Semantic Web Interest Group IRC Chat Logs for 2004-01-06

00:02:18 <CaptSolo>http://kaste.lv/~captsolo/info/blog_all.php/2004/01/06/p336

00:02:19 <dc_rdfig> A: http://kaste.lv/~captsolo/info/blog_all.php/2004/01/06/p336 from CaptSolo

00:02:40 <CaptSolo> A:|TO-DO :: Resume RDF Schema

00:02:40 <dc_rdfig> Titled item A.

00:03:24 <CaptSolo> A:Wishlist of improvement to be made to the Resume RDF Schema

00:03:25 <dc_rdfig> Added comment A1.

00:03:59 <CaptSolo> A:hopefully that will transform into a to-do list for me :)

00:03:59 <dc_rdfig> Added comment A2.

00:04:34 <CaptSolo> A:Please comment (in the blog entry comments) if you have ideas to add to it

00:04:35 <dc_rdfig> Added comment A3.

00:10:11 <libby> libby is now known as danbri_tmp

00:10:27 <danbri_tmp> danbri_tmp is now known as libby

00:23:50 <_deusx> _deusx is now known as deusx

01:05:15 <eaon|zZz> eaon|zZz is now known as eaon|ill

01:36:10 <edd>http://planet.rdfhack.com/

01:36:10 <dc_rdfig> B: http://planet.rdfhack.com/ from edd

01:36:14 <edd> B:|Planet RDF

01:36:14 <dc_rdfig> Titled item B.

01:36:22 <edd> B:Aggregated weblogs from semantic web hackers

01:36:22 <dc_rdfig> Added comment B1.

01:36:41 <edd> B:Think you should be on here but aren't? Ensure you have an RSS 1.0 feed and bug dajobe about it :)

01:36:42 <dc_rdfig> Added comment B2.

01:38:43 <danbri_dna> B:Very nice. Is there a way of indicating one's suitability in RDF/XML FOAF etc?

01:38:44 <dc_rdfig> Added comment B3.

01:38:56 <mattb> interesting

01:39:03 <mattb> a way to mark one's weblog in a foaf file?

01:39:13 <mattb> a job for foaf topics and suchlike perhaps

01:39:16 <danbri_dna> foaf:weblog does that

01:39:27 <danbri_dna> and foaf:topic of sw homepage or somesuch?

01:39:37 <mattb> i'd like it to be a use case for simple pubsub scuttering

01:39:47 <mattb> it should be able to gather thumbnails for people, etc

01:39:47 <edd> the list is editorially controlled and proplly ought to be kept that way tho

01:40:01 <mattb> yeah, but a scutter subscriber could suggest potential items to the editors

01:40:07 <mattb> it doesn't all have to be automatic

01:40:07 <edd> indeed

01:40:09 <mattb> discovery is the thing

01:40:13 <edd> you just want to say "I'm in this group"

01:41:33 <danbri_dna> B:How about adding the RDFIG Scratchpad RSS feed?

01:41:33 <dc_rdfig> Added comment B4.

01:42:02 <mattb> we've kept away from linklogs on the chumpologica

01:42:04 <mattmcc> You could add another subsection for 'People who say they blog about SW'

01:42:20 <mattb> editorial decision, of course

01:43:07 <mattb> rdfig is a human-created aggregator, in a sense

01:43:17 <danbri_dna> yes, its an odd mixing of levels

01:43:17 <mattb> cf captsolo noting an entry of his there today

01:43:19 <CaptSolo> planet is very nice :)

01:43:25 <mattb> whereas *every* captsolo post will be on planet

01:43:31 <mattb> ;)

01:43:38 <edd> dan dan daaaaah!

01:43:43 <mattb> the nice thing is that you get a mix of stuff, too

01:43:51 <CaptSolo> who mentions me again?

01:43:52 <edd> yes, as time will show

01:43:52 <mattb> most interesting writers don't just concentrate on a single topic

01:44:01 <mattb> eg norm's blossoms

01:44:11 <mattb> readers can branch out

01:44:23 <edd> oh dear, waxing horticultural again

01:44:31 <danbri_dna> I'm gonna put a blog at danbri.org I think.

01:44:39 <danbri_dna> it'll be about whatever the hell i like

01:44:45 <danbri_dna> including SW

01:44:47 <CaptSolo> heh ;>

01:44:50 <mattb> did i hear something about blogging every day?

01:44:56 * mattb counts the days in 2004

01:45:02 <edd> it's a good discipline

01:45:10 <mattb> if i blog every month i'll be surprised

01:45:10 <danbri_dna> yeah, but I made up a rule: I can store up blogs and write one long one ;)

01:45:18 <CaptSolo> mattb: is that supposed to be good that every my blog entry appears on planet?

01:45:19 <mattb> ah, blogging on layaway

01:45:21 <CaptSolo> ;>>>

01:45:29 <CaptSolo> i don't mind of course

01:45:42 <mattb> well, if it really didn't suit you, you could always produce a split rss feed

01:45:50 <mattb> some blogs have rss-per-topic for example

01:45:53 <CaptSolo> yep...

01:46:08 <mattb> it's *your* personal publishing choice

01:46:13 <mattb> isn't the web fun?

01:46:49 <CaptSolo> bet - IF somebody wanted to aggregate only those blog entries which ARE about SW

01:46:54 <CaptSolo> then -

01:47:08 <CaptSolo> either 1) make a separate rss for each category

01:47:37 <CaptSolo> or 2) add category (what ontology can we find for it? ;) to every feed entry

01:48:03 <CaptSolo> actually, my RSS has a <dc:subject> for each entry, but it's in plaintext

01:48:07 <mattmcc> Wordnet? And/Or a homepage of the topic where appropriate.

01:48:13 <danbri_dna> I think foaf:topic of http://www.w3.org/2001/sw/ is a nice way to say "this is about the semantic web"

01:48:45 <CaptSolo> subject could be 'per category' configurable uri

01:49:11 <CaptSolo> danbri: foaf:topic better than dc:subject ?

01:49:15 * mattmcc thinks it'd be great if blogs could be trusted to have a machine-readable topic for each entry. You could scutter on interest.

01:49:29 * edd needs rest

01:49:30 <CaptSolo> matt: that would be very valuable indeed

01:49:52 <CaptSolo> matt: a side note - blog entries may have more than one subject

01:50:03 <mattmcc> Sure.

01:50:07 <mattb> tools like movabletype have basic capability for this

01:50:08 <mattb> controlled lists

01:50:22 <CaptSolo> but then, again, we can add a couple of <dc:subject> or <foaf:topic> , right?

01:50:38 <CaptSolo> b2evolution has multicategories as well

01:51:02 <mattb> i have a basic rdf representation of each of my MT entries with http://hackdiary.com/categories/ prepended to the MT category name

01:51:05 <teefal> love the tagline for planet rdf

01:51:13 <teefal> it's inspired

01:51:20 <mattb> dajobe's idea, iirc

01:51:36 <mattb> (eg http://www.hackdiary.com/archives/000040rdf)

01:51:55 <mattb> i could theoretically publish some OWL equating my categories with those of others'

01:52:09 <danbri_dna> the HP labs guys did some work in this area in swad-e, haven't studied it for a while

01:52:15 <danbri_dna> .google swad-europe semantic blogging

01:52:17 <datum> swad-europe semantic blogging: http://www.w3.org/2001/sw/Europe/reports/open_demonstrators/hp-requirements-specification.html

01:52:26 <mattb> i'm trying to revisit that sort of thing

01:52:37 <mattb> i had an n3-based rewrite of hackdiary in progress on my stolen laptop

01:52:43 <danbri_dna> ouch

01:52:47 <mattb> exploring again using zope

01:53:02 <mattb> i like the idea of publishing systems like bloxsom

01:53:08 <mattb> that are largely filesystem-based

01:53:14 <mattb> drop files in the right place and the magic sorts it out for you

01:53:36 <CaptSolo> matt: you mean 000040.rdf

01:53:43 <mattb> it'd be nice to have a filesystem that could derive RDF from all the different filetypes on it, and aggregate them centrally

01:53:47 <mattb> CaptSolo: thanks :)

01:54:09 <mattb> drop in a jpeg, exif rdf is generated

01:54:09 <mattb> etc

01:54:10 <CaptSolo> http://www.hackdiary.com/categories/python -> 404 not found

01:54:10 <dc_rdfig> C: http://www.hackdiary.com/categories/python from CaptSolo

01:54:15 <danbri_dna> mattb, in macosx there's a notification callback for 'something changed in this dir'... was wondering about using that for rdf summarisers

01:54:17 <CaptSolo> ups

01:54:28 <mattb> C:=http://planet.rdfhack.com

01:54:28 <dc_rdfig> Replaced URL of C.

01:54:28 <CaptSolo> was not intended

01:54:30 <danbri_dna> dunno any linux-friendly filesystem can do that

01:54:31 <mattb> C:|so good we blogged it twice

01:54:32 <dc_rdfig> Titled item C.

01:54:35 <mattb> ;)

01:54:40 <CaptSolo> ;)))

01:54:44 <mattb> CaptSolo: yeah, there's no there there

01:54:52 <mattb> danbri_dna: similarly i'm interested in FAM on linux

01:54:59 <mattb> File Alteration Monitor daemon or somesuch

01:55:05 <danbri_dna> I dunno FAM, ah cool

01:55:11 <danbri_dna> how does it work?

01:55:13 <mattb> no idea

01:55:22 <mattb> there's some sort of api you can use to get notifications

01:55:25 <mattmcc> It listens to syscalls that relate to modifying files.

01:55:27 <mattb> i just see it starting every time my laptop boots

01:55:32 <mattmcc> SGI came up with it, it's pretty slick.

01:55:33 * danbri_dna hopes it isn't a replacement libc etc

01:55:46 <CaptSolo> matt: i wanted to say - categories/python -> 404 not found

01:55:59 <CaptSolo> not that uri should point to an actual web page, but still...

01:56:09 * mattb mutters about httprange-14

01:56:12 <mattb> perhaps i should have #'d it

01:56:23 <mattb> danbri_dna: you still feeling strongly about / vs #?

01:56:24 <mattmcc> With the right kernel patch (imon) FAM can hear about a file modification instantly (no polling needed)

01:56:27 <mattmcc> Otherwise it polls, iirc.

01:57:10 <danbri_dna> mattb: I feel strongly that both should be, in some way (redirects, avoiding conneg etc) acceptable.

