Semantic Web Interest Group IRC Chat Logs for 2005-05-19

02:34:06 <sbp> sbp is now known as |\

02:34:13 <|\> |\ is now known as sbp

02:36:57 <sbp> sbp is now known as |\

02:37:11 <|\> |\ is now known as sbp

03:27:59 <Indig0|sith4tw> Indig0|sith4tw is now known as Indig0

04:43:53 <Indig0> Indig0 is now known as Indig1

07:19:30 <jip> jip is now known as Jipp

07:26:39 <schepers> schepers is now known as scheperzzzzzzz

13:55:33 <MooonFog> MooonFog is now known as MoonFog

14:43:11 <scheperzzzzzzz> scheperzzzzzzz is now known as schepers

14:53:43 <crschmidt> phenny: tell DanC, regarding http://www.ilrt.bris.ac.uk/discovery/chatlogs/swig/2005-05-17.html#T22-33-13 , there is now http://wick.sourceforge.net/ . I've not used it, but sounds like what you were looking for/at.

14:53:46 <phenny> crschmidt: I'll pass that on for you when danc is around.

16:30:17 Topic now Semantic Web Interest Group hack-n-chat - UTF-8 charset please - Weblog: http://swig.xmlhack.com/ - Logs: http://ilrt.org/discovery/chatlogs/swig/ - please identify with NickServ.

16:30:17 Users on #swig: logger MoonFog larsbot tlr aharth balbinus Emmy karlcow bjoern_ erim bskahan shellac chrisc d2m CaptSolo libby Pike_ rich_holygoat ear1grey_ chaals schepers KjetilK phenny kpreid LotR sh1m julie sbp md-afk pjenvey deltab mattmcc ericP jang dmiles_afk EspeonEefi kjetilkWork evlist Mutiny workbench dngor aniasis Jipp crschmidt dandutk DanC Xavier waglo lorebot Ismael bitsko danbri iwaim__ Jibbler jsled GabeW chaven dc_swig marc_ MarkB edd

16:30:17 Users on #swig: earle benja_ kasei

16:30:17 <NickServ> This nickname is owned by someone else

16:30:17 <NickServ> If this is your nickname, type /msg NickServ IDENTIFY <password>

16:30:18 <NickServ> Password accepted - you are now recognized

16:30:18 <MemoServ> You have no new memos

17:18:29 * libby waves at jhendler

17:18:58 <jhendler> hi libby

17:19:55 <libby> sorry we lost you and the others on saturday - I was helping robin build his robot: http://swordfish.rdfweb.org/photos/2005/05/14/2005-05-14-Pages/Image37.html

17:20:14 <libby> good trip back?

17:21:20 <jhendler> cool robot. trip from Japan is never "good" but it was better than usual (I had a decent seat) so can't complain

17:22:22 * tlr waves, too

17:22:47 <libby> heya tlr!

17:23:53 <libby> yeah, I had 2 seats - luxury :)

17:24:22 <tlr> Well, I at least had an F seat in that KLM 777 -- which means aisle and full leg space. Not luxury, though.

17:30:43 <jhendler> I got a bulkhead seat so managed to stretch my legs

17:31:57 <libby> comfy!

17:32:19 <libby> I was on vigin so you get video on demand (on linux!) but the box takes up legroom

17:32:38 <libby> virgin

17:32:43 <sh1m> sh1m is now known as sh1mmer

17:32:48 <libby> fortunately I've got short legs

17:34:03 * jhendler has to run, just doing my usual hit and run to see what was up - ciao

21:00:23 <ows> evening

21:26:06 <dajobe>http://lists.w3.org/Archives/Public/semantic-web/2005May/0027.html

21:26:07 <dc_swig> A: http://lists.w3.org/Archives/Public/semantic-web/2005May/0027.html from dajobe

21:26:12 <dajobe> A:|Raptor 1.4.6 with GRDDL

21:26:12 <dc_swig> Titled item A.

21:45:06 <crschmidt> ooh, a raptor release

21:56:27 <DanC> A:!

