Semantic Web Interest Group IRC Chat Logs for 2004-11-09

00:00:29 <dmiles_afk> hi mdupont, Yeah I have been without internet conntion a little while now, I am at an wired cafe atm, i hope to be on more now.. i just bought this laptoip and geting iut ready to be able to styart coding agfain.. jeez i am kinda embarrased to not have written or contributed to anything in 2 years

00:00:59 <dmiles_afk> this is a p3-500-128mbram

00:01:48 <dmiles_afk> so i am going to have to set up a remote desktop *nix or win terminal server somewhere.. kinda homeless right n ow

00:02:08 <dmiles_afk> i gave up place to live to buy the laptop ;P

00:02:42 <dmiles_afk> i have a place to live in texas as soon as i drive out there (i am in seattle right now)

00:03:19 <dmiles_afk> i think i am going to stick arround seattle as long as i can bare living in car (1-2 mont5hs)

00:04:06 <dmiles_afk> have you been doing anything with introspector lately?

00:04:40 <dmiles_afk> dmiles_afk is now known as dmiles_nephrael

00:13:15 <dmiles_nephrael> dmiles_nephrael is now known as dmiles

00:20:06 <Pherl> dmiles: good to see you again

00:20:12 <Pherl> i am working for intel atm

00:20:38 <Pherl> working on itanium2 optimizations of the sap/r3 kernel on linux

00:20:58 <Pherl> lots of ideas for introspector and rdf

00:21:11 <Pherl> but right now alot more ideas on how to make fast programs

00:21:34 <Pherl> hopefully I will have some time soon to work on the introspector and implement my ideas

00:21:49 <bengee_> bengee_ is now known as bengee

00:21:51 <Pherl> i am thinking about how to store rdf triples directly in the object files

00:21:57 <Pherl> using the libbdf

00:22:21 <Pherl> it should be fast enough to translate the triples into data blocks

00:22:59 <dmiles> i bet you see performance boosteres to how unification of triples vould be procewssor based

00:23:32 <Pherl> This would be great for a profiler, for example, to store information about resources in a executable assocated to an execution context resources

00:23:43 <Pherl> thats right,

00:23:51 <Pherl> for example the entire idea of cache lines

00:24:12 <Pherl> the itanium has like 8 cache lines that you can use at a time

00:24:20 <libby> <libby>ISWC 2004 has started properly...just sitting in the introductory talks

00:24:23 <libby> <libby>48/205 research track acceptances/submissions

00:24:25 <libby> <libby>7/22 industrial track

00:24:29 <libby> <libby>48/68 posters

00:24:42 <Pherl> if you have for example an interation over a set of triples

00:24:54 <Pherl> you can fetch them sequentiall into a cache line

00:25:14 <Pherl> but if you want to join that with another set of triples, you can use another cache line

00:25:42 <Pherl> if you manage the cache propely, your queries can avoid direct memory accesses and only access cached data

00:26:07 <Pherl> the whole idea is to know what will be accessed next and prefetching that

00:27:56 <dmiles> even a CMP that compared say n n8umber of bytes and files a condition cade register with results 2 bits each

00:27:57 <dmiles> erm s/files/fills/g

00:27:57 <Pherl> ahh

00:27:57 <Pherl> yes, processing a bunch of bytes in a row

00:27:57 <Pherl> unrolling loops

00:27:57 <Pherl> exactly

00:27:57 <Pherl> that type of stuff is easy to do on the itanium

00:28:16 <dmiles> the itanium actually has a lower then machine level right?

00:28:47 <dmiles> erm lower the x86 std level

00:28:49 <Pherl> well, it gives you asm instr that allow you directly affect the parrallel processing features

00:28:57 <Pherl> that are automatic in the x86

00:29:08 <Pherl> like parallel execution

00:29:35 <Pherl> and speculative loading of data and instructions

00:29:48 <dmiles> *nod* now we just need a programable vlsi 0-7 machine instruct

00:30:18 <Pherl> basically they took all the hw logic and put that into the compiler

00:30:26 <Pherl> 0-7?

00:30:49 <dmiles> well since it would invade the opcode space for instructions

00:31:05 <Pherl> i dont understand

00:31:06 <dmiles> so like maybe 8 vlsoi virtual chips

00:31:25 <Pherl> very large scale OI?

00:31:41 <Pherl> you mean the new chips with many cores?

00:31:50 <libby> BLURB: first ISWC speaker: Edward Feigenbaum

00:31:52 <dc_rdfig> A: first ISWC speaker: Edward Feigenbaum from libby

00:32:11 <dmiles> well in the day of the IRQ triggering a JSR vector it would look up a physical targwet

00:32:35 <dmiles> so that the cpu would process IO

00:32:37 <dmiles> but...

00:32:55 <Pherl> interrupt vectors, sure

00:32:57 <dmiles> being able to define machine instructions is what i mean

00:33:01 <Pherl> ahh

00:33:19 <Pherl> well, it is interesting how machine instructions are defined.

00:33:27 <dmiles> i figure you could program some gate via a vlsi

00:33:43 <Pherl> vlsi, what do you mean

00:33:52 <Pherl> like some chip design language?

00:33:57 <dmiles> ok yeah.. if you can redefine them.. then you could make a nth lentgh compare

00:33:58 <libby> A:[http://wi-consortium.org/pdf/wi-hendler.pdf|Feigenbaum collaborated with Jim Hendler a while back on a paper, maybe this one]

00:34:00 <dc_rdfig> Added comment A1.

00:34:04 <Pherl> ahh

00:34:11 <Pherl> well, yes,

00:34:15 <dmiles> yes vlsi was like ia burnable cpu

00:34:45 <Pherl> but you can also execute like 16 compares in parallel in one clock cycle on the itanium

00:34:50 <libby> A:searching around he found slideshows but not killer apps

00:34:50 <dc_rdfig> Added comment A2.

00:34:53 <Pherl> and have 4 processors

00:35:20 <Pherl> or more, there are 64 way processors

00:35:37 <dmiles> ah , that beautifull

00:35:49 <Pherl> so you have the ability to program massive parallel programs without making your own chips

00:36:10 <Pherl> the asm language gives you that much control over the chip

00:36:17 <Pherl> and it does not change it at all

00:36:29 <Pherl> so it is up to you to program it well

00:36:46 <dmiles> you say in one clock cycle.. is it secretlyl more but in a lower level?

00:37:03 <Pherl> good question!

00:37:13 <dmiles> ok yerah .. so it does already do what i am thinking

00:37:29 <Pherl> from what I know, the instructions are in parallel

00:37:40 <Pherl> the machine has multiple pips

00:37:43 <Pherl> pipes

00:37:54 <dmiles> well i talked to a p3 desiner a few years ago.. and he said the language below asm executres much faster

00:37:59 <jeen> pft

00:38:11 <Pherl> yes on the opcode level

00:38:17 <Pherl> true

00:38:31 <Pherl> there is a lower level lang

00:39:19 <dmiles> right.. asm used to be equal to opcode like on the 6809

00:39:46 <dmiles> but now asm is more abstracted

00:39:51 <dmiles> (i am guessing()

00:39:52 <Pherl> yup

00:40:05 <Pherl> well, I guess we should go to #asm or something

00:40:08 <Pherl> :)

00:40:38 <Pherl> what I did want to say is that one should think about how to optimize a proof engine

00:40:47 <Pherl> using more advanced features

00:41:05 <Pherl> like cuts can be implemented using instruction speculation and predication

00:41:11 <Pherl> directly on the hardware

00:41:18 <dmiles> i hope in the next year to b e able to start thinking about that

00:41:41 <Pherl> and iterators can be implemented using cachelines

00:41:44 <dmiles> since a friend just tured in a proposal let us build oine

00:42:08 <Pherl> dmiles: i can tell you that intel is looking for good projects that use its itanium

00:42:18 <Pherl> if you want to push to get HW

00:42:33 <Pherl> i might be able to help, if you have a good proposal

00:42:49 <Pherl> Intel needs more applications that use the itanium

00:42:52 <libby> A:he interviewed sw gurus. is the sw 10 x better than google? is the benefit/cost ratio large? logic is a burden (complex, slow)

00:42:53 <dc_rdfig> Added comment A3.

00:43:05 <dmiles> neat.. it was a great prtoposal.. when he comes onto irc he'll point us to it

00:43:24 <crculver_Zzz> crculver_Zzz is now known as crculver

00:43:30 <Pherl> it think that the semantic web could benefit in the long run from being optimize to run quickly on a large machine

00:44:00 <Pherl> an itanium server with 16gbs of ram like I am working with is great for processing arrays of triples

00:44:11 <libby> A:revisiting 70-80s AI work - though AI and SW is not isomorphic, there's a big overlap

00:44:11 <dc_rdfig> Added comment A4.

00:44:23 <Pherl> with 4 processors and superfast disks

00:44:28 * libby a little disturbed at the direction of this conference

00:44:51 <dmiles> yes.. being pout of the loop i wonder if semantic web projects havew inspired more publiclly funded hardware projects

00:45:44 <Pherl> sure

00:46:06 <Pherl> Good question

00:47:02 <bengee> introductory talks, you said, libby? and they dare to mention AI?

00:47:15 <libby> this is the first keynote

00:47:32 <bengee> oh-oh..

00:47:39 <libby> A:|First keynote ISWC speaker: Edward Feigenbaum

00:47:39 <dc_rdfig> Titled item A.

00:48:02 <bengee> many people there?

00:48:06 <libby> yes...interesting choice (though he is interesting)

00:48:13 <libby> yep, I heard 800

00:48:18 <bengee> wow

00:48:19 <libby> room is pretty full

00:49:18 <libby> how's the site going benjee?

00:49:57 <bengee> put up a "coming soon" page today.

00:50:03 <libby> yay! :)

00:50:20 <bengee> had a hard time with photoshop, but am quite happy now.

00:51:12 <dmiles> so that machine iyou use.. thats nice, there are so manyt times when i think i haVE a better then normal representation and i have to convert large dataset .. then 10 minutes later completely change mind.. but the conversion process is awefulll

00:51:25 <libby> A:the SW as web scale expert system(s)

00:51:26 <dc_rdfig> Added comment A5.

00:52:35 <danbri_> morning libby

00:52:44 <danbri_> 800, cool

00:52:46 <dmiles> example might be that you decide the genlPreds/subProperty is squasable to the most general term

00:52:54 * danbri_ was worried would be underattended

00:52:54 <bengee> rdf tool development is ok, though. the gizmos start to sparkle (sparql?). but still too big a project for a single person, given the 2 months dev time..

00:53:06 <danbri_> any idea what % attendees are from outside Japan?

00:53:12 <libby> heya danbri_

00:53:26 <dmiles> hi danbri, ltns

00:53:29 <libby> more than 1/2 ? not sure

00:53:53 <libby> perhaps more than 70%

00:54:03 <libby> A:we live with systems which are not 100% correct: google!

00:54:04 <dc_rdfig> Added comment A6.

00:54:29 <danbri_> dmiles, welcome back! we missed you...

00:54:49 * DanC gathers libby is quoting; would prefer to see quote marks

00:55:13 <danbri_> yep, I didn't realise initially you were channeling for someone, lib

00:55:14 <libby> hm I'm paraphrasing

00:55:20 <dmiles> thank you, i have missed being here, i however like the timelessness

00:55:25 <libby> how do I indicate that?

00:55:43 <libby> A:"does logic give us the best set of tools for the practical sw? the evidence is mostly lacking"

00:55:44 <dc_rdfig> Added comment A7.