01:57:38 <danbri_dna> # seems safer option recently, in that timbl doesn't like /. but he's not been so anti / lately, so long as redirects are used.

01:57:53 <mattb> i prefer / aesthetically

01:58:03 <mattb> feels more hierarchical

01:58:03 <CaptSolo> matt: of similar subject to deriving rdf from a file dropped in a directory

01:58:19 <mattb> feels more webby, cf relative URLs etc

01:59:05 <CaptSolo> matt: -> a was looking for info about a full-text search engine http://www.mnogosearch.ru/index.html

01:59:21 * danbri_dna installs dnotify

01:59:23 <CaptSolo> so - it may index a number of different file formats.

01:59:48 <CaptSolo> the way it does it is pretty straightforward - i.e. for pdf it uses pdf2text and then indexes the resulting text

01:59:55 <mattb> ooh, dnotify interesting

02:00:55 <danbri_dna> hey, it works!!!

02:00:57 <mattb> CaptSolo: reminds me of product pitches for things like Verity i went to years ago

02:01:02 <danbri_dna> that was easy. i already had kernel support.

02:01:09 <mattb> this is why zope has interested me

02:01:19 <mattb> it's a bit like a filesystem, but you can attach methods to files and directories

02:01:20 <danbri_dna> jigsaw was interesting in this way too

02:01:23 <mattb> and it makes them all URL-accessible

02:01:39 <CaptSolo> hmmm, sounds interesting

02:01:41 <mattb> so i add an as_rdf method to a bunch of stuff

02:01:47 <mattb> then put some rss, jpegs and rdf in a directory

02:01:52 <CaptSolo> it is about zope? or about verity?

02:01:59 <mattb> and i can index them all using the as_rdf method i've added

02:02:01 <mattb> this is about zope

02:02:06 <CaptSolo> aha

02:02:12 <mattb> verity had automatic translation from many formats

02:02:17 <CaptSolo> should take a look at it

02:02:21 <mattb> it was always too expensive for companies i worked ofr

02:02:52 <CaptSolo> but that would be a good open-ended solution - have simple utilities that extract RDF from almost anything

02:03:03 <CaptSolo> pdf2rdf, mp32rdf, ...

02:03:26 <mattb> had interesting conversations with some people at www2003 about instrumenting operating systems in this way

02:03:37 <danbri_dna> +1

02:03:38 <mattb> edd and brian mcbride talking about a system that dumps triples about everything you do and every file you make

02:03:56 <mattb> that's what i'm trying to make in zope

02:03:57 <danbri_dna> whats the point of having rdf unless we use it for everything? ;)

02:04:02 <mattb> every URL will at least have a dc:title

02:04:06 <CaptSolo> the idea sounds good

02:04:59 <CaptSolo> would like to get to know more / see it in action

02:05:17 <CaptSolo> mozilla is keeping rdf of pages you visit...

02:05:25 <mattb> hopefully i'll make something out of it before too long

02:05:31 <CaptSolo> just i don't know how to get it dump rdf out of it into bookmarks.rdf

02:06:01 <CaptSolo> by keeping rdf i mean keeps page history in rdf

02:06:29 <CaptSolo> matt: will wait for news on this :>

02:07:00 <mattmcc> CaptSolo: Well, Moz can serialize your bookmarks to RDF/XML, but saving it's another matter.

02:07:58 <CaptSolo> matt: saving is a big problem?

02:08:32 <CaptSolo> matt: btw right category url should be http://www.hackdiary.com/cats.html#python

02:08:41 <mattmcc> Well, without being granted permission, the script is sandboxed from writing files.

02:08:41 <CaptSolo> instead of /categories/python

02:09:03 <CaptSolo> at least if you don't want people to get 404s when looking at there urls

02:09:18 <mattb> CaptSolo: yeah, i'm aware of the issues there

02:09:42 <danbri_dna> hmm i can't see how to get filename from dnotify, only a directory name.

02:09:48 <CaptSolo> matt: you should register the script in the chrome and then you can write a file i guess

02:10:01 <CaptSolo> ok, i leave you now

02:10:06 * CaptSolo asleep

02:10:12 <mattmcc> You actually have to register it as chrome just to get at the bookmarks. :)

02:14:07 <mattb> nn

02:31:55 <tester2004> testing

02:31:58 <tester2004> :)

03:37:12 <Arnia22> Arnia22 is now known as Arnia

03:43:19 <dmwaters> {global notice} Hi all! at 06:00 UTC 4 freenode servers need to be rebooted to upgrade their kernels to fix that kernel exploit. 3 of them are rotation servers, and 1 is a hub. I will do wallops as we go. if you would like to see the information about theese reboots and progress, please /mode your_nick +w to see the wallops. Thank you for your patience, and thank you for using freenode!

05:10:16 <Davey> Davey is now known as D[a]vey

05:37:47 <deusx> deusx is now known as _deusx

08:25:28 <golbeck___> golbeck___ is now known as golbeck

10:42:36 <swh_afk> swh_afk is now known as swh

10:48:16 <libby_> libby_ is now known as libby

11:06:11 <libby> hi AndyS

11:18:35 <D[a]vey> D[a]vey is now known as Davey

11:20:03 <Davey> Hmm, can't seems to convert this .ics file to RDF using ical2rdf.pl :/

11:20:23 <libby> is it online davey?

11:20:29 <Davey> libby: the .ics?

11:20:32 <libby> yeah

11:20:54 <Davey>http://davey.is-a-geek.com/php-cvs/PEAR/XML_Rdfical/CalendarDataFile.ics

11:20:55 <dc_rdfig> D: http://davey.is-a-geek.com/php-cvs/PEAR/XML_Rdfical/CalendarDataFile.ics from Davey

11:21:18 <libby> did you generate it using something?

11:21:22 <Davey> D:|A complex iCalendar file generated by Mozilla Calendar

11:21:22 <dc_rdfig> Titled item D.

11:21:31 <Davey> libby: that answer your question?

11:21:39 <libby> yep

11:21:43 <Davey> :)

11:22:09 <libby> hm it should work although there was some aort of issue with moz putting many vcalendars in one file I think

11:22:51 <libby> ah, also we never looked at todos

11:22:52 <Davey> hmmm

11:23:06 <Davey> I have removed the todos and it still dies :/

11:23:46 <Davey> How does VCalendar and iCalendar differ? and why do they have the same name in the data?

11:24:54 <libby> vcalendar is a precursor to icalendar. but where it says 'vcalendar' in the data, that's an icalendar file!

11:25:37 <Davey> oh well that makes sense. *grumbles*

11:25:53 <libby> I do;t have time to check right now but I strongly sustpect it's because several vcalendars in one file. THat's a bug since multiple vcalendars are allowed in one file

11:26:19 <libby> can I use your file as sample data for our ytestcases, or is it too private?

11:26:21 <Davey> it does have more than one BEGIN:VCALENDAR in it.

11:26:39 <Davey> no, thats the php|cruise conference schedule and some other test data only :)

11:26:44 <libby> yep which is fine for RFC 2445 but is not in our testcases

11:27:05 <libby> so that's ok then? to put it with thesre here? http://www.w3.org/2002/12/cal/test/

11:27:12 <Davey> yup, thats fine :)

11:27:28 <Davey> but you will need to actually host it there, as its on my local machine only atm :)

11:27:47 <libby> that's cool, thanks Davey

11:27:51 <Davey> all I need is some example complex Rdfical RDF/XML so I can learn hwo to create :)

11:28:08 <Davey> so if you have some I could use, that would be excellent :)

11:28:12 <libby> that link I just put in has rdf and ics files

11:28:25 <Davey> excellent :)

11:28:30 <libby> there's also the online versio of ical2rdf: http://www.kanzaki.com/courier/ical2rdf

11:29:04 <Davey> I'm using a slightly old version of ical2rdf I'd expect, a month or two, so I'll test with the new one too

11:29:26 <libby> not been updated much recently I don;t think

11:29:53 <Davey> hmm, or not, its what is here: http://www.w3.org/2002/12/cal/ical2rdf.pl

11:30:57 <Davey> http://www.kanzaki.com/courier/ical2rdf?u=http%3A%2F%2Fdavey.is-a-geek.com%2Fphp-cvs%2FPEAR%2FXML_Rdfical%2FCalendarDataFile.ics&xsl=on <--- seems to work fine....

11:30:57 <dc_rdfig> E: http://www.kanzaki.com/courier/ical2rdf?u=http%3A%2F%2Fdavey.is-a-geek.com%2Fphp-cvs%2FPEAR%2FXML_Rdfical%2FCalendarDataFile.ics&xsl=on from Davey

11:31:43 <Davey> E:|Conversion of a complex iCalendar file to Rdfical and shown as (X)HTML using XSL

11:31:43 <dc_rdfig> Titled item E.

11:32:50 <libby> oh that's interesting. I wonderif masahide has updated ical2rdf at all?

11:32:51 <Davey> hmmm

11:33:03 * Davey is looking

11:33:38 <Davey> no way to see his version

11:33:42 <libby> although the component section at the end seems to assume only one vcalendar

11:33:48 * libby will ask

11:34:06 <Davey> libby: look at it without xsl on, its multiple VCalendars, and it also parses VTodo

11:34:21 <libby> cool :)

11:34:32 <Davey> here: http://www.kanzaki.com/courier/ical2rdf?u=http%3A%2F%2Fdavey.is-a-geek.com%2Fphp-cvs%2FPEAR%2FXML_Rdfical%2FCalendarDataFile.ics

11:35:12 <Davey> hmm, how to tell ical2rdf where to put the output?

12:33:50 <libby> heya GregElin

13:20:05 <chaalsMEL> chaalsMEL is now known as chaalzzz

13:23:05 <bryce> unfortunately, i have a meeting today that will conflict with the upcoming www2004 photoRDF gathering in this channel today. :(

13:24:02 <libby> aw :(

13:24:24 <libby> I'll chump any descisions we make etc, and the logs will be here

13:24:24 <bryce> it's kind of hard to explain to the boss "oh, i was actually wanting to sit around and chat with some people i've never met."

13:24:29 <libby> heh, sure

13:25:04 <bryce> i'll read the logs, and hopefully i can volunteer for something, also.

13:25:38 <libby> cool, thanks bryce

13:25:49 <libby> there will be other meetings...