21:56:28 <dc_swig> Added comment A1.

21:56:30 <phenny> DanC: 14:55Z <crschmidt> tell DanC regarding http://www.ilrt.bris.ac.uk/discovery/chatlogs/swig/2005-05-17.html#T22-33-13 , there is now http://wick.sourceforge.net/ . I've not used it, but sounds like what you were looking for/at.

21:56:54 <DanC> I couldn't reach http://www.ilrt.bris.ac.uk/ to get the context

21:57:17 <dajobe> the site should be back now

21:57:41 <DanC> ah... auto-complete ... cool.

21:57:55 * DanC looks at A to see how much of grddl is supported...

22:03:29 * DanC has trouble reaching swig.xmlhack.com :( skips to lists.w3.org...

22:06:43 <dajobe>http://lists.w3.org/Archives/Public/semantic-web/2005May/0028.html

22:06:43 <dc_swig> B: http://lists.w3.org/Archives/Public/semantic-web/2005May/0028.html from dajobe

22:06:52 <dajobe> B:|Rasqal 0.9.9 with SPARQL 2005-04-19 syntax

22:06:53 <dc_swig> Titled item B.

22:06:54 <crschmidt> yay!

22:07:10 <crschmidt> oof, it's late for you

22:07:19 <dajobe> barely 23:07

22:07:25 <crschmidt> well, that's late for me!

22:07:26 <DanC> dajobe, you're a madman!

22:08:09 <DanC> and crschmidt yes, wick is just what I'm after. cool. where did you find it?

22:08:27 <crschmidt> DanC: someone from #swhack is the maintainer, apparently

22:08:35 <crschmidt> he dropped a link to us this morning

22:08:53 * crschmidt wonders if dajobe plans to build .debs before going to bed tonight

22:08:58 <dajobe> no

22:09:00 <crschmidt> k

22:11:02 <DanC> raptor downloads are on sourceforge... was that hard to set up, dajobe? what does it cost?

22:11:26 <crschmidt> yay, all rasqal tests pass

22:11:29 <dajobe> there was a small registration form and q&a mostly about the description and license

22:11:29 <crschmidt> now to try and make install

22:11:43 <dajobe> no cost, for the normal service.

22:11:55 <DanC> I meant cost in time, hassle, etc.

22:12:20 <dajobe> the upload web forms a rather tiresome, as I upload ~5-8 files per release and each needs 3 fields filling in

22:13:03 <DanC> sounds like a job for the ClientForm python stuff. great fun.

22:13:31 * DanC starts ./configure

22:13:43 <DanC> do I get GRDDL config by default?

22:13:43 <crschmidt> hm

22:13:57 <crschmidt> q = RDF.SPARQLQuery("SELECT ?a where { ?a <http://xmlns.com/foaf/0.1/nick> ?b. }") <- that gives me an "unexpected '?'"

22:14:05 <dajobe> A:apparently this is the 27th release of raptor

22:14:05 <dc_swig> Added comment A2.

22:14:10 <DanC> checking GRDDL parser requirements... no - libxml2 and libxslt are both not available

22:14:33 <dajobe> B:and 10th rasqal. that one's easier to work out

22:14:34 <dc_swig> Added comment B1.

22:14:42 * DanC installs libxslt1-dev

22:14:55 <crschmidt> oops

22:15:00 <crschmidt> looks like i'm still running on the old rasqal

22:15:01 <dajobe> yeah there should be a line above about not finding something

22:15:09 <dajobe> crschmidt: let's not start that game :)

22:15:12 <crschmidt> yep

22:15:18 <crschmidt> stupidity on my part

22:15:32 <dajobe> and there have been 20 releases of redland

22:15:53 * crschmidt needs to learn configure

22:16:08 <DanC> RDF parsers available : rdfxml ntriples turtle rss-tag-soup grddl

22:16:30 <DanC> what's to learn? you type ./configure and that's it. Well, you cross your fingers and pray.

