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

00:50:30 <sethl> danbri: have you compiled ruby 1.8 on mac os x?

02:09:56 <jordan> does anyone know where I can find a copy of the Description Logic Handbook online? googling didn't find any results

02:10:11 <jordan> a friend told me an online version exists

02:24:26 <Michael^2> Michael^2 is now known as Michael^

05:11:15 <DanC>http://www.oreilly.com/catalog/xsltckbk/

05:11:16 <dc_rdfig> A: http://www.oreilly.com/catalog/xsltckbk/ from DanC

05:11:33 <DanC> A:|XSLT Cookbook -- Solutions and Examples for XML and XSLT Developers

05:11:33 <dc_rdfig> Titled item A.

05:11:57 <DanC> A:arrived in a care package from O'Reilly today

05:11:57 <dc_rdfig> Added comment A1.

05:12:37 <DanC> A:IB5 978059600K83890000 (via cuecat)

05:12:37 <dc_rdfig> Added comment A2.

05:13:18 <DanC> A2:IB5 978059600372290000

05:13:18 <dc_rdfig> Replaced comment A2.

05:13:49 <DanC>http://www.oreilly.com/catalog/consynrss/

05:13:49 <dc_rdfig> B: http://www.oreilly.com/catalog/consynrss/ from DanC

05:13:57 <DanC> B:|Content Syndication with RSS

05:13:57 <dc_rdfig> Titled item B.

05:14:09 <DanC> B:By Ben Hammersley March 2003

05:14:09 <dc_rdfig> Added comment B1.

05:14:26 <DanC> A:by By Sal Mangano

05:14:26 <dc_rdfig> Added comment A3.

05:14:33 <DanC> A:December 2002

05:14:33 <dc_rdfig> Added comment A4.

05:14:50 <DanC> B:IB5 978059600K83890000 (via cuecat)

05:14:50 <dc_rdfig> Added comment B2.

05:16:55 <DanC>http://conferences.oreillynet.com/etcon/

05:16:55 <dc_rdfig> C: http://conferences.oreillynet.com/etcon/ from DanC

05:18:39 <DanC> C:|O'Reilly Emerging Technology Conference -- April 22-25, 2003

05:18:39 <dc_rdfig> Titled item C.

05:19:00 <DanC> C:this looks like fun. bummer I'm not going

05:19:00 <dc_rdfig> Added comment C1.

05:44:29 <DanC> A:good stuff on date calculations and formatting

05:44:29 <dc_rdfig> Added comment A5.

05:51:23 <DanC> A:what's the license for all the code in this book? can I download it somewhere?

05:51:24 <dc_rdfig> Added comment A6.

05:52:32 <DanC> A:yes, [examples|http://examples.oreilly.com/xsltckbk/] are available

05:52:32 <dc_rdfig> Added comment A7.

05:52:47 <DanC> A:3.6M! compressed!

05:52:47 <dc_rdfig> Added comment A8.

06:20:57 <DanC> .google TRN airport

06:20:59 <datum> TRN airport: http://www.airport-car-deals.com/car-rental/TRN_torino/airport-car-rental.html

06:21:11 <DanC> no, no, silly...

06:25:43 <chaalsTRN>http://jibbering.com/2002/9/airports.svg?IT

06:25:44 <dc_rdfig> D: http://jibbering.com/2002/9/airports.svg?IT from chaalsTRN

06:26:00 <chaalsTRN> D:Map of airports in IT for which Jim has data.

06:26:01 <dc_rdfig> Added comment D1.

06:26:13 <chaalsTRN> D:|Map of airports in IT for which Jim has data.

06:26:13 <dc_rdfig> Titled item D.

06:26:26 * DanC tries to load it

06:26:32 <chaalsTRN> D1: Would be nice if it could accept IATA coads too.

06:26:33 <dc_rdfig> Replaced comment D1.

06:27:52 <chaalsTRN> D: An SVG that gets the data and puts spots on things. Known to work in IE using Adobe plugin. I think it works in Batik too. Don't know about Mozilla

06:27:53 <dc_rdfig> Added comment D2.

06:28:53 * chaalsTRN wonders if there is a public example of Max F's animated IRC record

06:29:20 <DanC> operation timed out.

06:32:59 <eikeon> .google TRN

06:33:00 <datum> TRN: http://trn.sourceforge.net/

06:33:00 <dc_rdfig> Label TRN not found.

06:35:21 <DanC> Name: Torino

06:35:22 <DanC> IATA: TRN

06:35:22 <DanC> ICAO: LIMF

06:35:22 <DanC> Turin/Torino, Italy

06:35:22 <DanC> Latitude: 45\x{00B0}12' 2" N

06:35:22 <dc_rdfig> Label IATA not found.

06:35:22 <DanC> Longitude: 7\x{00B0}38' 58" E

06:35:23 <dc_rdfig> Label ICAO not found.

06:35:57 <DanC> -- http://www.ar-group.com/Airport-Locator.asp?RequestCity=turino&RequestLocation=&1049438117116

06:36:45 <DanC> my galeon installation is hosed (debian mozilla/gtk2 conflict). What browser do you use, eikeon?

06:37:15 <DanC> I'm trying mozilla again; it's faster than what I remember.

06:47:26 <eikeon> DanC: Mostly galeon these days.

06:48:00 <eikeon> IE if I swivel my chair over to XP.

06:49:00 <eikeon> Latitude: 45°12' 2" N

06:49:00 <eikeon> Longitude: 7°38' 58" E

09:09:21 <golbeck> golbeck is now known as golbeck_zzz

09:20:43 <arnarl> arnarl is now known as arnarl|away

10:34:36 <danbri_lap> libby? re rdfq testing, http://www.w3.org/2001/12/rubyrdf/pack/tests/gqt.rb

10:34:44 <danbri_lap> ...i switched over to generating a test script from another script

10:35:04 <danbri_lap> -> http://www.w3.org/2001/12/rubyrdf/pack/tests/_genq.rb

10:35:16 <danbri_lap> needs a bit of tidying up, but works better than the old approach

10:36:33 <libby> nice one :)

10:52:00 <danbri_lap> do you have some queries + manifest i can use? its time to migrate to the new format(s)...

