Semantic Web Interest Group IRC Chat Logs for 2003-03-26

00:00:14 <DanC> I:TAG issue namespaceDocument-8

00:00:15 <dc_rdfig> Added comment I5.

00:00:25 <DanC> I:|TAG issue namespaceDocument-8

00:00:25 <dc_rdfig> Titled item I.

00:00:29 <DanC> I5:""

00:00:29 <dc_rdfig> Deleted comment I5.

00:15:11 <danbri_> <DanC> the most straightforward way to talk with all those parties, danbri, is to call for review of a Proposed Rec, no?

00:15:11 <danbri_> <DanC> or perhaps talk with them in CR.

00:15:19 <danbri_> or Last Call?

00:15:28 <danbri_> never too late to talk...?

00:15:34 <danbri_> <DanC> Of the people who are using it, perhaps you're right, none of them gave de-re/de-dicto any thought. they just ran the code and walked away fat-dumb-and-happy.

00:15:38 <danbri_> I think that's likely

00:26:28 <danbri_>http://www.itworld.com/nl/ebiz_ent/03182003/

00:26:29 <dc_rdfig> A: http://www.itworld.com/nl/ebiz_ent/03182003/ from danbri_

00:26:55 <danbri_> A:RDF and other monkey wrenches, Sean McGrath, 2003-03-18

00:26:55 <dc_rdfig> Added comment A1.

00:26:59 <danbri_> A:|RDF and other monkey wrenches, Sean McGrath, 2003-03-18

00:26:59 <dc_rdfig> Titled item A.

00:27:08 <danbri_> A1:"Somebody, I think it was Adam Bosworth of BEA, once said that every layer of abstraction costs you 50% of your audience. Or words to that effect."

00:27:08 <dc_rdfig> Replaced comment A1.

00:27:41 <danbri_> A:"... Abstraction creates a high priest environment in which only a few can ever hope to really understand the "vision" buried in all the abstraction. In the hands of the chosen few, the abstractions are a precision tool wielded to powerful effect. In the hands of the other 94%, the tool is more like a monkey wrench. A tool that can be used for every job but is the *wrong* tool for every job."

00:27:41 <dc_rdfig> Added comment A2.

00:27:56 <danbri_> A:"I fear that RDF - the uber-abstraction that underlies the Semantic Web - is one such monkey wrench. Given the undoubted benefits to all of us, if Tim Berners Lee's vision of the Semantic Web[2] becomes a reality, how can we make the abstractions it depends on, usable by the 94% of the world's system developers? The population for whom the abstractions are a step too far for them to comfortably follow?"

00:27:57 <dc_rdfig> Added comment A3.

00:28:24 <danbri_> A:I've had similar concerns: it's easy for something that's applicable to all tasks to be ideally suited to none of them.

00:28:24 <dc_rdfig> Added comment A4.

00:29:03 <danbri_> A:Nearby: [http://www.oreillynet.com/pub/wlg/2961|Uche Ogbuji's reply].

00:29:03 <dc_rdfig> Added comment A5.

00:30:39 <bijan> A:I'm not finding this view of abstraction, and the argument from the costs thereof, persuasive.

00:30:40 <dc_rdfig> Added comment A6.

00:31:20 <bijan> A: As Russell pointed out, we start our understanding from "middle sized dry goods"...which are abstractions (at least, relative to fundemental particles)!

00:31:20 <dc_rdfig> Added comment A7.

00:31:46 <danbri_> A:Yup, I wouldn't couch it in terms of abstraction laters, more in terms of whether a technology is generalist vs attempts to solve a specific family of problems.

00:31:46 <dc_rdfig> Added comment A8.

00:31:56 <danbri_> A8:Yup, I wouldn't couch it in terms of abstraction layers, more in terms of whether a technology is generalist vs attempts to solve a specific family of problems.

00:31:56 <dc_rdfig> Replaced comment A8.

00:32:30 <danbri_> A:RDF is pretty generalist, hmm but so is XML, I'm not even convincing myself here. Scrap that argument!

00:32:30 <dc_rdfig> Added comment A9.

00:32:30 <bijan> A:Not even clear about that, though that's a useful point. XML seems driven by the generalist desire.

00:32:31 <dc_rdfig> Added comment A10.

