.

15:58:10 * DanC reads n*m again...

15:58:13 * danbri nearly always forgets

15:58:35 * crysflame exposes http://www.crystalflame.net/archives.rdf to the world, and wonders if it's dangerous.

15:58:50 <crysflame> all posts ever, as rdf, with foaf data.

15:59:03 <chrisgoad> While you're reading, a comment. Points with different projections are entities

15:59:27 <chrisgoad> that make sense, and asserting that they denote the same point on the earth

15:59:43 <libby> wow crysflame

15:59:51 <chrisgoad> is a coherent thing to do - so I wouldn't necessarily call this just a workaround.

15:59:56 <jben> cool crysflame

16:00:09 <DanC> the n*m argument is an unfortunate fact of life, documented in http://esw.w3.org/topic/NaryRelations

16:00:15 <earle> earle is now known as grault

16:00:26 <zool> how do i work annotation into this crysflame scheme

16:00:36 <zool> re urgen

16:00:46 <urgwerk> re

16:01:00 <chrisgoad> Yes, it is a fact of life IF you take the interpretation properties approach,

16:01:21 <chrisgoad> but doesn't come up under interpreted values. I'll have a look at NaryRelations

16:01:23 <chrisgoad> now...

16:01:40 <xovver> xovver is now known as xover

16:01:48 <crysflame> zool: trackbacks

16:01:59 <crysflame> look at the per-post trackback information

16:02:09 <crysflame> if you want to annotate an article, write about it somewhere else and trackback.

16:02:24 <crysflame> also, adding comments to this feed would help. anyone got an RSS 1.0 feed that has comments in it? i'd like to learn by example.

16:02:29 <DanC> well, as long as we're agreed that your owl:sameAs example doesn't say what you meant, I can understand if you go the cyc:cospatial route, and I can understand if you're not convinced about InterpretationProperties. But I'm holding one "I told you so" card in reserve in case my intutions pan out and you live to regret it ;-)

16:02:56 <zool> damn, now i have to understand trackbacks and decide about them

16:03:08 * DanC too, re trackbacks

16:03:08 <arnarl> hi

16:03:15 * danbri three re trackbacks

16:03:32 <zool> how do they look in terms of service centralisation

16:03:40 <chrisgoad> Thanks for explaining the objection. I'll think further on this.

16:03:42 <danbri> btw I wrote to Ben @sixapart / MT / TypePad offereing to help check their RDF/FOAF emissions...

16:03:49 <danbri> am waiting on some samples from them now

16:03:53 <zool> yay

16:03:54 <DanC> welcome, chrisgoad

16:04:10 <_joshua> danbri: TypePad may include GeoURL support

16:04:21 <danbri> whee

16:04:27 <crysflame> oo.

16:04:42 <danbri> in shiny futuristic bandwidth using RDF format, or retrochic but easy to type META tags?

16:05:00 <_joshua> Dunno. They asked me about it

16:05:19 <zool> should be easy to rdfise geourl though? and plausibly desirable?

16:05:47 <crysflame> is!

16:05:50 <zool> abstraction/translation interface between them, in similar way convert rss2->1 in client?

16:05:52 <zool> cool

16:06:00 <_joshua> Depends. If you find a document via geourl, I think the expectation si taht it'd be viewable by humans in a web browser, and most RDF isn't

16:06:12 <_joshua> if RDF were embedded in the HTML doc, I'd consider parsing that

16:06:25 * mattb envisages a FOAFalytic converter for regulating rdf emissions

16:06:26 <zool> rdf was designed with html embedding in mind as i understand it

16:06:35 <zool> that is one of the main reasons the syntax is so modal

16:06:45 <danbri> I talked to ben a bit about their image gallery app, that would be nice to get into Wordnet and geo markup for...

16:06:50 <zool> grubstreet has rdf-in-html embedded geometadata

16:07:01 <danbri> typically a whole batch of photos can be tagged with roughly same geo info

16:07:05 <_joshua> nod. I'm not sure how to go about that but I could do that.

16:07:13 <_joshua> jo, you do perl?

16:07:14 * danbri uploads his pre-lunch pics of rainy bristol

16:07:30 <zool> _joshua: as if my life depended on it

16:07:39 <grault> grault is now known as earle

16:07:44 <_joshua> I can't tell if that is sarcastic :)

16:07:50 * zool card-carrying london perlmonger etc etc

16:07:51 <danbri> jo does perl

16:08:00 <jben> jo does perl

16:08:00 <_joshua> I could show you the spidering script...

16:08:05 <earle> _joshua: that's what I meant when I mentioned embedded RDF earlier - in HTML

16:08:10 <_joshua> Oh, ta.

16:08:17 * zool would be interested in that yes

16:08:28 <zool> earle can you show us a gs2 uri illustrating this?

16:08:33 <_joshua> URL with something including RDF embedding?

16:08:42 * zool can't remember how much made it out of your wiki prototype

16:08:47 <danbri> does it spider as in collect further links to crawl? or just grabs each of a known list of urls?

16:09:13 <zool> we could use mattb's pub/sub aggregating crawler for this

16:09:20 <danbri> earle, i forget there was an rdf embedded version too. i just the ;format=rdf version for plain rdf

16:09:32 <_joshua> It doesn't actually spider.

16:09:50 <zool> _joshua, let's see it?

16:10:00 <_joshua> sent

16:10:15 <DanC> hmm... geourl is the epitome of http://esw.w3.org/topic/EmbeddingRDFinHTML use cases.

