Semantic Web Interest Group IRC Chat Logs for 2002-05-14

01:08:08 <deltab> deltab is now known as or

01:08:28 <or> or is now known as deltab

03:36:03 <gkayotic> Is someone alive here who can answer a question on formatting RSS with XSLT?

03:36:34 <deltab> if it's a very simple question

03:37:11 <gkayotic> ok, I'm using XSLT to format some RSS files, but they are different versions (1.0, 0.9, 0.91) and have different namespaces.

03:37:32 <gkayotic> Do you know if there is some way of making my templates work with all three namespaces?

03:38:32 <deltab> I don't, sorry

03:39:03 <gkayotic> that's ok

03:39:07 <deltab> but I think there are already some RSS XSLT files out there

03:39:11 <gkayotic> it's kinda a weird thing

05:01:52 <mnot> earthquake

05:02:21 <mnot> holy shit that scared me

05:03:12 <burtonator> yeah... just felt it!

05:03:24 <burtonator> that was at least 5.5

05:03:32 <mnot> checking the quake sites now

05:04:04 <mnot> that's the first time I've felt the building move back and forth *several* times...

05:05:27 * mnot thinks about moving into a more modern building...

05:10:23 <epsas> aloha peoples of the earth

07:44:54 <ChanServ> [#rdfig] This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome

08:08:05 <lilo> [GlobalNotice] Hi all. Just a reminder that user mode +w displays WALLOPS, containing contain non-critical comments and announcements and detailed server admin information. On most clients, "/mode <yournick> +w" will turn them on. We just experienced an intermittent fault with one of the new main hubs. The other hub picked up the load. Thanks for your patience.

10:09:27 <JibberJim> Dajobe? How does your list of people match the photos I can't seem to make sense of it? and don't know the people enough - I was going to RDF and SVG it!

10:51:24 * dajobe back

10:51:31 <dajobe> left to right

10:57:12 <JibberJim> Ah! I was thinking you normally has left-right back, then left-right front or something.

10:57:37 <JibberJim> Anyone know of any RDF which says someone memberOf RDFCore ?

10:58:00 <dajobe> some of DanC's w3c structure scraping stuff must do that

11:07:32 <dajobe> JibberJim: see http://www.w3.org/2001/sw/WebOnt/webont-mem-pub.rdf ?

11:41:40 <JibberJim> I really need the inverse of that dajobe, a property of the person with foaf:mbox rather than a property of the organisation.

11:43:34 <dajobe> incidently, which pic did you choose?

11:43:39 <dajobe> there are 3 images, at 2 scales

11:44:09 <JibberJim> I picked 2, at the middle scale, as it better fits with the other images I've got in the DB, and is a reasonable size in KB aswell.

11:44:30 <dajobe> ah, I thought #1 had a better view of Ralph (far right) in particular

11:46:36 <dajobe> is this SVG thing likely to work in mozilla + adobe plugin?

11:46:40 <dajobe> on linux that is

11:47:28 * edd laments that even x86 SVG Linux plugin is dead

11:47:41 <JibberJim> It'll work with the adobe plugin and Batik 1.5b2, the with Mozilla as long as it's not win32 Mozilla.

11:48:29 * edd tests out his www2002 coffee mug. full marks for keeping coffee warm. unsure about its other attributes

11:48:53 <JibberJim> Do coffee cups have other attributes?

11:49:15 <dajobe> hmm, yeah adobe svg for linux is gone, not in download area

11:51:29 <JibberJim> - http://www.adobe.com/svg/viewer/install/old.html still has it ?

11:51:53 <JibberJim> KSVG is getting better if you're a KDE kind of person...

11:52:20 <dajobe> I have the old one

11:52:37 <dajobe> don't see ksvg on http://www.w3.org/Graphics/SVG/SVG-Implementations

11:52:40 <dajobe> anyway

11:52:43 * dajobe -> lunch

11:53:24 <JibberJim> The "old" one is also the newest one, they just never officially released the Linux build.