00:32:35 <bijan> HEhehehehhehe

00:32:39 <danbri_> A:Quite

00:32:40 <dc_rdfig> Added comment A11.

00:33:06 <soccos|away> soccos|away is now known as soccos

00:33:11 <bijan> A:Reuse is, at some level of abstraction, about abstraction (or generality; these aren't equivalent).

00:33:11 <dc_rdfig> Added comment A12.

00:33:16 * danbri_ grins at Bijan, heads off to look for some middle-sized edible goods

00:33:21 <bijan> Hehehehe

00:33:41 <danbri_> (I'm reverse engineering the recipie for cup-a-soup...)

00:35:51 <bitsko> A:Uche is wrong on one very important point: there's a lot of people that get relational databases and code that don't get RDF.

00:35:51 <dc_rdfig> Added comment A13.

00:38:47 <bitsko> A:One need only look at the RDF mailing lists vs. the RSS-DEV mailing list to see the numerical difference in people who "get" RDF. Most of the RSS-DEV folk really want to "get" RDF, but don't, as evidenced by the near zero usage of RDF tools for RSS.

00:38:48 <dc_rdfig> Added comment A14.

00:40:21 <danbri_> A:I think RDF query has something to offer on that front. I realise that "we just need one more layer of facilitating technology" can (and should) ring alarm bells. But...

00:40:22 <dc_rdfig> Added comment A15.

00:40:58 <danbri_> A:...many RDF query tools offer a reassuringly SQL-like mode of interaction. Send a textual query, get back a table of results, with a row for each 'hit'.

00:40:58 <dc_rdfig> Added comment A16.

00:41:34 <bijan> A:And, let's add the fact taht RDBMS have had, what, 30 years of effort, tool building, graduate and undergrad and industry course, etc. etc. etc.

00:41:34 <dc_rdfig> Added comment A17.

00:41:43 <bijan> A: And loads of folks STILL don't get it!

00:41:43 <dc_rdfig> Added comment A18.

00:43:13 <danbri_>http://lists.w3.org/Archives/Public/www-zig/2003Mar/0022.html

00:43:13 <dc_rdfig> B: http://lists.w3.org/Archives/Public/www-zig/2003Mar/0022.html from danbri_

00:43:34 <danbri_> B:|Discussion of XML record syntaxes in the Z39.50 Implementors Group

00:43:34 <dc_rdfig> Titled item B.

00:44:54 <danbri_> B:I've not read the whole thread, but discussion touches on XML schema(s) for Dublin Core, 404s at namespace URIs, etc.

00:44:54 <dc_rdfig> Added comment B1.

00:53:33 <soccos> Ooops, I killed dc_rdfig

01:08:35 <danbri_> urk!

01:19:53 <soccos> soccos is now known as soccos|away

01:34:33 <tav> tav is now known as tav|offline

02:15:42 <sandro> Hrm. Any owlers around? Am I understanding correctly that an owl ontology cannot have rdfs:comment values about classes or properties?

02:16:07 <sandro> I mean OWL DL and OWL Lite. It would be okay in OWL Full, I guess.

02:16:57 * danbri_ heard something in that vein

02:17:40 <sandro> Eeek.

02:17:57 <sandro> It makes perfect sense, but..... ouch.

02:18:11 * danbri_ ignoring non-Full OWL, mostly

02:18:39 <sandro> Really? I anyone expected to implement OWL Full?

02:19:27 <sandro> Maybe one can write OWL-DL and OWL-Lite extractors from an OWL-Full ontology....

02:21:19 <danbri_> I'm not implementing a reasoner, I'm writing reasonable things in it

02:22:04 <sandro> kind of like programming in pseudo-code, I guess.

02:22:17 <sandro> I prefer to have the computer able to run it, myself. :-]

03:18:31 <DanC> I don't bother with the OWL lite/DL constraints much. I just write horn clause axioms about the terms in the OWL vocabulary and use them.

03:19:33 <DanC> I don't think much of what I do fits in OWL/DL

03:35:32 * danbri_ saves http://www.w3.org/2001/12/rubyrdf/pack/tests/ts_rdftests.rb

03:35:48 <danbri_> finally hooking up ad-hoc tests to Ruby TestUnit machinery

03:35:56 <danbri_> not much yet, but I know what to do now

03:35:58 * danbri_ -> zzz

06:28:37 <oc|gone> oc|gone is now known as oc

07:46:58 <oc> oc is now known as oc|gone

07:51:40 <sbp-> sbp- is now known as sbp

08:38:09 <mariyo> tansaku: hi, from the looks of your ip, we must live in the same neighborhood. nice to meet you. are you working on any rdf projects in japan?