00:55:52 <DanC> journalistic standards allow for a certain amount of paraphrasing in quoted text, libby

00:55:57 <jeen> it's about 1/3 third japanese attendees according to frank

00:55:59 <danbri_> john doe: computers work better with flibflobs

00:56:14 <danbri_> ...however flibflobs require a tradeoff against flabflims

00:56:30 <libby> hm I would ahve just stuck it in the logs but dmiles was talking

00:56:35 <danbri_> ...further highly funded research is needed into flibflob/flamflim transflobulation

00:56:36 <danbri_> etc

00:57:00 <DanC> yes, play notation works pretty well... we use it in W3C meeting records, but...

00:57:08 <DanC> ... I wonder how well play notation mixes with chump syntax

00:57:12 <libby> A:|First keynote ISWC speaker: Edward Feigenbaum (the text below is paraphrased from his talk and slides. apologies for innacurracies etc

00:57:12 <dc_rdfig> Titled item A.

00:57:16 <libby> A:|First keynote ISWC speaker: Edward Feigenbaum (the text below is paraphrased from his talk and slides. apologies for innacurracies etc)

00:57:17 <dc_rdfig> Titled item A.

00:57:22 <danbri_> (I don't think there are universally established notations in IRC-land. It yep as danc says, depends on tool support)

00:57:46 <danbri_> chump, bookmark A:

00:57:53 * danbri_ forgets the notation

00:58:05 <DanC> can I scrub out that british apology?

00:58:05 <crschmidt> logger: chump a

00:58:05 <crschmidt> A:See [http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/2004-11-09#T00-58-05-1|discussion]

00:58:05 <dc_rdfig> Added comment A8.

00:58:10 <crschmidt> is that what you meant?

00:58:20 <crschmidt> I'm sorry, I meant to prefix that so it wouldn't actually do that :/

00:58:31 <danbri_> that's fine I think

00:58:47 <danbri_> British apology fine as-is I think

00:59:02 <DanC> yeah, but you're a brit. ;-)

00:59:04 <bengee> libby++ for community coverage

00:59:08 <danbri_> yep

00:59:12 <bijanp> "I'm sorry for my slumping upper lip"

00:59:15 <bijanp> Sorry,

00:59:17 <danbri_> think of the airfares you've saved

00:59:25 <bijanp> "I'm *terribly* sorry for my slumping upper lip"

00:59:39 <libby> well you only have my word that it's an accurate representation

00:59:43 <DanC> libby++ for... exactly; no need to apologize for doing stuff above and beyond the call of duty.

00:59:51 <deltab> I'm sorry you elected Bush again.

01:00:01 <bijanp> Not sorrier than me

01:00:06 <libby> (esp as I just missed a bucnh)

01:00:19 <bijanp> But this is perhaps the wrong forum

01:00:39 <DanC> I'd just like to keep titles as titles. feel free to apologize in a comment.

01:00:57 <danbri_> election ranting has been declared inscope for #foaf

01:01:11 <danbri_> inscope here if you've got rdf tools relating to it...

01:01:13 <dmiles> not many forums here on this irc server for politicvs ;P..

01:01:15 <dmiles> yay

01:01:33 <bijanp> Kendall scraped electoral-vote.com into rdf

01:01:36 <bijanp> And repubbed it

01:01:43 <bijanp> and thinks bush is the devil

01:01:47 <bijanp> You mean like that?

01:02:44 <DanC> re policitcs and IRC, the supreme irony is that I went to #politics while I was preparing to vote and got into a debate about XML standards!

01:03:59 <JibberJim> XML standardisation seems to have been pretty politicised from my seeing...

01:04:48 <dmiles> DanC, to vote in america?

01:05:14 <danbri_> I find people overuse the word 'politics' in W3C WG context, when they mean 'subtle, intergroup considerations'

01:05:45 <dmiles> (oh n/m,i thought you were in dublin)

01:07:50 <libby> A:"build a little, test a little" "find a task domain that has a real world importance, is rich with posibilities, then do a series of experiments, ascending the sprial"

01:07:51 <dc_rdfig> Added comment A9.

01:08:09 <jhendler> A: the paper libby mentions is at [http://springerlink.metapress.com/app/home/contribution.asp?wasp=3d325dwwrqdkpjdv9xaw&referrer=parent&backto=issue,2,80;journal,1024,1770;linkingpublicationresults,1:105633,1| this springer link]

01:08:11 <dc_rdfig> Added comment A10.

01:08:13 <libby> A:this community has a tendency to work top-down rather than bottom up"

01:08:14 <dc_rdfig> Added comment A11.

01:08:19 <libby> chers jim!

01:08:22 <jhendler> A: which takes you to a download page.

01:08:23 <dc_rdfig> Added comment A12.

01:08:40 * jhendler supposed to be free, hope they are still giving it away -- hi Libby

01:10:23 <danbri_> yep, full text pdf online

01:11:05 * jhendler not my best SW paper, but at least a couple of provocative ideas (and a chance to coauthor with Ed)

01:11:12 <DanC> yes, in america. I had my voter registration card, and a page from jocoelection.org showing all the candidates from state representative to president, and I wanted to use #policics to see what web sites people were using to evalutate the issues.

01:11:53 <danbri_> what's your state?

01:11:57 <DanC> KS

01:12:08 <danbri_> out of the Canada 2.0 zone?

01:12:11 <jhendler> danbri - my state since the election is depressed

01:12:45 <DanC> yes, KS is part of the United Sates of Texas; voted for Bush 60:40.

01:13:29 * DanC kept http://dm93.org/z2001/CivicDuty somewhat up to date as he researched things; hasn't really orgainzed it well though

01:14:49 * danbri_ rereads, likes the lead-in parag

01:16:29 <libby> A:applications to help you answer archival questions: wants stuff like how did X meet Y to collaborate on the Z project (in science). other interesting questions you might want to ask

01:16:29 <dc_rdfig> Added comment A13.

01:17:16 * jhendler wonders if it is coincidence that Tim lives in a blue state, Dan in a red state, and Eric M. is in the crucial swing state between them :->

01:17:27 * DanC struggles to see what "paper libby mentions" refers to

01:17:52 <DanC> it's clearly all EricM's fault. ;-)

01:18:19 * jhendler DanC I meant her reference in A1, I don't know how to do links to other chump comments in chump

01:18:31 * DanC is just being lazy

01:18:40 <libby> A13: he talks about example applications - to help you answer archival questions: Feigenbaum suggets stuff like - how did X meet Y to collaborate on the Z project (in science), and other interesting questions

01:18:41 <dc_rdfig> Replaced comment A13.

01:18:51 * libby lost the ability to paraphrase

01:18:54 <Talliesin> What did the presidential candidates think about httpRange-14, that's what I want to know.

01:19:15 * jhendler yeah, but if chump would give you a URI return, instead of "A1" or such, then you could link to it in other chump comments

01:19:55 <jhendler> tallliesin: Kerry had someone on his staff who could of understood the question - I think that's as close as any of them got :->

01:20:13 <DanC>http://iswc2004.semanticweb.org/

01:20:13 <dc_rdfig> B: http://iswc2004.semanticweb.org/ from DanC

01:20:23 <DanC> B:|3rd International Semantic Web Conference (ISWC2004)

01:20:23 <dc_rdfig> Titled item B.

01:20:25 <libby> A:"cybergraves" for scientists

01:20:26 <dc_rdfig> Added comment A14.

01:20:36 <DanC> B:7-11 November 2004 Hiroshima, Japan

01:20:36 <dc_rdfig> Added comment B1.

01:21:11 <DanC> B:thanks to libby et. al for [http://esw.w3.org/topic/ConnectingAudiences|ConnectingAudiences]

01:21:11 <dc_rdfig> Added comment B2.

01:21:28 <danbri_> +1

01:22:18 <DanC> A:among the [http://iswc2004.semanticweb.org/program/index.html#invited|ISWC invited speakers]

01:22:18 <dc_rdfig> Added comment A15.

01:22:43 <DanC> B:animated GIFs. to laugh or to cry?

01:22:43 <dc_rdfig> Added comment B3.

01:23:20 <libby> A14:"cybergraves" for scientists..and also the non-famous

01:23:20 <dc_rdfig> Replaced comment A14.

01:24:36 <sbp> hmm, ISWC 2005's in Galway. minor chance of getting there

01:24:44 <libby> :)

01:25:11 <jhendler> B: or see the [http://www.siderean.com/iswc2004/iswc.jsp| Semantic Metadata] for the conference (Seamark(TM) viewer?)

01:25:12 <dc_rdfig> Added comment B4.

01:26:19 <libby> A:"when a distinguished and elderly scientist he is almost certainly correct; when he says something is impossible, he is almost certainly wrong" - arthis C clarke

01:26:20 <dc_rdfig> Added comment A16.

01:26:26 <libby> gah

01:26:44 <libby> A16:"when a distinguished and elderly scientist he is almost certainly correct; when he says something is impossible, he is almost certainly wrong" - Arthor C clarke

01:26:45 <dc_rdfig> Replaced comment A16.

01:27:26 <libby> how do you spell arthur?

01:27:41 <libby> A:"stop so much talking and start more doing"

01:27:42 <dc_rdfig> Added comment A17.

01:27:50 <libby> hehe. I like this guy

01:28:33 <Talliesin> Arthur is spelt the second way you tried.

01:28:41 <sbp> third

01:29:29 <libby> sigh, thanks

01:29:39 <libby> A16:"when a distinguished and elderly scientist he is almost certainly correct; when he says something is impossible, he is almost certainly wrong" - Arthur C clarke

01:29:40 <dc_rdfig> Replaced comment A16.

01:29:58 <libby> I've crapped up the quote too. sigh

01:30:13 <libby> A16:"when a distinguished and elderly scientist says somethign is possible he is almost certainly correct; when he says something is impossible, he is almost certainly wrong" - Arthur C clarke

01:30:14 <dc_rdfig> Replaced comment A16.

01:31:18 <jhendler> oh, oh = here's another CLarke quote relevant to our previous discussion:

01:31:45 <jhendler> I have a fantasy where Ted Turner is elected president but refuses because he doesn't want to give up power.

01:31:46 <jhendler> Arthur C. Clarke

01:31:52 <DanC> B:some of the workshops and tutorials in the [http://iswc2004.semanticweb.org/program/index.html|program] look pretty interesting. Pointers to workshop notes appreciated.

01:31:52 <dc_rdfig> Added comment B5.

01:31:53 <libby> heh

01:32:09 <danbri_> any outcomes from Trust WS?

01:32:12 <DanC> B:any chance ISWC offers or endorses a wiki or blogroll?

01:32:12 <dc_rdfig> Added comment B6.

01:33:42 <jhendler> Oh, and here's another one by Clarke: New ideas pass through three periods: 1) It can't be done. 2) It probably can be done, but it's not worth doing. 3) I knew it was a good idea all along!

01:34:11 * jhendler notes Ed F. seems like he has reached number 2 -- too bad I'm not in the audience to quote this to him ...

01:34:39 <danbri_> no 2?

01:34:39 <libby> I think he's being more positive than that actually, perhaps I didn't make that clear...

01:34:46 <danbri_> ah right, sorry

01:35:10 <jhendler> no, that's okay libby, I'm relating to my discussions with him, not to his talk (which I can't hear)

01:35:16 <libby> it's more like: pulll your socks up and get doing it

01:35:18 <libby> right

01:36:19 <DanC>http://www.ontopia.net/topicmaps/materials/identitycrisis.html

01:36:20 <dc_rdfig> C: http://www.ontopia.net/topicmaps/materials/identitycrisis.html from DanC

01:36:27 <DanC> C:|Curing the Web's Identity Crisis

01:36:27 <dc_rdfig> Titled item C.

01:36:35 <DanC> C:|Curing the Web's Identity Crisis: Subject Indicators for RDF

01:36:36 <dc_rdfig> Titled item C.