16:10:21 * danbri runs ../../../bin/depiction.rb --dir=1 --gen-pics

16:10:35 <danbri> ...with 0% rdf-added goodness, yet

16:10:36 <DanC> logger_1, pointer?

16:10:36 <DanC> See http://ilrt.org/discovery/chatlogs/rdfig/2003-04-30#T16-10-36

16:11:30 <danbri>http://frot.org/rdfweb/ayf.html

16:11:30 <dc_rdfig> M: http://frot.org/rdfweb/ayf.html from danbri

16:11:39 <danbri> M:|Jo's perl RDF scutter

16:11:40 <dc_rdfig> Titled item M.

16:11:50 <danbri> M:For harvesting rdf via seeAlso traversal

16:11:50 <dc_rdfig> Added comment M1.

16:12:05 <danbri> M:See also [http://rdfweb.org/topic/ScutterSpec|ScutterSpec]

16:12:05 <dc_rdfig> Added comment M2.

16:12:09 <_joshua> zool: forgive the ugly code

16:12:30 <danbri> Oh, I should finish the etags support I started...

16:12:46 <earle> _joshua: http://downlode.org/perl/gs_rdf/wiki.cgi?Penderel%27s_Oak

16:13:04 <zool> short variable names hurt my head :)

16:13:14 <_joshua> Sorry

16:13:14 <zool> makes sense though more or less

16:13:21 <_joshua> i write dense code

16:13:49 <zool> so it would be nice to be able to offer pointers to other geourls in the files you're pointing at, and really spider them

16:14:06 <danbri> I find scutters really useful for loading up a little multi-file dataset: ./ayftest.rb 'http://un.earth.li/~kake/cgi-bin/wiki.cgi?action=index;format=rdf'

16:14:15 <_joshua> I guess. I figure if they can do that, they could also just ping me with all the URLs

16:14:16 <danbri> ...its close to sitemappery too

16:14:27 <_joshua> I'm trying to be very polite

16:14:32 * DanC considers mailing www-tag about the geourl RDF/HTML case

16:15:02 <zool> re. slurping embedded RDF out, convert to xhtml using tidy or something like it, then just run that through an rdf parser and scoop the triples right out; easier

16:15:05 <danbri> ooooooooooooooooooooooooooooooh

16:15:09 <zool> (this makes sense?)

16:15:12 <danbri> my new phone just arrived at shop. bbiab!

16:15:15 <_joshua> Hm

16:15:29 <_joshua> It's an idea.

16:15:31 <zool> the thing about mattb's aggregator is it is non-violent so to speak

16:15:40 <zool> does some quite smart stuff with etags etc

16:15:50 <_joshua> The problem is that some large % of people can't even get <meta name=ICBM value="X,Y"> correct

16:15:57 <mattb> trying to minimise the impact of large numbers of scutters running...

16:16:00 <_joshua> so I have zero hopes of more complicated stuff

16:16:14 <zool> you could support both geourl format with this, and also rdf format, in a dual-purpose scutter that would incidentally pick up other triples and forward them to an aggregator service

16:16:28 <mattb> yup

16:16:34 <zool> well, metar foo can be fiddly

16:16:34 <mattb> rdf-in-html scuttering is on the list :)

16:16:39 * zool smiles

16:17:02 <dajobe> how iN?

16:17:07 <dajobe> in <!-- -->

16:17:14 <zool> has anyone sketched out an RDFisation of geourl format? should be pretty straightforward looking... is there a spec kicking about

16:17:25 <zool> dajobe: in the <HEAD> surely?

16:17:32 <crysflame> right on the front page?

16:17:35 <_joshua> there's the wgs_84 thingy

16:17:59 <libby> zool, the tricky thing I guess is spcifying what you are referring to

16:18:21 <zool> libby in what way, the document versus concept thing, or something deeper?

16:18:36 <zool> as i understand it geourl is designed to talk about documents

16:18:45 <libby> I was thinking of that - doc/concept

16:18:56 <_joshua> realistically geourl is just a toy technology

16:19:12 <zool> it's just [[ <meta name="ICBM" content="XX.XXXXX, XX.XXXXX">

16:19:12 <zool> <meta name="DC.title" content="THE NAME OF YOUR SITE"> ]]

16:19:31 <zool> right, i'm confusing myself remembering all the ENT-like stuff that headmap wanted to do on top of it

16:19:46 <libby> maybe not too tricky at all

16:20:23 * zool shakes head not in this case

16:20:36 <crysflame> the big master rdf has been updated with comments.

16:21:30 * zool searches for the reference to the headmap stuff

16:21:41 <zool> i need some kind of filtering proxy bot search interface of some kind

16:22:08 <_joshua> Heh

16:22:12 <jben>http://www.headmap.org/wiki/index.php?RdfGeoSchema

16:22:12 <dc_rdfig> N: http://www.headmap.org/wiki/index.php?RdfGeoSchema from jben

16:22:25 <jben> ..old email

16:22:27 <_joshua> write a web proxy that fulltext indexes every page you've visited

16:23:06 <zool> ekk at email :) there was another one of anselms i had in mind, perhaps the one with the forward from joshua at the top

16:23:10 * zool brain broken

16:23:30 <_joshua> heh

16:24:04 <jben>http://blogosphere.headmap.com/pipermail/headmap-code/2003-February/000043.html