13:27:45 <bryce> right. there is a good chance i could get my job to fund my trip to the conf, but i am planning to attend anyway, as my brother lives in NYC, so it's a pretty cheap trip.

13:28:06 <libby> neat :)

13:29:14 <bryce> bryce is now known as bryce_afk

13:29:14 <Davey> Hmm, don't suppose you guys also cover XSLT too do you? :)

13:34:40 <danbri> ask away, y'never know

13:34:54 <dajobe> it better be an easy question :)

13:40:19 <Davey> hehe, its *not* an easy question :)

13:41:10 <Davey> I have an XHTML document with some non-XHTML XML in it, I want to transform it so that the non-XHTML XML is transformed into XHTML and the rest stays the same

13:42:22 <dajobe> you mean non-XML HTML?

13:42:31 <Davey> no

13:42:48 <Davey> http://paws.davey.is-a-geek.com/ <-- view source to see what I mean :)

13:42:48 <dc_rdfig> F: http://paws.davey.is-a-geek.com/ from Davey

13:43:04 <dajobe> view what?

13:43:10 <Davey> F:|PHP and Web Standards Conference development site

13:43:10 <dc_rdfig> Titled item F.

13:43:18 <Davey> dajobe: view source on it

13:43:24 <dajobe> what in the source?

13:43:31 <Davey> <paws:*> near the bottom

13:43:41 <Davey> hmm... wait

13:43:46 <Davey> no, its there :)

13:44:13 <Davey> The XSLT is here: http://paws.davey.is-a-geek.com/paws.xsl

13:44:20 <dajobe> so you want to pass everything unchanged except edit the <paws:related> subtree of elements

13:44:48 <Davey> yup

13:45:30 <Davey> but it needs to display in the #related8x6 <div>

13:46:39 <Davey> it can't be in that part of the document because of the way I've got it templated.

13:49:42 <dajobe> beyond my xslt skills, I'ma fraid

13:50:36 <_deusx> _deusx is now known as deusx

13:51:42 <Davey> dajobe: :/

13:52:03 <dajobe> I've never done copying, just select & construct stuff

14:17:09 <yalaforge> hi! I'm currently trying to document the musicbrainz mm: vocabulary using RDF Schema. Is anyone around who wants to help?

14:20:07 <libby> isn;t it alreday done?

14:20:16 * libby doesn;t know, just wondered

14:20:26 <yalaforge> there's just an informal description on the website

14:20:52 <evlist> Davey, what's the problem with your XSLT transformation?

14:21:05 <Davey> evlist: depends on the platform right now. Hehe

14:21:38 <evlist> depends on the platform?

14:21:47 <libby> yalaforge: one thing we did was generate an rdf schema using rdf data files, using CWM. there's an online tool that does something similar I think so

14:21:53 <evlist> you mean on the XSLT processor?

14:21:59 <Davey> evlist: IE and serverside aren't applying a template I think it should be, Mozilla isn't transforming right at all :/

14:22:11 <yalaforge> libby: do you have a link for me?

14:23:03 <evlist> What is your source document? is it http://paws.davey.is-a-geek.com/ ?

14:23:18 <Davey> evlist: indeed

14:23:23 <libby> the progos tools are the online thing: http://lists.w3.org/Archives/Public/www-rdf-interest/2004Jan/0007.html - not tried it yet though

14:23:28 <Davey> evlist: /paws is the XSLT

14:23:39 <yalaforge> thanks

14:23:59 <Davey> evlist: you have to see it in Mozilla 1.2.1 or equivelent or IE6 to see the proper source

14:24:21 <evlist> is it filtered server side?

14:24:41 <libby> the CWM stuff - and example is here: http://esw.w3.org/topic/RdfCalendarSchema

14:24:58 <Davey> evlist: if you don't have those browsers, yes

14:25:10 <libby> it might get you started anyway yalaforge

14:25:21 <yalaforge> ok, thanks!

14:25:26 <evlist> That's what I missed it then (first tried with konqueror then did a wget)!

14:25:43 <Davey> evlist: yeah, both of those can't transform client-side, so I do it serverside :)

14:30:50 <evlist> your XSLT could be simplified, but it seems OK with both libxslt and Saxon.

14:31:44 <[GNU]> hello more general people

14:31:59 <Davey> evlist: hmmm, I'm using libxslt serverside

14:31:59 <[GNU]> i got tha RDFAPI for PHP, and and figure out how to generate FOAF files with it

14:32:13 <[GNU]> some hints?

14:32:27 <evlist> Hmmm... suddently, it displays something in epiphany (moz based)

14:32:39 <evlist> with a PHP error...

14:33:15 <Davey> yeah, my bad :)

14:33:17 <dajobe> Davey's the php guy [GNU]

14:33:29 <[GNU]> :) thanks

14:33:30 <libby> BLURB: WWW2004 image and conference description meeting, 2004-01-06, 1500UTC, here

14:33:30 <dc_rdfig> G: WWW2004 image and conference description meeting, 2004-01-06, 1500UTC, here from libby

14:33:40 <[GNU]> Davey: got a secound or two?

14:33:54 <Davey> [GNU]: I'm about to add an XML_FOAF package to PEAR if you can wait a day? :)

14:34:01 <bryce> .time 1500utc

14:34:01 <datum> Sorry, I don't know about time zone 1500UTC.

14:34:11 <bryce> .time utc

14:34:12 <datum> 2004-01-06T14:34:11Z

14:34:17 <[GNU]> hmm :) how much secounds has a day?

14:34:26 <libby> G:[http://esw.w3.org/topic/WWW2004|more info] [http://esw.w3.org/topic/ImageDescriptionRdfExamples|more info about image description]

14:34:27 <dc_rdfig> Added comment G1.

14:34:35 <Davey> [GNU]: 86400 :)

14:34:44 <libby> G:in 25 mins or so. all welcome

14:34:44 <dc_rdfig> Added comment G2.

14:34:51 <libby> hope that's ok guys

14:35:04 <libby> should last about an hour

14:35:05 <[GNU]> Davey: ok, so i will wait until you got that online

14:35:11 <evlist> Davey since Saxon likes it, I'd say it's a Mozilla bug unless maybe, a problem with the mime type sent to the browser or a similar issue.

14:35:17 <[GNU]> or if you need beta testers... here i am

14:35:20 <Davey> [GNU]: http://davey.is-a-geek.com/php-cvs/pear/XML_FOAF/docs/examples/example1.php for an example of using it :)

14:35:38 <libby> G:[http://rdfig.xmlhack.com/2003/12/19/2003-12-19.html#1071846235.658126|previous discussions]

14:35:38 <dc_rdfig> Added comment G3.

14:35:38 * Davey personally things the comments in the example make it look more complex

14:36:07 <Davey> [GNU]: I'm also going to be writing XML_FOAF_Parser and XML_FOAF_Lite, the latter will be like foaf-a-matic

14:36:14 <[GNU]> Davey: .phps looks like what i am looking form

14:36:30 <Davey> [GNU]: yes, the top is the output, the lower part is the source of that file.

14:36:46 <[GNU]> ja

14:37:17 <[GNU]> @Davey.registered.freenode

14:37:20 <[GNU]> hmm, whats that?!

14:37:54 <Davey> [GNU]: its the new cloaking

14:38:20 <Davey> all users who were cloaked with I:line before, are beta testing the new IRCd cloaking

14:38:42 <[GNU]> hmm :)

14:39:43 <Davey> [GNU]: some apidoc for you if you want it, http://davey.is-a-geek.com/php-cvs/pear/XML_FOAF/docs/

14:40:04 * Davey lix phpDocumentor

14:40:23 <[GNU]> from my point of view the api ist very straight fwd

14:40:41 <[GNU]> seeAlsos aint working?

14:40:55 <danbri> Davey, I msg'd you (nothing major), not sure if you saw it though...

14:41:02 * danbri bbiab

14:41:11 <Davey> danbri: oh, no, I didn't

14:41:15 <Davey> you're not identified

14:41:32 <Davey> danbri: if you try again when you get back, it'll get through

14:41:37 <Davey> [GNU]: sure they are...

14:41:38 <danbri> I am now. Shall I retype?

14:41:49 <Davey> danbri: please

14:42:01 <danbri> done

14:42:03 <danbri> did that work?

14:42:48 <Davey> yup

14:45:17 <Davey> [GNU]: Still trying to decipher what you asked...

14:45:59 <[GNU]> i wondered why the seelAlso stuff is commented out in your example

14:46:07 <[GNU]> and i guessed the wouldnt work

14:46:30 <Davey> they're not...

14:46:41 <[GNU]> ok

14:46:54 <Davey> they're not comment out

14:46:56 <Davey> commented out

14:47:28 <[GNU]> //$matt->seeAlso('http://mmcc.cx/foaf.rdf');

14:47:32 <[GNU]>http://davey.is-a-geek.com/php-cvs/pear/XML_FOAF/docs/examples/example1.phps

14:47:32 <dc_rdfig> H: http://davey.is-a-geek.com/php-cvs/pear/XML_FOAF/docs/examples/example1.phps from [GNU]

14:47:47 <Davey> oh, don't look at the phps ;)

14:47:56 <[GNU]> H:please delete that link

14:47:56 <Davey> the source for example1.php is *in* example1.php

14:47:56 <dc_rdfig> Added comment H1.

14:48:07 <[GNU]> :)

14:48:08 <[GNU]> ok

14:49:29 <dajobe> H:=http://example.org/

14:49:29 <dc_rdfig> Replaced URL of H.

14:49:42 <Davey> oooh... neat :)

14:49:44 <dajobe> H1:replace me

14:49:44 <dc_rdfig> Replaced comment H1.

14:50:02 <Davey> H:=http://davey.is-a-geek.com/php-cvs/pear/XML_FOAF/docs/examples/example1.php

14:50:02 <dc_rdfig> Replaced URL of H.

14:50:21 <Davey> H1:Example Usage of PEAR::XML_FOAF

14:50:21 <dc_rdfig> Replaced comment H1.

14:50:30 <Davey> H:|Example Usage of PEAR::XML_FOAF

14:50:30 <dc_rdfig> Titled item H.

14:50:34 <dajobe> then H1:""

14:51:50 <Davey> H1:XML_FOAF will be added to the PEAR Repository in the next 24hrs. It allows creation of complex FOAF files from scratch, and can be extended to add other XML/RDF data easily through inheritance.