10:52:14 <danbri_lap> ...and test data

10:55:07 <libby> not quite yet....

10:55:23 <libby> alberto's format example is what I've been using at the moment.

10:55:52 <danbri_lap> logger_1, pointer?

10:55:52 <danbri_lap> See http://ilrt.org/discovery/chatlogs/rdfig/2003-04-04#T10-55-52

10:56:01 <libby> all my old format testcases are here: http://swordfish.rdfweb.org/rdfquery/tests/, http://swordfish.rdfweb.org/rdfquery/tests/readme.txt

11:03:10 <danbri_lap> thanks

11:03:32 <danbri_lap> some stuff in my manifest is to do w/ checking syntax of the squish, don't think that's in albertos format

11:05:21 <libby> yeah, think you're right

11:05:34 <libby> though he was some stuff in rdf store that does that too.

11:05:34 <danbri_lap> I need to convert my rdfcore testing thingy too

11:05:49 <libby> cool...

11:16:10 <danbri_lap> hi sandro

12:18:12 <libby> hi jeen

12:19:39 <Jeen> hey libby

12:26:20 <libby> jeen, do you hsaave any sample swrql queries to paste in to the demo?

12:26:39 <Jeen> not yet, but I can come up with some if you like.

12:26:56 <Jeen> there's a couple of sample queries in the user manual, also.

12:27:22 <Jeen> (but those don't work on the demo repositories directly, different dataset)

12:32:47 <sbp-> sbp- is now known as sbp

12:32:49 <libby> I haven't looked at the manual yet...a few sample queries near here: http://sesame.aidministrator.nl/sesame/ would be useful

12:33:50 <Jeen> yeah, I agree. and so does arjohn. we're working on it right now...

12:35:02 <libby> neat :)

12:35:10 <libby> I'll check out the manual

12:44:25 <libby> I'm a big examples fan. cool there are some in the manual.

12:44:45 <libby> did you alreday add to Alberto and Andy's usecases doc jeen btw?

12:47:11 <JibberJim> D: Hmm, that should be easy, but I've not plugged it together anywhere, best chance to get something similar is to mark up some [http://jibbering.com/trip/ams.rdf|RDF for your journey] and have it [http://jibbering.com/trip/display.svg?url=ams.rdf|displayed].

12:47:17 <dc_rdfig> Added comment D3.

12:47:56 <Jeen> libby: er, for SeRQL you mean? again, not yet, we just finished implementing the engine today :)

12:48:09 <Jeen> but thanks for reminding me, I'll do that.

12:49:32 <libby> sorry, didnt mean to bug you. just be good to see teh comparison.

12:49:53 <libby> nice work :))

12:50:22 <Jeen> if I didn't want to be bugged I wouldn't hang out here ;)

12:54:34 <libby> heheh

13:14:20 * danbri_goneout finishing up delayed rdfcore minutes from last week

13:14:23 <danbri_goneout> danbri_goneout is now known as danbri

13:19:27 <Jeen> alright, I put some example queries online for our museum repository.

13:21:26 <Jeen>http://sesame.aidministrator.nl/sesame/serql/serql-examples-museum.txt

13:24:29 <sandro_> sandro_ is now known as sandri

13:24:32 <sandri> sandri is now known as sandro

13:25:31 <dc_rdfig> E: http://sesame.aidministrator.nl/sesame/serql/serql-examples-museum.txt from Jeen

13:25:42 <_Wack> _Wack is now known as Wack

13:32:57 <sandro> is it just me getting hosed by this irc server?

13:33:05 <danbri> mostly, yes

13:33:23 <danbri> a lot of: sandro has quit ("Getting off stoned server - dircproxy 1.0.3")

13:34:53 <sandro> fooey

13:35:05 <sandro> care to read http://esw.w3.org/topic/MultipleCorrectBehaviors ?

13:39:49 <danbri> stuck in 1st sentence, could you clarify what 'intended' binds to? whose intention?

13:40:27 <danbri> the creators of documents and services might intend things; URIs, as textual names, don't...

13:41:51 <danbri> is http://www.w3.org:80/ vs http://www.w3.org/ useful eg for MCB?

13:46:06 <sandro> no, they produce the same user-agent behavior.

13:46:59 <sandro> intended (by the URIs controlling authority)

13:47:31 <danbri> do all URIs have controlling authorities?

13:47:54 <sandro> Not sure, but all http ones do.

13:51:31 <danbri> hmm http://c2.com/cgi/wiki?HowToDeletePages I tried this in ESW wiki, didn't work...

13:54:01 <danbri> dajobe, could you add a link <a href="http://esw.w3.org/topic/FrontPage">ESW Wiki</a> or similar from http://rdfig.xmlhack.com/ ?

13:54:38 <dajobe> what's the label

13:54:50 <dajobe> ESW wiki isn't really very descriptive

13:55:05 <dajobe> is there an events link yet?

13:56:40 <danbri> ScheduledTopicChats is the thing to link to, though needs more content

13:57:55 <danbri> It's name is the 'ESW Wiki'; for explanation, you could say 'Semantic Web developer collaborations...', perhaps

13:58:44 <danbri> this -- http://jibbering.com/2002/9/airports.svg?IT -- is so cool :)

13:59:28 <dajobe> at the top or sidebar?

13:59:44 <danbri> top please

14:00:24 <dajobe> reload

14:00:40 <dajobe> anything else while I'm here?

14:00:53 <danbri> thanks... thinking...

14:01:00 <danbri> "Semantic Web developer collaborations are made at the ESW Wiki."

14:01:16 * danbri wonders about the word 'document' in there somewhere

14:01:24 <danbri> nah its fine f'now

14:01:45 <JibberJim> -- source RDF data is available from http://jibbering.com/rdf/airports-country.1?IT;FI ...