16:24:04 <dc_rdfig> O: http://blogosphere.headmap.com/pipermail/headmap-code/2003-February/000043.html from jben

16:24:37 <zool> anyway, something like <rdf:Description rdf:about="[ this uri ]"><geo:lat>51.23213</geo:lat><geo:long>-0.01212</geo:long><dc:title>wibble</dc:title><dc:lang>whatever</dc:lang></rdf:Description>

16:24:41 <zool> wouldnt be more complex

16:24:49 <zool> then you could bolt in any triples you like

16:25:05 <_joshua> That looks a lot more complex :)

16:25:21 <zool> tie that in to a richer rdf representation of space like my toy one, or into a foaf network of interconnected annotations or whetever it was

16:25:28 <zool> that is only because it is all on one line

16:25:30 <crysflame> <a xmlns:geo="http://...." geo:lat="" geo:long="">

16:25:41 <zool> also :)

16:25:41 <crysflame> opinion="negative"

16:25:48 <_joshua> about 20% of people that submit urls to geourl to be spidered get the ICBM tag wrong

16:26:03 <zool> can't you let them know via trackback? :)

16:26:12 <crysflame> zool: nope

16:26:17 * zool pouts

16:26:18 <crysflame> also, trackback is semiautomatic sometimes

16:26:30 <crysflame> also, theregister doesn't support it, the one we developed it for

16:26:33 <zool> you could get in bad loops

16:26:43 <_joshua> zool: go to geourl and look at the top 10 sites that just submitted

16:26:56 <danbri> <zool> has anyone sketched out an RDFisation of geourl format? should be pretty straightforward looking... is there a spec kicking about

16:27:01 <crysflame> you can always get in loops :)

16:27:08 <_joshua> oh yeah

16:27:09 <_joshua> anyway

16:27:13 <danbri> the RDF emmissions from geourl, which me and danc helped spec, should do it, shouldn't they?

16:27:23 <zool> cant you refuse to let them submit unless they pass an initial parse phase, or do you worry that will put too many people off? thought about a foaf-a-matic style tool to autogenerate them ?

16:27:30 * crysflame is using them, and the foaf explorer recognizes them.

16:27:32 <danbri> see http://esw.w3.org/topic/GeoInfo example up front

16:27:34 <zool> danbri: where are those

16:27:36 <zool> ah cool

16:27:38 <crysflame> -> danbri

16:28:11 <danbri> eg http://geourl.org/rdf/?lat=37.75&lon=-122.45

16:28:24 <danbri> (morten in site comments asked why this didn't seem to support the dist parameter)

16:28:25 <zool> what do we gain with the foaf:topic?

16:28:34 <_joshua> It's probably broken

16:28:38 <danbri> you gain a clearer distinction between things and documents about them

16:28:58 * zool wonders

16:28:59 <danbri> the document is (we posit) in some sense about the topic of (some place)

16:29:06 <crysflame> i dropped the foaf:topic

16:29:13 <crysflame> and used pos:isWithin100m

16:29:24 <danbri> prob is we don't really know that much about what the lat/long means, for geourl

16:29:34 <danbri> it could be what the page is about, or where its author lives, or where server is.

16:29:52 <danbri> topic works well for restaurant reviews too, links a review to the thing its about.

16:29:55 <_joshua> The intention is "the contents of the page"

16:29:57 <danbri> well, but not ideally

16:30:22 <danbri> if the page was a photo of a painting of a ... etc it gets murky, but maybe there's an 80%/20% payoff there...

16:30:23 <zool> do you think that accepting the geourl model will cause later problems, danbri?

16:30:42 <danbri> what geourl model? its a nice if bit vague dataset...

16:30:55 <zool> can this not be viewed as a bootstrapping stage towards something more granualr or do you think that's dangeroute

16:30:56 <danbri> 'page content ... has something to do with ... location'

16:31:06 <danbri> yeah, i think its bootstrap friendly :)

16:31:08 <zool> by model i mean, way of stating things about the world

16:31:10 <_joshua> In some sense you can't really know how people will use something until you see them use it.

16:31:14 * danbri takes a break to play with new phone

16:32:11 <jben> top level : sectionlevel : specific instance level

16:35:19 <nwalsh> Quick geo-related question?

16:35:29 <_joshua> it's RIGHT BEHIND YOU

16:35:45 * nwalsh looks behind

16:36:59 <nwalsh> Among all the GeoInfo, I'm surprised there isn't a reasonably common way of saying "this is a city name", "this is a country name" etc. I can invent my own, of course, but that seems...less interoperable. Though I suppose everything can be mapped.

16:37:18 <nwalsh> Is that really an open problem?

16:38:37 <_joshua> because it's Hard. go check out the TGN

16:38:41 <zool> 'page has url and lat and long and title'

16:38:41 <zool> i dont want to break the semantic web though :)

16:38:44 <zool> _joshua: has the fact that you havent gone for anything more granular (e.g does this geourl refer to the person, place or event) prevent you from drawing conclusions about how people use it?

16:38:53 <zool> anselm ran across similar problem scraping indymedia news reports to map onto their blogosphere; is this article about cafe baghdad in toronto, or talking about toronto in baghdad, or talking about baghdad in toronto?

16:39:07 <nwalsh> I peeked at TGN. Fair enough.