01:37:12 <DanC> C:found via [http://www.ontopia.net/ontopia/events/iswc-2004-tutorial.html|ISWC tutorial materials] by Pepper and Garshol

01:37:13 <dc_rdfig> Added comment C1.

01:37:27 * jhendler thinks they'd be better off adding URIs to topic maps...

01:38:07 * danbri_ wonders if jhendler has any time to spend on the TM/RDF TF of SWBPD WG

01:38:23 <DanC> C:relative to earlier drafts by the same title, it seems to have many more contemporary references and some problems fixed, but the argument still seems flawed

01:38:23 <dc_rdfig> Added comment C2.

01:38:34 <libby> some of my photos are here: http://swordfish.rdfweb.org/photos/2004/11/08/ - tourist stuff mostly

01:38:51 * jhendler wonders if jhendler even has the time to spend on this irc tonight...

01:39:01 <danbri_> C:There is no crisis. Move along. Nothing to see here.

01:39:01 <dc_rdfig> Added comment C3.

01:39:07 <DanC> C:this version is fine until the section "Using the same URI to identify a different subject". To which the answer is, simply: don't do that.

01:39:08 <dc_rdfig> Added comment C4.

01:40:09 <DanC> well, this paper reflects non-trivial thought, and they have running code and they're giving a tutorial at ISWC, which is more than I can say. I think "nothing to see here" is, well, not respectful.

01:40:12 <libby> breaktime here, then a choice of semweb services or rdf query

01:40:23 <DanC> semweb services, please.

01:40:35 <DanC> I doubt you have much to learn about rdf query

01:41:11 <danbri_> C:Talk of crisis is pure melodrama. RDF allows you to model things crudely, or with subtlety. The problem is not with RDF or the Web, but with particular schemas and datasets. So I resent the perpetuation of the FUD that we're in "crisis" because there are scruffy schemas and murky datasets in the Web.

01:41:11 <dc_rdfig> Added comment C5.

01:41:17 <jhendler> DanC - seems to me that WebArch section 2 is violated in this a fair amount...

01:41:32 <DanC> hmm... on the other hand... maybe RDF query is maturing, soaking up DB optimization research and such.

01:42:06 <danbri_> I'd love to see an RDF query system that used OWL info to optimise queries...

01:42:24 <DanC> logger, chump C

01:42:24 <DanC> C:See [http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/2004-11-09#T01-42-24|discussion]

01:42:25 <dc_rdfig> Added comment C6.

01:43:49 <DanC> resent FUD... yes, I'm not pleased that they chose to include this paper among the pointers from their ISWC tutorial

01:45:00 * bengee would like to see/hear patrick stickler attending that tutorial ;)

01:45:14 <DanC> C:"However, the TAG has no solution to offer, other than rather pathetic hand-waving". Hmm. speaking of dis-respect! I don't recall comments on webarch directly from Mr. Pepper.

01:45:14 <dc_rdfig> Added comment C7.

01:45:49 * jhendler thinks they sound desperate - wonder how much longer ontopia will be around ...

01:47:49 <Talliesin> C:Crisis, What Crisis?

01:47:49 <dc_rdfig> Added comment C8.

01:47:57 * DanC thinks he'd better leave his computer alone for a while, lest he do something regrettable

01:48:26 <jhendler> Talliesin, probably the crisis with their sales...

01:48:34 <bengee> heh

01:48:53 <Talliesin> C:We've been here before, and the best we can get from it is a third voice in the httpRange-14 debate that satisfies neither of the existing camps.

01:48:53 <dc_rdfig> Added comment C9.

01:50:11 <LatinPrince> hi

01:52:55 * jhendler who was that masked man?

01:55:51 <LatinPrince> 12«11jhendler12» excuse me I just need a li informatio nabout .rdf files

01:56:25 * jhendler LatinPrince - I was referring to TIMBL who came and went in an instance...

01:56:36 <LatinPrince> 12«11jhendler12» ahh ok

01:57:51 <LatinPrince> 12«11jhendler12» Do you possilbly know where I can get a decent calendar plugin in rdf/rss format for php-nuke web portal?

01:58:30 <LatinPrince> 12«11jhendler12» I was directed there threw a google search I did just a moment ago

01:58:51 <jhendler> I suspect everyone else on this irc is more likely to be able to answer that then I am...

01:59:15 <jhendler> when libby gets back from break, she may know - or DanC

01:59:26 <danbri_> i expect we'd mostly google for it

01:59:45 <jhendler> Google? Is that a web thing?

02:00:10 <LatinPrince> 12«11jhendler12» Is DanC on now? I tried to google for it but this is basically the first portal im tring to build so my expierence is little to none at all

02:00:21 <LatinPrince> 12«11jhendler12» lol

02:00:28 <libby> I don;lt know of anything like that lationprince sorry

02:00:38 <libby>http://www.aifb.uni-karlsruhe.de/WBS/pha/rdf-query/index.html

02:00:39 <dc_rdfig> D: http://www.aifb.uni-karlsruhe.de/WBS/pha/rdf-query/index.html from libby

02:00:52 <LatinPrince> 12«11jhendler12» heh thx for responding at least

02:01:27 <libby> D:|A survey of RDF query languages

02:01:28 <dc_rdfig> Titled item D.

02:01:38 <jhendler> LP - I'm not an implementor/hacker as much as most of thesefolks - good luck in finding what you need.

02:02:02 <LatinPrince> heh thx for responding at least I appreciate it

02:02:05 <libby> D:ISWC 2004 talk in rdf query section

02:02:06 <dc_rdfig> Added comment D1.

02:02:27 <libby> D:papers don't seem to be online (yet?)

02:02:28 <dc_rdfig> Added comment D2.

02:03:58 <LatinPrince> 12«11libby12» Hi, I'm sorry to bother you I'm kinda new at this could I possibly ask you a question? I was told you might know the answer or at least direct me in the right direction.

02:04:30 <libby> sorry, I don;pt know the answer - I saw your question

02:05:07 <libby> the rdf calendar stuff is here: http://www.w3.org/2002/12/cal but I don't think there is php software supporting it

02:05:10 <LatinPrince> 12«11libby12» Ahh ok thx for responding

02:05:37 <libby> (I'm not sure if that's what you were asking though)

02:06:10 <LatinPrince> 12«11libby12» Humm... I guess I'll try a different portal was kind of tring to go for that one because its high on features and low on banners and ads

02:06:24 <LatinPrince> 12«11libby12» Yea thats pretty much it

02:06:43 <LatinPrince> 12«11libby12» thx though

02:07:12 <libby> you might be best looking at a php channel maybe? the rdf stuff people have mostly implemented in xslt I think - there may well not be any portal software that supports it. probbaly none actually

02:08:07 <libby> ah: http://futureshare.lip6.fr/iCal.html ?

02:08:18 <LatinPrince> 12«11libby12» Do you know any PHP channel with ppl that might be able to help? Again I'm totally new at this kinda disoriented

02:08:34 <libby> rdf icalendar is besically a stright transformation of icalendar to rdf

02:08:44 <libby> nope, I don't use php

02:08:54 <danbri_> you might find http://www.wiwiss.fu-berlin.de/suhl/bizer/rdfapi/ interesting, latinprince

02:08:54 <libby> you can /list I think, find channels

02:09:17 <LatinPrince> 12«11libby12» ahh ok cool ill play with that see if I can implement it in some way thx again

02:14:56 * jhendler waves at bijanp

02:15:16 * jhendler wonders if bijanp evaluates to T or to NIL

02:16:09 * libby wonders how the iswc sws session is going

02:20:08 <libby> D:fortunately Peter Haase, the speaker/coauthor has put it all onlione anyway. other authors: Jeen Broekstra, Andreas Eberhart, Raphael Volz

02:20:08 <dc_rdfig> Added comment D3.

02:22:07 * jeen wonders how obvious it is to ppl

02:22:23 <jeen> ...that extending rdf qls to owl support is not a promising way

02:22:48 * jhendler wonders what jeen means - seems like a promising way to me

02:23:11 <jhendler> Owl graphs are just rdf graphs, and can be exploited in that way quite well

02:23:43 <libby> B:[http://www.siderean.com/iswc2004/iswc.jsp|metadata for the papers] - uses skos! (jeen says there's some character encoding problems though)

02:23:44 <dc_rdfig> Added comment B7.

02:23:59 <libby> B7:[http://www.siderean.com/iswc2004/iswc.jsp|metadata for the papers] - uses skos! (jeen says there's some (minor) character encoding problems though)

02:23:59 <dc_rdfig> Replaced comment B7.

02:24:08 <jeen> I'd think that most people who model in owl would prefer a ql that looks at it in a more abstract level. querying a nested owl property restriction in triple format is a b*tch

02:25:16 <jhendler> depends what you are trying to do -- if working with large ontologies like NCI, it's a lot easier to query thegraph directly for properties than it is to load the whole ontology and use a reasoner of any kind

02:25:26 <DanC> php-nuke and RDF calendar sounds cool, LatinPr... er... where'd he go? oh well.

02:25:43 <libby> D:ericp asks a q about expressivity and logical entailment - separate or together?

02:25:44 <dc_rdfig> Added comment D4.

02:27:12 <libby> D:ericp asks a q about expressivity and logical entailment - separate or together? [dawg separates them, speaker says dependent; missed answer though...]

02:27:13 <dc_rdfig> Added comment D5.

02:27:23 <libby> D4:ericp asks a q about expressivity and logical entailment - separate or together? [dawg separates them, speaker says dependent; missed answer though...]

02:27:23 <dc_rdfig> Replaced comment D4.

02:27:43 <jeen> jhendler, sounds fine to me, but in those case probably a non-extend rdf ql will even suffice. I think Peter's (and my) point is that an OWL QL specifically tailored for owl would need to be 'beyond' the triple level. But I must admit that in my case it's more a gut feeling.

02:28:16 <jeen> D: answer is that in Peter's opinion, formal semantics of the QL should be part of the specification of that QL and not be left to the implementation

02:28:16 <dc_rdfig> Added comment D6.

02:28:17 <libby> D%:question about why different serializations of RDF have not been considered in the survey: speaker says QL acts on the data model so not relevant

02:28:33 <libby> sigh

02:28:41 <libby> D5:question about why different serializations of RDF have not been considered in the survey: speaker says QL acts on the data model so not relevant

02:28:42 <dc_rdfig> Replaced comment D5.

02:28:43 <jhendler> I guess my feeling is that it would be a mistake to put that in the query language per se -- that is if I say "return all the animals" and the system

02:29:08 <jhendler> gives me some, why do I care whether it is via query or not (or maybe I can turn on a "plus reasoning" property

02:29:32 <jhendler> I've spent a lot of years in the AI community, and to be honest I think we've forgotten that a lot of what we are doing is just querying ...

02:29:59 <jhendler> if you want to know what the restrictions are, either check the document where the links go, or query the graph - whyt create a specifal langauge for it?

02:30:51 <libby> D:q from frank v-h - who are you so tolerant of qls that don't support rdfs? speaker: you can still be an rdf ql

02:30:52 <dc_rdfig> Added comment D7.

02:31:41 <libby>http://www.csd.uoc.gr/~kotzino/iswc2004/grql-abstract.html

02:31:42 <dc_rdfig> E: http://www.csd.uoc.gr/~kotzino/iswc2004/grql-abstract.html from libby

02:32:08 <bengee> B: I wonder if the iswc2004 rdf schema is ever going to be re-used.. (ns uriref=http://iswc2004.semanticweb.org/ns#, no mappings to deployed terms, ..)

02:32:09 <dc_rdfig> Added comment B8.