14:51:50 <dc_rdfig> Replaced comment H1.

14:52:17 <Davey> ok, gonna go grab a shower, bbs

14:52:24 <[GNU]> have fun

14:52:35 <dajobe> ETOOMUCHINFO

14:52:52 <Davey> dajobe: pffft :P

15:00:37 <libby> hey bwm

15:00:39 <libby> .time

15:00:39 <datum> Tue, 06 Jan 2004 15:00:39 GMT

15:00:50 * bwm waves to libbry

15:00:57 <bwm> s/bry/by/

15:01:23 <GregElin> Afternoon (UTC).

15:01:33 <libby> heya GregElin

15:01:34 <bwm> G'day greg

15:01:45 <lyndonwong> hello Greg

15:01:47 <GregElin> G'day Brian, Libby.

15:01:53 <GregElin> Hey Lyndon.

15:02:09 <kota> hello all.

15:02:10 <GregElin> Lyndon, you should introduce yourself.

15:02:15 <masaka> evening (midnight, JST)

15:02:39 <libby> heya masaka :)

15:02:49 <kota> good morning from snowing seattle :(

15:02:50 <masaka> hey libby :)

15:02:56 <mc> have we started?

15:02:56 <GregElin> Hi kota, masaka.

15:03:05 <libby> no mc, just intros

15:03:08 <GregElin> We are just saying hello to each other, mc.

15:03:19 <libby> hey JibberJim

15:03:19 * bwm wonders if there is an agenda

15:03:19 <GregElin> Hey Jim.

15:03:38 <bryce> g'morning from brisk but sunny Denton, Texas.

15:03:48 <mc> mornign texas

15:03:55 <lyndonwong> I know Greg from his work with FotoNotes, and was invited to join this IRC. I was formerly a director of prod mgmt for the Webshots unit of ExciteAtHome, and now consult for a rich internet app platform startup called Laszlo Systems. (that suffice?)

15:04:37 <lyndonwong> I look forward to getting to know everybody! thx greg for inviting me :-)

15:04:38 <bwm> does "rich internet app platform startup" mean you have loads'a'dosh?

15:04:39 <[GNU]> Davey: any option to add xmlns:lang with the add/set methods??

15:04:42 <GregElin> excellent, Lyndon.

15:05:29 <GregElin> Maska, bryce, kota - are you here for the discussion WWW Photo History project?

15:05:44 <kota> yep.

15:05:50 <masaka> yes

15:05:54 <bryce> yes

15:06:06 <lyndonwong> ;-) sorry, 'rich internet apps' is just marketing mumbo jumbo... it's a startup that enables developers to create apps with client-like UI attributes via XML & JavaScript, and targets the Flash player as a run-time (think alternative to DHTML or Java applets)

15:06:31 <GregElin> I think Agenda is pretty loose for our discussion of the WWW Photo History project: report on our basic infrastructure, discuss timeline, discuss UIs (wrapping up the emails).

15:06:43 <Davey> [GNU]: I will add that, yes. But not in the release tomorrow

15:06:53 <mc> can we set a time on the meeting? like an hour from now?

15:06:56 <GregElin> Kota, masaka, bryce can you do a sentence intro, your interests?

15:07:02 <dajobe> Dave Beckett, rdf bits and pieces. did some www2003 community sites, offered to do a www2004 something blog setup (& irc too)

15:07:06 <yalaforge> ok, I've got a first version for the musicbrainz stuff on http://www.stud.uni-karlsruhe.de/~uy7l/mm-schema.html. Anyone who's interested can join #musicbrainz. Bye!

15:07:10 <[GNU]> Davey: sure :) thanks

15:07:10 * bwm is disappointed about loose agenda - feels meeting would benefit from a little structure - maybe greg could quickly sketch one

15:07:29 * libby has invited a few poeple

15:07:50 <Davey> [GNU]: if you have other questions PM me, rather than clutter up this meeting :)

15:07:52 * libby hoping for some tech discussios about http://esw.w3.org/topic/ImageDescriptionRdfExamples

15:08:14 <libby> can I have that as an agenda?

15:08:17 <GregElin> mc - an hour from now is fine to call it quits. Should we make the agenda tighter?

15:08:19 <libby> item

15:08:20 * bwm wonders whether we should be in another room - not disrupting rdf interest

15:08:30 <dajobe> nah

15:08:32 <golbeck> I added a section to the Wiki on that subject that presents the Image ontology we are using at Mindswap

15:08:40 <libby> it's ok bwm, there was plenty o0f notice, and people can switch to #foaf for many things

15:08:52 <libby> thanks golbeck

15:08:57 <bwm> libby: ok

15:08:58 <bryce> Bryce Benton, experience with php/mysql, accessibility, blogging, front end stuff, interested in photo metadata.

15:09:05 <mc> could we review here, for the log, what's been done and what's going on and by which sites?

15:10:00 <kota> kota, interested in the semantic web, now trying to catch up with rdf stuffs, including this image description thing.

15:10:31 <GregElin> Good suggestion, mc. Let's start with quick reports of what's been done and where some key action items stand. Then let's put our meta data and semantic web data into some distinct buckets. Then let's discuss the imagedescription metadata concretely.

15:11:12 <ChanServ> [#rdfig] This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome

15:11:12 <libby> is it ok if I use the weblog for agenda items?

15:11:12 <golbeck> well, i wrote an ontology for conferences and events that I believe Jim mailed out to the group. I don't remember seeing much feedback about it, though I would be happy for some.

15:11:24 <GregElin> Hmmm... that was interesting. It would appear niven just went down. THough I'm still on.

15:11:28 <libby> G:attending [libby Miller|http://ilrt.org/people/libby/]

15:11:28 <dc_rdfig> Added comment G4.

15:11:33 <dajobe> carry on GregElin

15:11:42 <libby> I think everyone got back greg

15:11:46 <JibberJim> G:attending [Jim Ley|http://jibbering.org/]

15:11:46 <dc_rdfig> Added comment G5.

15:11:59 <GregElin> Log as you will, Libby.

15:12:14 <GregElin> Item 1: What's been done. I'll start.

15:12:14 <bwm> G:attending bwm

15:12:15 <dc_rdfig> Added comment G6.

15:12:19 <libby> BLURB:WWW2004 images/conferences agenda item I: quick reports of what's been done and where some key action items stand

15:12:19 <dc_rdfig> I: WWW2004 images/conferences agenda item I: quick reports of what's been done and where some key action items stand from libby

15:12:22 <swh> G:attending [Steve Harris|http://inanna.ecs.soton.ac.uk/]

15:12:23 <dc_rdfig> Added comment G7.

15:12:24 <bryce> G:attending [Bryce Benton|http://tenyearsof.us/]

15:12:26 <nmg> golbeck: apologies - I've only given it a cursory read so far, and need to write some more detailed notes. It's basically okay and fit for purpose, though ;)

15:12:27 <GregElin> G: attending GregElin

15:12:27 <dc_rdfig> Added comment G8.

15:12:30 <golbeck> G:attending [Jen Golbeck|http://owl.mindswap.org/]

15:12:30 <dc_rdfig> Added comment G9.

15:12:33 <dajobe> G:attending Dave Beckett

15:12:34 <dc_rdfig> Added comment G10.

15:12:38 <mc> may i suggest that the meta data/ ontology stuff is mission critical, and that we may need to fork off the UI discussion to a sub group and report back next meeting?

15:12:47 <mattb> G:attending [Matt Biddulph|http://www.hackdiary.com]

15:12:47 <dc_rdfig> Added comment G11.

15:12:49 <libby> BLURB:WWW2004 images/conferences agenda item J: let's put our meta data and semantic web data into some distinct buckets

15:12:49 <dc_rdfig> J: WWW2004 images/conferences agenda item J: let's put our meta data and semantic web data into some distinct buckets from libby

15:12:57 <nmg> G: attending [Nick Gibbins|http://www.ecs.soton.ac.uk/~nmg/]

15:12:57 <dc_rdfig> Added comment G12.

15:13:12 <libby> BLURB:WWW2004 images/conferences agenda item K: discuss the imagedescription metadata concretely

15:13:12 <dc_rdfig> K: WWW2004 images/conferences agenda item K: discuss the imagedescription metadata concretely from libby

15:13:23 <GregElin> I revised the planning document and would like to freeze it. Also proposed and sent out a detailed timeline.

15:13:42 <libby> can some or all of that be made public GReg?

15:13:48 <lyndonwong> G:attending [Lyndon Wong http://lyndonwong.blogspot.com]

15:13:48 <dc_rdfig> Added comment G13.

15:13:59 <GregElin> I haven't yet gotten around to pulling material from the document and fleshing out the public info website I mocked up in December.

15:14:13 <libby> can I put you donw for an action GregElin?

15:14:14 <dajobe> with ref to item 1 - what's been done, there are actions in http://esw.w3.org/topic/WWW2004

15:14:22 <GregElin> I think it can all be made public if it is okay with you all. I'd like to keep the budget private, though.

15:14:31 <libby> make ssense to me, cheers greg

15:15:05 <GregElin> I think the Word document has a high level description of the project.

15:15:29 <GregElin> But not technology.

15:15:39 <GregElin> I don't have any news on the hosting or URL.

15:15:42 <masaka> G:attending [masahide kanzaki http://kanzaki.com]

15:15:42 <dc_rdfig> Added comment G14.

15:16:14 <GregElin> That's it for me.

15:16:24 <GregElin> Dan, where do we stand on the email list?

15:16:50 * libby thinks danbri is running a bit late

15:17:02 <Davey> he's at the post office

15:17:25 <libby> golbeck have you gotthe url for the ontoloyg for conferences?

15:17:39 <GregElin> We'll come back to Dan. Libby, want to bring us up to speed? Any word from creative commons?

15:18:09 <libby> no, nothing greg. prhaps a bad time to ask. I'll bug them again. but basically their site helps you with choice of licensing

15:18:19 <nmg> libby: http://www.mindswap.org/~golbeck/web/www04photo.owl

15:18:21 <golbeck>http://www.mindswap.org/~golbeck/web/www04photo.owl

15:18:21 <dc_rdfig> L: http://www.mindswap.org/~golbeck/web/www04photo.owl from golbeck

15:18:27 <libby> ta guys

15:18:58 <libby> I suggested this license: http://creativecommons.org/licenses/by-nc-sa/1.0/

15:19:08 <bryce> Not sure if golbeck's ontology may be better, but there is also ebiquity's ontology: http://www.schemaweb.info/schema/SchemaDetails.aspx?id=116

15:19:12 <libby> L:|an ontology for describing conferences

15:19:13 <dc_rdfig> Titled item L.