14:01:45 <dajobe> ok

14:01:48 * dajobe :wq

14:01:48 <danbri> Maybe you could like 'Semantic Web' to http://www.w3.org/2001/sw/

14:01:50 <danbri> doh

14:01:55 <dajobe> ok

14:01:58 <sandro> re the name"ESW", I was recently asked what was up with this new "Extended Semantic Web" or something.... I rephrased the page on the wiki to downplay the word "extended".

14:02:22 <dajobe> reload

14:02:50 <em> danbri, I'd suggest something like 'Semantic Web Advanced Development Collaboratory' (or something) in a larger font than the text on the frontpage of the wiki as well

14:02:56 <dajobe> +1

14:03:11 <danbri> " This wiki can also link to InterWiki space, and is usually known as ESW from the outside (moin) world. ESW can stand for Evolving, European, Experimental, Extended, Enthousiastic, ... Semantic Web, reflecting its origins in the SWAD-Europe project, and affiliation with the wider RDF / Semantic Web Interest Group." -- http://esw.w3.org/topic/FrontPage

14:03:21 <danbri> E* :)

14:03:23 * em notes he could do this now that its a wiki

14:04:03 <em> as long as 'E' dosent stand for 'Eric', i'm happy :)

14:05:11 <danbri> I'm finding it pretty hand, seems to remove a *lot* of the "Can I be bothered to go in and edit that web page, yeah I'll do it tommorrow" lethargy...

14:05:12 <dajobe> do you want to do any RDF -> RDF/SW gentle renaming

14:05:52 <libby> like the differnet versions of E* danbri :)

14:05:59 <danbri> I that was sandro i think

14:06:18 <danbri> dave, could you link 'IRC' to: http://esw.w3.org/topic/InternetRelayChat

14:06:20 <dajobe> does the wiki have a cvs-like annotate thing? who wrote what line?

14:06:23 <dajobe> ok

14:06:35 <dajobe> reload

14:07:15 <sandro> click on the "?" icon to get into the change history. It's not as easy to use as I'd like. danbri wrote most of the line, I added a few more E's.

14:07:41 * em changes s/Enthousiastic/Enthusiastic on frontpage

14:07:56 * danbri happens to be listening to the shaman, hence "ezergood, eberneezergood" pops into his head

14:08:08 <danbri> i linked the scratchpad from FrontPage

14:08:09 <dajobe> once more with spelling? :)

14:08:20 <dajobe> is scratchpad ok

14:08:26 <dajobe> we tend to call it the chump

14:08:28 <danbri> em, do you have some SW developer logo could use?

14:08:44 * dajobe notes the chump is *not* powered by RDF. XML+XSLT

14:08:59 <danbri> Yeah, I'm not sure... Chump is the tool really, and a verb too (one 'chumps' a link...), scratchpad works for me

14:09:04 <dajobe> np

14:10:46 <danbri> sandro, can you advise on which version of MoinMoin is best to install? I tihnk i'll switch over my other wiki to use same tools

14:10:58 <dajobe> oh, events link?

14:12:19 <em> there is the activity icon on the /2001/sw/ page ( http://www.w3.org/2001/sw/sw ) and RDF icons http://www.w3.org/RDF/icons but not a Semweb Developers specific logo... i've asked the media team that did the rdf logos to take a crack at this but havent had time to follow-up ...

14:12:21 <danbri> good idea, yes please... to ScheduledTopicChat

14:12:49 <dajobe> the semweb activity too w3c specific, em I think

14:13:12 <danbri> yep, given the site is not a w3.org one, though this is the channel of a W3C group, so is some defense

14:13:26 * dajobe phrased it badly

14:13:31 <em> really? thought it was fairly generic

14:13:47 <em> not very exciting mind you; but generic

14:13:53 <dajobe> the "RDF developer" one might fly ?

14:13:55 <danbri> em, I heard a rumour you'd scheduled an #rdfig ScheduledTopicChat on XTM & RDF for next week...

14:14:03 <danbri> could you document it in http://esw.w3.org/topic/ScheduledTopicChat please!

14:14:25 <danbri> libby, could you summarise the upcoming calendar and query ones there too?

14:14:38 * DanC tunes in to discussion of what to call http://rdfig.xmlhack.com/

14:14:58 <DanC> it calls itself the RDF IG Scratchpad. So for consistency, I tend to go with that.

14:15:13 <danbri> me too

14:15:14 <em> yes, but the focus was a review of peppers document not a XTM & RDF generic discussion (though I suspect education on both sides will in part result)

14:15:32 * dajobe stuck events in the sidebar, maybe not?

14:15:44 <danbri> em, i'm happy with whatever scope, as long as someone writes it down somewhere!!!

14:15:56 <em> danbri, will do

14:16:17 <danbri> Seriosuly, things are picking up with topic chats... DanC and I spoke to Bijan+JimH about a WSDL/RDF/OWL one, Bijan has the ball...

14:16:28 <danbri> ...and talking w/ Charles about an ImageDescription one, etc...

14:16:37 * dajobe notes "ScheduledTopicChats" in wiki page "ScheduledTopicChat"

14:16:51 <danbri> Is it plural, i can never remember...

14:17:14 <JibberJim> Could you organise a date for that one soon danbri, as I'm losing regular access shortly, and won't see things...

14:17:29 <arnarl|away> arnarl|away is now known as arnarl

14:17:30 <danbri> the image one?

14:17:36 <danbri> ok i'll mail you too

14:18:39 * dajobe notes "W3C" appears nowhere on this page ;)

14:18:41 <JibberJim> yes the image one - mind you'll I'll try and make the geo calendaring one next week, so can find out then.

14:18:48 * dajobe notes that might be a feature :)

14:20:33 <danbri>http://www.autoidcenter.org/main.asp

14:20:34 <dc_rdfig> F: http://www.autoidcenter.org/main.asp from danbri

14:20:43 <DanC> re ScheduledTopicChat: remember automatic back-links. i.e. make a page for the topic your chatting about and talk about having a ScheduledTopicChat there. then we get backlinks for free

14:20:55 <danbri> F:|Autho-ID center, "Identify any object anywhere automatically"

14:20:55 <dc_rdfig> Titled item F.