22:16:42 <crschmidt> no, i need to learn the options

22:16:48 <crschmidt> since i need to install over my old rasqal

22:16:48 <sbp> dajobe: logger gives me "Unknown option: userhosts"

22:16:55 <crschmidt> and it doesn't do that by default

22:17:01 <sbp> but "-userhosts Record user@host in /join messages" is clearly in the --help

22:17:05 <sbp> any hints before I trawl the source?

22:17:40 <dajobe> dunno, check the source

22:17:46 <sbp> latest CVS, same with both -userhosts and --userhosts

22:17:49 <sbp> m'kay

22:18:00 <dajobe> it's probably not been enabled for years

22:18:20 * DanC finishes make; starts make check ...

22:18:30 <DanC> one failure...

22:18:35 <sbp> ah: 'userhost!' => \$::LogUserHosts,

22:18:37 <crschmidt> woot

22:18:38 <sbp> no plural, it seems

22:18:44 <DanC> Testing rdf/xml (may fail due to XML parser bugs)

22:18:44 <DanC> Checking ex-55.rdf FAILED

22:18:44 <DanC> 2c2

22:18:44 <DanC> < <http://example.org/resource> <http://www.w3.org/1999/02/22-rdf-syntax-ns#value> "def" .

22:18:44 <DanC> ---

22:18:44 <DanC> > <http://example.org/resource> <http://www.w3.org/1999/02/22-rdf-syntax-ns#value> "defdef" .

22:18:51 <dajobe> see the title

22:18:57 <dajobe> it works with expat. it's an libxml bug

22:19:13 * DanC sweeps that test back under the rug, carries on...

22:20:27 <DanC> All 13 tests passed

22:20:33 <DanC> hmm... there were lost more than 13 tests

22:20:41 <dajobe> that's the top level dir tests

22:20:57 <DanC> hmm... rdfdiff; I don't recall seeing that

22:20:58 <dajobe> there's some super-dooper auto test thing somewhere but it scared me

22:21:14 <dajobe> it's an rdf graph compare thing, somewhat

22:21:19 <dajobe> I don't entirely trust it

22:21:23 <DanC> ok

22:21:27 <dajobe> =I've seen it fail

22:21:33 * DanC tries to figure out how to play with grddl...

22:21:44 <dajobe> rapper -i grddl <url>

22:22:11 * dajobe beer get

22:22:24 <DanC> lt-rapper: Parsing returned 60 statements

22:22:25 <DanC> from my homepage

22:22:51 <crschmidt> hm

22:25:08 <crschmidt> bleh

22:25:22 * crschmidt apparently doesn't know SPARQL syntax anymore :)

22:28:02 <crschmidt> ^sparqlquery select ?a where { ?a todo:text ?t . }

22:28:07 <crschmidt> oops, bot's not here

22:28:38 <DanC> hm... <http://www.w3.org/2003/g/po-ex> <http://www.w3.org/2003/g/data-view#namespaceTransformation> <http://www.w3.org/2003/g/grokPO.xsl> .

22:28:38 <DanC> lt-rapper: Parsing returned 1 statements

22:28:47 <DanC> that's what I get from

22:28:49 <DanC> ./rapper -i grddl 'http://www.w3.org/2003/g/po-ex'

22:28:57 <crschmidt> eep

22:28:58 <crschmidt> segfault

22:29:20 <dajobe> DanC: hmm... I might have missed that one

22:29:55 <dajobe> I didn't try all the tests for sure.

22:29:59 <DanC> yeah, there's a recursive aspect to GRDDL that I was wondering if you had implemented

22:30:13 <dajobe> I think it goes 1 step

22:30:53 <dajobe> but I don't think I did that particular style either

22:31:01 <DanC> ok, that's cool, but I suggest you add to your release notes that your GRDDL client is different from some of the others

22:33:05 * DanC wonders if rapper will try to figure out what the input is automatically if it's not told...

22:33:16 <dajobe> with -g it will guess

22:33:42 <DanC> hmm...