15:19:14 <golbeck> L: Draft ontology for events and, specifically, conferences

15:19:15 <dc_rdfig> Added comment L1.

15:19:41 <GregElin> That looks pretty good at first glance, Libby.

15:19:45 * JibberJim isn't interested in non-commercial licences.

15:19:53 <libby> L:see also [http://www.schemaweb.info/schema/SchemaDetails.aspx?id=116|ebiquity's ontology] and [http://www.w3.org/2002/12/cal|RDFiCal]

15:19:53 <dc_rdfig> Added comment L2.

15:20:00 <nmg> bryce: the ontologies cover differrent areas - jen's is about conference attendance rather than the submission process (etc)

15:20:02 <libby> I think we probably need to have some discussion

15:20:06 <bryce> ok

15:20:08 * JibberJim doesn't like to limit what people can do, including selling it!

15:20:34 <libby>http://creativecommons.org/licenses/by-nc-sa/1.0/

15:20:34 <dc_rdfig> M: http://creativecommons.org/licenses/by-nc-sa/1.0/ from libby

15:20:50 <libby> M:|CC Attribution-NonCommercial-ShareAlike 1.0 license

15:20:51 <dc_rdfig> Titled item M.

15:21:12 <libby> M:what do people think? for WWW2004 conference photos and metadata?

15:21:12 <dc_rdfig> Added comment M1.

15:21:28 <golbeck> This draft is linked into FOAF, but I think it is reasonable (and web healthy) to link to /map to other ontologies we find interesting

15:21:30 <GregElin> (I'm noting issues to discuss: ontology, Jim's point)

15:21:33 <dajobe> maybe suggest (not require) that one, and the one without the non commercial

15:21:48 * danbri back

15:21:59 * libby thinks would be ebtter to ahve one license if possible

15:21:59 <nmg> I concur with dajobe - attribution and sharealike are the crucial bits

15:22:02 <danbri> (sorry late, errand)

15:22:24 * danbri will paste some thoughts on image desc mailing list(s) into wiki, bbiab

15:22:36 <dajobe> M:or maybe also without the non-commercial restriction [http://creativecommons.org/licenses/by-sa/1.0/|attribution-sharealike]

15:22:36 <dc_rdfig> Added comment M2.

15:22:53 <GregElin> Jim, is your point a general philosophical one, or you have specific thoughts regarding this situation?

15:23:16 <bryce> it's still possible to use the data commercially, it just requires prior consent.

15:23:25 <JibberJim> General philosophical one.

15:24:00 * libby doesn;t really mind, except I wonder what people would be more likely to agree to

15:24:17 <GregElin> "use the data commercially" is a big catch all, bryce. Especially since we are talking about photographs of individuals.

15:24:39 <GregElin> And the data ranges from what's in the photo to conference material.

15:24:39 <libby> actaully a variety of licenses would be ok - we coudl just select the subset of images taht were free to use for any circumstances using the CC markup

15:25:01 <bryce> that's why i vote for http://creativecommons.org/licenses/by-nc-sa/1.0/

15:25:03 <JibberJim> as long as it makes it into the RDF yep, no problem with a variety

15:25:09 <nmg> will the same license also apply to the metadata as well as the images?

15:25:13 <GregElin> +1 libby. But people donating images need to understand what they are agreeing.

15:25:29 <libby> yes and will compelixify the UI

15:25:38 <libby> nmg, I was thinking so. though they could differ

15:25:40 <JibberJim> bryce having to contact the people to get consent isn't really feasible for mass authored content - I certainly wouldn't want it, I get enough mail as it is.

15:25:57 <bryce> ic

15:25:57 <dajobe> +1 same license for metadata; +1 for a small set of recommended licenses (2?)

15:26:37 <libby> how can we resolve this? take it to the list?

15:26:39 <GregElin> I think the final answer regarding the license will depend on the specific data we define.

15:26:44 <bryce> +1 dajobe

15:26:45 <libby> (when we have one)

15:26:53 <dajobe> move on to ontology chat?

15:27:01 * JibberJim sees nothing but votes for a small set of recommended licences.

15:27:03 <GregElin> I think we've made a dent on this path. Let's move on.

15:27:10 <GregElin> First...email list update...Dan?

15:27:15 <golbeck> we should probably pick some default, though, so people don't HAVE to really pay attention to the license. i know i really wouldn't care, and having to read about/ think about it would actually discourage me from submitting.

15:27:37 <GregElin> (agreed JibberJim - final license will be discussed on list)

15:27:48 <libby> M:votes seem to be for 2ish recommended licenses; will discuss on list

15:27:48 <dc_rdfig> Added comment M3.

15:27:56 <GregElin> Yep.

15:28:06 <dajobe> logger, chump M

15:28:06 <dajobe> M:See [http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/2004-01-06#T15-28-06|discussion]

15:28:07 <dc_rdfig> Added comment M4.

15:28:18 <libby> M:golbeck suggests having a default, so people don;t have to care too much

15:28:19 <dc_rdfig> Added comment M5.

15:28:23 <GregElin> (thanks for looking through licenses and making recommendation, Libby)

15:28:36 <libby> np, the site makes it easy

15:28:55 <GregElin> DanBri, where does email list stand?

15:29:58 * libby agrees w dajobe - move on to ontologies/vocabs?

15:30:15 <danbri> email list: we're (W3C folks) just back from the winter break... need to discuss with colleagues per my msg a few days back

15:30:29 <GregElin> Okay.

15:30:41 <libby> danbri do you have an eta? cc lists a nuisance...

15:30:43 <GregElin> onto ontology/vocab...

15:30:52 <danbri> I wanted to ask you folks about scope. What I want to propose it a list for semantic web / rdf image description... but introduce it as being set up initially for 3 particular collaborations

15:31:10 <danbri> (i) www2004 image project (ii) the exif rdf vocab collab (iii) the image area vocab collab

15:31:26 <danbri> all of which have links with the #rdfig occasional IRC chats we've had on ImageDescription

15:31:42 * libby would like that

15:31:50 <danbri> ie. have a midway point between a www2004-only list, and a "everyone join who is into image metadata"

15:31:53 <libby> however, the WWW2004 project uis not a bit woder than images

15:31:56 <libby> wider

15:32:05 <libby> is now a bit wider (sorry)

15:32:07 <danbri> ...put it under the wing of the W3C Interest Group, and open it up gradually in scope

15:32:17 <danbri> yes, all image description projects turn out that way lib :)

15:32:27 <danbri> images are, almost, a distraction

15:32:46 <GregElin> Any time we discuss photos...the discussion will inevitably get a bit wider...that's the idea...make photos wider...

15:32:47 <danbri> if nobody objects to this approach, I'll have something reasonably coherent to propose to my colleagues

15:32:51 <danbri> +1

15:33:14 <mc> would you mind including the notion of UIs in your description?

15:33:22 <danbri> I don't want to have to propose another list for the exif/rdf collab, which has also sometimes dissapeared into private CC: discuss (masaka, karl, myself...)

15:33:34 <GregElin> Sounds good...want to keep email focused (mostly) on project up until the conference.

15:33:36 <bwm> bwm: feels we need to be careful to get some focus if we are to have something for www 2004

15:33:37 <danbri> sure... UIs for navigation and for description...

15:33:52 <danbri> OK, I'll ask Exif folks to put [exif] in the subject lines, etc.

15:33:59 <danbri> I doubt there'll be huge amounts of traffic on that

15:34:00 <GregElin> Agree with Brian.

15:34:05 <libby> bwm, about what specifically?

15:34:21 <bwm> bwm is happy with shared list, just worried about talk of widening scope

15:35:11 * libby was only talking about current scope, which includes the conference ontology

15:35:12 <bwm> bwm would like to see a system component block diagram so we know what we are building

15:35:39 <GregElin> I think the list needs to focus on what people need to know for the project, and for doing these other things. Small 2-3 person conversations, especially preliminary brainstorming conversations about some angle/detail/tech approach, should be minimized on the list.

15:35:49 <danbri> brian, that's why I suggest scoping to just these three image-descriptoin collaborations

15:36:16 <danbri> the other two haven't generated much traffic, and are close anyway (exif schema, content region schema -- could easily be subtasks of this collab anyway)

15:36:23 * libby wants to use thelist for the technical details....

15:36:25 <bwm> danbri: +1

15:37:00 <danbri> I don't think its cost effective to have one list for each image-related schema/namespace that spins out of these collaborations

15:37:11 <tav> tav is now known as tav|offline

15:37:18 <GregElin> We can always divide the list further if we need, right?

15:37:31 <GregElin> I agree Dan.

15:37:41 <dajobe> can we just go with it, danbri to do asap?

15:37:50 <bwm> bwm suggests chair (Greg) call for objections to danbri's proposal

15:38:06 <bwm> and on hearing none approve it

15:38:14 <GregElin> ANY OBJECTIONS TO DANBRI'S PROPOSAL?

15:38:18 <libby> heh

15:38:43 * GregElin bangs gavel

15:38:49 <bwm> :)

15:38:57 <danbri> OK, I'll take proposal to my colleagues asap. Thanks.

15:39:00 <GregElin> Okay. ontologies, vocabs.

15:39:14 <libby> I:action danbri ask for a list asap

15:39:14 <dc_rdfig> Added comment I1.

15:39:18 * bwm suggets from now on greg change nick to gavelbanger

15:39:20 <dajobe> logger, chump I

15:39:20 <dajobe> I:See [http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/2004-01-06#T15-39-20|discussion]

15:39:20 <dc_rdfig> Added comment I2.

15:39:24 <golbeck> I'll point out this one, which I just put on the wiki yesterday

15:39:28 <golbeck>http://www.mindswap.org/~glapizco/technical.owl

15:39:28 <dc_rdfig> N: http://www.mindswap.org/~glapizco/technical.owl from golbeck

15:39:34 <GregElin> speaking for self, I'm a bit confused between conference material ontology, conference ontology, and conference photo ontology...

15:40:05 <golbeck> N: Ontology for Images, image regions (SVG), videos, frames, segments, and what they depict

15:40:05 <dc_rdfig> Added comment N1.