14:21:10 <dajobe> there are still no chats scheduled ;(

14:21:12 <danbri> F:More RFID interestingness

14:21:12 <dc_rdfig> Added comment F1.

14:21:45 <DanC> calendar chats are scheduled every other week thru WWW2003

14:21:49 * dajobe walks away from the editor

14:22:23 <DanC> pls don't start pages with headings. If you want to make a SemanticWebCollaboratory page, do so. But the FrontPage's name is FrontPage

14:23:03 <DanC> I think the default (/esw/) page could even be changed to SemanticWebCollaboratory.

14:31:54 <em> danbri, ttp://esw.w3.org/topic/ScheduledTopicChat updated (looks like by more than me as well)

14:32:08 <dajobe> . http://esw.w3.org/topic/ScheduledTopicChat

14:33:36 * DanC moved collaboratory heading into running text in FrontPage

14:34:54 <DanC> eric, could I get you to change "the RDF standards" to "RDF, XML, and URI technologies" on http://www.w3.org/2001/sw/?

14:35:02 <DanC> (or: agree to the change so I can make it?)

14:35:22 <DanC> and s/Overview/Contents/ or In this page:

14:35:39 * DanC noodles on http://www.w3.org/2000/01/sw again / some more

14:36:05 <em> yes (i aree to change do your suggested changes and ok with you doing this)

14:36:32 * em thinks delete key stopped working...

14:37:06 <em> +1 to <DanC> I think the default (/esw/) page could even be changed to SemanticWebCollaboratory.

14:37:06 <DanC> can "abstract representation" go?

14:37:43 <danbri> can "in collaboration with a large number of researchers and industrial partners." go? (it's redundant: being developed by W3C implies that...)

14:37:44 <em> yes to abstract representation

14:38:24 <DanC> hmm... "other standards to be defined"... would be nice to exploit the Framework idea... "other technologies in the Resource Description Framework". that's not quite right, but do you get the idea?

14:38:24 <danbri> ...or are there 'industrial partners' who are not in the Consortium that you're referring to?

14:38:33 * danbri nods, likes

14:38:50 * em has a new paragraph he was planning on getting on there... tim's intro paragraph didnt work for me but havent had tiome to fix

14:39:24 * em has to do 2 things before rdfcore call... has to run

14:39:57 <danbri> can you post your draft here? (after rdfcore...?)

14:40:11 <em> yes

14:41:02 <danbri> ta

14:44:42 <DanC> pls review http://www.w3.org/2001/sw/ v 1.150 2003/04/04 14:44:07 connolly

14:45:12 <danbri> anyone seen Bijan today?

14:45:37 <DanC> danbri? /2001/sw/ para v1.150 ok by you?

14:46:52 <danbri> looks good, wondering about '_the_ representation'

14:47:25 <danbri> Anyone representing data in the Web (eg. MS Access .mdb files; csv, random XML...) has put data into the Semantic Web?

14:48:35 <DanC> no, but anyone doing semantic web has put data on the web.

14:50:06 <DanC> I can see that it could be better, but I'm not further inspired. As long as we agree that it's better than it was before, I won't back this change out.

14:50:08 * danbri ponders "The Semantic Web is the representation of data" -> "is a strategy for representing data {on|in} the Web"

14:50:36 <DanC> well, the semantic web is the result of the strategy

14:51:06 <danbri> Is the Semantic Web, as a thing... a practice, vs a dataset... yeah, its the aspect of the Web that does data the RDFish way...

14:51:14 <danbri> yup, strategy wasn't right.

14:51:45 <danbri> Your current text allows both readings 'representation' can be read as 'act of representing' and also 'result of representing'

14:51:49 <danbri> hey chaals

14:52:24 * DanC focuses on the ailing http://www.w3.org/2000/01/sw/

14:54:51 <DanC> btw, libby, danbri, I'm at risk, at best, for RdfCalendar next weds. I'll be at DAML PI.

14:55:03 <libby> bummer...

14:55:17 <danbri> pity.

14:55:23 * danbri wonders what to do by then, on calendar...

14:55:56 <libby> we dont have an agenda yet, but geo stuff was mooted, esp since jibberjim can some at that time

14:56:20 <libby> I've got a bunch more testcases I found to try out

14:56:35 <danbri> i want to do more on opening hours use case

14:56:49 <danbri> try show link to restaurant review scenario

14:56:54 <libby> I want to do timetables usecase

15:00:37 <libby> timestables are tricky. the local buus one has 0700, 0800, every 30 mins until 1800, 1900 etc etc; then different saturday, sunday and bank holidays....

15:01:03 <DanC> mooted... that's like tabled. I think it means the opposite thing in the US/UK

15:01:55 <DanC> why is the semweb news in a font so small I can't read it? http://www.w3.org/2001/sw/#events

15:05:05 <em> cause div.item has font-size: 80%; .. i though i removed this

15:05:40 <em> if you havent made mods to page yet; please add removing that from css

15:07:22 <libby> perhaps I'm using mooted wrong: thought it was something like 'argued for' from vague memories of student lawyers' moots. probably totally wrong!

15:07:57 <danbri> anbri@fireball:~$ wn mooted -simsv

15:07:58 <danbri> Synonyms (Grouped by Similarity of Meaning) of verb moot

15:07:58 <danbri> Sense 1

15:07:58 <danbri> consider, debate, moot, turn over, deliberate

15:07:58 <danbri> => discuss, talk over

15:07:58 <danbri> --------------

15:16:45 * DanC commits v 1.151 2003/04/04 15:16:11 http://www.w3.org/2001/sw/ , releases write lock

15:16:48 <DanC> ah... I can read the news now.