16:39:10 * zool listens quietly

16:39:16 <zool> yes as far as i know, but others prolly have better context

16:39:22 <zool> in my schema i would put a city in rdf:type space:City or space:Populated_Place then give it a name string... but i am just making everything up... and hoping that precedents will form through connections in the world

16:39:25 <zool> i've only been playing around within one city though really

16:39:31 <_joshua> GeoURL is intended to be about whatever the page is about

16:39:37 <zool> what's the TGN?

16:39:38 * danbri boots new phone... it's imprinting on me now... (i'll be first face it'll see...)

16:39:40 <_joshua> My page on my trip to Cancun should be marked up as Cancun

16:39:46 <zool> then the foaf:topic is the way to go, i guess

16:39:52 <danbri> aghast at number of charger packs under my desk...

16:39:54 <_joshua> Lots of bloggers entered their stuff so presumably the page is about them

16:40:13 <zool> lol danbri

16:41:15 <zool> will a compliant RDF parser get upset by default if it just sees the triples and not the headers... if the rdf is embedded in html

16:41:27 <zool> that would make it *look* more straightforward at least

16:41:57 <deltab> zool: Getty Thesaurus of Geopgraphical Names:

16:42:10 <nwalsh> zool: TGN = http://www.getty.edu/

16:43:21 * danbri inserts squishy bluetooth headset into ear; is creeped out by reminder of http://movieweb.com/movie/existenz/

16:43:28 <_joshua> Never saw that

16:43:43 <danbri> TGN is so big and good (and unavailable) that I think its existence can take energy away from doing a free one...

16:43:46 <mattb> euw

16:43:51 <_joshua> nod

16:43:51 <danbri> seems sorta pointless, a solved problem.

16:43:52 * nwalsh thinks danbri's having way too much fun with his new toy

16:44:24 * wkearney_away wonders if tgn will ever 'get a grip' and recognize opening access to it's repository...

16:44:30 <danbri> libby's just sent it a photo...

16:44:32 <wkearney_away> wkearney_away is now known as wkearney99

16:44:35 <danbri> hi bill

16:44:36 * urgwerk jumps out from under a pile of collapsed cardboard boxes

16:44:41 <urgwerk> oh, here you all are

16:44:47 <danbri> sheesh, don't scare us like that!

16:44:51 <urgwerk> discussion still up?

16:44:53 * danbri tidies up the boxes

16:44:56 <danbri> 24x7 ;)

16:45:30 * wkearney99 has often been tempted to do a low-and-slow scraping of TGN against other db records...

16:45:37 <urgwerk> so, I was following application of the product we build here at work ( www.remtrol.com ) and started to explore into sensorML directions

16:46:34 <urgwerk> sensorML seemingly is being developed mostly for Geo type applications, making geo-data agr(i)gatable or something

16:46:41 * danbri hears a beep, apparently my earpiece is now paired with the phone

16:46:53 <_joshua> BAD TOUCH BAD TOUCH

16:47:13 <urgwerk> if this data generation point is already becoming XML-ized,, is the "geo-talk" discussion here wrt rdf then related?

16:48:53 <jhendler> danbri - when you get a second, look at http://esw.w3.org/topic/AnRdfHarvesterStartingPoint and see my note there

16:48:57 <urgwerk> is there an rdf to sensorML link?

16:49:31 <jhendler> There is a sensor ontology in OWL that we are mapping to sensorML, we have not been able to find a sensorML RDF to date

16:50:15 <danbri> Jim, http://mindswap.org/cgi-bin/2003/ret/ret.rdf?uri=http%3A%2F%2Fowl.mindswap.org%2F is great...

16:50:41 <danbri> I think your note suggests a miscommunication: we don't really care how the RDF is generated, whether in batch mode, by hand or GUI or on the fly or XSLT or prolog...

16:50:45 <danbri> just about what it says!

16:51:47 <jhendler> ok, but the is no "link" to the RDF, we would need an http-get(RDF) or something like that to really do it right

16:52:05 * danbri doesn't quite understand

16:52:18 <danbri> the url i gave is for an rdf description of your group and its site, isn't it?

16:52:31 <jhendler> the page says "link to your RDF files" - but there is no file (uri) called xxx.rdf for our group

16:52:57 <jhendler> rather we query for classes and their instances, and that could be served up remotely by RDF

16:53:19 <danbri> so you don't have an aggregate of rdf about your group?

16:53:32 <danbri> is it spread across several urls we should harvest?

16:53:46 <jhendler> not in a single document, it is crawled into a database, and then queried

16:53:57 <jhendler> which is an idea I stole from you, by the way :->

16:54:05 <danbri> :)

16:54:14 <danbri> so just a starter url we could scutter from would be fine...

16:54:39 <danbri> OUCH!

16:54:47 <danbri> this direct ear/brain connection is LOUD

16:54:51 <jhendler> yeah, but what would it be? that's the problem...

16:54:51 <danbri> (libby phoned me...)

16:55:02 <zool> i'd like to make an ontology to describe community wireless networks, unless i can find one existing

16:55:06 <jhendler> (tell her I said hi :->)

16:55:43 <jhendler> zool, I don't know of one, but if you create one, please submit to the ontology library at daml.org so the next person doesnt' have to create one...

16:56:18 <jhendler> danbri - we can dump our people page at some moment in time (I'll do that and link it to that page), but I think this is a good example

16:56:18 <urgwerk> jhendler, I did a quickscan google but didn't find a ref to OWL's sensor ontology, is it behind the works or can public view it?