15:40:15 <GregElin> Nick, monica, can you help clear that up?

15:40:25 <GregElin> (looks good golbeck!)

15:40:26 <danbri> are there any distinguishing features of 'conference photos'?

15:40:33 <golbeck> N: Currently the default ontology for images in MINDSWAP's Photostuff tool

15:40:33 <dc_rdfig> Added comment N2.

15:40:39 <nmg> will do

15:40:42 <danbri> I haven't reviewed these links yet btw, maybe they explain

15:41:24 <golbeck> this ontology does not distinguish confernece from other photos, though i don't think it necessarily should. i think that should be a distinction in how we describe what is depicted

15:41:26 * libby would like to see integrated with existing vocabs if possible, lkie with http://esw.w3.org/topic/ImageDescriptionRdfExamples

15:42:06 <libby> - also the events vocab with RDFical

15:42:20 <dajobe> [technical point the urls in that http://www.mindswap.org/~glapizco/technical.owl will turn out funny, the xml:base ends in ".owl"]

15:43:38 <DanC> I gather there's an interesting photo RDF meeting happening; if there's an attendance record, please let it show my regrets.

15:43:43 <libby> seems a bit odd to repeat the foaf properties depcts and depiction rather than reusing foaf....

15:43:53 <libby> danc, yep, under G:

15:43:59 <bwm> G:regrets DanC

15:43:59 <dc_rdfig> Added comment G15.

15:44:30 <danbri> ...esp if we're using this to showcase SemWeb technology, which makes a big selling point of this mix'n'match capability of RDF/XML/OWL...

15:44:48 <golbeck> the depicts in this ontology is not restricted to an image, though. it can also be used on a region of the image

15:45:02 <libby> It seems to me that it's not the ontology that so important as guidence about how to use vocabs together, including various image vocabs, conference vocabs etc

15:45:05 <nmg> libby: re: repeating foaf properties, I think that there is some value in a (very) small ontology for depiction that only contains the depicts/Depiction vocab

15:45:09 <golbeck> i haven't looked at foaf:depicts domain, but is it restricted to foaf:Image

15:45:20 <danbri> ah, we backed out of that design, it was too hairy... made application code have to do all sorts of checks.

15:45:28 <libby> there's foaf:regiondepicts. we had some discussion about the distinction

15:45:30 <golbeck> In either case, it seems reasonable to make Image in this ontology an equivalent class to foaf:Image

15:45:30 <GregElin> Well let's remember the repository will be open. We could start with vocab "A" and another person could transform into vocab "B", right?

15:45:35 <nmg> we get better modularity that way (and the existing foaf vocab can be treated as a post-fact extension of this tiny ontology)

15:45:39 <danbri> similarly re domain/range, there was question of using them with MPEG videos etc.

15:46:37 <GregElin> I'm leaning toward nmg's POV (if I understand this all, that is. heh)

15:46:51 <GregElin> Did I see geocode info in any of these vocabs?

15:47:05 <golbeck> this ontology has a separate class for videos, segments, etc, but depicts can be used on those as well

15:47:14 <golbeck> well, it can be used on anything, actually. it has no domain or range

15:47:21 <dajobe> geo lang & long in http://esw.w3.org/topic/ImageDescriptionRdfExamples 3.1

15:47:25 <GregElin> No videos will be in the collection prior to the conference in May. Need to start simple.

15:48:15 <golbeck> See also http

15:48:24 <golbeck> See also http://owl.mindswap.org/images/

15:48:30 <bwm> bwm wonders if the range of depicts is Depiction and we can evolve what a depiction is over time

15:48:31 * danbri nods re videos

15:48:40 <golbeck> one of our demos using this ontology for photo markup

15:48:49 <GregElin> It's okay if vocab supports video, but I don't want the added headaches of the bandwidth of storing them in the beginning.

15:49:17 <danbri> bwm, yeah its tempting to take that almost tautological approach.... "what can be depicted? why, a depictable thing, or course...". But leaving the range wide open isn't so different...

15:49:19 <bwm> bwm doesn't hear anyone advocating storing video in May 2004 system

15:49:56 <GregElin> Okay. I need a little conversation here on conference v. conference material ontologies.

15:50:24 <golbeck> I think (this is a gut reaction, not a well thought out theory) putting a range on depiction may be the wrong way to go

15:50:36 <golbeck> i, personally, may want to say anything for depicts

15:50:36 * nmg agrees with golbeck

15:50:54 <golbeck> it could be an event, for example

15:51:07 <nmg> "this is a photo of the presentation of this paper", etc

15:51:13 <golbeck> or some complex interaction of intances and relationships that i can point to somewhere

15:51:15 <GregElin> Yep

15:51:26 <bwm> oops bwm may have meant domain not range

15:51:31 <GregElin> We've got 10 minutes left.

15:51:44 <GregElin> What else do we need to discuss as group?

15:51:51 <danbri> (in the foaf case we've depiction<->depicts, each with oppsite domain/range to its inverse)

15:52:10 <libby> yeah you can depict anything

15:52:12 <danbri> I'm still interested in the aspects of this problem that require conference-specific vocabulary...

15:52:31 <golbeck> i'm not sure for image markup, there is a need for conference specific vocab

15:52:35 <GregElin> Yes. Let's pick up Dan's point.

15:52:43 <golbeck> i think a good image markup ontology, that is general, is the best way to go

15:52:51 <golbeck> with a conference specific ontology for what is depected

15:52:55 <nmg> danbri: describing the images and their depictions doesn't. describing the things that they depict may do.

15:53:08 <bwm> bwm would like someone to pick up an action to produce a description of the May 2004 system - i.e. a block diagram of the system components

15:53:16 <nmg> +1 golbeck

15:53:17 <GregElin> Well, I think there is a vocab related to conferences as opposed to say, sporting events. Their are panelists at a conference and team members at sporting event.

15:53:44 * golbeck and nmg are on the same wavelength

15:53:45 <GregElin> +1 golbeck

15:53:50 <libby> GregElin: yep but there are common elements, such as times, dates

15:54:00 <GregElin> agreed, libby

15:54:17 <libby> if we use golbeck's ontologies, I would like to see how they relate to existing vocabs

15:54:32 * GregElin wondering if anyone heard the conference v. conference material question earlier.

15:54:57 <nmg> GregElin: sorry, missed that

15:54:58 <danbri> the kind of thing I feel specific need for re conferences, is in the ical-ish space; representation of conference sessions, substructure of larger meetings, notions of things like panel-session, BOF etc.

15:55:10 <golbeck> i would actually like it related to more existing vocabs, if we go that way

15:55:10 <GregElin> +1 dan

15:55:13 <golbeck> equivalences, etc

15:55:23 <danbri> I have quite a lot of photos of panel sessions, for example

15:55:46 <bwm> bwm suggests we should aim to remove equivalences

15:55:53 <nmg> conferences are a kind of event, so most of that will be temporal and organizational info. conference material is all about authorship and affiliations

15:56:08 <golbeck> bwm - why remove equivalences?

15:56:24 * bwm don't duplicate existing vocab if we don't have to - resuse it

15:56:32 * libby agrees

15:56:37 <asdfgh> asdfgh is now known as GNU][

15:57:30 <bwm> bwm is keen folks leave with actions

15:57:31 <GregElin> okay nmg. That makes sense...but who in this group is tasked with those vocabs?

15:57:40 <golbeck> equivalences don't necessarily duplicate vocab. it i way , for example, that my Conference is equivalent to Conferences described in 2 other ontologies, i'm creating a mapping that connects the two

15:57:54 <bwm> bwm proposes golbeck accept action to identify and remove equivalences for the existing vocab

15:58:09 <libby> only for owl-compliant tools golbeck

15:58:24 <GregElin> bmw, I think folks already have actions from before. If there are new actions that have come up, we should identify.

15:58:40 <libby> "<Jhendler> libby, our tool is inherently multi-ontology -- any tool built to a single schema will fail on the SW IMHO"

15:58:41 <bwm> bwm accepts ruling of chair

15:59:24 <golbeck> bwm's proposal seems to imply that my ontology is going to be what we start from. i'm happy to do that, if he's correct in his assumption

15:59:25 * GregElin chair places a gold star next to brian's name.

16:00:30 <GregElin> Your ontology is regarding photos, golbeck, or conferences or both?

16:00:31 <golbeck> Conferneces

16:00:31 <GregElin> I think mc's group is working on conference material ontology, correct?

16:00:34 <danbri> Are we all back now?

16:00:44 <libby> this is not a criticism of you golbeck, but I strongly dislike the top-down onology-creation approach

16:00:44 <danbri> I was typing away to a much smaller audience. What was the last thing the rest of you heard?

16:00:45 <nmg> GregElin: that's correct

16:00:53 <GregElin> Right. Libby's list is the starting point for photo vocabs.

16:01:09 <bwm> last thing: bwm got gold star from gavelbanger

16:01:18 <GregElin> heh

16:01:21 <danbri> I'm going to repost my recent typings, if that's OK. Sorry for the noise.

16:01:24 <danbri> [[

16:01:32 <danbri> <danbri> I have quite a lot of photos of panel sessions, for example

16:01:32 <danbri> <danbri> and there are IRC transcripts of many of these nowadays, audio logs etc. Now *even if* we don't have audio/video/etc for this specific www2004 project, having categories defined for "Panel session" and related conferency concepts is a minimal but useful hook for future extensions

16:01:32 <danbri> <danbri> so for may, we have a way of saying "pic43.jpg represents panelsession_353 whose title was 'semantic web versus web services' and whose chair was ...'"...

16:01:32 <danbri> <danbri> ...folk elsewhere, or later, or us if we find cycles, could add the extra statement "and foo.html is a textual transcript and bar.mp3 is an audio transcript"

16:01:35 <danbri> <danbri> having the notion that (a) there is a class of things "Panel Session" is the first key component... but, trickier (b) how do we identify panel sessions, such that data about them can be merged?

16:01:38 <danbri> <danbri> re (b) I would claim we use OWL and an owl:InverseFunctionalProperty, foaf:homepage, and state that each panel session has a homepage.

16:01:41 <danbri> <danbri> I think we'll need to do this exercise for each category (class, type etc) being used, ie. not only define a class, but figure out practical conventions for individiduation/identification of particular members of that class.

16:01:45 <danbri> <danbri> ugh, netsplit. or did I bore everyone away?