15:17:07 <em> :)

15:17:19 <em> thanks danc

15:17:57 * DanC returns to /People/Connolly/events/ , RDF formalization thereof

15:18:42 <DanC> hmm... contact:Person or foaf:Person? both is doable, but sorta a pain

15:24:27 <danbri> http://xmlns.com/foaf/0.1/ [[

15:24:28 <danbri> <rdfs:subClassOf rdf:resource="http://xmlns.com/wordnet/1.6/Person"/>

15:24:28 <danbri> <rdfs:subClassOf rdf:resource="http://www.w3.org/2000/10/swap/pim/contact#Person"/>

15:24:30 <danbri> ]]

15:24:52 <DanC> hm... RDFApplicationsAndProjects ... too awkward to expect lots of backlinks... are they RDF applications, RDF schemas, or RDF vocabularies?

15:25:17 <dc_rdfig> G: http://xmlns.com/foaf/0.1/ from danbri

15:26:07 <DanC> CategoryFaq

15:26:14 <DanC> CategoryQandA

15:31:24 <DanC> bummer, CategoryQandA isn't a MoinMoin:WikiWord

15:34:35 <DanC> hmm... MoinMoin uses FooMarket... let's try that...

15:34:40 <DanC> VocabularyMarket

15:34:46 <DanC> SchemaMarket

15:35:10 <DanC> danbri, thoughts on investing in "RDF Schema" vs. "RDF Vocabulary"?

15:35:31 <DanC> is a schema a vocabulary description?

15:36:27 <danbri> yes

15:36:32 * danbri prefers Vocabulary

15:36:53 <em> +1 to vocab as well

15:36:59 <danbri> ... neutral ground re OWL

15:38:27 <DanC> ok, VocabularyMarket

15:38:39 <dajobe> +1 vocab, schema must die :)

16:05:27 <golbeck_zzz> golbeck_zzz is now known as golbeck

16:24:44 <aBhaid> hello

16:25:12 <aBhaid> Anybody here know much about RDF?

16:27:06 <libby> some....

16:27:11 <libby> anything in particular?

16:27:54 <aBhaid> Well I am just getting into RDF coming from a Client/Server background and so far my head hurts!

16:29:01 <libby> any particular aspect of hurtyness?

16:29:06 <Morbus> perhaps http://disobey.com/d/2002/sw123/

16:29:50 <aBhaid> One thing that really puzzles me is - where does 'normal' SQL fit in to all this

16:30:38 <libby> you can store RDF in SQL databases; or you could output RDf from an SQL database...

16:31:24 <aBhaid> Yes, I realise this but the data has to be put and gotten with this special SQL, eg. SELECT ?x, ?y kind of lingo

16:32:25 <libby> not sure what you mean - RDF data is just data, you can store it in SQL just like you can any data, using sql...

16:33:41 <aBhaid> OK, let me express this a differnt way

16:35:13 <aBhaid> RDF is triple based, so S,P,O gets stored in a table S in Col1, P in Col 2 and O in Col3. Correct?

16:35:55 <libby> there are various ways of storing, and that's probbal;y the simplest, yep

16:37:10 <aBhaid> So in effect the semantics of the relational model is not available physically, but it is available through the assertions made in RDF and stored in the RDF schema. Is this so?

16:37:23 <libby> there is some more info about various SQL schemas for RDF here I think: http://www.w3.org/2001/sw/Europe/reports/scalable_rdbms_mapping_report/

16:37:27 <libby> (section 4)

16:37:48 <aBhaid> I like the sound of that

16:38:01 <libby> it depends on the sophistication of the relational mapping...

16:38:06 <libby> but often, yep

16:38:56 <libby> - this is because in general you don;t know the shape of your rdf data

16:39:06 <aBhaid> Do you mean that you can have what I would call hybrid mixed of pure RDF triple data and relational style data?

16:39:57 <libby> well eric prud'hommeaux has been working on a system that analyses a simple triplestable for regularities that can be fitted into a normal relational db. just looking for the ref for you

16:40:27 <libby> here we go: http://dev.w3.org/cvsweb/perl/modules/W3C/Rdf/bin/compileTables

16:41:23 <libby> there are probably more, which peple use for specific purposes. if you know what your data is going to look like, no reason why you shoudl not put it in a realational database I guess, though I don;t know about any apis or query language interfaces to that

16:41:35 <libby> doesn't mean there arent any....

16:44:20 <libby> there's also eric work on mappig RDF query languages to relational dbs. Jan grant has done ome on that too.

16:45:15 <aBhaid> I will have to read some more but the impression I have so far is that and what I know from SQL is that once you store data in a triple fashion, even attribute-value pairs, you lose the capability of querying the data 'easily' in SQL.

16:45:28 <libby> here are some links for that: http://www.w3.org/2001/sw/Europe/reports/rdf_ql_comparison_report/#faq24

16:45:55 <libby> and here: http://www.w3.org/2001/sw/Europe/reports/rdf_ql_comparison_report/#faq22

16:46:15 <libby> hm, as far as I know you are correct.

16:46:55 <libby> all the schemas I know about for generic storgae are slower than ordinary data

16:47:09 <libby> (this is not true of Berkeley DB-based systems though)

16:47:31 <aBhaid> Berkeley DB?

16:48:32 <libby> yeah, I've never had much success with it, but it's fast, and 9for example) reland and jena systems use it (or can be made to)

16:48:40 <libby> redland rather

16:49:05 <libby> this is all in my opinion. there's a great deal out there...

16:49:20 * jhendler sorry if already topic of discussion - but is the chump server hosed (can't check logs as we can't get to server)

16:50:17 <libby> logger_1, pointer?

16:50:17 <libby> See http://ilrt.org/discovery/chatlogs/rdfig/2003-04-04#T16-50-17

16:50:29 <libby> I can;t get through either jhendler

16:51:08 <libby> BLURB:chatting with aBhaid about relational databases and RDF

16:51:12 <libby> hm

16:51:29 <aBhaid> Is it Ok to continue?

16:51:34 * jhendler thanks - scratchpad just loaded again - suspect it was just something temproary - sorry to both folks

16:52:35 <libby> it's fine anyway aBhaid, just our weblog: http://rdfig.xmlhack.com

16:53:45 <jhendler>http://www1.timesofindia.indiatimes.com/cms.dll/html/uncomp/articleshow?msid=42151091

16:54:13 * jhendler hmm...