10:32:17 <tansaku> hi mariyo

10:32:49 <tansaku> I'm not sure if what I just typed came through before I got disconnected

10:33:01 <tansaku> anyway, apologies if I reiterate

10:33:04 <tav> tansaku: nope

10:33:18 <tav> (haven't seen you type anything)

10:33:21 <tansaku> aha

10:33:29 <tansaku> then this is for mariyo

10:33:33 <tansaku> I work on NeuroGrid http://www.neurogrid.net

10:33:38 <tansaku> it uses rdf

10:33:43 <tansaku> well it uses triples

10:33:47 <tansaku> mariyo: nice to meet you too - sorry for the lag

10:33:58 <tansaku> tav: thanks

10:34:24 <tav> woo! 2 hour lag! =)

10:35:03 <tav> anyways, congrats btw. don't think i congratulated you back in august.

10:35:11 <tav> (september? october?)

10:35:14 <tansaku> it's called a sandwich break :-)

10:35:29 <tansaku> August and September (I got married twice)

10:35:35 <tansaku> thanks for the congrat

10:35:50 * tav beams at the 6 month lag

10:36:39 <tansaku> you came out of hibernation then?

10:36:46 <tav> yes =)

10:36:56 <tansaku> good sleep?

10:37:14 <tav> very. good dreams too.

10:37:20 <tansaku> excellent

10:37:29 <tav> btw, what do you think of http://tav.espians.com/pedipeace/1 ?

10:37:59 <tav> it's (imo) a viable proposal to stop this war (or at least our participation in it).

10:38:28 <tav> there is also http://tav.espians.com/pedipeace/2 -- a followup

10:40:06 <tansaku> sounds good

10:40:37 <tansaku> unfortunately I'm working against a deadline and so can't take you up on specific points right now

10:41:17 <tav> sure. but if any thoughts strik you, please let me know. tav@espians.com as usual.

10:41:24 <tav> ^strike

10:41:33 <tansaku> I agree there has to be a better way to achieve peace than just marching

10:41:40 <tansaku> I'll be in touch

10:42:08 <tav> need to fly myself. you wake up from hibernation and find out that a lot of work has to be done =)

10:52:40 <tansaku> :)

12:45:31 <mariyo> tansaku: thanks for the info. i will look into it. i was at the ipsj conference today in hachiyoji. seems to be lots of grad students working on ontology systems with rdfs, daml+oil. U. of Hokkaido work was really interesting saito-san, tanaka-san et. al. described integration of WordNet and TGN ontologies. I will be there tomorrow presenting with SC 34. Do you know anyone who will be going? seems to be lots of talks on semantic web ser

12:48:19 <danbri> 'our participation' == UK?

12:52:02 <mariyo> danbri: ok, if you want to come tomorrow you are most welcome :)

12:52:32 <danbri> I was replying to tav> btw, what do you think of http://tav.espians.com/pedipeace/1 ?

12:52:32 <danbri> [05:39] <tav> it's (imo) a viable proposal to stop this war (or at least our participation in it).

12:52:36 * mariyo guesses that tansaku lives down the block from me.

12:52:42 <danbri> before I realised this was #rdfig and not #war-news

12:52:55 * danbri doesn't, unfortunately!

12:54:14 <danbri> The Wordnet<->TGN integration does sound very interesting. Do you have any further info on that?

12:54:51 <mariyo> danbri: i have the copy of the paper but it's in japanese and the prgram on the web is also japanese

12:55:25 <mariyo> do you know the research going on at the U. of hokkaido?

12:57:05 <mariyo> yes, you were cited and there were references to rdl and dql he also talked about the Sesame European ITS ONTOknowledge

12:57:52 <mariyo> i can see if i can get an abstract in English tomorrow.

12:58:18 <mariyo> rdl/rql that is

12:59:33 <danbri> mariyo, any urls and (ideally) translations would be much appreciated

12:59:59 <danbri> But I'm also keen to make sure the english-speaking rdf crowd know this work is going on, even if there is no translation yet

13:00:16 <danbri> I didn't know about the U. of hokkaido work, for example

13:00:52 <mariyo> we have got to get these brilliant japanese researchers to publish in english sometimes. this paper would be great for the sematic web conference.