16:56:34 <zool> jhendler, k cool... how does that intersect with the list at rdfschema.info... that is where i tend to look first.. dont know how much of the daml.org stuff is experimental/current... or whather owl content belongs at daml.org ?

16:56:40 <jhendler> of why rdf remote querying is needed.

16:56:55 * zool tries to stop askign questions

16:57:33 <jhendler> urgwerk - link is currently in an unstable site, I just asked my students to move it to our ontologies link, and then we'll chump it

16:58:03 <jhendler> zool, the daml.org site has open submission ontology library and RDF/DAML (now OWL) toolkit page. Anyone can submit

16:58:13 <danbri> jim, that'd be neat (re people info dump)

16:58:41 <danbri> zool, i'm interested re wireless node markup. wonder how much more is needed...

16:58:53 <jhendler> They accept DAML, RDFS and OWL - although mostly DAML at this point - OWL starting to come in

16:59:00 <danbri> a link to an admin from it, its name, its location...

16:59:22 <jhendler>http://www.mindswap.org/cgi-bin/2002/searchdamlont.pl

16:59:22 <dc_rdfig> P: http://www.mindswap.org/cgi-bin/2002/searchdamlont.pl from jhendler

16:59:40 <jhendler> P:| search engine for daml ontology library (bit flaky)

16:59:40 <dc_rdfig> Titled item P.

16:59:59 <zool> danbri, also status, essid, adhoc or managed... unsure what else

17:00:02 <_joshua> seriously considering kidnapping an iPaq, installing linux, and getting a GPS unit for annotation

17:00:06 <jhendler> P: type the name of a class or property, get a string match to all such in the library (about 200+ ontologies)

17:00:06 <dc_rdfig> Added comment P1.

17:00:20 * danbri nods, zool maybe scribble those in a wiki?

17:00:31 <danbri> the idea of using a wiki page as a ns URI intrigues me

17:00:37 <jhendler> P: various tools to see the ontologies are there -- the OWL button runs convertToOWL which is slgithyl out of date, but close enough for now

17:00:37 <dc_rdfig> Added comment P2.

17:00:41 <danbri> bit _too_ scruffy perhaps? :)

17:01:12 * zool nods, versioning, craziness, adhoc interesting vibes, but ultimately probably not :)

17:02:46 <libby> b**ger, timetables are really hard to do in icalendar. I think

17:03:01 <libby> s/b**ger/b*gger/

17:03:19 <shellac> that extra 'g' offends me

17:03:36 <libby> heh

17:03:59 * danbri saves photos into web, (pub lunch pics and rain) http://rdfweb.org/people/danbri/media/2003/04/30/1/

17:04:53 <earle> earle is now known as grault

17:05:14 <danbri> libby, shellac, chaasBRS, danbri are within100M of 'probably the best pub food in bristol' (according to their own advertising, which I believe)

17:05:22 * libby too

17:05:24 <libby> yum

17:05:28 * shellac hand model with burger and spring rolls

17:05:47 <danbri> nice nails

17:05:48 * jhendler says "hoist one for me gang"

17:06:18 <jben>http://rdfweb.org/people/danbri/media/2003/04/30/1/images/img_1321.jpg

17:06:19 <dc_rdfig> Q: http://rdfweb.org/people/danbri/media/2003/04/30/1/images/img_1321.jpg from jben

17:06:31 * danbri has another lime'n'soda for Jim

17:06:47 <danbri> that was libby's lunch.

17:06:47 <jhendler> danbri - another interesting thing - we get the people page as query result (i.e. Ntriples) but without the organizing info (the various schemas and ontology's) which aren't returned by the query

17:06:54 <jhendler> bear's some thinking about.

17:06:56 <shellac> images don't have meaningful alts (accessibilty ed)

17:06:58 <danbri> the flash makes it look less appetising

17:07:07 <libby> it's all organic too :)

17:07:18 <libby> lovely

17:07:18 <jhendler> nice wather!

17:07:31 * shellac still drying out

17:07:54 <danbri> pub is almost line of site from here... (802.11 hmm :)

17:07:58 * jhendler thinks that the one of Chaas(?) behind the flower should be labeled as a picture of bijan :->

17:08:17 <libby> heh

17:08:33 * danbri was saving viewers from the sight of some serious eatin'

17:09:04 <shellac> didn't you snap the curry?

17:09:36 <danbri> no :(

17:09:49 <danbri> anyways, we need markup for all this...

17:10:28 * chaasBRS snapped the curry

17:13:14 * zool makes http://esw.w3.org/topic/WirelessOntology

17:14:19 <jhendler>http://www.w3.org/2001/sw/WebOnt/guide-src/food.owl

17:14:19 <dc_rdfig> R: http://www.w3.org/2001/sw/WebOnt/guide-src/food.owl from jhendler

17:14:36 <jhendler> R:| The WebOnt WG OWL Food Ontology

17:14:36 <dc_rdfig> Titled item R.

17:14:47 <jhendler> R: perhaps this will help in marking up the photos of Libby's lunch

17:14:47 <dc_rdfig> Added comment R1.

17:15:10 <jhendler> danbri - added query result of our people page to the SW groups wiki.

17:15:25 <DanC> norm, re "this is a city name", "this is a country name" ... cf 'What property should I use to name other ordinary places and things?' in http://esw.w3.org/topic/PropertiesForNaming

17:15:53 <DanC> I made up map:cityName one day.