11:55:29 <edd> but Mozilla's plugin API is now incompatible with that SVG plugin

11:55:31 <edd> so it won't work.

11:59:16 * ramin is away: I'm busy

12:09:04 <JibberJim> F:[codepiction RDF with paths|http://jibbering.com/rdfweb/1021377838531.rdf] and [simple SVG|http://jibbering.com/rdf/codepiction-clientside.svg]

12:09:04 <dc_rdfig> Label F not found.

12:09:10 <JibberJim> E:[codepiction RDF with paths|http://jibbering.com/rdfweb/1021377838531.rdf] and [simple SVG|http://jibbering.com/rdf/codepiction-clientside.svg]

12:09:10 <dc_rdfig> added comment E5

12:46:26 <JibberJim> E:[People highlighted, all generated from the RDF|http://jibbering.com/rdf/highlight.svg]

12:46:26 <dc_rdfig> added comment E6

12:56:15 <JibberJim> Oopss, I had Aaron, and Dave Beckett swapped around!

13:06:06 <sandro__> they're daml:equivalentTo each other anyway, aren't they? :-)

13:06:28 <sandro__> sandro__ is now known as sandro

13:09:30 <rreck> anyone interested in talking about the project gutenberg ontology with me ?

13:12:56 <JibberJim> - DanC's Dog: http://jibbering.com/rdfsvg/search.1?noun=Dog

13:19:06 * edd tries libby's demo, but it won't run as swordfish serves SVG as image/svg-xml rather than image/svg+xml

13:25:47 <JibberJim> better than lists.w3.org which serves it as image/svg

13:26:46 <edd> JibberJim: have you complained to w3 about that?

13:27:01 <JibberJim> Yeah

13:27:10 * edd won't duplicate, then

13:34:37 <dajobe> JibberJim: good job!

13:35:33 <JibberJim> All generated from http://jibbering.com/svg/codepiction.html

13:40:10 <dajobe> E:good job!

13:40:10 <dc_rdfig> added comment E7

13:40:44 <dajobe> E:generated from Jim's [http://jibbering.com/svg/codepiction.html|SVG paths to co-depiction RDF pages]

13:40:44 <dc_rdfig> added comment E8

13:41:09 <dajobe> hmm, maybe you chopped off a bit of libby's hair :)

13:41:53 <JibberJim> Yeah, I chopped off a bit of a view people...there's no delete, you're commited once you've clicked.

13:41:59 <JibberJim> s/view/few

13:44:08 * edd notes that x86 SVG viewer works for standalone SVG, just not embedded.

13:44:43 * edd enjoys jibberjim's foaf screensaver games

13:44:59 * JibberJim needs a picture of edd for them...

13:45:16 <edd> i need to update my aged foaf file

13:45:35 <edd> i imagine i'm depicted in one or two pics from www2002

13:46:25 <dajobe> libby has some pics too

13:46:51 * edd lol at dajobe's disembodied head bouncing around his screen

13:47:01 <dajobe> lol

13:53:54 <JibberJim> - http://jibbering.com/rdf/highlight-2.svg?url=http://jibbering.com/rdfweb/combined.rdf - another screensaver

14:00:16 <dajobe> JibberJim: scary

14:00:53 <JibberJim> I didn't know I was that scary, that's one of the better photos of me :-(

14:01:04 <dajobe> you? I didn't see that

14:57:35 <JibberJim>http://www.bbc.co.uk/radio4/news/inbusiness.shtml

14:57:35 <dc_rdfig> A: http://www.bbc.co.uk/radio4/news/inbusiness.shtml from JibberJim

14:57:47 <JibberJim> A|Google on the Radio

14:58:09 <JibberJim> A:|Google on the Radio

14:58:09 <dc_rdfig> titled item A

14:58:43 <JibberJim> A: BBC radio news program focusing on how google might make money from iself.