16:01:47 <danbri> ]]

16:01:59 <golbeck> i don't know that i am correctly classified in favor of top-down approach

16:02:12 <libby> soryr, perhaps unfair

16:02:15 <GregElin> very useful danbri. Makes up for you being late ;-)

16:02:30 * danbri blushes (queue at post office<-why)

16:03:06 <GregElin> golbeck...we are taking a narrow top down approach to have something to begin with, to have a starting point we know works. But everyone is welcome to attack the shared repository as they see fit.

16:03:23 <danbri> for each class of things we think we want in our ontology, let's think...ooOO("If I had descriptions of two of these, how would I know if they referred to the same thing?"). Hopefully for most/all, it'll be a simple identifying property like homepage, mailbox, checkSum etc.

16:03:31 <libby> so do we reckon that 'depicts' of some kind is useful for talking about events as well as people, beer etc

16:03:45 <libby> ?

16:04:03 <danbri> not clear to me, yet.

16:04:36 <danbri> being able to capture images that represent events, carved up by verb or whatever, is useful. But the 'does light bounce off it' case is useful too.

16:04:47 <danbri> may be two strongly related notions jumbled up together

16:04:50 <libby> I agree re identifying property, though I'm doubtful that panels page homepages

16:05:00 <libby> panels have hompages

16:05:03 <GregElin> I have to think about that question for a couple days, Libby.

16:05:04 <golbeck> i think my one resistance here is to make FOAF somehow a default. i love foaf. i just don't know if it is sufficient for everything we are trying to do. it seems that every proposal seems to be countered with "we have this n foaf and we do it like this."

16:05:30 <GregElin> Well said, Danbri.

16:05:47 <GregElin> +1 golbeck.

16:05:56 <libby> hm, not at all, just the depcits stuff golbeck.mindswap work sometimes comes acrross like that too....

16:06:01 * danbri nods, foaf's meant for mixing with other stuff

16:06:29 <danbri> we haven't talked much about Dublin Core... there are some qualifiers in DC relating to images that I'm not familiar enough with

16:06:32 <nmg> +0.5 golbeck - I agree in general, but I also see libby and danbri's side of it

16:06:40 * libby is very interested in mixing stuff, which is why a bit resistant to new vocabs

16:06:48 <GregElin> Yes, and yet I think foaf is inherintly about relationships where as sometimes we just want to say this is X and leave it at that.

16:07:12 <nmg> libby: mixing is the Right Thing *provided* that the existing vocabularies actually do what we want

16:07:18 <danbri> yep

16:07:38 <nmg> moreover, there may be assumptions in those existing vocabularies that are incompatible with our requirements

16:07:40 <libby> ok, and I do take the point about wanting depicts to be braoder than images maybe

16:08:24 <golbeck> however, for people using a tool based approach to authoring, having one ontology that points to others can be very useful. this is partially why i'm in favor of equivalences. it means i can look to one place that is equivalent to other ontologies, instead of having to look for and know the 10 different places that have concepts i want to use.

16:08:28 <libby> perhaps nmg...it is just interesting that so difficult to mix them... :)

16:08:31 <nmg> this is the reason I'd like to see a micro-ontology just for depiction - foaf can be retrofitted on this (so that foaf:depicts has a Person range) without breaking existing applications

16:09:02 <nmg> libby: yes - many thorny issues here...

16:09:44 <golbeck> one caveat on all the things i'm saying is that i come from an owl compliant philosophy.

16:09:52 <GregElin> yes, many thorny issues, and not all need to be immediately resolved.

16:10:06 <bryce> i was thinking along these lines recently (micro ontologies with equivalences), nmg... it's similar to the varieties of linux distros.

16:10:12 <bwm> golbeck: "owl compliant" or "owl required"

16:10:32 <GregElin> One purpose of pulling together a big enough archive "in the wild" is so we have something real against which to imagine/try these things.

16:10:58 <golbeck> maybe this owl issue is something to discuss.

16:11:25 <danbri> <golbeck> one caveat on all the things i'm saying is that i come from an owl compliant philosophy.

16:11:34 <danbri> do you mean OWL DL, OWL Full, OWL Lite?

16:11:45 <GregElin> I vote we work with smallest, most forgiving vocabs in the beginning.

16:11:47 * danbri hopes we're all OWL compliant in at least one of those senses

16:11:55 <danbri> small is good

16:12:00 <danbri> existing imho is good

16:12:08 <danbri> adaptable to feedback is good

16:12:12 <golbeck> i try always to stay in owl LITE because it's more widely supported

16:12:15 <GregElin> In other words: pursue a "do least amount as harm as possible" approach.

16:12:32 <golbeck> re bwm's point: i don't think anything is "owl required". if you're using anything in owl you lose semanticsin a tool that doesn't know owl

16:12:48 <GregElin> Any objections to small, existing, LITE, approach as a general guiding principle?

16:12:52 <golbeck> but i don't think that's a good reason to avoid owl

16:13:24 * libby not sure what's in LITE

16:13:26 <GregElin> okay...small, existing, LITE, compatible rather than required...

16:13:28 <nmg> golbeck: conversely, if you write RDF in an OWL-friendly fashion (ie. not in OWL-Full), you don't disadvantage the OWL folk and don't take a great hit to expressivitiy

16:13:43 <libby> would like to avoid depending on owl if poss

16:13:56 <libby> compatible is good tho

16:14:04 <GregElin> The semantic newbie wants to ask a big question.

16:14:26 <bwm> bwm would like to be compatable with owl but as little dependent on it as poss

16:14:36 <danbri> +1

16:14:37 <libby> +1 bwm

16:14:41 <nmg> bwm: could you elaborate?

16:15:08 <golbeck> any rdfs is compatible with owl

16:15:17 <bwm> specifically - reuse existing vocabs where possible rather than relying on inference for equivalence - reason is runtime performance

16:15:28 <nmg> agreed +1 bwm

16:15:33 <bwm> golbeck: any rdfs is compatible with owl full

16:15:46 <GregElin> Is it possible/useful to put forth a vocab/ontology as just plain ascii text (e.g., not formatted for RDF or OWL, etc) as an agreed upon set of terms for people to try and be compatible?

16:15:53 <libby> also lots of tools don;t support it. I gues that's where the foaf objections are sorta coming from

16:15:57 <bwm> but bwm is happy that owl tools can do more

16:16:32 <golbeck> if we're talking about not being relaint on owl, then we are essentially saying use rdfs

16:16:43 <nmg> GregElin: yes, but that may not be especially useful; writing the ontology in RDF or OWL allows you to make deductions from the data you've expressed

16:16:58 <GregElin> I kinda of feel like we are discussing the database schema in Oracle table descriptions rather than a entity relationship diagram...

16:17:11 <libby> ...and also define it more precisedly which is really useful too

16:17:14 <bwm> golbeck: its not a black/white thing; its don't rely on owl unless you have to

16:17:35 <bwm> but if you have to, then owl shows of what value it adds

16:17:38 <GregElin> Okay. Then I need some pointers to some newbie tools for reading this sh-t. I keep having to do view source.

16:17:43 <libby> I think golbeck and nmg have made a case for depicts being such as case

16:18:45 <golbeck> if we are not "relying" on owl, it means i couldn't rely on my local restrictions on range being considered at all

16:18:52 <GregElin> BTW...I think the official part of this chat has ended. Chair calls for any last minute comments, observations, objections....

16:19:01 <GregElin> Any big problems with the timeline?

16:19:01 <libby> next chat maybe?

16:19:26 <bwm> bwm request chair take action to issue agenda 24 hours in advance of next chat

16:20:13 * GregElin Chair grumbles and agrees to action item.

16:20:14 <bwm> don't want to be heavy about process, but a little helps

16:20:27 <bwm> thank you

16:21:50 <GregElin> I have the next chat down as Jan 13, so we have time to do some work.

16:22:04 <libby> it seems there are substantial technical discussions to be had here...

16:22:11 <GregElin> I'm going to chat with monica tomorrow at 4pm GMT regarding UI's. Any interested party welcome.

16:22:46 <bryce> re: UI -- is the chat here?

16:22:57 <lyndonwong> re: jan 13, same time?

16:23:02 <GregElin> Yes. Bryce. (let me double check...)

16:23:23 <GregElin> Can we do 16:00 UTC Jan 13?

16:23:25 <bryce> .time gmt

16:23:26 <datum> Tue, 06 Jan 2004 16:23:25 GMT

16:24:00 <bryce> bryce is now known as bryce_atmeeting

16:24:13 <libby> sounds ok

16:25:14 <GregElin> .time utc

16:25:14 <datum> 2004-01-06T16:25:14Z

16:25:35 <GregElin> Next chat, jan 13, this channel, 16:00 UTC/GMT.

16:25:48 <libby> cool

16:25:54 <lyndonwong> thx!

16:26:11 <libby> yeah thanks greg, and bwm...

16:26:37 <GregElin> UI Chat for those interested, this channel, Jan 7 (tomorrow) 16:00 UTC/GMT.

16:27:08 <GregElin> Signing off. Thanks everyone! DanBri, please let us know if the list is a go!

16:27:16 <libby> bye greg

16:29:09 <GregElin> bye libby, all.

16:29:36 <libby> grep, did you say 1600 for next meet on 13th?

16:29:39 <GregElin> Actually. I lied. I'm not signing off. I'm just going to start paying attention to other tasks.

16:29:44 <libby> heh

16:29:58 <danbri> thanks all. Greg, will do.

16:30:01 <GregElin> aka grep

16:30:06 <libby> sorry!

16:30:11 <GregElin> yes. 1600.

16:30:16 <libby> cool

16:30:17 <GregElin> Naw. I like it.

16:30:18 <libby> ta

16:30:29 <libby> GrepElin

16:33:08 <libby> G:next meet [http://www.timeanddate.com/worldclock/fixedtime.html?day=13&month=1&year=2004&hour=16&min=0&sec=0&p1=0|2004-01-13, 1600 UTC]

16:33:08 <dc_rdfig> Added comment G16.

17:05:15 <tav|offline> tav|offline is now known as tav

17:29:50 <_kasei> _kasei is now known as kasei

17:42:27 <teefal> any links to namespace naming conventions or thoughts?