02:32:25 <libby> E:|Generating On the Fly Queries for the Semantic Web: The ICS-FORTH Graphical RQL Interface (GRQL) - abstract for rdfql section talk at ISWC

02:32:26 <dc_rdfig> Titled item E.

02:32:56 <libby> E:by Nikos Athanasis, Vassilis Christophides, and Dimitris Kotzinos (who couldn't make it :(

02:32:56 <dc_rdfig> Added comment E1.

02:33:24 <libby> E:[thbe full paper|http://139.91.183.30:9090/RDF/publications/grql-final.pdf] (pdf)

02:33:24 <dc_rdfig> Added comment E2.

02:33:33 <libby> E2:[the full paper|http://139.91.183.30:9090/RDF/publications/grql-final.pdf] (pdf)

02:33:34 <dc_rdfig> Replaced comment E2.

02:34:06 <libby> E:isn't it great that lots of the papers are online despite not being on the site? :)

02:34:08 <dc_rdfig> Added comment E3.

02:34:28 <crschmidt> nnnnnnnnnnnnnnnnnn222222222222222222222222222h`

02:34:33 <jeen> jhendler, because, I guess, a lot of ppl who use OWL are not prepared to think in terms of a graph, but like to think in 'bigger blocks', like class definitions, etc.

02:35:20 <crschmidt> AGH, KITTY, SORRY

02:35:34 <jhendler> yes, that is their mistake as I keep trying to explain to them

02:35:38 * libby thinks owl is rdfql is a bit like rdf in xquery (though not an exact analogy because of the issues with rdf/xml and canonicalization)

02:35:57 <jeen> libby, I agree

02:35:59 <crschmidt> ^q ?paper where (?author rdfs:label "Kotzinos, Dimitris") (?paper dc:creator ?author)

02:35:59 <julie>http://iswc2004.semanticweb.org/papers/pdf/32980486.pdf

02:36:00 <dc_rdfig> F: http://iswc2004.semanticweb.org/papers/pdf/32980486.pdf from julie

02:36:30 <jeen> it's possible but not exactly easy.

02:36:30 <crschmidt> F:| Paper by Dimitris Kotzinos

02:36:31 <dc_rdfig> Titled item F.

02:36:31 <jhendler> libby/jeen - I don't see the analogy at all.

02:36:38 <crschmidt> logger: chump F

02:36:38 <crschmidt> F:See [http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/2004-11-09#T02-36-38|discussion]

02:36:38 <dc_rdfig> Added comment F1.

02:36:45 <libby> F:that url doesn't exist though....I guess later

02:36:51 <dc_rdfig> Added comment F2.

02:36:54 <crschmidt> F: That's the paper described in the papers.rdf

02:36:54 <dc_rdfig> Added comment F3.

02:37:13 <libby> do you want julie to chump everything like that chris (in this case makes sense :)

02:37:19 <crschmidt> (that wasn't my intention, heh)

02:37:23 <jhendler> can someone give me an example of an OWL query that would be hard to express in sparql or similar?

02:37:47 <crschmidt> I was just testing to see if hte papers.rdf had a URL that might not be on the site, as usual I didn't think about the chump

02:38:30 <libby> E:looks like ithe paper will eventually be [http://iswc2004.semanticweb.org/papers/pdf/32980486.pdf|here] (thanks crschmidt/julie!)

02:38:30 <dc_rdfig> Added comment E4.

02:41:13 <libby> jhendler - when querying rdf using say xslt, one whinge is that the syntax-based query doesn't allow you to make nice intuitive rdf queries, just syntactic ones. in some cases at least rdf looks like a carrier syntax for owl to me. though that's not true always by any means, e.g. using inversefunctionalproperty

02:41:49 <jhendler> OK - but i ask my question again - give me an example of what you want to ask OWL that is hard to ask in Sparql or the like

02:42:07 <libby> you could make simplar points about rdfs I guess, it's just less complex

02:42:28 <libby> s/simplar/similar/

02:42:42 <libby> not that it's hard but that it's not intuitive

02:42:48 <jhendler> i.e. if you want to know whether some property is inversefunctional, that is an easy query. If you want to know whether two people are the same (which gets caused by an inversefunctional) then you get that for free if you say "+reasoning" or whatever

02:43:07 <jhendler> libby - i'll settle for an RDF example

02:43:10 <jhendler> (RDFS)

02:43:34 * libby potters around looking

02:44:39 <jeen> right, something like 'give me all classes that are subsumed by the class of things that are made from at most one type of grape'. As an example.

02:44:47 <jeen> on the wine ontology for instance.

02:45:13 * libby was just looking at the wine ontology

02:45:17 <jeen> I'm sure it's possible to express this in triple format, but it seems a bit unnatural to do so to me.

02:45:44 <jhendler> but why in the world are you asking it that way? Just ask for the those things that have single type of grape - that doesn't seem odd and is something that Sparql has to allow

02:46:26 <jhendler> seems to me the problem is people are thinking you query ontology qua ontology, and it seems quite rare to me that I would want to do that

02:47:35 <jhendler> We've had lots of rule-based systems for a long time, and the queries asked of them are things like "what disease does this guy have" (based on the rules) not "what class of disease subsumes the symptoms expressed pairwise in a DL scheme"

02:48:15 <jeen> interesting. perhaps you have a point, as I said, it is a gut feeling on my part. I'll have to think about this (it would certainly make my life as a rdf/owl system developer a lot easier ;) )

02:49:18 <jhendler> let me say it was my gut feeling too for the first fifteen years of my AI career. It was when I finally realized that I might be wrong that Parka and Shoe (and later DAML and OWL) occured to me...

02:49:51 <jhendler> that said - there is stuff that an RDFS or OWL querier may want to do in special ways or with a special syntax - but what i think is we can get a lot further with

02:50:07 <jhendler> a triples syntax than many people have explored to date

02:50:13 <jeen> only stumbling block i see is the absense of formal semantics for sparql. though I would be happy with a 'reasoning switch', if part of the QL spec itself. I think it is important that given a query, the answer is unambiguous

02:50:53 <jhendler> for example, the PIQ interface (the triple-based querier for Parka) turned out to be way more powerful than we thought

02:51:43 <libby> E:[http://athena.ics.forth.gr:8999/RQLdemo/|demo!]

02:51:43 <dc_rdfig> Added comment E5.

02:52:32 <jhendler> unambiguous, probably - but not necessarily common - that is, if I ask DB 1 for its animals and it includes "bear" and I ask DB2 and it doesn't, then how am I to know (and why should I be allowed to know) whether it is because DB2 doesn't have any, or doesn't kow that bears are animals (because it is an RDF store and not an inferencing store)

02:53:01 <jhendler> it would be useful to know, but that is like saying the web would be nice if those stupid 404s would just go away...

02:53:09 <libby> E:nice interface. uses the schema to generate the possible form-like queries by the looks of it

02:53:10 <dc_rdfig> Added comment E6.

02:53:36 <jhendler> we might need a lot of uniformity that can't be enforced (and might not be desirted) on a distributed, dynamic KB like the Web

02:54:16 * jhendler realizes he should get off his soapbox and go back to work -- I'd so much rather be having these discussions in Japan than doing what I'm doing, that I'm running on way too long...

02:55:12 <libby>http://www.science.uva.nl/~mdr/Publications/Files/iswc2004.pdf

02:55:12 <dc_rdfig> G: http://www.science.uva.nl/~mdr/Publications/Files/iswc2004.pdf from libby

02:55:33 <libby> G:|Information Retrieval Support for Ontology Construction and Use (pdf)

02:55:34 <dc_rdfig> Titled item G.

02:55:49 <libby> G:by Willem Robert van Hage, Maarten de Rijke, and Maarten Marx

02:55:50 <dc_rdfig> Added comment G1.

02:56:08 <libby> G:another paper from ISWC, in the rdf query section

02:56:09 <dc_rdfig> Added comment G2.

02:57:14 <libby> G:an approach to searching that uses both statics and ontologies

02:57:15 <dc_rdfig> Added comment G3.

02:57:43 <libby> oh a picture of a monkey :)

02:58:20 <jeen> that always works as a crowd puller eh?

02:58:27 <libby> I reckon

02:58:41 <libby> every talk should have monkey pictures in it

02:58:46 <jhendler>http://money.cnn.com/2003/09/04/pf/saving/pepsi_monkey_game/monkey.03.jpg

02:58:46 <dc_rdfig> H: http://money.cnn.com/2003/09/04/pf/saving/pepsi_monkey_game/monkey.03.jpg from jhendler

02:58:48 <libby> or soem sort of animal, anyway

02:58:58 <jhendler> H:| a picture of a monkey

02:59:00 <dc_rdfig> Titled item H.

02:59:51 <libby> H:+[http://money.cnn.com/2003/09/04/pf/saving/pepsi_monkey_game/monkey.03.jpg|a picture of a monkey]

02:59:51 * jhendler has to learn how libby puts those image links in the irc

02:59:52 <dc_rdfig> Added comment H1.

02:59:56 <libby> heh, done

02:59:57 <libby> :)

03:00:14 <jhendler> thanks!

03:01:10 * libby wonders if I should be adding links to the future location of the papers on the site :/

03:01:37 <libby> I wonder if it's deliberate policy or just takes a qhile to get them all put on the web

03:02:05 <jeen> i think it has to do with the publisher actually

03:02:55 <libby> yeah I was wondering that. mind you it's good they get put on the web at all

03:03:02 <jhendler> H:+[http://hometown.aol.com/coondane/images/monkey%20face.jpg|a monkey picture?] -- found via image search on Google(tm) for monkey

03:03:03 <dc_rdfig> Added comment H2.

03:03:35 <libby> G3:an approach to searching that uses both statistcs and ontologies

03:03:36 <dc_rdfig> Replaced comment G3.

03:03:39 * jhendler I get it - but I guess I should learn how to scale these before I mess up the chump way too much...

03:04:06 <libby> hm, wonder if can scale them

03:04:09 <libby> cute picture

03:05:02 <jhendler> finding things Google image search does wrong has become something of a hobby - necessary if we're going to get people to pay attention to the Sem Web image work IMO

03:05:09 <JibberJim> that's the cat monkey yeah?

03:05:40 <jhendler> hmm, there's no monkey photos on the mindswap site -- anyone have a good monkey ontology?

03:05:42 <libby> hm I don;t see anything about scaling images in the chump

03:06:31 <jeen> jhendler, if that is so, a big potential audience is the fark.com photoshoppers community, they regularly use gis (google image search) for 'stock photos' to use in these contests. and they are always moaning about how poor it performs :)

03:07:09 <jhendler> H: swoogle to the rescue - [http://protege.stanford.edu/plugins/owl/owl-library/koala.owl#Monkey| a monkey ontology] (actually one of 13 classes found for "monkey" - yay swoogle!)

03:07:10 <dc_rdfig> Added comment H3.

03:09:07 <jeen> btw is it possible to get back a label for an older entry in the chump, so that you can add a comment to it?

03:09:10 * CloCkWeRX changes some of his contacts from foaf:Person to ex:Monkey :)

03:09:22 <libby> nope, I don't think so jeen

03:09:23 <bengee> H:could be handy for the rdf world if it contained codemonkey (intersectionOf coder, monkey?)

03:09:23 <dc_rdfig> Added comment H4.