14:58:43 <dc_rdfig> added comment A1

14:59:16 <JibberJim> A:[Real Audio version|http://www.bbc.co.uk/radio4/news/audio/google_business.ram]

14:59:16 <dc_rdfig> added comment A2

16:11:27 <edd> eikeon: hi :)

16:45:47 <larsbot> if I want to say: "web page A is written in language B", what's the common way to do it in RDF?

16:46:16 <edd> you mean link xml:lang ?

16:46:25 * edd prods dajobe

16:46:34 <edd> ah, he's not here.

16:46:37 <larsbot> well, most of all I'm asking, *how* it would be done

16:46:58 <jang> timely

16:47:03 * edd suspects dajobe knows

16:47:03 <larsbot> I presume using an RDF triple where A is the subject, but...?

16:47:12 <dajobe> what?

16:47:36 <larsbot> see my question about page A and language B

16:47:42 <deltab> hmm, does xml:lang have a URI?

16:47:46 <dajobe> human language?

16:47:49 <larsbot> yes

16:47:53 <dajobe> xml:lang - not really

16:48:01 <JibberJim> Isn't there a dc:language?

16:48:04 <dajobe> yes

16:48:08 * edd makes a noise that sounds like 'doh'

16:48:23 <larsbot> that settles the predicate, but what about the object?

16:48:34 <dajobe> so the statement is <web page URI> dc:language "en". or something

16:48:44 <larsbot> you'd use a literal?

16:48:57 <dajobe> yes, that is the answer

16:49:21 <dajobe> see the RDF primer section 6.1 http://www.w3.org/TR/rdf-primer/

16:49:25 <larsbot> how reliable is that? what if I want to use Ethnologue codes?

16:49:36 <dajobe> never heard of them

16:49:39 <dajobe> do they have URIs?

16:49:44 <larsbot> nope

16:49:49 <larsbot> nor do ISO 639/RFC 3066 ones

16:49:49 <deltab> then define a term that refers to them

16:49:51 <AaronSw-CA-US> iso8996874673:en

16:50:06 <dajobe> if you can make a uri, then use that as the value

16:50:13 <JibberJim> 3166 ?

16:50:17 <larsbot> yeah, but won't everyone then use *different* URIs?

16:50:20 <larsbot> that's for languages

16:50:20 <dajobe> if not, you'll need to introduce a blank node, and use rdf:value, rdf:type

16:50:26 <larsbot> grrr, I meant countries

16:50:50 <dajobe> isn't topic maps published subjects make URIs for these things?

16:51:01 <jang> larsbot: using different URIs to represent languages...

16:51:11 <larsbot> different URIs to represent the *same* language, yes

16:51:20 <larsbot> dajobe, that's what I am getting at

16:51:28 <jang> that's my cue to do my bit for european unity and suggest that maybe if we all adopted the queen's english, the problem would go away

16:51:29 <larsbot> I'm trying to find out if RDF people are interested in URIs for languages at all

16:51:32 <eikeon> Hi edd

16:51:34 <dajobe> sure

16:51:38 <deltab> RFC 3066 obsoletes RFC 1766, which was the recommended best practice for Dublin Core's Language

16:51:47 <dajobe> I looked at the TM PS page this morning, didn't see any published docs

16:51:58 <jang> which, unfortuantely, is an attitude common amongst some of the library/ontology-writing community

16:52:00 <dajobe> deltab: yes but DC hasn't updated its reference in the RFC

16:52:08 <dajobe> that's too terse

16:52:12 <deltab> I know

16:52:17 <larsbot> dajobe, there's a general recommendation doc out, but it's very rough so far

16:52:17 <dajobe> deltab: yes but the DCMI hasn't updated its reference in the Dublin Core element set RFC

16:52:29 <larsbot> there's a TC working on PS sets for languages

16:52:38 <dajobe> I know, but as I said, I saw nothing yet

16:52:42 <deltab> 3166 is the number of the International Standard for country codes

16:52:45 <larsbot> yeah

16:53:01 <larsbot> here it is: http://www.oasis-open.org/committees/tm-pubsubj/docs/recommendations/psdoc.htm