17:15:57 <DanC> and usps:cityName another day

17:18:16 <jhendler> map:cityName owl:equivalentClass usps:CityName.

17:18:22 <jhendler> ?

17:19:09 <jhendler> R: see also the [http://www.w3.org/2001/sw/WebOnt/guide-src/wine.owl | OWL wine ontology]

17:19:09 <dc_rdfig> Added comment R2.

17:19:38 <danbri> jim, thanks!

17:19:55 <danbri> I think common naming properties are very appealing

17:20:10 <libby> ok, mine of the same event...http://swordfish.rdfweb.org/photos/2003/04/30/

17:20:26 <jhendler> danbri - I like both the ideas of common naming properties and mappings - we could call it -- umm, I know, the Semantic Web!

17:21:50 * jhendler thinks libby's photo of danbri out in the rain taking a picture of a puddle could be submitted to the "there will always be an England" photo competition

17:22:04 * DanC thinks there's something really valuable in the topicmap naming/scoping stuff; is noodling on mapping it to RDF

17:22:40 <DanC> I dunno if map:cityName and usps:cityName are equivalent yet. the usps property has constraints like "16 characters max"

17:22:42 <DanC> from the post office.

17:23:18 <DanC> ah; in fact, I know they're not equivalent; the domain of map:cityName is world-wide; usps:cityName'd things are only us citities

17:23:38 <DanC> so perhaps { usps:cityName rdfs:subPropertyOf map:cityName }

17:25:07 <zool> would you say there was an advantage to this approach other than having a {url rdf:type map:City} and it havign a generic map:name ?

17:26:56 <DanC> umm... yes, I think so...

17:27:24 <DanC> consider Boston... it might have map:name "Boston, MA, US" and "Boston" and "beantown". but only "Boston" is its cityName

17:27:37 <DanC> at least, thats it's only usps:cityName

17:28:09 * zool nods

17:28:15 <jhendler> many years ago I consulted for a NOAA group trying to create a KB of mapping - they wanted not only these sortso f map:names, but a whole set of them

17:28:26 <jhendler> which could be selected by a cartographer based on scale of the map

17:28:35 <jhendler> ie. Bos, Boston, Boston Mass, etc.

17:29:35 <DanC> yup... map:name3emsWide, etc. hmm...

17:33:06 <_joshua> have you guys seen friendster.com? it's social network crack.

17:34:07 <libby> yeah, I'm in it as a friend of matt and zool

17:34:27 * zool tried not to get too caught up in it networkwise, makes me nervous

17:34:40 * DanC hunts for emacs/efs http/webdav support

17:34:48 * _joshua scarfs curry

17:35:22 * DanC finds http://www.zope.org/Members/nemeth/usingEmacsWithZope

17:56:05 <Dorward> Dorward is now known as DorwAFK

18:10:26 <nmg> DanC: I was looking for that (and emacs/efs scp support) the week before last, and couldn't find anything that fit the bill

18:10:49 * DanC settled for cadaver to /tmp

18:12:33 * jhendler wonders if friendster has met TrustBot

18:12:45 <nmg> fair enough - it's probably unreasonable to expect that emacs expand to encompass *all* protocols...

18:13:57 <jhendler>http://www.mindswap.org/~evren/services/sensor-jpa.daml

18:13:58 <dc_rdfig> S: http://www.mindswap.org/~evren/services/sensor-jpa.daml from jhendler

18:14:14 <jhendler> S:| sensor ontology in DAML

18:14:14 <dc_rdfig> Titled item S.

18:14:24 <jhendler> S: based on some DARPA sensor work

18:14:24 <dc_rdfig> Added comment S1.

18:15:05 <jhendler> S: specialized for accoustics sensors in [http://www.mindswap.org/~evren/services/acoustic-jpa.daml acoustic sensor ontology]

18:15:05 <dc_rdfig> Added comment S2.

18:16:55 <_joshua> anselm

18:17:01 <anselm> hey

18:18:22 <urgwerk> :-)

18:18:36 * urgwerk was off lost in onto-land

18:19:38 <urgwerk> ah good, that's what I was looking for

18:22:09 <jhendler> urgwerk, the sensor onts are not real exciting, but they were used in some real applications work, so they're not totally bogus

18:22:53 <urgwerk> we connect a packet cellular data relay to all kinds of sensors

18:23:21 <urgwerk> that seems kind of integral to Geo stuff

18:23:27 <urgwerk> it is the content generation engine

18:24:06 <urgwerk> but translating something like 4-20mA to an engineering unit, when all kinds of 'customer' preferences,, like Kg vs. lb come into play...

18:24:15 <jhendler> cool, in our case the sensors actually live offline, the ontology is used to filter for relevant sensors of a pareticular type in a particular goloc

18:24:40 <jhendler> and then messages are sent to them via some other means (pointers not in this ontology because it is publicly releasable)

18:26:35 <urgwerk> so we've had to ground break a lot of this already,, just trying to translate my experience into a contribution

18:27:24 <urgwerk> there is even a gps on the "macaw" as we call it

18:27:36 <urgwerk> so you can tag coordinates to data too

18:28:47 <_joshua> sweet

18:29:02 <jhendler> way cool

18:29:35 <jhendler> we've done some work on linking web services to locations via 802.11 locating, are looking for interesting ways to take that further

18:30:15 <jhendler> idea we had w/tagging data to location was that we could do things like leave messages for people in particular places

18:30:22 <urgwerk> we use a similar network to pagers, blackberries, and cell phones

18:30:35 <jhendler> i.e. if you are driving to MINDSWAP channel would have "turn left here" located at some point in space

18:30:38 <urgwerk> but the G3 stuff is looking good too

18:31:02 <urgwerk> US national only atm

18:32:36 <urgwerk> chatzilla :-)

18:33:26 <anselm> virtual breadcrumbs...