22:33:43 <DanC> $ ./rapper -g http://www.w3.org/People/Connolly/

22:33:43 <DanC> lt-rapper: Guessed parser name 'rdfxml'

22:34:16 <dajobe> not much of a guess...

22:34:38 <dajobe> one day I'll give it mime types and some initial data

22:34:39 <DanC> what heuristic did I use in glean.py, I wonder...

22:35:08 <crschmidt> bleh

22:35:14 <crschmidt> can't figure out how to reproduce segfault

22:35:20 <crschmidt> ooh

22:35:21 <crschmidt> there it goes

22:35:35 <dajobe> you probably should rebuild redland and python bindings

22:35:39 <crschmidt> yep

22:35:44 <crschmidt> just realized that

22:36:31 <crschmidt> hm

22:36:35 <crschmidt> i never built them in the first place :)

22:36:36 <DanC> if meta["content-type"].startswith(RDF_mediaType):

22:36:36 <DanC> return parseRDF(tmp, addr)

22:36:36 <DanC> else:

22:36:36 <DanC> return grddl(addr, tmp, already

22:37:13 <ows> hi there... where can I find something to search folks hotmails that are researching in AI (namely semantic web) using foaf technologies or not.. crschmidt you said something about that using RDQL can you remeber how to do that

22:37:15 <dajobe> yeah; it's the way I did the guessing, I don't have access to it yet. I need to make a meta-parser which then starts the fetch and 'execs' the right parser

22:39:00 <DanC> yeah... the way glean.py works, it's recursive, and some levels use the XSLT stuff and some levels just read RDF/XML

22:41:27 <DanC> man... WICK is brand new... released today. I'm gonna wish into this channel more often!

22:41:39 <crschmidt> heh, heh

22:41:45 * crschmidt rebuilds redland

22:42:24 <ows> crschmidt: hi fellow :) did you read what I said? :P

22:42:26 * DanC got invited to play volleyball... in about 20 minutes... if I leave now, I'd be on time...

22:42:36 <crschmidt> Yep

22:43:20 <ows> and can you remind me what kind of query I should do?

22:43:36 <DanC> is there a query service set up with the data, ows?

22:44:03 <DanC> hotmails... mail messages of hotmail users?

22:46:45 <crschmidt> ^sparqlquery select ?p where {?p foaf:nick "DanC". }

22:46:45 <julie> (r1114689381r261), http://www.w3.org/People/Connolly/#me

22:47:25 <DanC> cooool. wicked coool.

22:48:05 <crschmidt> ^sparqlquery select ?p where {?p foaf:nick "DanC"; a foaf:Person. }

22:48:05 <julie> Query returned no results

22:48:21 <libby> danc, go play volleyball, the internet'll still be here later ;)

22:48:25 <DanC> ^sparqlquery select ?pg where { [ foaf:nick "DanC"] is cyc:socialParticipants of [ contact:homePage ?pg] }

22:48:25 <julie> Bad query (SPARQL syntax error at 'i').

22:48:38 <dajobe> that's not in sparql - is, of

22:48:38 <DanC> oh frap. no is/of in SPARQL

22:49:05 <DanC> I suppose with named bnodes, it's not required.

22:49:40 <DanC> ^sparqlquery select ?pg where { [ contact:homePage ?pg] cyc:socialParticipants [ foaf:nick "DanC"] }

22:49:45 <DanC> does turtle have is/of ?

22:49:47 <julie> http://www.xtech-conference.org/, http://www.w3.org/2004/12/allgroupoverview.html, http://www.w3.org/2001/tag/ (x2), http://www.w3.org/2004/12/rules-ws/

22:49:48 <dajobe> no

22:49:54 <dajobe> phew, something else worked

22:50:30 * DanC gets vball at T-10 notice from gizmo; heads out... hasta!

22:52:31 <ows> DanC: no, hotmails of AI researchers (namely Semantic web)

22:56:00 <ows> DanC: main objective is to try to share ideas with AI researchers in real time using instant messaging (instead of mail)