16:54:48 <libby> when you put a url in on its own it's supposed to write to the page aBhaid...

16:55:20 <jhendler> supposed to, being the currently operative word...

16:55:37 <aBhaid> you mean it wasn't working a moment ago?

16:56:13 <jhendler> yes, and now doesn't seem to be either --

16:56:17 <jhendler>http://www1.timesofindia.indiatimes.com/cms.dll/html/uncomp/articleshow?msid=42151091

16:56:32 <Soultaker> funny; are we talking about Berkeley DB for storing RDF data?

16:56:48 <libby> kinda, yep

16:56:58 <libby> I don;t know much aboutit though

16:57:13 <aBhaid> Yes, is it an RDBMS or a 'special' DB for RDF type data

16:57:22 <dc_rdfig> H: chatting with aBhaid about relational databases and RDF from libby

16:57:32 <libby> is special, but not specialized for RDF though

16:57:40 <Soultaker> ah, I happen to be implementing an RDF storage system with it as we speak :)

16:57:47 <libby> ooh, check it out, chump's back

16:58:01 <aBhaid> What is the actaul name 'Berkeley DB'?

16:58:06 <libby> H:[logs|http://ilrt.org/discovery/chatlogs/rdfig/2003-04-04#T16-50-17]

16:58:18 <Soultaker> yes, that's the actual name

16:58:27 <dc_rdfig> I: http://www1.timesofindia.indiatimes.com/cms.dll/html/uncomp/articleshow?msid=42151091 from jhendler

16:58:29 <dc_rdfig> J: http://www1.timesofindia.indiatimes.com/cms.dll/html/uncomp/articleshow?msid=42151091 from jhendler

16:58:34 <dc_rdfig> Added comment H1.

16:59:46 <jhendler> grrr, now I got it twice

17:00:10 <jhendler> I:| chump stutter, ignore me

17:01:02 <jhendler> J:| Al-Jazeera outscores sex on the net (from the Times of India)

17:01:42 <jhendler> J: not strictly SemWeb, but timbl has been heard to complain about the amount of sex stuff on the web

17:01:45 <dc_rdfig> Titled item I.

17:01:46 <dc_rdfig> Titled item J.

17:01:59 <jhendler> J: thought this one would make him feel a little better...

17:02:55 <dc_rdfig> Added comment J1.

17:03:05 <aBhaid> On the difference between realtional storage and triple style storage. If data is stored in triple style, does one "re-relationalise" the data so that it can be SQL'ed set style or is the data accessed purely via APIs

17:03:47 <dc_rdfig> Added comment J2.

17:04:07 <libby> you can do it via queries for example, they just look a bit peculiar, and aren;t very efficient

17:04:25 <aBhaid> Is this the SquishSQL for example?

17:04:34 <libby> and here: http://www.w3.org/2001/sw/Europe/reports/rdf_ql_comparison_report/#faq22?

17:04:56 <aBhaid> Isn't efficiency a problem?

17:05:21 <libby> squishql does have a mapping to realational db, a simpel schema. the original was by Matt biddulph. think various others have mappings too (e.g eric prud'hommeaux)

17:05:29 <libby> efficiency, yep

17:05:31 <libby> lack of

17:05:36 <aBhaid> Not to mention integrity

17:05:45 <libby> but, I'm not looking at huge scales at the moment

17:05:59 <libby> did you have any huge apps in mind?

17:06:55 <libby> the berkeley db stuff is very scalable and fast, in contrast

17:07:40 <aBhaid> I must check this one out.

17:08:12 <libby> I think that it is very low-level. probbaly what you want it things built on top of it, like REdland

17:08:42 <aBhaid> What is REdland?

17:09:13 <libby> as well as a part of bristol, UK, it's also a RDF db. shoudl be able to google for it

17:11:12 <danbri> see also http://lists.w3.org/Archives/Public/www-rdf-interest/2002Aug/0129.html

17:11:49 <aBhaid> I can imagine a layer on top of triple'ly stored data in an RDBMS (if it is OK to say it like that) where the triple data is transformed into relation sets (lots of temp tables) that can then be operated on with 'normal' SQL. Sounds like a lot of legwork though.

17:12:37 <libby> ericp or jang are your people for that sort of thing (Eric Prud'hommeaux and Jan Grant)

17:13:15 <Soultaker> why would you want to use 'normal' SQL for RDF queries anyway?

17:13:35 <aBhaid> Does what I am saying make sense somehow?

17:15:19 <libby> I think so. I'm not an expert at sql

17:15:43 <libby> mormal sql querieswould be nice and fast

17:16:42 <Soultaker> I am not convinced SQL is really suited to the kind of queries one would like to perform on an RDF database

17:17:08 <libby> I think, aBhaid, that ericp is doing something like that: http://rdfig.xmlhack.com/2003/03/25/2003-03-25.html#1048622430.568575

17:17:21 <Soultaker> I'm not saying it isn't, but I can imagine some better specialized database system working better

17:17:22 <libby> though not temporary tables, as it's not trivial to generate them

17:18:48 <aBhaid> I think I would want to use 'normal' SQL because although RDF seems to have been concieved for web resource, from what I read it can be used to store ordinary data as well, e.g. a product catalog. Now that catalog must be integrated with other catalogs on heterogenous systems for example and here is where RDF and XML come in. But to query the original catalog on a single DB it (RDF queries) seems like a big hammer for a small nut

17:21:10 * DanC tries to figure out how to tell gnome to use evolution for mailto: support

17:23:15 <libby> aBhaid, I think you're right. RDF's useful when you don;t control your data, e.g. if it may change over time substantially

17:23:54 <libby> to an extent, all data changes of course...