18:34:49 <zool> hi anselm!

18:35:50 <urgwerk> hmn so sensor as collector where there are other collectors

18:36:08 <urgwerk> guess I need to head into the collector ontology

18:36:31 <urgwerk> s/other collectors/other collector types/

18:36:34 <zool> the breadcrumb onramp...

18:37:27 <urgwerk> another great band name

18:38:33 <anselm> rich gibson mentioned rfid tags for his keys... and his cat... was a great example of tagging _moving_ things.

18:38:57 <urgwerk> yes we looked into hooking the macaw to rfid's

18:39:06 <_joshua> rfid has abasically no range sadly

18:39:08 <urgwerk> it would chirp out a roll call

18:39:11 <_joshua> i tried to get a rfid devkit once

18:40:24 <urgwerk> if an rfid was in the macaw zone, then the macaw would uplink so you could get distance

18:40:31 <anselm> hey zool! nice flight?

18:51:59 <anselm> i helped write a bluetooth stack once for a small embedded device... we had a very very tiny webserver on it... something like that could be a smart rfid with more range. maybe cost effective for a cat (kitty blog?) maybe not for a box of soap tho.

18:54:11 <_joshua> kinda pricey

18:54:47 <danbri>http://bluez.sourceforge.net/

18:54:48 <dc_rdfig> T: http://bluez.sourceforge.net/ from danbri

18:54:57 <danbri> T:|Linux Bluetooth protocol stack

18:54:57 <dc_rdfig> Titled item T.

18:59:56 <_joshua> I wish I had enough time for embedded systems

19:05:15 <urgwerk> re

19:08:51 <anselm> embedded systems are fun to use but hardware limits make it unfun to cut code for them sometimes.

19:09:57 <wkearney99> sonny, you ain't coded until you've had to be bit stingy with 4k of RAM. Or run the logic only during vertical blanking intervals. </geezer rant>

19:10:03 <anselm> http://www.wireless-designer.com/ubicom_bluetooth.html < that was it.

19:10:04 <dc_rdfig> U: http://www.wireless-designer.com/ubicom_bluetooth.html from anselm

19:10:25 <anselm> try again: http://www.wireless-designer.com/ubicom_bluetooth.html

19:11:09 <anselm> 4k of ram eh?

19:11:22 <urgwerk> yup

19:13:14 <qmacro> 4K? you were lucky! :-)

19:13:30 <_joshua> I've done stuff on machine with 256 bytes of ram

19:14:51 <urgwerk> urgwerk is now known as urglunch

19:19:07 <anselm> HeapManager* manager = HeapManagerAllocate(256)... :-)

19:19:47 <_joshua> Ha.

19:20:08 <_joshua> PICs are neat chips

19:23:26 * zool wakes up

19:23:45 <_joshua> Damn, not enough demerol

19:39:14 <anselm> anselm is now known as andyfood

19:41:20 <urglunch> urglunch is now known as urgwerk

20:09:08 <dfg_olin> newbie question about Jena RDFQL: is there syntax to support multiple sources of RDF, say multiple FROM statements?

20:12:43 <Morbus> danbri: i had a question to ask you this morning, and now i just can't remember it.

20:12:48 <Morbus> it's going to drive me absolutely bonkers.

20:13:19 <dajobe> if only there was a way to record knowledge ;)

20:24:39 <andyfood> andyfood is now known as anselm

21:01:46 <DorwAFK> DorwAFK is now known as Dorward

22:23:42 <jhendler>http://www.cogx.com/kt2002/

22:23:42 <dc_rdfig> V: http://www.cogx.com/kt2002/ from jhendler

22:23:57 <jhendler> V: |DAML and topic maps stuff

22:23:57 <dc_rdfig> Added comment V1.

22:24:26 <jhendler> V: Niklita Ogievetsky, Cogitech, Inc. presentation at knowledge technologies 02

22:24:26 <dc_rdfig> Added comment V2.

22:24:39 <jhendler> V: probably been chumped before, but this is first I saw it

22:24:39 <dc_rdfig> Added comment V3.

22:25:34 <deltab> jhendler: thanks for reminding me that La Femme Nikita starts in half an hour

22:26:01 * jhendler always glad to offer useful Sem Web services :->

22:49:55 <_joshua> WHY does chump not log in RDF?

22:50:21 <wkearney99> is there not one running here?

22:50:25 <wkearney99> used to be, didn't there?

22:50:46 <wkearney99> it's dc_rdfig, isn't it?

22:50:49 <_joshua> there's a chump here

22:50:59 <_joshua> I mean, why are the logs chump emits not in RDF :)

22:51:25 <wkearney99> they're available in RDF and a number of other formats, aren't they?