16:53:04 <dajobe> I asked a TM PS member at www2002 about it, he said they meet after xml europe

16:53:06 <larsbot> note that it sucks at the moment

16:53:17 <larsbot> that's right, we're meeting May 24

16:53:25 <dajobe> was that the thing that ... yes ... "too controversial parts of the previous version have been removed from this version." - lol

16:53:36 <dajobe>http://www.oasis-open.org/committees/tm-pubsubj/docs/recommendations/psdoc.htm

16:53:37 <dc_rdfig> B: http://www.oasis-open.org/committees/tm-pubsubj/docs/recommendations/psdoc.htm from dajobe

16:54:13 <dajobe> B:|OASIS Topic Maps Published Subjects TC - Recommendations for Documentation of Published Subjects

16:54:13 <dc_rdfig> titled item B

16:54:56 <dajobe> B:<larsbot> note that it sucks at the moment

16:55:03 <dc_rdfig> added comment B1

16:55:36 <larsbot> B1:<larsbot> note that it is a *very* early draft

16:55:36 <dc_rdfig> replaced comment B1

16:55:52 <larsbot> I'm meeting these people Thursday next week :-§

16:56:06 <dajobe> well, I'll be at xml europe too

16:56:13 <larsbot> that's very good

16:56:18 <larsbot> this TC: http://www.oasis-open.org/committees/geolang/index.shtml

16:56:21 <larsbot> is meeting on Thursday as well

16:56:58 <larsbot> we're currently working on published subject sets for languages and countries

16:57:15 <dajobe> hmm are Bernard V and Steve P. on all TM lists :)

16:57:25 <larsbot> pretty much :)

16:57:37 <larsbot> well, Bernard is not in ISO...

16:57:52 <larsbot> I would really like someone from RDF land to participate in that TC

16:57:57 <larsbot> the question is: is there any interest?

16:58:20 <dajobe> dunno

16:58:27 <larsbot> I'd be happy for us to just do this and for the RDF folks to use it, but I'd hate for us to do it and for the RDF folks to *then* say they hate it for some reason

16:58:31 <dajobe> maybe posting some of this to www-rdf-interest ?

16:58:45 <dajobe> i.e. ask that Q there, rather than on irc

16:58:53 <larsbot> I could do that

16:59:11 <larsbot> what's your non-IRC identity, BTW?

16:59:15 <dajobe> but I'm happy for TC PS, geolang etc to do it

16:59:21 <dajobe> try /whois dajobe

16:59:41 <larsbot> ah, got it; pleased to mee you :)

16:59:46 <larsbot> s/mee/meet

17:01:31 <dajobe> I think the W3C is trying to get the IETF to make URIs for some lists of things

17:01:48 <dajobe> such as things registered under IANA - content types

17:01:58 <dajobe> well, it is discussed on www-tag anyway

17:02:26 <larsbot> that's interesting, because it's pretty much the same thing

17:02:38 <larsbot> I'll have a look in the www-tag archives

17:04:01 <larsbot> thanks for the recommendations, anyway

17:04:04 <larsbot> I'll go off and try that

17:04:35 <dajobe> hmm

17:04:41 <dajobe>http://www.topicmaps.org/xtm/1.0/#desc-psis

17:04:41 <dc_rdfig> C: http://www.topicmaps.org/xtm/1.0/#desc-psis from dajobe

17:04:47 <dajobe> C:|Topic Maps - Published Subjects

17:04:48 <dc_rdfig> titled item C

17:04:56 <dajobe> C:"A published subject is any subject for which a subject indicator has been made available for public use and is accessible online via a URI. A published subject indicator is therefore any resource that has been published in order to provide a positive, unambiguous indication of the identity of a subject for the purpose of facilitating topic map interchange and mergeability."

17:04:56 <dc_rdfig> added comment C1

17:05:01 <dajobe> C:*online* ?