03:09:31 <libby> heh

03:09:53 <jeen> ok, good to know

03:10:27 <jhendler> bengee - but that is easy -- swoogle finds "hacker" to be http://xmlns.com/wordnet/1.6/Hacker-2

03:11:10 <jhendler> so just create an ontology linking these -- see, that's the semantic web at its best (well, would be if the monkey class was actually in that ontology from protege - but it isn't)

03:11:17 <bengee> but I don't wanna be rdf:type a wordnet synset entry... ;)

03:11:35 <jhendler> a semantic 404 error

03:12:12 <jhendler> ahh, bengee, then it is easy, use http://www.mindswap.org/2003/owl/mindswap#Programmer instead

03:12:38 <bengee> hehe, no way to argue with jhendler ;)

03:13:32 <jhendler> bengee proves to be amazingly perceptive :->

03:14:30 <jhendler> actually, /me is amazed at how often swoogle finds something useful - and often multiple things - I've certainly been able to stump it, but some of the thigns I've tried to stump it with were there

03:15:05 <jhendler> I tried "bicycle" and it found 29 classes - many of which were relecant

03:15:08 * bengee should better finish his resurce browser now. don't want to end my intern with a "dear semweb community, I tried hard, but all I came up with is the formal description of a codemonkey" ;)

03:15:10 <jhendler> s/relecant/relevant

03:15:21 <bengee> s/resurce/resource/

03:17:19 <bengee> yeah, swoogle is cool. not sure how long they'll be allowed to use that name, though..

03:19:59 <bengee> btw, jhendler, is the mindswap conf ontology more or less stable?

03:20:46 * bengee is looking for one for a conf editor/annotator at semanticweb.org..

03:21:05 <jhendler> stable - yes; particular well designed - no... that is, we're no longer modifying it, but it could use some work

03:21:20 * libby thinks (sorry) that you could make a conf ontology with icalendar + a couple of properties e.g. subevevent

03:21:35 <libby> mind you, icalendar isn't that great

03:21:40 <libby> hm

03:21:52 <libby> (but it is intereoperable)

03:22:36 <jhendler> bengee - the aktive portal folks did some work on ontologies for this sort of stuff - let's see it may be http://www.mindswap.org/2004/SSSW04/aktive-portal-ontology-latest.owl#workshop

03:23:04 <bengee> yeah, think I saw that.

03:23:22 <jhendler> or else http://www.aktors.org/ontology/portal

03:23:28 <libby> lunchtime here

03:23:42 <bengee> a mapping to rdf ical stuff could be interesting..

03:24:05 <bengee> schemaweb publishes events as vevents (from eventsherpa)

03:24:06 <jhendler> not terribly expressive - but they were useful -- adding ical would be interesting - we've fooled around with it a bit and it looks like something that someone should do...

03:24:40 * jhendler waves bye to libby - have a good lunch

03:26:03 <shammah> jhendler, I know we have plans to support ical.

03:27:06 * jhendler shammah - who is "we" ?

03:27:24 <shammah> jhendler, kowari/tks

03:27:56 <jhendler> ahh, cool - didn't recognize the nick... kowari with ical would be nice for a lot of things

03:28:31 <shammah> well we certainly agree :)

03:28:52 <shammah> it should also go nicely with our current support for mbox.

03:32:40 * bengee will probably check swoogle for a reasonably deployed vocab for confs and workshops and try to use rdf ical as generic super-terms..

03:34:00 <bengee> would be nice to have a link to the w3photo annotations..

04:57:08 <jsled> H: Ooooh! A Monkey!

04:57:11 <dc_rdfig> Added comment H5.

05:02:51 * libby waves at keck - you're at ISWC right? who are you?

05:05:10 <libby>http://annotation.semanticweb.org/iswc2004/annotated_docs/PreistISWC04Abstract.htm

05:05:10 <dc_rdfig> I: http://annotation.semanticweb.org/iswc2004/annotated_docs/PreistISWC04Abstract.htm from libby

05:05:38 <libby> I:|A Conceptual Architecture for Semantic Web Services, Chris Preist

05:05:39 <dc_rdfig> Titled item I.

05:06:07 <libby> I:paper presented at ISWC in semantic web services session

05:06:08 <dc_rdfig> Added comment I1.

05:06:14 <libby> I:|A Conceptual Architecture for Semantic Web Services, Chris Preist (abstract only)

05:06:15 <dc_rdfig> Titled item I.

05:07:46 <libby> I:[http://swws.semanticweb.org/public_doc/D4.2.pdf|a related document from the SWWS project]

05:07:47 <dc_rdfig> Added comment I2.

05:12:15 <libby> I:"what is a service - 1) provision of value in some domain; 2) software entity to provide something of value; 3) a set of WSDL operations which implement some behaviours (webservice)"

05:12:15 <dc_rdfig> Added comment I3.

05:13:55 <libby> I:"3 phases: discovery, contract agreement, service delivery"

05:13:55 <dc_rdfig> Added comment I4.

05:32:23 <libby> I'm having a lot of trouble distinguishing between web services and semantic web services. must be missing something

05:36:10 <libby> SWS has explicit semantics, WS implicit....? I bet WS people don't think of it like that tho...

05:37:30 * bengee thought it would just be some kind of a "wsdl in rdf vocab" that would allow enhancing certain parts of a service description while still maintaining a certain degree of compatibility to wsdl..

05:38:34 <libby>http://www.aiai.ed.ac.uk/project/ix/documents/2004/2004-iswc-uszok-kaos.pdf

05:38:35 <dc_rdfig> J: http://www.aiai.ed.ac.uk/project/ix/documents/2004/2004-iswc-uszok-kaos.pdf from libby

05:38:51 <libby> J:|Applying KAoS Services to Ensure Policy Compliance for Semantic Web Services Workflow Composition and Enactment

05:38:52 <dc_rdfig> Titled item J.

05:39:06 <libby> yeah makes sense bengee

05:39:27 <libby> have you been working all night? :(

05:39:35 <libby> J:by Andrzej Uszok, Jeff M. Bradshaw, Renia Jeffers, Austin Tate, Jeff Dalton

05:39:36 <dc_rdfig> Added comment J1.

05:39:47 <bengee> yeah. second night now :(

05:39:47 <libby> J:paper presented at ISWC in semantic web services session

05:39:48 <dc_rdfig> Added comment J2.

05:39:52 <libby> aww :(

05:40:21 <bengee> but having an active irc window keeps me awake ;)

05:40:44 <libby> I:[http://ontology.ihmc.us/|more information here]

05:40:45 <dc_rdfig> Added comment I5.

05:40:53 <libby> up against a deadline?

05:41:16 * bengee bought a fan heater and CDs for cold/lonely deri nights ;)

05:41:35 <bengee> kinda. self-set deadline..

05:42:04 <libby> self-motivated!

05:42:14 <bengee> always! :)

05:44:06 <libby> bengee is self-motivated and hardworking and would make a great addition to any semantic web team :)

05:44:13 <bengee> and seeing the rdf bits slowly fall into place motivates a lot...

05:44:26 <bengee> hehe, thanx

05:45:12 <libby> can we have a look yet?

05:46:20 <libby> I:[http://www.aiai.ed.ac.uk/project/cosar-ts/|Coalition Search and Rescue - Task Support (CoSAR-TS), a related project]

05:46:21 <dc_rdfig> Added comment I6.

05:47:01 <bengee> not really. a part of the design is on semanticweb.org (coming-soon page), very early (last week) parts are on the labs site (link from http://www.semanticweb.org/)

05:47:25 * libby has a peek

05:47:58 * bengee wants to publish a rdf-driven project resource browser tonight..

05:48:06 <libby> ooh

05:48:30 <bengee> hope the css works on linux machines..

05:48:37 <libby> whee! I like the boomerang thingies

05:49:05 <bengee> hey, those are w3c rdf mini graphs!

05:49:16 <libby> I know, just teasing

05:49:34 * bengee not a designer..

05:49:49 <libby> it looks nice

05:49:51 <libby> really!

05:50:17 <bengee> heh, thanx. did you say my foaflets I did with the same tool?

05:50:21 <bengee> s/say/see/

05:50:31 <libby> oh yeah, foaflets in the sea?

05:50:36 <libby> they were cool

05:50:55 <bengee> really easy tool :)

05:58:08 <libby> I:"KAos provides necessary capabilities to verify and enforce user defined policy in the automated process of planning and executing workflows"

05:58:08 <dc_rdfig> Added comment I7.

06:02:39 <libby>http://www.russet.org.uk/download/publications/biomoby-comparison-iswc2004.pdf

06:02:40 <dc_rdfig> K: http://www.russet.org.uk/download/publications/biomoby-comparison-iswc2004.pdf from libby

06:03:15 <libby> K:|Applying Semantic Web Services to bioinformatics: Experiences gained, lessons learnt

06:03:15 <dc_rdfig> Titled item K.

06:03:38 <libby> K: by Phillip Lord, Sean Bechhofer, Mark D. Wilkinson, Gary Schiltz, Damian Gessler, Duncan Hull1, Carole Goble and Lincoln Stein4

06:03:39 <dc_rdfig> Added comment K1.

06:03:53 <libby> K:paper presented at ISWC in semantic web services session

06:03:54 <dc_rdfig> Added comment K2.

06:05:23 <libby> K:bioinformatics: distributed data, data-type intensive; currently you use html; service composition is copy and paste. Automation is via perl and screenscraping

06:05:24 <dc_rdfig> Added comment K3.

06:07:43 <libby> K:future solution - sws?

06:07:44 <dc_rdfig> Added comment K4.

06:10:55 <libby> K:the text formats produced are very complex, but strings only; no xml, big legacy problem

06:10:55 <dc_rdfig> Added comment K5.

06:12:00 <libby> K:'thin semantics' - a simple wrapper to the text saying 'this is a blast file'"

06:12:01 <dc_rdfig> Added comment K6.

06:14:07 <libby> K:this can still help a lot with searching

06:14:08 <dc_rdfig> Added comment K7.

06:14:31 <libby> K:lots of data is represented in lots of says, e.g. 26 text formats for representing dna sequences

06:14:31 <dc_rdfig> Added comment K8.

06:15:21 <libby> K:use 'shims' pipe connectors - thin semantics again

06:15:21 <dc_rdfig> Added comment K9.

06:23:43 <libby> K:for services - can;t get any semantics from the wsdls, only from the websites; these user-orientated services have to be written by the users (

06:23:44 <dc_rdfig> Added comment K10.

06:24:16 <libby> K10:for services - can;t get any semantics from the wsdls, only from the websites; these user-orientated services have to be written by the users (e.g. pedro), could be a bottleneck - 600+ services

06:24:17 <dc_rdfig> Replaced comment K10.

06:26:46 <libby> K:ontologies are tricky without a lot of money, time and community involvement: do your own one (risk diverging form the biologist's descriptions); have an api where anyone can just add terms (seems to work quite well); use ontologies whereever you can find them (risks islands of interoprtability)

06:26:46 <dc_rdfig> Added comment K11.

06:27:19 <dngor> ontowiki

07:13:33 <libby>http://www.aifb.uni-karlsruhe.de/WBS/pci/FromTablesToFramesTech.pdf

07:13:36 <dc_rdfig> L: http://www.aifb.uni-karlsruhe.de/WBS/pci/FromTablesToFramesTech.pdf from libby

07:14:07 <libby> L: by Aleksander Pivk, Philipp Cimiano, York Sure

07:14:08 <dc_rdfig> Added comment L1.

07:14:17 <libby> L:|From Tables to Frames

07:14:18 <dc_rdfig> Titled item L.

07:14:41 <libby> L:paper presented at ISWC in semantic web services session

07:14:42 <dc_rdfig> Added comment L2.

07:14:58 <libby> L2:paper presented at ISWC in SW mining session

07:14:58 <dc_rdfig> Replaced comment L2.

07:17:12 <libby> L:"an approach for automatic generation of frames out of tables which subsequently supports the automatic population of ontologies from table-like structures"

07:17:12 <dc_rdfig> Added comment L3.