17:24:48 <aBhaid> Even then, isn't there an issue with changing the 'shape' of the 'ontology' in RDF? something I remember about "instances"

17:25:00 * libby gets rdf query manifest and resultset testing working roughly :)

17:25:14 <Soultaker> you have a point there, but I think it's not all that useful to have an RDF database and then use SQL on it because you don't plan to do 'real' RDF queries

17:25:42 <libby> most of my query stuff doesn't use schemas, so changes in the schema dont affect querying

17:27:04 <aBhaid> Yes, I agree but when a DB is integrated with heterogenous systems in an Extended Exterprise, then RDF move up the "Top 10 List", does it not. And you one has their data in a 'shape' that is inefficient for SQL but highly efficient for interoperability.

17:28:49 <bijan> Even then, isn't there an issue with changing the 'shape' of the 'ontology' in RDF?

17:28:54 <bijan> (Sorry, quote)

17:29:06 <bijan> Answer: No. RDF instances aren't like OOP language instances.

17:29:55 <bijan> Or rather, there are many instances, but they are nothing like those found in the OOP world.

17:29:58 <bijan> er...

17:30:02 <bijan> s/instances/issues/ :)

17:34:50 <aBhaid> Isn't there a stage where changing "things" in an RDF schema or something or other, gets problematic?

17:39:51 <bijan> Well, in some sense yes, in some sense no

17:40:48 <bijan> Whenever you change something, obviously, there's an opportunity for difficulty

17:40:55 <aBhaid> Got to go, dinners up

17:40:56 <bijan> But the difficulties are of a very different sort

17:41:11 <aBhaid> Thanks

17:44:08 <_joshua> hi

18:21:00 <DanC> wow... that took a while... I'm resuming work on formalizing my index of trips and talks, documenting the issues and design patterns in the ESW Wiki (http://esw.w3.org/topic/) along the way... I just spent an hour or so getting the right context for PropertiesForNaming

19:48:59 <arnarl-home> hi

19:57:00 <tav> tav is now known as tav|offline

19:58:49 <tav|offline> tav|offline is now known as tav

20:45:38 <DanC> argh! it's really frustrating that RDF property elements can't be syntactically inverted.

20:47:38 <danbri_lap> yup... depicts vs depiction has been annoying

20:50:23 <DanC> aha... I figured out a work-around using rdf:value as a sort of noop.

20:50:46 <DanC> and I have to connect a couple things with names.

20:58:25 <sandro> It's tough to workaround when you literals involved, though..... owl:sameAs and owl full, I guess...?

21:01:31 <danbri_lap> DanC, thanks for the massive sort out of http://esw.w3.org/topic/GeoInfo

21:02:04 <danbri_lap> I was feeling a bit swamped by the number of (uncatalogued) links I'd hoarded...

21:04:03 <danbri_lap>http://www.xulplanet.com/tutorials/xultu/rdfds/

21:04:07 <dc_rdfig> K: http://www.xulplanet.com/tutorials/xultu/rdfds/ from danbri_lap

21:04:12 <danbri_lap> K:|http://www.xulplanet.com/tutorials/xultu/rdfds/

21:04:12 <dc_rdfig> Titled item K.

21:04:37 <danbri_lap> K:|rdfds - RDF Datasource Library in JavaScript (for Mozilla RDF)

21:04:37 <dc_rdfig> Titled item K.

21:06:53 <sandro>http://esw.w3.org/topic/MultipleCorrectBehaviors

21:06:53 <dc_rdfig> L: http://esw.w3.org/topic/MultipleCorrectBehaviors from sandro

21:07:20 <sandro> L:|There are two kinds of resources....

21:07:21 <dc_rdfig> Titled item L.

21:09:50 <danbri_lap> L:Only two?

21:09:50 <dc_rdfig> Added comment L1.

21:18:34 <sandro> L:| The MultipleCorrectBehaviors Test: Defining RDF's "New Breed" of URIs

21:18:34 <dc_rdfig> Titled item L.

21:21:44 <sandro> L1:

21:21:44 <dc_rdfig> (danbri_lap) Only two?

21:21:47 <DanC> ok, scraping my index of events for foaf:knowsWell is going well: {{{

21:21:49 <DanC> :id2594090 k:endingDate [

21:21:49 <DanC> dt:date "2000-07-20" ];

21:21:49 <DanC> k:eventOccursAt :place_MA_Cambridge;

21:21:49 <DanC> k:socialParticipants <http://www.w3.org/People/Connolly/home-smart#me>; k:startingDate [

21:21:49 <DanC> dt:date "2000-07-19" ];

21:21:50 <DanC> k:subEvents [

21:21:52 <DanC> a k:SocialOccurence;

21:21:54 <DanC> k:socialParticipants [

21:21:58 <DanC> foaf:homepage <http://www.ksl.stanford.edu/people/dlm/>;

21:22:00 <DanC> foaf:name "Deborah McGuinness" ],

21:22:02 <DanC> <http://www.w3.org/People/Connolly/home-smart#me> ];

21:22:04 <DanC> s:label "DARPA DAML jumpstart workshop";

21:22:06 <DanC> foaf:homepage <http://www.w3.org/2000/07/19-DAML> .

21:22:08 <DanC> }}}

21:22:46 <danbri_lap> re <DanC> ok, scraping my index of events for foaf:knowsWell is going well:

21:22:59 <danbri_lap> ...we stopped using foaf:knowsWell (and foaf:friend) a while back

21:23:18 <danbri_lap> just foaf:knows or more concrete claims now, since 'well' was impossible to define clearly

21:24:32 <DanC> ok, well, I didn't actually use foaf:knowsWell; I used subevents and socialparticipants.

21:24:55 <libby> neat :)

21:25:21 <DanC> lemme check in the .rdf version...

21:25:42 <danbri_lap> yep, that makes sense (re subevents/participants)

21:27:06 <DanC>http://www.w3.org/People/Connolly/events/events-smart.rdf

21:27:06 <dc_rdfig> M: http://www.w3.org/People/Connolly/events/events-smart.rdf from DanC

21:27:16 <DanC> M:|events-smart.rdf

21:27:16 <dc_rdfig> Titled item M.