17:05:01 <dc_rdfig> added comment C2

17:07:19 <larsbot> C:That definition is being [replaced|http://www.y12.doe.gov/sgml/sc34/document/0299.htm#published-subject]

17:07:20 <dc_rdfig> added comment C3

17:07:54 <larsbot> so the term "online" is no longer used

17:11:09 <dajobe> C:in that case the [http://www.oasis-open.org/committees/tm-pubsubj/index.shtml|TC page] needs updating to show that

17:11:09 <dc_rdfig> added comment C4

17:12:20 <larsbot> C:true; N0299 is very recent

17:12:21 <dc_rdfig> added comment C5

17:14:21 <JibberJim> Nick search with codepiction data e.g. http://jibbering.com/rdf/nick-search.1?person=dajobe

17:14:34 <dajobe> search of what?

17:15:05 <JibberJim> The same Codepiction data, you enter the nick, it returns picture, name and email address.

17:15:20 <dajobe> not everyone in the pic has a nick

17:15:22 <dajobe> iyswim

17:15:39 <JibberJim> I know! they don't have a foaf:nick property...

18:02:17 <bijan> Anyone know if there's anonymous node support in Redfoot that I'm just missing?

18:03:55 <edd> bijan: eikeon would, but it seems he's away from his keyboard right now.

18:04:14 <bijan> k.

18:04:22 <bijan> I'll just leave that thought hanging in the logs....

18:23:05 <eikeon> We have some support for anonymous resources... should be better when we implement latest working draft soon.

18:24:37 <eikeon> Is that what you are meaning by anonymous node support?

19:52:14 <bijan> Pretty much.

20:22:01 <rreck> what is a good ontology editor ?

20:24:57 <dajobe> how about protege?

20:29:01 <rreck> it seem like the only alternative, the dl is very slow

20:32:36 <tav`> tav` is now known as tav

20:39:48 <dajobe> hey Guha

20:39:59 <Guha> hi

20:40:29 <edd> hey guha

20:40:44 <dajobe> I wonder, if you are the guha in the pic I took at www2002, linked at http://rdfig.xmlhack.com/2002/05/13/2002-05-13.html#1021324548.262422

20:40:45 <Guha> wow, not everyone is still in Maui ;-)

20:41:12 <Guha> yup, the same one.

20:42:10 * edd regrets he never got chance to talk to Guha in hawaii

20:42:31 <Guha> I didn't know you were there ...

20:43:10 * DanCon catches up, waves

20:43:30 <Guha> hi dan, any luck with getting your stuff back?

20:43:55 <DanCon> well, I had a dream where I found it in my hotel room.

20:44:29 <rreck> i dont know if that qualifies as "luck"

20:44:58 <dajobe> I'm still amazed I made my connection in NY; I got off one plane listening to the boardcall for plane #2

20:45:25 <sat> anyone aware of tool to validate RDF doc against a RDF schema?

20:45:55 <dajobe> sat: try the sesame stuff, rdf suite from ics-forth

20:46:19 <sat> ok...thank you

20:46:55 <dajobe> BLURB:Tools to validate an RDF doc against an RDF schema

20:46:56 <dc_rdfig> D: Tools to validate an RDF doc against an RDF schema from dajobe

20:47:02 <dajobe> D:[http://sesame.aidministrator.nl/|sesame]

20:47:02 <dc_rdfig> added comment D1

20:47:25 <DanCon> say, dajobe, can you hack your gallery installation? ala... "if you're in this picture, enter your email address here: ___"

20:47:26 <dajobe> D:[http://www.ics.forth.gr/proj/isst/RDF/|RDF Suite] from ICS-FORTH

20:47:27 <dc_rdfig> added comment D2

20:47:53 <DanCon> sat, timbl wrote an RDF schema validator thingy... http://www.w3.org/2000/10/swap/util/validate.n3

20:48:04 <edd> is it chumped?

20:48:07 <dajobe> DanCon: that would be tricky I think

20:48:07 <DanCon> it pretty much works. run ala: cwm validate.n3 -think -purge --with myfile.n3

20:48:20 <dajobe> the gallery templates are installation wide IIRC

20:49:03 <DanCon> hmm... so the foaf convention is that each person keeps a "foaf file" with pointers to the pictures they're in?