07:18:18 <libby> L:html tables, that is

07:18:19 <dc_rdfig> Added comment L4.

07:22:55 <libby> J:[http://ontology.ihmc.us/|more information here]

07:22:56 <dc_rdfig> Added comment J3.

07:23:13 <libby> J:[http://www.aiai.ed.ac.uk/project/cosar-ts/|Coalition Search and Rescue - Task Support (CoSAR-TS), a related project]

07:23:14 <dc_rdfig> Added comment J4.

07:23:42 <libby> J:"KAos provides necessary capabilities to verify and enforce user defined policy in the automated process of planning and executing workflows"

07:23:43 <dc_rdfig> Added comment J5.

07:24:23 <libby> I5:""

07:24:23 <dc_rdfig> Deleted comment I5.

07:24:27 <libby> I6:""

07:24:28 <dc_rdfig> Deleted comment I6.

07:24:30 <libby> I7:""

07:24:31 <dc_rdfig> Comment I7 not found.

07:24:51 <libby> I5:""

07:24:52 <dc_rdfig> Deleted comment I5.

07:26:04 <libby> K:I forgot to use quotes but these comments are paraphrased from the talk and slides

07:26:04 <dc_rdfig> Added comment K12.

07:29:55 <libby> J:"we use wordnet and googlesets to find synonyms for cell labels"

07:29:56 <dc_rdfig> Added comment J6.

07:32:00 <libby> .t EST

07:32:02 <phenny> Tue, 09 Nov 2004 02:32:00 EST

07:36:29 <libby> is anyone having problems submitting stuff to www2005?

07:36:40 <CaptSolo> libby: blogging from ISWC? :)

07:36:55 <libby> I've not personal-blogged but I am here :)

07:37:51 <libby> the daedline was initially midnight 8th nov, haawaii time; now it's been 'extended' to 5pm EST. which is before the first time

07:38:04 <libby> (it's not for me, I forgot to write anything :(

07:38:42 <bengee> does something not work re www2005

07:38:44 <bengee> ?

07:38:49 <libby> yeah

07:38:58 <bengee> aharth plans to..

07:39:05 <libby> it wont allow submissions cos it claims the daedline has passed - but it hasn't

07:39:21 <libby> .t EST

07:39:23 <phenny> Tue, 09 Nov 2004 02:39:21 EST

07:40:35 <libby> .t Hawaii

07:40:37 <phenny> Sorry, I don't know about time zone HAWAII.

07:40:40 <libby> hm

07:40:55 <libby> .t HST

07:40:57 <phenny> Sorry, I don't know about time zone HST.

07:41:02 <libby> hm

07:41:32 <libby> anyway, it's 9.30 PM which leaves 2.5 hours left till daedline

07:42:06 <libby> .t HAST

07:42:08 <phenny> Sorry, I don't know about time zone HAST.

07:42:17 <libby> well, you should, phenny

07:43:07 <larsbot> don't be so severe with her, libby. bots have feelings, too, you know :-)

07:43:22 * libby so mean

07:43:26 <larsbot> yeah :)

07:43:57 <libby> .t HADT

07:44:00 <phenny> Sorry, I don't know about time zone HADT.

07:44:23 <bengee> libby, you mean, the submission form does not work any more although the deadline shouldn't have passed yet?

07:44:32 <libby> yeah

07:44:41 <libby> :/

07:44:59 <bengee> hm, aharth hasn't submitted yet..

07:45:11 <libby> unless I've missed something, they've got their timezones mixed up

07:45:23 <libby> no I think there may be a few more like that

07:46:22 * libby knows of one or two people here finishing theirs' off

07:48:28 <libby>http://www.csd.abdn.ac.uk/%7Eggrimnes/pubs/LearningFOAFDesc.pdf

07:48:29 <dc_rdfig> M: http://www.csd.abdn.ac.uk/%7Eggrimnes/pubs/LearningFOAFDesc.pdf from libby

07:48:44 <libby> M:|Learning Meta-Descriptions of the FOAF Network, Gunnar AA. Grimnes, Pete Edwards and Alun Preece

07:48:44 <dc_rdfig> Titled item M.

07:49:12 <libby> M:a preprint of the paper Gunnar's giving now at ISWC 2004 in the sw mining track

07:49:13 <dc_rdfig> Added comment M1.

07:51:02 <libby> M:"SW topologies: SW forests - xml will do; semantic webs are different, interconnected graphs with no clear top level node - like foaf"

07:51:02 <dc_rdfig> Added comment M2.

07:52:31 <libby> M:"foaf 6.5 m triples; 259298 known foaf files as of sept 2004; many hand generated but e.g. ecademy, livejournal also producing many files"

07:52:31 <dc_rdfig> Added comment M3.

07:53:25 <shammah> libby: cool. what rdf-store are you using?

07:53:43 <libby> M1:a preprint of the paper Gunnar's giving now at ISWC 2004 in the sw mining track (found by [http://blog.ilrt.org/price/|Simon Price] - thanks Simon!)

07:53:46 <dc_rdfig> Replaced comment M1.

07:53:52 <libby> shammah - not me, gunnar :)

07:54:00 <libby> I don;t know what he's using....

07:54:14 <libby> handy to ahve the stats though

07:55:31 <shammah> would be even nicer to get 6.5m triples of foaf data for testing :)

07:55:52 <libby> he got it from someone else he said

07:56:06 <libby> - might ahve to wait for the final paper to see

07:56:58 <libby> M:"foaf is scruffy - human errors, interest is a problem - some people just use a string and people use different urls for the same concept"

07:56:58 <dc_rdfig> Added comment M4.

07:58:10 <libby> M:cleaned it up. wanted to find groiups of people; then concert to prolog and use with ILP. then want to feed it through again with generated rules but not done that yet"

07:58:11 <dc_rdfig> Added comment M5.

07:59:17 <libby> M:"use hierarchical aggolomerative cluster algorithm (HAC), building a tree by recursively merging most similar clusters"

07:59:18 <dc_rdfig> Added comment M6.

08:01:20 <libby> M:"need similarity measures for RDF grphs. "hamming distance" didn;t work very well (immediate properties only); tried conceptual graph comparision - Montes-y-Gomez 2000; must extract subgraph around the person to use it"

08:01:20 <dc_rdfig> Added comment M7.

08:03:04 <libby> M:"trial and error - extract subgraph by 2 steps forward and one step backwards (any property) - works quite well for foaf; other data (e.g. imdb) didn't work with this heuristic"

08:03:05 <dc_rdfig> Added comment M8.

08:04:55 <libby> M:"used aleph ILP system. Uses 100 most frequent preds but not 'knows' - or else clusters are all based just on knows. the rules generateed were also constrained by the foaf schema"

08:04:55 <dc_rdfig> Added comment M9.

08:07:44 <CaptSolo> libby: had to write for www05 so soon?

08:07:51 <CaptSolo> then i forgot to write anything as well :)

08:07:58 <CaptSolo> not that i had time ...

08:09:05 <CaptSolo> hi bijan :)

08:09:14 <libby> M:"rules found: (1) people who have written a particular paper together; (2) people who have group homepage www.aktors.org (3) people with nearest airport ABZ (i.e. aberdeen agents group) (4) people who trusthighly someone else (only used by 8 people though :) (5) people who created a paper of type application/postscript"

08:09:15 <dc_rdfig> Added comment M10.

08:09:31 <libby> hi capt, yeah daedline is pretty far ahead

08:11:05 <CaptSolo> :)

08:12:08 <libby> M:"filtering of pred/value pairs might be a good idea. foaf is too sparse so get weird results, same preds are not used enough. now trying IMDB which is much denser. trying to make a movie-trivia learner - e.g. did you know that all the films you like have someone in them who dated julia roberts? use common sense and ontologies to filter predicates"

08:12:08 <dc_rdfig> Added comment M11.

08:14:11 <libby> M:"conclusions: lots of preprocessing required, scruffiness is a problem. evaluation of rules is difficult - end up with lots of them but few are interesting. scaling up is the biggest problem of the sw - some of these experiments took weeks..."

08:14:12 <dc_rdfig> Added comment M12.

08:14:49 <libby> sorry that's ffar too many comments for the chump

08:15:20 * libby jsut pleased that foaf got a mention here :)

08:15:51 * libby wonders if masahide's here

08:17:01 <libby> ---end of the day, apart from demos and things

08:17:20 <libby> and snacks :)

08:17:48 * bengee dares an upload..

08:18:39 <libby> bummer, masahide's not on the attendee list...I thought he was coming

08:18:51 * libby looks

08:19:15 <libby> where is it bengee?

08:19:47 <bengee> only the project documentation site (http://labs.semanticweb.org/relaunch2004-prj_browser)

08:20:16 <bengee> incomplete, etc. but onto-backed and somehow working.

08:20:50 <bengee> and created with an (almost) generic rdf editor

08:21:49 <libby> gosh, nice

08:21:55 <bengee> not really cool on the front-end, but for me, it's an important proof of concept..

08:22:05 <libby> yeah, it workie :)

08:23:28 * libby gotta go, bye bengee - dont work too hard

08:23:43 * bengee off, too, now

08:23:59 <bengee> CU, ta for the community coverage

09:47:14 <Jccq> Jccq is now known as GiovanniT

10:32:54 <d2m_> d2m_ is now known as d2m

11:22:48 <karlScrewed> karlScrewed is now known as karlcow

13:24:59 <bblfish> I have an OWL file and would like to turn it into some nice easy to read html format (like a javadoc) so people less familiar can browse it. Are there any good solutions to this?

13:27:54 <rowland> bblfish: www.schemaweb.info allows you to view an ontology's classes and properties in html

13:28:19 * bblfish looking

13:29:07 <bblfish> Should I just submit a schema?

13:29:49 <rowland> could do I suppose - it would save you the trouble of hosting it yourself

13:30:16 <rowland> Plus schemaweb.info has some useful ReST and SOAP interfaces

13:30:33 <rowland> So you can access those to get hold of the ontology

13:31:04 <rowland> bblfish: Your other option is to put something together yourself

13:32:07 <bblfish> thanks. I have placed my ontology here http://bblfish.net/work/atom-owl/2004-11-09/

13:32:08 <bblfish> But it is not finished, so I am not sure I want to upload it yet somewhere...

13:33:42 * bblfish ok I am starting to see how it works...

13:34:44 <dajobe> rowland: you might try using swoop from mindlab. it is an owl browser/editor and may have some kind of export

13:34:50 <phenny> dajobe: 02:09Z <sbp> ask dajobe if he could possibly set mode +c for #rdfig please (strips colour codes)

13:35:21 <dajobe>http://www.mindswap.org/2004/SWOOP/

13:35:24 <dc_rdfig> N: http://www.mindswap.org/2004/SWOOP/ from dajobe

13:35:28 <dajobe> N:|SWOOP - A Hypermedia-based Featherweight OWL Ontology Editor

13:35:29 <dc_rdfig> Titled item N.

13:35:32 <bblfish> yes the schemaweb looks good, but the class browser is a little minimal

13:35:51 <rowland> bblfish: Yeah, I thought that as well.

13:36:13 <bblfish> I need it to help people who are skeptical overcome their skepticism.

13:36:43 <rowland> bblfish: Have you shown them any designs of how you intend to present classes and properties?

13:36:58 <rowland> dajobe: Sorry, yeah, I noticed that a few days ago

13:37:19 <rowland> dajobe: I was having a little play with TrustMail at the time

13:38:05 <bblfish> Well I wrote my own UML diagrams previousl. Eg: http://bblfish.net/work/atom-owl/2004-08-12/blogexample.html

13:38:52 <bblfish> I'll look at swoop. Protégé has an html export facility but it looks like it was written for the OWL predecessor.