13:01:36 <danbri> It's sad that that's the way things work, but I fear you're right. Unless work is at least summarised in English, it misses out on a huge audience...

13:01:37 <mariyo> there was another on Fuzzy logic in OWL

13:02:29 <mariyo> yeah, we only had about 12 people in the room and 4 were the presenters and another was the chair.

13:03:20 <mariyo> from Tohoku University.

13:05:22 <mariyo> i guessed that most of you only know the people at Keio since they are in the W3C semantic web group. i will get more info from the Hokkaido group since they will come to the SC34 presentation tomorrow. oops, it's getting late and i need to check my demo. i'll report back.

13:05:37 <danbri> OK, thanks, that'd be great

13:06:04 <mariyo> good, see you soon and bye for now.

13:06:10 <danbri> I hear there's some interest in Japan in "PICS-2" efforts, content filtering etc using RDF/XML... any news on that?

13:06:12 <danbri> ok bye!

13:07:05 <mariyo> saw something like that on the program will check again tomorrow. thanks for the tip.

13:19:58 <tav> heh

13:20:19 <tav> danbri: just realised that i had discussed with tansaku in here. thought i was talking to him in #infoanarchy

13:20:30 <danbri> heh, no worries

13:20:33 <tav> anyways, will reply to you in #war-news

13:21:08 * danbri considers 'semweb for peace' (and ...for evil) in scope for #rdfig, fwiw

13:21:57 <tav> i intend to use some p2p, foaf tech for that sms thing btw

13:22:12 <JibberJim> sms thing? /me is working on an SMS thing at the moment...

13:22:15 <tav> (the sms thing to organise the people)

13:22:26 * danbri intrigued

13:22:35 * tav points JibberJim to http://tav.espians.com/pedipeace/1

13:22:50 <JibberJim> SMS organising demos - have you seen the work done by Amnesty in .NL ?

13:23:03 <tav> JibberJim: nope.. enlighten me

13:23:34 <JibberJim> at a Mobile conference, there was a presentation on it...

13:23:40 <danbri>http://www.smartmobs.com/

13:23:41 <dc_rdfig> C: http://www.smartmobs.com/ from danbri

13:23:55 <danbri> C:|Smart Mobs: The Next Social Revolution, by Howard Rheingold

13:23:56 <dc_rdfig> Titled item C.

13:24:20 <danbri> C:Website accompanying the book. Couched in terms of mobile devices rather than the (Semantic) Web...

13:24:20 <dc_rdfig> Added comment C1.

13:24:33 <danbri> C:...but raises questions about how the SW might transform activism.

13:24:34 <dc_rdfig> Added comment C2.

13:24:57 <danbri> C:He's very taken with the way SMS has been used to organise protests, for example.

13:24:57 <dc_rdfig> Added comment C3.

13:25:05 <tav> someone want to buy me that book? ;p

13:25:15 <JibberJim> Yeah (I'm looking for the presentation... we recorded it for the web.)

13:25:25 <danbri> C:I found that suprising, since my phone barely worked during the 15 Feb 2002 anti-war demo in London.

13:25:25 <dc_rdfig> Added comment C4.

13:25:35 <tav> danbri: oh?

13:26:00 <danbri> C:Orange, fwiw. All the SMS messages from that day arrived in batch, 2 days later, as I left London.

13:26:00 <dc_rdfig> Added comment C5.

13:26:03 <tav> i found sms'ing on 15feb to be the only thing you could do... (couldn't really talk with all the background...)

13:26:18 <bitsko> am I just paranoid to think that using SMS to organize protests must be one of the most technically stupid things to do?

13:26:30 <danbri> Re privacy?

13:26:34 <tav> bitsko: why? ;p

13:26:41 <bitsko> re. government snooping

13:26:44 <tav> (privacy, centralisation...?)

13:27:13 <tav> ah, this is where i assume that everyone is being monitored in means that they cannot even realise and therefore suggest we just be transparent about everything

13:27:40 <tav> besides, transparency is empowering

13:27:45 * danbri nods, sometimes

13:27:50 <JibberJim> Hmm, I can't find the presentation, maybe it was one of the people, who didn't let us re-broadcast it.