21:27:40 <DanC> M:a formalized (ok, well, scraped) version of [my index of travel/talks|http://www.w3.org/People/Connolly/events/]

21:27:40 <dc_rdfig> Added comment M1.

21:28:24 <DanC> M:I have rules to turn the city names into lat/long; haven't done that yet

21:28:24 <dc_rdfig> Added comment M2.

21:28:55 <DanC> M:I'm considering using real RDF datatypes for the dates; haven't done that yet. using InterpretationProperties for now

21:28:55 <dc_rdfig> Added comment M3.

21:29:29 <DanC> M:worked around lack of syntactic inverse properties in RDF using rdf:value. hope to document that in the ESW wiki

21:29:29 <dc_rdfig> Added comment M4.

21:29:54 <DanC> M:at least a few of the events are working; not sure about all of them. it's a lot of data to debug.

21:29:55 <dc_rdfig> Added comment M5.

21:31:27 <DanC> M:still feeling schizophrenic about PropertiesForNaming

21:31:27 <dc_rdfig> Added comment M6.

21:36:59 <libby> talking of scraping...

21:37:04 <libby>http://swordfish.rdfweb.org/discovery/2003/04/www2003/

21:37:05 <dc_rdfig> N: http://swordfish.rdfweb.org/discovery/2003/04/www2003/ from libby

21:37:24 <libby> N:|scraped www2003 programme data

21:37:25 <dc_rdfig> Titled item N.

21:37:40 <libby> N:not all checked but things starting t_ seem to be ok

21:37:40 <dc_rdfig> Added comment N1.

21:40:24 <libby> wow that's a lot of meetings danc

21:42:08 <DanC> it's pretty much all the travel I've done in my WWW career, except for confidential meetings

21:42:27 <DanC> I wanna do an SVG animation of it

21:43:23 <libby> neat :)

21:43:49 <DanC> M:anybody wanna do an SVG animation of it for me?

21:43:49 <dc_rdfig> Added comment M7.

21:45:30 <libby> a maxf-style animated danc whizzing across the globe...

21:46:23 * sandro is amused by "except for confidential meetings".... "I could tell you I was in San Francisco last year, but then I'd have to kill you."

21:47:25 * DanC wonders if sandro has thought about the "A wrote section 3" problem with HashURIs

21:48:06 <sandro> is that not on the drawbacks list.....? /me looks

21:48:19 <DanC> yes, it's on the drawbacks list.

21:48:41 <libby> N:use with care. And there is a bunch of data missing at the moment, because my regexes aren;t too hot.

21:48:41 <dc_rdfig> Added comment N2.

21:49:07 <sandro> it's kind of in #3, but not so clearly

21:50:30 <sandro> I guess I don't actually see that as a new problem. You can still say "A wrote section 3" as long as you dont mean section-3 of an out-of-the-page-fragment-semantics document.

21:50:38 <sandro> s/new//

21:51:46 <pixel_2> pixel_2 is now known as pixel

21:52:40 <DanC> yes, I saw it as #3, but SteveP added another bullet.

21:53:29 <DanC> out-of-the-page-fragment-semantics?

22:33:27 <golbeck_> golbeck_ is now known as golbeck

23:38:01 <DanC>http://esw.w3.org/topic/CityLookup

23:38:02 <dc_rdfig> O: http://esw.w3.org/topic/CityLookup from DanC

23:38:17 <DanC> C:|CityLookup -- mapping city names to lat/long via weather underground

23:38:17 <dc_rdfig> Titled item C.

23:38:59 <DanC> C:please try it yourself and let me know how it goes

23:38:59 <dc_rdfig> Added comment C2.

23:39:34 <DanC> O:|CityLookup -- mapping city names to lat/long via weather underground

23:39:34 <dc_rdfig> Titled item O.

23:39:39 <DanC> C2:""

23:39:39 <dc_rdfig> Deleted comment C2.

23:39:44 <DanC> O:please try it yourself and let me know how it goes

23:39:44 <dc_rdfig> Added comment O1.

23:40:11 <DanC> O1:please try it yourself and let us know how it goes. (it's a Wiki page. you can scribble on it.)

23:40:11 <dc_rdfig> Replaced comment O1.

23:40:15 <danbri_lap> O:Is it just US cities?

23:40:15 <dc_rdfig> Added comment O2.

23:40:33 <DanC> O:no, weather underground covers the globe

23:40:33 <dc_rdfig> Added comment O3.

23:41:24 <DanC> C:Emerging Technologies, an O'Reilly conference Apr 22-25, Santa Clara, CA

23:41:24 <dc_rdfig> Added comment C2.

23:41:55 <danbri_lap> O:re input data, <usps:stateAbbr>TX</usps:stateAbbr> doesn't make sense for non-US. Do we just omit it?

23:41:55 <dc_rdfig> Added comment O4.

23:42:26 <DanC> no, you need the country name... hang on...

23:42:46 <DanC> e.g. <map:countryName>Canada</map:countryName>

23:42:58 * DanC would be much obliged if you'd add a non-us example

23:45:20 <danbri_lap> <map:countryName>UK</map:countryName>

23:45:20 <danbri_lap> ?

23:45:21 <DanC> O:bummer... Toronto,Canada doesn't work... weather underground indirects thru a set of choices

23:45:21 <dc_rdfig> Added comment O5.

23:45:32 <danbri_lap> or england, britain, united kingdom???

23:45:38 * danbri_lap lives somewhere with too many names

23:46:07 <DanC> whatever names the form at http://www.wunderground.com/ groks will work

23:47:11 <DanC> bummer... Bristol, United Kingdom suffers the same indirection problem

23:47:47 <danbri_lap> crap, python/cwm barfs, TypeError: 'intern' is an invalid keyword argument for this function

23:47:51 <danbri_lap> can't fix it tonight...

23:48:01 <DanC> be brave! cvs update! ;-)

23:48:37 <DanC> stick that error message in CwmTips or CityLookup, meanwhile

23:52:33 * danbri_lap adds http://esw.w3.org/topic/CwmTips "It ain't workin'" section