13:39:40 <bblfish> the DAML+OIL think I think

13:39:51 <rowland> yeah, probably

13:40:52 <bblfish> At the foaf foaf conference in Galway there was someone that shown an interesting demo of this, but I am not sure if he ever released it. I can't quite remember who it was.

13:41:09 <bblfish> this: something that transformed OWL into html

13:42:02 <rowland> hehe

13:42:05 <rowland> OWL2html

13:42:08 <rowland> using XSLT

13:42:19 * mortenf has xslt that generates stuff like this: http://purl.org/net/vocab/2004/07/visit

13:42:59 <rowland> mortenf: That looks good

13:43:01 <mortenf> not complete, etc...

13:47:10 <bblfish> the person I was in Galway had done it in java... But yes, it looks good.

13:47:24 <bblfish> SWOOP does not seem to have an export to html functionality

13:47:40 <mortenf> bengee has something iirc

13:48:57 <bblfish> do you think (s)he has published it?

13:49:10 <mortenf> no, i don't he has (yet)

13:49:20 <mortenf> ... think ...

13:50:04 <rowland> bblfish: XSLT would be simpler than writing a servlet that used jena

13:50:25 <rowland> Although I find XSLT a bit fiddly;-)

13:50:31 <bblfish> yes. :) All depends on what you know of course.

13:50:41 <rowland> :-)

13:51:15 <bblfish> there is this http://powl.sourceforge.net/

13:51:39 <bblfish> put it seems a bit fiddly to set up too. And I am not sure my server will give me access to php4 and all the rest.

13:53:34 <rowland> Hmm, it looks less fearful than Protege

13:54:01 <bblfish> fearful =? featureful

13:54:08 <rowland> hehe

13:54:17 <rowland> A mannerism of mine

13:54:47 <rowland> Protege has improved a lot, but it still does my head in

13:55:24 <bblfish> it is quite easy now. It is really easy to install. The powl looks a lot more difficult to install, though it may be easier to use

13:56:07 <rowland> True. Powl is online - no need for a client other than a browser. I find that quite attractive.

13:57:43 <bblfish> yes. but I can't find a place to use it. If they had a demo site, then at least I could first find out if it is what I want. But I have not seen one. Or have you?

13:58:04 <rowland> Did you not try the demo?

13:58:14 <bblfish> it does not link to anything

13:58:21 <rowland>http://powl.contxserver.de/powl-cvs/

13:58:22 <dc_rdfig> O: http://powl.contxserver.de/powl-cvs/ from rowland

13:58:22 <bblfish> it links to the source, as far as I can tell

13:58:48 <rowland> I've been having a play - you can create a new ontology or load one of their examples

13:58:50 <bblfish> oh. I see. I thought that was the CVS reporsitory

13:59:13 <bblfish> there are so many links, one sometimes misses the tree for the forst

13:59:34 <rowland> Not to worry

13:59:43 <bblfish> thanks :-)

13:59:55 <rowland> No probs;-)

14:02:42 <bblfish> yes. you can export your ontology with that.

14:03:41 <rowland> Cool

14:04:13 <crschmidt> It doesn't use proper Accept: headers when downloading

14:05:06 <crschmidt> (so using http://xmlns.com/foaf/0.1/ doesn't work, while http://xmlns.com/foaf/0.1/index.rdf does)

14:06:00 <bblfish> I have seen a few other tools that have that problem. I reported the problem like that to Protégé, who told me that was a problem with Jena. Never knew if this was fixed.

14:06:37 <bblfish> (though that is java library and this is probably a C library)

14:06:58 <crschmidt> This? powl is all PHP, I think?

14:07:38 <crschmidt> ah, nevermind, the library si probably curl or somethign similar.

14:07:39 <rowland> Yeah

14:07:41 <bblfish> probably using redland at the back, no?

14:07:45 <crschmidt> Nope

14:07:45 <crschmidt> RAP

14:07:52 <rowland> No, it's a PHP RDF parser

14:08:00 <bblfish> heh :-)

14:08:15 <crschmidt> But I'm pretty sure RAP doesn't do downloading from URIs, which means that you have to do it with curl.

14:08:27 <crschmidt> (or something similar)

14:08:29 <rowland> Ah

14:08:39 <crschmidt> And it's possible to set up Accept: headers in curl, but an extra step.

14:08:53 <crschmidt> (I could be totally wrong on this, I haven't worked with RAP heavily ever.)

14:12:31 <crschmidt> This: http://chxo.com/rdfworld/index.htm seems to support the fact that RAP doesn't have a download_from_url function: file_get_contents() is a php builtin (rather than a RAP function).

14:13:43 <bblfish> O: returns a very nice html presentation of my OWL file

14:13:45 <dc_rdfig> Added comment O1.

14:14:24 <crschmidt> BLURB: RDF Tool Download Support

14:14:26 <dc_rdfig> P: RDF Tool Download Support from crschmidt

14:14:45 <crschmidt> P: In looking at RAP, it seems that it doesn't have a "download model from URI" function.

14:14:45 <dc_rdfig> Added comment P1.

14:14:51 <Talliesin> <bblfish> I have seen a few other tools that have that problem.

14:15:04 <Talliesin> Can you tell me which, if your info on those is still current.

14:15:24 <crschmidt> P: Although the tool that I use (Redland) offers this support, I didn't end up using it for reasons I can't remember at the moment.

14:15:25 <dc_rdfig> Added comment P2.

14:15:32 <Talliesin> (suddenly remembered I was trying to do a survey of tools as to how they are in the particular way)

14:15:47 <crschmidt> P: Which RDF tools do you use? Do they have Download support? If so, do you use it or something else like curl?

14:15:47 <dc_rdfig> Added comment P3.

14:16:01 <bblfish> Talliesin: oh. It was sésamé. I had exactly the same problem importing foaf into sesame

14:16:21 <Talliesin> Do you know if that's still the case.

14:16:30 <Talliesin> ?

14:16:41 <bblfish> ie: http://xmlns.com/foaf/0.1/ doesn't work, you needed to use the index.rdf

14:16:43 <bblfish> no

14:16:52 <bblfish> that was perhaps 2 months ago. I could check

14:17:37 <Talliesin> Cheers

14:18:12 <crschmidt> P: Talliesin did a survey on this on one mailing list or another recently: Do your tools use Accept: headers to specify a preference for application/rdf+xml ?

14:18:18 <dc_rdfig> Added comment P4.

14:19:43 <Talliesin> P: FOAF Explorer was the only tool reported as not specifying application/rdf+xml, but I think I may need to ask again actively seeking tools which fail to do this.

14:19:43 <dc_rdfig> Added comment P5.

14:21:28 <mortenf> P: FE does now (application/rdf+xml, */*)

14:21:29 <dc_rdfig> Added comment P6.

14:21:39 <dajobe> P:redland/raptor/rasqal all prefer application/rdf+xml (higher q) but then accept others at lower q

14:21:40 <dc_rdfig> Added comment P7.

14:22:25 <crschmidt> P: julie does the same as FE (application/rdf+xml, */*)

14:22:26 <dc_rdfig> Added comment P8.

14:23:37 <MacIntyre> Does anyone have a decent example of Photo RDF that includes everything from foaf depictions to image regions? I'm testing our Photoshop CS's XMP RDF to see if it will auto-detect the new markup.

14:24:20 <mortenf> see http://www.wasab.dk/morten/2004/08/photos/foafcamp/3/image-6.html

14:24:33 <MacIntyre> Thanks mortenf

14:24:52 <JibberJim> see http://jibbering.com/rdfsvg/example.rdf for regions

14:25:04 <JibberJim> ah, although that might be old syntax :-(

14:26:11 <MacIntyre> Yeah - wasn't their some new stuff that came out of the recent WWW* ?

14:26:47 <bblfish> Talliesin: i think Sésame does not do this. I tried the following with the latest Sesame:

14:26:48 <bblfish> File -> Build New Project -> Select "Owl File"

14:26:50 <bblfish> if I then enter http://xmlns.com/foaf/0.1/ it does not load anything, but uf I fetch http://xmlns.com/foaf/0.1/index.rdf then it does it right.

14:27:30 <Talliesin> P: Main thing is does it have application/rdf+xml at a higher q than */*, text/html, application/xhtml+xml, text/xml, application/xml (unless it's a tool which has some good behaviour for them, and for RDF - i.e. it isn't just an RDF tool and it won't be "upset" if it doesn't get RDF)

14:27:31 <dc_rdfig> Added comment P9.

14:34:05 <MacIntyre> Hmmmm. They showed up in the Photoshop CS File Info Window and I could browse all the values....but it didn't save.

14:35:24 <bblfish> I sent a mail to the protégé owl mailing list (protege-owl AT SMI.Stanford.EDU) on this on the 27 June 2004 entitled "foaf inclusion"

14:35:25 <bblfish> Holger Knublauch responded that this was a problem with Jena

14:35:27 <bblfish> On the 6 August I sent a mail to the jena mailing list (jena-dev AT yahoogroups.com) entitled "content negotiation bug"

14:36:16 <danbri_>http://www.w3.org/TR/2004/WD-ttaf1-dfxp-20041101/

14:36:17 <dc_rdfig> Q: http://www.w3.org/TR/2004/WD-ttaf1-dfxp-20041101/ from danbri_

14:36:43 <danbri_> Q:|W3C Timed Text (TT) Authoring Format 1.0 (Dist'n Format Exchange Profile (DFXP))

14:36:43 <dc_rdfig> Titled item Q.

14:37:25 <bblfish> Anyway, I did not follow up, as I had other things to do... :-)

14:38:13 <danbri_> Q:As far as I can tell, this is being presented as a profile of a spec that is as-yet unpublished/unwritten. It's a draft of a language we could use for annotating media files (eg. spoken word) with subtitles etc.

14:38:13 <dc_rdfig> Added comment Q1.

14:41:40 <chaalsNCE> Ciao Giovanni

14:45:16 <CaptSolo> danbri - hi :)

14:45:22 <CaptSolo> hi bblfish, chaals :)

14:45:38 <bblfish> hi. CaptSolo

14:46:27 <bblfish> hi Uldis :-)

14:59:50 <CaptSolo> yes :)

15:00:04 <CaptSolo> how did you find out my real name, mmm? ;)

15:00:17 <crschmidt> ^name CaptSolo

15:00:17 <julie> Uldis Bojars

15:00:29 <CaptSolo> WOOOWWWWW!!!

15:00:38 <CaptSolo> pleasure to meet you, Julie! :)

15:00:54 <CaptSolo> not she recognises me. which is nice.

15:00:59 <CaptSolo> thanks, chris

15:01:04 <CaptSolo> and hello :)

15:01:23 <CaptSolo> ^whereisthemoney CaptSolo

15:01:37 <crschmidt> ^icbm CaptSolo

15:01:37 <julie> No results.

15:01:43 <CaptSolo> along the same lines as Google Keys parody ;)

15:01:56 <CaptSolo> well - i thought so. i don't know where it is either...

15:02:36 <CaptSolo> chris: did you load more data / add new functionality to Julie? she did not recognise me before.

15:02:44 <CaptSolo> and what is 'icbm'?

15:02:54 <crschmidt> ^icbm crschmidt

15:02:54 <julie> 42.9813 -71.4369(x8)

15:03:00 <crschmidt> I don't know, people are always adding more data.

15:05:03 <KjetilK> intercontiental ballistic missile?

15:05:16 <crschmidt> Yep

15:05:24 <crschmidt> Just a geo:lat and geo:long

15:05:26 <KjetilK> bearing and thrust....

15:05:30 <KjetilK> :-)

15:05:33 <monkeyiq> a charmer of a geospatial predicate

15:05:51 <rowland> ^icbm rowland

15:05:52 <julie> No results.