13:28:17 <tav> the centralisation thing *is* worrying... they could just tell the network operators to prevent provision of service

13:28:34 <tav> this is where i'd like to leverage the community wireless projects (in london)

13:28:38 <danbri> C:There's a [http://www.smartmobs.com/book/book_bib_ch_8_0_30.html|detailed bibliography] online too.

13:28:39 <dc_rdfig> Added comment C6.

13:28:48 <bitsko> I recall reading a short story (later turned into a book?) that used a smartmob technique called "walkabouts"

13:28:53 <danbri> tav, are you based in london?

13:28:56 <tav> danbri: yes

13:29:08 <tav> bitsko: tell me more

13:29:24 <danbri>http://www.independent.co.uk/story.jsp?story=51748

13:29:24 <dc_rdfig> D: http://www.independent.co.uk/story.jsp?story=51748 from danbri

13:29:24 <tav> JibberJim: could you give me an overview if the webcast cannot be found pls?

13:29:34 <danbri> D:|Home > Independent

13:29:34 <danbri> Across the site

13:29:34 <danbri> Also from this section

13:29:34 <danbri> The best writing. analysis and comment

13:29:34 <dc_rdfig> Titled item D.

13:29:34 <danbri> Pay up

13:29:34 <danbri> Saddam puts Iraq on war footing

13:29:36 <danbri> UN weapons team searches Saddam palace

13:29:38 <danbri> The TXT MSG Revolution

13:29:50 <danbri> ugh, cut/pasting from independent site screwed up

13:29:55 <danbri> D:|The TXT MSG Revolution

13:29:55 <dc_rdfig> Titled item D.

13:29:59 <JibberJim> - http://www.showcaster.com/info.asp?venue=PentonSMS02&presentation=2567354&f=1048685283121

13:30:09 <danbri> D:"In revolutions people used to say, 'keep your powder dry.' Now they say, 'keep your cellphone charged'"

13:30:10 <dc_rdfig> Added comment D1.

13:30:21 <tav> danbri: that gives me a 404 -- the independed

13:30:31 <tav> ^independent

13:30:33 <danbri> I think their site is having troubles

13:30:36 <JibberJim> I'm not sure I remember the jist, just remembered the presentation (not even sure that's it, but it's related) It'll be another presentation in that conference.

13:30:43 <tav> ah

13:30:51 <danbri> HTTP 403.9 - Access Forbidden: Too many users are connected

13:30:51 <danbri> Internet Information Services

13:31:10 <bitsko> in the story, 10s-100s-1000s of people would quietly gather in locations "spontaneously" (called walkabouts, as when any one person was questioned they were "just walking about"), chat for a bit (usually issue based), then disperse

13:31:29 <JibberJim> The general idea though was that Amnesty.nl had done lots of work on collecting mobile numbers from various sources, and then using SMS to control how it worked.

13:31:47 <danbri> D:The [http://news.independent.co.uk/world/pacific_rim/story.jsp?story=51748|full story] is on a different server.

13:31:47 <dc_rdfig> Added comment D2.

13:31:59 <tav> bitsko: oooh! this is exactly what i'm talking about. but with a bit more direct action emphasis

13:33:37 <bitsko> the story/book had a more interesting angle that I'd been interested in called a "decision duel", using a directed brainstorming technique, similar to "mind mapping"

13:34:24 <tav> which book is this? smartmobs?

13:34:37 <bitsko> David's Sling, Marc Stiegler, Baen Books, 1988, pb, 346 pp

13:34:38 <danbri> no

13:34:53 * bitsko praises google yet again

13:36:06 * tav adds book #373 to his wishlist =(

13:36:26 <tav> how did decision duels take place?