22:51:36 <dajobe> yes

22:51:42 <dajobe> the chump bot writes XML

22:51:58 <dajobe> it's XSLT'ed into HTML and RDF(RSS1.0) on demand

22:52:03 <_joshua> ah

22:52:29 <wkearney99>http://rdfig.xmlhack.com/index.rss

22:52:29 <dc_rdfig> W: http://rdfig.xmlhack.com/index.rss from wkearney99

22:52:47 <_joshua> that's not quite logging in RDF

22:52:49 <dajobe> similarly the logger bot writes rdf/xml logs and XSLTs into HTML

22:52:55 <wkearney99> W:

22:52:55 <dc_rdfig>http://rdfig.xmlhack.com/index.rss

22:53:15 <_joshua> nod

22:53:22 <dajobe> logger_1, where am i?

22:53:22 <dajobe> See http://ilrt.org/discovery/chatlogs/rdfig/2003-04-30#T22-53-22

22:53:26 <wkearney99> W:""

22:53:26 <dc_rdfig> Added comment W1.

22:53:40 <dajobe> look at http://ilrt.org/discovery/chatlogs/rdfig/2003-04-30.rdf for this very sentence

22:54:02 * wkearney99 dagnabbit, how does one delete a chumped url.... ugh, nevermind...

22:54:13 <dajobe> W:""

22:54:13 <dc_rdfig> Added comment W2.

22:54:20 <dajobe> that's odd

22:54:30 <wkearney99> uh, nope, yeah, I *tried* that

22:55:09 <dajobe> W:=http://www.w3.org/

22:55:09 <dc_rdfig> Replaced URL of W.

22:55:17 <dajobe> W:|w3c

22:55:17 <dc_rdfig> Titled item W.

22:55:37 <dajobe> W2:""

22:55:37 <dc_rdfig> Deleted comment W2.

22:55:39 <dajobe> W1:""

22:55:40 <dc_rdfig> Deleted comment W1.

22:55:46 <dajobe> that's it

22:56:20 <wkearney99> actually, that very sentence would be: http://ilrt.org/discovery/chatlogs/rdfig/2003-04-30.rdf#T22-53-40 wouldn't it?

22:56:47 <dajobe> that would be a URI

22:57:07 <dajobe> which has properties that contain the sentence

22:57:23 <dajobe> so... yes

22:57:27 * wkearney99 forgot about the cool rdf format from the chump...

22:57:43 <dajobe> it's in the sidebar on the right

22:57:51 <dajobe> maybe I should add an icon

22:58:19 * wkearney99 vaguely recalls his failure to use png instead of gifs somewhere...

22:58:24 <dajobe> yes

22:58:28 <dajobe> didn't you switch to css

22:59:10 <wkearney99> ah, right, for the [rdf] button, yes, css instead (but still haven't gotten around to fixing that either)

23:04:43 <dajobe> hmm, you have 2 blogs

23:05:21 <dajobe> well... who doesn't

23:06:38 <_joshua> i have zero blogs

23:06:51 <dajobe> you have something

23:06:58 <dajobe> I didn't even know it was you I was reading

23:07:18 <_joshua> hrr?

23:07:24 <dajobe> muxway

23:07:28 <_joshua> ha

23:07:30 <_joshua> that's not a blog

23:07:46 * dajobe shrugs

23:08:03 <_joshua> that's actually the deltas to a big database

23:08:06 <_joshua> changes in time order

23:51:58 <_joshua> dan moniz

23:52:39 <dnm> joshua schachter

23:52:50 <_joshua> do you still want a symbolics lisp machine?

23:52:53 <dnm> Yes.

23:53:03 <dnm> s/a/another/;

23:53:20 <dnm> All Your LispM Are Belong To Me.

23:54:53 <dnm> Have one?

23:55:01 <_joshua> no but

23:55:02 <_joshua> A Symbolics 3650 Lisp Machine. This is one of the coolest computers

23:55:02 <_joshua> ever made. I just have too much hardware and have to start thinning the

23:55:02 <_joshua> herd a bit. While some of the gear has been given away, the lispm is one

23:55:02 <_joshua> i paid a good bit of money for (i actually bought it surplus from Symbolics!)

23:55:02 <_joshua> and so this one is for sale. It'll go on ebay soon if there's no local

23:55:04 <_joshua> interest. The machine is fully operational , has a 750 MB disk, 4 megawords

23:55:06 <_joshua> (36 bit word size) of memory, and has Genera 8.3 (the latest version as of

23:55:08 <_joshua> 2001) installed. Some pictures of it can be found at the url listed on the

23:55:10 <_joshua> bottom.. just scroll about 2/3 of the way down. I'm asking $800.00.

23:55:12 <_joshua> (for those who dont know what a lisp machine is, it is an implementation of

23:55:14 <_joshua> the Lisp programming language _in hardware_ :)

23:55:19 <_joshua> Is that an interesting piece of hardware? Is that a good price? Should I forward to you?

23:55:31 <dnm> 3650. Hrm.

23:56:01 <_joshua> Why has nobody made an emulator? IP issues?

23:56:27 <dnm> If it has a console, keyboard, and all that, with all the cables, I'd consider it. The 3650 is slowish compared to other LispM models. I still have a currently non-functioning but was once functional XL1200 in Toronto.

23:56:40 <dnm> Anyway, I'd certainly be interested in talking and/or negotiating with the seller.

23:57:05 <dnm> Forwarding would be appreciated.

23:57:27 <dnm> As for emulators, IP issues yes, also, Symbolics still exists and sells/maintains Genera, OpenGenera, and old LispM hardware.

23:57:46 <dnm> Open Genera itself is a VM for Genera on top of OSF/1

23:57:52 <dnm> Runs on Alphas.

23:58:12 <_joshua> Ah