20:49:16 * edd decides it's probably not chumped then...

20:49:44 <DanCon> timbl has an action item to announce validate.n3 to www-rdf-interest, I think.

20:49:57 <edd> cool.

20:50:31 <sat> thanks for the link...although, I am looking for java tool

20:51:12 <DanCon> well, in theory, cwm should run under jython. I'm interested to see that theory tested.

20:51:38 <dajobe> D:more can probably be found at the [http://www.daml.org/|DAML] site

20:51:38 <dc_rdfig> added comment D3

20:52:18 * DanCon starts to think that hacking RDF support into gallery might have a pretty high return on investment.

20:52:40 <sat> what are others doing to validate that the RDF file is conformant to the schema?

20:52:46 <dajobe> the ".dat" files that gallery makes are plain text

20:53:11 <DanCon> gallery doesn't keep its stuff in an RDB? I thought it was a php thing.

20:53:51 <edd> you can choose backend, iirc. puts mine in DB files.

20:53:52 <dajobe> it is php, but plain text for image metadata

20:53:57 <dajobe> oh

20:54:04 <DanCon> hmm... "We are seriously considering not supporting a flat-file database solution (like what is currently used in Gallery 1.x) in Gallery 2.0

20:54:10 <DanCon> " -- http://www.menalto.com/projects/gallery/modules.php?op=modload&name=News&file=index

20:54:37 <dajobe> sat: validation of rdfs is something that can be interpreted many ways since it isn't defined in the RDF specs

20:55:31 <dajobe> sat: the daml+oil/owl tools have/will have a much more precise defn

20:56:00 <sat> recentely we made a typo to say DC:tite and that kinda passed off as a valid RDF file

20:56:07 <sat> although DC:tite is undefined

20:56:15 <sat> in the DublinCore

20:56:26 <DanCon> typo: that's exactly the sort of thing timbl's tool catches.

20:57:06 <JibberJim> DanCon - I'd be very interested in working on Photo/Image RDF projects...

20:59:03 <JibberJim> I've been looking at ways to describe spatial descriptions of images (SVG and photos) in RDF for more detailed describing for voice browsers.

21:03:43 * DanCon is looking for WWW2002 photos to show his family, mostly.

21:03:56 <JibberJim> And a chopped out dog?

21:03:58 <dajobe> sat: rdfsuite is java BTW

21:04:50 <dajobe> I forgot to mention mine: http://photos.dajobe.org/gallery/www2002-hawaii

21:05:21 <dajobe> libby has some more

21:05:49 <JibberJim> Yeah it looks like you had a really hard working conference...

21:06:20 <dajobe> I took out all the pics where it was raining :)

21:06:42 <JibberJim> That must've taken hours...

21:08:08 * JibberJim has to bring back Libby's hair, she looks silly...

21:08:14 <dajobe> pls

21:10:25 <rreck> looks like fun

21:21:48 * DanCon hunts for TAG photos...

21:23:54 <DeSharq> tsk poor hound

21:42:58 <connolly> connolly is now known as DanC_

23:30:45 <AaronSw-CA-US> AaronSw-CA-US is now known as AaronSw

23:34:28 <AaronSw> Hi rael.

23:35:05 <dajobe> aren't you both at etcon?

23:37:10 <raelity> indeed. aren't you? ;-)

23:37:26 <dajobe> i wish

23:37:59 <dajobe> all I can do is get my raelity in bytes

23:38:19 <raelity> hardy har har

23:38:39 <dajobe> you junked MT for bloxxom now?

23:39:09 <raelity> not junked. just wanted to do something a little different. love mt. love blosxom more... for me... having fun experimenting

23:41:51 <dajobe> were you also sleeping in that apple server launch?

23:43:06 * dajobe wanders over to http://www.actuallyworks.com/panopticon/