13:36:46 <bitsko> the "sling" in question ends up being "smart" metal rods capable of destroying nuclear missiles in their bunkers when dropped from orbit. decision duels and walkabouts are just part of the story line

13:39:36 <bitsko> "the potential of public-access computer networks to transform politics." -- ESR on David's Sling

13:40:03 <tav> basically, i want to give people the ability to define their "friends" (foaf, etc.), and then through some simple trust metrics (see raph's thesis) evaluate what's happening where, so that they can meet up at particular crossroads and discuss specific issues

13:42:35 * bitsko notes Marc Stiegler participates in Xanadu/Sunless Sea http://www.sunless-sea.net/forum

13:43:49 <danbri>http://umbriel.dcs.gla.ac.uk/NeSC/general/esi/events/163/index.html

13:43:49 <dc_rdfig> E: http://umbriel.dcs.gla.ac.uk/NeSC/general/esi/events/163/index.html from danbri

13:43:58 <danbri> E:|Schemas and Ontologies: Building a Semantic Infrastructure for the GRID and Digital Libraries

13:43:59 <dc_rdfig> Titled item E.

13:44:16 <danbri> E:"(...) The role of schema registries will be considered, in particular looking at recent initiatives within the digital library world such as the DCMI registry, and SCHEMAS registry. Availability of appropriate schema creation tools is essential to complement the registration process, as will be illustrated by a demonstration of the recently developed Metadata for Education Group (MEG) schema creation and registration tool."

13:44:16 <dc_rdfig> Added comment E1.

13:44:44 * danbri wonders if dajobe and shellac know more about this (given their involvement in MEG)

13:46:04 <tav> anyways, got to shoot. please drop me a line -- tav@espians.com -- about any further thoughts/things to look into

14:32:37 * bitsko randomly notes that E also abitrarily overloads addition and concatenation <sigh>

15:18:03 <danbri_> BLURB:RDF query testing puzzle...

15:18:04 <dc_rdfig> F: RDF query testing puzzle... from danbri_

15:18:24 <danbri_> F:I want to do automated RDF query testing for my Ruby stuff. And have a design choice I'm puzzling over.

15:18:24 <dc_rdfig> Added comment F1.

15:18:34 * sandro wonders about making a public-webont-comment on the idea that OWL validation has two forms, and the "lint" form is equivalent to imposing CWA and UNA on the deductive closure of the ontologies.

15:18:48 <danbri_> F:Either I can write a script that reads the RDF manifest and does code-generation of Ruby test scripts.

15:18:49 <dc_rdfig> Added comment F2.

15:19:35 <danbri_> F:Or I can write a smarter generic test script that reads the manifest at runtime and just loops through each test, with assertions based on what it's just read.

15:19:35 <dc_rdfig> Added comment F3.

15:19:51 <danbri_> F:Previously I went the code-gen route, but now I'm not so sure. Advice welcomed...

15:19:52 <dc_rdfig> Added comment F4.

15:20:09 <danbri_> (does that make sense?)

15:22:30 <bitsko> F:Reading the manifest at runtime sounds simpler

15:22:31 <dc_rdfig> Added comment F5.

15:23:22 <bitsko> F:also fewer steps for anyone who comes after to follow

15:23:23 <dc_rdfig> Added comment F6.

15:24:22 <bitsko> danbri_: what does the manifest and tests look like?

15:24:31 <danbri_> F:Now I've got an RDF parser more reliably integrated, I'm inclined to agree. I remember now that one issue was need to test query stuff in an environment without RDF/XML parsers

15:24:31 <dc_rdfig> Added comment F7.

15:24:48 <Morbus> anyone experimented with parsing RDF with regexp?

15:24:49 * Morbus ducks.

15:25:25 <bitsko> isn't that how most RSS parsers work

15:25:27 * bitsko ducks now.

15:25:51 <danbri_> F:Context: [http://www.w3.org/2003/03/rdfqr-tests/|rdf query testcases] work. See [http://www.w3.org/2001/12/rubyrdf/pack/tests/|RubyRDF tests dir] (messy) and [http://lists.w3.org/Archives/Public/www-rdf-rules/2003Jan/0010.html|brief explanation].

15:25:53 <dc_rdfig> Added comment F8.

15:26:12 <Morbus> bitsko: heh, heh.

15:26:19 <Morbus> if you believe pilgrim, sure.

15:26:43 <DanC> argh! type-ahead-find gets in the way of filling out forms, in galeon. Clues on how to stop it?

15:27:00 <Morbus> you could turn off type ahead, or restrict it to links only.

15:27:12 <Morbus> wait. that's moz. dunno how much galeon is like moz.

15:27:29 <bitsko> probably the same in that case

15:27:57 <DanC> how to turn it off?