17:42:43 <teefal> for rdf

17:50:31 <karlcow> teefal, I guess it could be one of the topic of the Best Practices of Semantic Web group

17:50:58 <Ankh> ericP - I wondered how you were connected there for a minute :-)

18:12:05 <sandro> Hmmm. Anyone know of a good RDF checksum algorithm? (the tricky part is giving canonical names to the bnodes, I think.)

18:30:28 <QArl> QArl is now known as karlcow

18:32:45 <ericP> sandro, do you want to know if two documents are the same? or if a particular serialization of some RDF matches a hash?

18:33:25 <ericP> for the latter, you can pick any serialization in, say, ntriples, as long as you keep it around in that serialization

18:35:09 <ericP> but then, that isn't a very interesting case

18:35:40 <Davey> Davey is now known as D[a]vey

18:37:19 <sandro> No, I want a small handle to store to refer to an RDF graph -- if the graph changes, I want to know about it.

18:39:53 <ericP> i think you need some ordering of statements because otherwise you can do a sort of XOR attack

18:40:25 <ericP> plus the whole bnods prob...

18:41:29 <ericP> ("attack" implies someone wants to pretend that the graph didn't change, when, in fact, it did)

20:23:16 <bryce_atmeeting> bryce_atmeeting is now known as bryce

20:48:46 <setre> setre is now known as Anita-Skoglund

20:49:14 <Anita-Skoglund> Anita-Skoglund is now known as setre

21:18:13 <golbeck___> golbeck___ is now known as golbeck

21:40:05 <mimeme>http://visus.mit.edu/bibtex/0.1/

21:40:06 <dc_rdfig> O: http://visus.mit.edu/bibtex/0.1/ from mimeme

21:40:31 <mimeme> O: a short specifications document for bibTeX expressed in OWL

21:40:31 <dc_rdfig> Added comment O1.

21:41:10 <mimeme> O: I'd be glad to hear others comments, especially if you have ideas of an alternate bibliographic vocabulary to use.

21:41:10 <dc_rdfig> Added comment O2.

21:41:44 <mimeme> O: I chose bibTeX since it's what I'm familiar with, but it's old, and maybe there's something better out there that I don't know of.

21:41:44 <dc_rdfig> Added comment O3.

21:47:30 <evlist> mimeme: it's a detail and I may be too much picky, but I use to recommend using xs:token instead of xs:string except when whitespace must be preserved.

21:50:13 <mimeme> ahh, okay. I guess I'm a little rusty on the definition of xs:token. Does that mean that something like "foo bar" could be read as "foobar" if it were defined as xs:token versus xs:string?

21:50:18 <evlist> Also, <bibtex:hasNumber rdf:datatype="&xsd;nonNegativeInteger">S2</bibtex:hasNumber> looks weird.

21:50:27 <evlist> (S2 isn't an integer)

21:51:08 <mimeme> evlist: hehe, yeah, you caught that :-) That type of "number" was in the bibTeX example and I was going to change the OWL definition before it went live. gonna do that now.

21:51:23 <evlist> "foo bar" is a valid token that will be interpreted (for validation) as "foo bar"

21:51:40 <evlist> " foo bar " is also a valid token that will be interpreted (for validation) as "foo bar"

21:52:17 <evlist> That makes a difference if you want to restrict the datatype later on.

21:53:41 <mimeme> ahh, okay, I understand now. That's a good but subtle distinction, and it makes sense to me.

21:54:05 <evlist> the name (token) is very misleading!

21:54:47 <mimeme> indeed it is! I just glossed over that part of the XML Schema spec :-)

22:08:49 <DouweOsinga> Hi

22:13:21 <mortenf> hi there

22:14:00 <DouweOsinga> I have an interest in RDF, saw a reference to this channel somewhere and just wandered in

22:14:08 <mortenf> welcome! :)

22:14:39 <DouweOsinga> Thanks.

22:14:40 <bryce> welcome Douwe0singa. ;)

22:15:39 <DouweOsinga> If I would look for an ontology to describe information you'd find in a travel guide, where would I start?

22:16:03 <mortenf> try schemaweb.info

22:16:17 <bryce> try www.schemaweb.info ... hey, mortenf beat me to it!

22:16:31 <mortenf> also http://rdfschema.info/

22:16:37 <mortenf> :)

22:16:52 <mortenf> and http://www.mindswap.org/cgi-bin/2002/searchdamlont.pl

22:21:58 * mortenf curses at esw.w3.org - again...

22:22:24 <mortenf> hmm, or rather some faulty router on the way there.

22:23:35 <DouweOsinga> Thanks for the links

22:24:02 <mortenf> sure

22:28:46 <DouweOsinga> At http://www.world66.com we set up an open content travel site. I want to see if we can put it in a RDF structure to make it possible to synchronize data with other open content travel sites.

22:29:03 <mortenf> great idea

22:30:54 * mortenf browses the denmark section :)

22:31:41 <bryce> not sure if this is your area, Douwe, but the General Info on http://www.world66.com/about/copyright_policy looks wacky on winXP firebird 0.7

22:31:58 <bryce> it's off the page on the left

22:32:38 <DouweOsinga> It does, yes. The guy who worked on it has been stuggling with it. I don't know why really

22:33:06 <mortenf> same here in opera

22:33:08 <bryce> probably because he wants it to look nice, that's why. ;)

22:33:20 <DouweOsinga> Yes. We used to have a very simple layout

22:33:27 <DouweOsinga> Now it is much prettier

22:33:30 <DouweOsinga> And more buggy

22:33:33 <DouweOsinga> So it goes I guess

22:33:40 * DanC loses contact with irc.w3.org

22:35:30 <DouweOsinga> So has RDF been used to facilitate data exchange between open content sites?

22:35:47 <DouweOsinga> I think open source partly works because of the tools

22:36:11 <mortenf> i think some has been done re restaurant reviews, not sure though

22:36:16 <DouweOsinga> With open content, if I take something from one site, set it up on my own and start editting that, it almost automatically forks

22:36:29 <DouweOsinga> Restaurant reviews would be very interesting

22:36:34 <mortenf> true, it's probably not easy

22:36:52 <evlist> hasn't it be done by the ODP (Restaurant reviews)

22:38:07 <mortenf> you may want to have a look at http://esw.w3.org/topic/RestaurantRecommendation (can't see it myself though atm)

22:38:48 <chaalzzz> chaalzzz is now known as chaalsMEL

22:39:43 * chaalsMEL notes that some W3C servers are offline at the moment (and I think esw is one of those that doesn't have redundancy to cover downtime)

22:39:47 <evlist>http://chefmoz.dmoz.org/

22:39:47 <dc_rdfig> P: http://chefmoz.dmoz.org/ from evlist

22:40:06 <evlist> P:|Welcome to the ChefMoz dining guide beta!

22:40:06 <dc_rdfig> Titled item P.

22:40:24 <evlist> P:Yes, you can read and write restaurant reviews here, but you'll also find direct links to reviews on other sites

22:40:24 <dc_rdfig> Added comment P1.

22:40:47 <DouweOsinga> Great. That looks like a start.

22:41:23 <bryce> i found it at http://chefmoz.org/ (minus the .dmoz from prev link)

22:41:23 <DouweOsinga> I can see the w3 server btw

22:42:32 <evlist> P:=http://chefmoz.org/

22:42:32 <dc_rdfig> Replaced URL of P.

22:42:55 <evlist> Thanks dajobe for the tip :-)

22:43:08 <evlist> yes, the other site seems to be down

22:43:46 <DouweOsinga> Then there is of course always the thing about compatible share-alike licenses, but that needed working out anyway.

22:47:32 <deusx> deusx is now known as _deusx

22:50:01 <DouweOsinga> But say at world66 we implement the same scheme as at chefmoz.org and we sync our data...

22:51:01 <DouweOsinga> if they change their data and somebody change a record at our site, we still have a fork

22:52:46 <mortenf> well, given good identifiers (uris) you'd just have multiple "descriptions"

22:53:15 <mortenf> with some decent granularity, it shouldn't be a problem

22:53:29 <mortenf> i.e. don't have an entire page be a single record

22:53:36 <DouweOsinga> Yes.

22:53:54 <DouweOsinga> It would work in the current setup, unless the change the phone number or something

22:54:12 <mortenf> hmm, they use phones for ids?

22:54:16 <DouweOsinga> but if you have two sources that tell you a different phone number, you have a problem anyway

22:54:45 <DouweOsinga> No, I mean if chefmoz says the phone number is now this and a random visitor says it is that

22:55:10 <mortenf> with some date properties on the "properties" (that would then have to be full-blown resources) you could handle that

22:55:24 <DouweOsinga> Not a very good example as they could have two different phone numbers

22:56:10 <DouweOsinga> Yes. Actually it is no different from two visitors who enter two different phone numbers

23:03:00 <DouweOsinga> Okay, thanks for the help. I'll work on it for a bit and probably be back.

23:03:14 <mortenf> great, see you around

23:17:06 <chaalsMEL> Hmmm. What IS identity in the case of a restaurant?

23:17:34 <mortenf> i don't know the answer to that one :)

23:17:41 <chaalsMEL> change of address? change of name? change of chef?

23:18:24 <chaalsMEL> mortenf claims this is daml:sameAs that?

23:18:59 <mortenf> erh, i'm not sure how to parse that question?

23:19:06 <mortenf> ah, now i get it

23:19:13 <mortenf> yeah, that may be a way to do it

23:19:41 <mortenf> seems to be the answer quite often - add another level of indirection...

23:24:50 <chaalsMEL> There is a pretty well-agreed concept of what is a Person.

23:25:27 <chaalsMEL> So when someone says "this is the same person", it's unliekly that people will disagree about the truth of the statement.

23:25:46 <chaalsMEL> Which probably isn't the case for restaurants under any of the changed conditions I listed above.

23:26:47 <chaalsMEL> The semantic Web isn't yet too good at handling the idea that people disagree - price of having it built by decent people :-)

23:26:56 <mortenf> :)

23:27:28 <chaalsMEL> (let alone the possibility that people would delierately tell lies)

23:27:39 <mortenf> nooooooh

23:27:44 <chaalsMEL> s/delierately/deliberately/

23:31:10 <deltab> we'll invent electric monks soon enough :-)

23:33:55 <deltab>http://www.mat.upm.es/~jcm/adams.html

23:33:55 <dc_rdfig> Q: http://www.mat.upm.es/~jcm/adams.html from deltab

23:34:04 <deltab> Q:|The Electric Monk

23:34:04 <dc_rdfig> Titled item Q.