15:05:54 <rowland> :(

15:06:01 <KjetilK> ^icbm KjetilK

15:06:02 <julie> 59.9496 10.8235

15:06:16 <crschmidt> rowland: got a FOAF file?

15:06:21 <crschmidt> ^available rowland

15:06:21 <julie> No results.

15:07:22 <rowland> crschmidt: heh, yeah

15:07:50 <KjetilK> rowland: go ^add URL for it

15:07:54 <crschmidt> rowland: ^add <url>

15:08:03 <rowland> cool

15:08:04 <rowland> ok

15:08:16 <crschmidt> <> not needed, just an indicator that it'ss a url ;)

15:08:31 <rowland> cheers

15:10:05 <rowland> ^add http://grid.cx/foaf.rdf

15:10:06 <julie> Adding that to my database...

15:10:11 <rowland> julie: cheers

15:10:12 <julie> Added 16 statements from http://grid.cx/foaf.rdf. Model size is 651067.

15:10:31 <crschmidt> ^agentknows Rowland

15:10:31 <julie> No results.

15:10:39 <crschmidt> hm.

15:10:43 <rowland> oh, mine appears a little sparse

15:11:07 <crschmidt> ^contact rowland

15:11:08 <julie> No results.

15:11:12 <crschmidt> ^contact Rowland

15:11:12 <julie> tel:+44-23-8059-8346 foaf:phone

15:11:21 <rowland> at least that works

15:11:22 <rowland> hmm

15:11:35 <crschmidt> I'm not sure why agentknows doesn't work

15:11:36 <crschmidt> it should

15:11:52 <crschmidt> ^homepage Rowland

15:11:52 <julie>http://grid.cx

15:11:55 <dc_rdfig> R: http://grid.cx from julie

15:13:47 <rowland> crschmidt: can Julie replace foaf files?

15:14:02 <rowland> or at least replace statements

15:14:36 <crschmidt> rowland: if you add the file again, the duplicates will be gotten rid of next time I smush

15:14:43 <crschmidt> Assuming you have an IFP, which you do

15:15:16 <rowland> ok

15:17:06 <CaptSolo> ^avalable CaptSolo

15:17:11 <CaptSolo> ^available CaptSolo

15:17:11 <julie> foaf:homepage, foaf:name, foaf:firstName, srw1.1:masters(x3), srw1.1:reads, foaf:weblog(x2), rdf:type, foaf:surname, rdfs:seeAlso(x2), foaf:mbox_sha1sum, foaf:knows(x4), foaf:title, foaf:schoolHomepage, foaf:nick

15:17:29 <rowland> ^available rowland

15:17:29 <julie> No results.

15:17:33 <rowland> ^available Rowland

15:17:34 <julie> No results.

15:17:36 <rowland> oi

15:17:53 <crschmidt> hm.

15:17:59 <crschmidt> ^available Rowland

15:17:59 <julie> foaf:homepage, foaf:name, foaf:firstName, foaf:phone, rdf:type, foaf:surname, foaf:workplaceHomepage, foaf:mbox_sha1sum, foaf:knows, foaf:title, foaf:workInfoHomepage, foaf:schoolHomepage, foaf:nick

15:18:10 <rowland> eh?

15:18:15 <crschmidt> you probably had a space on the end?

15:18:22 <rowland> oh, quite possibly

15:19:08 <crschmidt> The bot is very literal :)

15:19:45 <rowland> heh

15:22:52 <chaalsNCE> ^available Chaals

15:22:53 <julie> No results.

15:22:59 <chaalsNCE> ^available Charles

15:23:00 <julie> No results.

15:23:41 <chaalsNCE> ^available charles@w3.org

15:23:41 <julie> No results.

15:24:19 <chaalsNCE> ^add http://www.w3.org/People/Charles/foaf.rdf

15:24:19 <julie> Adding that to my database...

15:24:26 <julie> Added 317 statements from http://www.w3.org/People/Charles/foaf.rdf. Model size is 651384.

15:24:30 <chaalsNCE> ^available charles@w3.org

15:24:31 <julie> No results.

15:24:35 <chaalsNCE> ^available Charles

15:24:36 <julie> No results.

15:24:40 <JibberJim> you're not available charles.... face it

15:24:43 <chaalsNCE> ^available Chaals

15:24:43 <julie> No results.

15:24:56 <chaalsNCE> Hey, I am as available as I get!!

15:25:04 <mortenf> ^available chaals

15:25:05 <julie> foaf:nick(x2), foaf:mbox_sha1sum(x2), foaf:name(x2), rdf:type(x2)

15:25:14 * mortenf waves and smiles

15:25:29 * chaalsNCE waves to mortenf and frowns a bit

15:25:59 <chaalsNCE> and wonders what the rdf:type of chaals is :-)

15:26:02 * crschmidt smushes database.

15:26:22 <chaalsNCE> ^available chaals

15:26:23 <julie> foaf:nick(x2), foaf:mbox_sha1sum(x2), foaf:name(x2), rdf:type(x2)

15:26:24 <crschmidt> ^q ?type where (?p foaf:nick "chaals") (?p rdf:type ?type)

15:26:24 <julie> foaf:Person(x2)

15:26:42 <mortenf> crschmidt, two requests: add a space before results, and say "No results (remember, I'm case sensitive)" when no results are found

15:27:00 <crschmidt> mortenf: why the space before results?

15:27:09 <mortenf> R:

15:27:10 <dc_rdfig>http://grid.cx

15:27:10 <crschmidt> chumping-stuff?

15:27:12 <mortenf> yep

15:27:15 * crschmidt nods.

15:27:23 <crschmidt> I think someone else made the same request.

15:30:53 <crschmidt> ^homepage Rowland

15:30:53 <julie>http://grid.cx

15:30:56 <dc_rdfig> S: http://grid.cx from julie

15:30:58 <crschmidt> dang it.

15:31:16 <crschmidt> I shoudl have tested that in a non-chumping channel, heh

15:32:06 <crschmidt> okay, fixed

15:33:28 <rowland> ^homepage Rowland

15:33:28 <julie> http://grid.cx

15:35:29 <crschmidt> ^available chaals

15:35:29 <julie> foaf:nick, foaf:mbox_sha1sum, foaf:name, rdf:type

15:35:33 <crschmidt> much better

15:35:39 <crschmidt> ^name chaals

15:35:40 <julie> Charles McCathieNevile

15:36:38 <CaptSolo> ^homepage CaptSolo

15:36:39 <julie> http://kaste.lv/~captsolo

15:36:47 <jsled> ^homepage jsled

15:36:48 <julie> http://www.asynchronous.org/jsled

15:37:24 <CaptSolo> crschmidt: is more info about julie available online?

15:38:14 <crschmidt> CaptSolo: not really.I need to make a webpage for her. I have a blog post about her somewhere, I think...

15:38:16 <crschmidt> .g rdf tools and why I like them

15:38:19 <phenny> rdf tools and why I like them: http://www.simonstl.com/articles/RDF.html

15:38:22 <crschmidt> hm.

15:38:25 <crschmidt> .g rdf tools and why I like them crschmidt

15:38:28 <phenny> rdf tools and why I like them crschmidt: http://crschmidt.livejournal.com/260054.html

15:42:36 <CaptSolo> hi cloud :)

15:43:19 <Cloud> Hi CaptSolo

15:43:30 <CaptSolo> chris: nice to hear there will be a webpage

15:43:45 <CaptSolo> thanks for the pointer to the blog post - will take a look at it

15:44:10 <CaptSolo> i noticed your livejournal before - linked to an article in it, i guess

15:44:31 <CaptSolo> ^info

15:44:31 <julie> I'm a Redland/Python based RDF query bot. Source in SVN at <http://crschmidt.net/svn/rdfpython/trunk/>. Commands are ^add <url>, which adds statements, ^newcommand <name> is <RDQL Query>, ^runcommand <commandname> <arguments> . ^commandlist lists current commands, ^commandlist <command> shows info on a command, ^part will have me part a channel, ^join <#chan> will have me join a channel. Talk to crschmidt for more. @@ More here

15:48:27 <crschmidt> You can always play with the bot in #julie too

15:48:33 <crschmidt> with no fear of retribution for botspam

15:48:38 <crschmidt> (some of the commands get very spammy)

17:49:56 <DanC>http://esw.w3.org/topic/DawgShows

17:49:59 <dc_rdfig> T: http://esw.w3.org/topic/DawgShows from DanC

17:50:06 <DanC> T:|DawgShows

17:50:07 <dc_rdfig> Titled item T.

17:50:29 <DanC> T:a playground for SPARQL demo services and related events

17:50:29 <dc_rdfig> Added comment T1.

17:58:11 <chaalsNCE> ye all

17:58:22 <balbinus> hi

18:26:05 <PhilT> msg sh1m That you Tom?

18:26:32 <crschmidt> It is, although he hasn't talked in a while.

19:48:49 <sh1m> sh1m is now known as sh1mmer

20:43:49 <balbinus> does anyone know of an hires version of the RDF logo (http://www.w3.org/RDF/icons/)?

20:45:49 <mortenf> i've got a mockup: http://www.wasab.dk/morten/2004/02/rdf.svg

20:47:02 <balbinus> perfect :)

20:47:07 <balbinus> thanks a lot, mortenf :)

21:04:41 <JibberJim> why's that only a mockup mortenf?

21:05:06 <mortenf> it's not perfect? :)

21:05:42 <JibberJim> what's wrong with it?

21:06:05 <mortenf> it doesn't look exactly like the original

21:08:11 <balbinus> mortenf: any rights / licence / ... ?

21:08:27 <mortenf> i was just thinking about that

21:09:16 * mortenf looks for a cc license

21:09:28 <balbinus> by-sa?

21:10:00 <JibberJim> ah, where's the original?

21:10:13 <balbinus> the original logo?

21:10:18 <balbinus>http://www.w3.org/RDF/icons/

21:10:22 <dc_rdfig> U: http://www.w3.org/RDF/icons/ from balbinus

21:10:44 <balbinus> :(

21:10:49 <balbinus> U:|RDF Icons

21:10:49 <dc_rdfig> Titled item U.

21:11:14 <balbinus> U:[SVG version here|http://www.wasab.dk/morten/2004/02/rdf.svg] by mortenf

21:11:14 <dc_rdfig> Added comment U1.

21:11:20 <JibberJim> U: The example HTML links are broken!

21:11:21 <dc_rdfig> Added comment U2.

21:11:31 <JibberJim> U: references http://www.w3.org/RDF/icons/rdf_developer_button.64

21:11:32 <dc_rdfig> Added comment U3.

21:13:32 <balbinus6> U:W3C's link checker says 301 for Eric Miller's homepage

21:13:33 <dc_rdfig> Added comment U4.

21:13:40 <mortenf> license added, by-sa/2.0

21:14:08 <balbinus6> U:Thanks mortenf for the SVG (by-sa 2.0, btw)

21:14:09 <dc_rdfig> Added comment U5.

21:15:23 <balbinus6> well, good night everyone!

22:00:06 <nephrael> nephrael is now known as dmiles

22:00:29 <dmiles> dmiles is now known as dmiles_nephrael

22:36:16 <CaptSolo> hi chris-p!

22:36:24 <CaptSolo> how're things? :)

22:37:02 <CaptSolo> what's new?

23:01:46 <chris-p> Hi CaptSolo! You still here? (Sorry for the delay)

23:11:17 <nephrael> nephrael is now known as dmiles

23:11:36 <dmiles> dmiles is now known as dmiles_nephrael

23:13:14 <eva_mendez> eva_mendez is now known as evamen_