15:28:16 <danbri_> F:See also [http://lists.w3.org/Archives/Public/www-rdf-rules/2003Mar/0013.html|Alberto's work on an improved manifest schema] and http://www.w3.org/2003/03/rdfqr-tests/recording-query-results.html

15:28:16 <dc_rdfig> Added comment F9.

15:28:41 <danbri_> F9:See also [http://lists.w3.org/Archives/Public/www-rdf-rules/2003Mar/0013.html|Alberto's work on an improved manifest schema] and [http://www.w3.org/2003/03/rdfqr-tests/recording-query-results.html|Andy's draft of a resultset testing format].

15:28:41 <dc_rdfig> Replaced comment F9.

15:29:01 <Morbus> in Moz, it's Edit > Prefs > Advanced > Keyboard Navigation > "Find As You Type"

15:29:57 <bitsko> I don't see a keyboard option in Galeon's settings

15:36:01 <sandro> Has someone done a Venn Diagram of which rdf/rdfs/owl terms and usage styles fit into RDFS, OWL Lite, OWL DL, and OWL Full?

15:36:41 <danbri_> not that I know of

15:36:45 <danbri_> good idea though

15:38:56 <sandro> any venn tools better than "dia"?

15:43:29 <sandro> maybe a table of features vs the four ontology languages would be easier.

15:44:26 <danbri_> a table would at least do the prep work for a prettier rendering...

16:05:19 <arnarl> hi

16:24:53 * eikeon thinks there is an rdfig calendar meeting in 34 minutes?

16:28:35 * libby is in a meeting until then, would appreciate people chumping the agenda.....

16:28:46 <libby> danbri_ maybe...?

16:29:01 <danbri_> how do you like the agenda chumped? one per item?

16:29:10 * danbri_ wished zakim bot hung out here

16:29:40 <libby> there's only one I think :)

16:35:31 <danbri_> ./qt.rb

16:35:31 <danbri_> Loaded suite ./qt

16:35:31 <danbri_> Started

16:35:31 <danbri_> Query: examples/squish/wot

16:35:31 <danbri_> Query: examples/squish/test0

16:35:32 <danbri_> Query: examples/squish/test1

16:35:34 <danbri_> Query: examples/squish/test3

16:35:36 <danbri_> Query: examples/squish/test2

16:35:40 <danbri_> Query: examples/squish/test0c

16:35:42 <danbri_> Query: examples/squish/test4

16:35:44 <danbri_> Query: examples/squish/test1-bogusclause

16:35:46 <danbri_> Query: examples/squish/test0b

16:35:48 <danbri_> .

16:35:50 <danbri_> Finished in 0.137315 seconds.

16:35:52 <danbri_> 1 tests, 18 assertions, 0 failures, 0 errors

16:35:54 * danbri_ smiles

16:35:56 <danbri_> only testing squish syntax, for now

16:35:58 <danbri_> and not even doing that properly, but i know how to make incremental progress

16:36:52 <danbri_> I'll chump the agenda in a few mins

16:41:22 <libby> thanks man

16:43:51 <danbri_>http://www.methodize.org/nntprss

16:43:52 <dc_rdfig> G: http://www.methodize.org/nntprss from danbri_

16:44:03 <danbri_> G:|NNTP/RSS

16:44:03 <dc_rdfig> Titled item G.

16:44:12 <danbri_> G:"Bridging the worlds of NNTP clients and RSS feeds, nntp//rss is an application that will enable you to use your existing favorite NNTP newsreader to read your information channels."

16:44:12 <dc_rdfig> Added comment G1.

16:44:41 <danbri_> G:Popped up in context of discussing ways of keeping up with various interesting IRC chump'd feeds...

16:44:41 <dc_rdfig> Added comment G2.

16:57:40 <DanC> BLURB: inconsistencies in meeting records, an OWL use case

16:57:41 <dc_rdfig> H: inconsistencies in meeting records, an OWL use case from DanC

16:58:01 <danbri_>http://lists.w3.org/Archives/Public/www-rdf-calendar/2003Mar/0038.html

16:58:01 <dc_rdfig> I: http://lists.w3.org/Archives/Public/www-rdf-calendar/2003Mar/0038.html from danbri_

16:58:11 <DanC> H:suppose, for the sake of meeting records, surname is an owl:FunctionalProperty

16:58:11 <dc_rdfig> Added comment H1.

16:58:22 <danbri_> I:|RDF calendar meeting Wednesday March 26th 2003, 17:00 UTC

16:58:22 <dc_rdfig> Titled item I.

16:58:37 <danbri_> I:See [http://lists.w3.org/Archives/Public/www-rdf-calendar/2003Mar/0038.html|agenda announcement]

16:58:38 <dc_rdfig> Added comment I1.

16:58:39 <DanC> H:and every action owner has to be an attendee of the meeting

16:58:40 <dc_rdfig> Added comment H2.

16:59:05 <danbri> danbri is now known as danb_lap

16:59:12 <DanC> H:and suppose the attendees are [ owl:oneOf ( [:surname "Smith"] [:surname "Jones"])]

16:59:12 <dc_rdfig> Added comment H3.

16:59:24 <DanC> H:and we see ACTION: Williams.

16:59:24 <dc_rdfig> Added comment H4.

16:59:41 <DanC> H:there's an inconsistency there that OWL tools can find, yes?

16:59:42 <dc_rdfig> Added comment H5.

16:59:48 <DanC> H:I hope to do this with cwm today.

16:59:48 <dc_rdfig> Added comment H6.

17:00:01 <danbri_> danbri_ has changed the topic to: RDF calendar meeting - 15:00 UTC for 90 mins

17:00:19 <danbri_> danbri_ has changed the topic to: RDF calendar meeting - 15:00 UTC for 90 mins - http://rdfig.xmlhack.com/

17:00:27 * DanC just typed /topic RDF calendar meeting - 15:00 UTC for 90 mins http://rdfig.xmlhack.com/

17:00:35 <DanC> :)

17:00:48 <danbri_> heh

17:00:49 * DanC will be right there... gotta clear tax accounting clutter off desk...

17:01:30 <danbri_> I:Agenda: "We will spend the time collaboratively working on documentation for the

17:01:30 <danbri_> project in the ESW wiki."

17:01:30 <dc_rdfig> Added comment I2.

17:01:45 <danbri_> I2:Agenda: "We will spend the time collaboratively working on documentation for the project in the ESW wiki."

17:01:45 <dc_rdfig> Replaced comment I2.

17:01:58 <libby> thanks danbri_ :)

17:02:07 <danbri_> I:See [http://lists.w3.org/Archives/Public/www-rdf-calendar/2003Mar/0036.html|DanC's message] for elaboration.

17:02:07 <dc_rdfig> Added comment I3.

17:02:13 <libby> ------ RDF calendar meeting start --------

17:02:52 <libby> does anyone have any news or other agenda item,s they want to say anything about?

17:03:02 * danbri_ thanks non-calendar folks for 90 minutes of patience

17:03:28 <libby> I've *nearly* turned the www2003 data into RDF

17:03:31 <libby> ...but not quite

17:03:47 * DanC wonders if freenet has policies about reserving channels for meetings

17:04:12 <danbri_> semi-related news: I've been working on packaging RubyRDF, including RDF parser, tests, query stuff. It is looking more stable, so I hope to find a little calendar app to test it with.

17:04:27 <danbri_> ...I think I'll focus on the small business / opening hours scenario,

17:04:34 <danbri_> .

17:04:52 <libby> that'd be cool

17:04:58 * danbri_ doesn't think freenode care at that level of detail, so long as folk are reasonable civilised

17:05:25 <danbri_> I liked DanC's list of 'topics that come to mind',

17:05:27 <danbri_> - how to convert .ics to .rdf

17:05:27 <danbri_> - how to convert .rdf to .ics

17:05:27 <danbri_> - how to use rules to convert itineraries to calendars

17:05:27 <danbri_> - how to scrape calendars from HTML documents

17:05:41 <libby> apps are tricky. I just spent 2 hours explaining how great the SW will be to people...but I didnt really have anythign to show them that'd grab them at that level

17:06:10 <DanC> re how to scrape calendars from HTML documents ... did you just say you did that for WWW2003, libby?

17:06:16 <libby> yeah :)

17:06:29 <libby> I asked them for the data, but no reasponse (probbaly askedd wrong person)

17:06:52 <libby> I probably did it wrong though. I didnt use an dom-based stuff, and probably should have

17:07:02 <libby> intead, I used a nasty perl regex

17:07:04 <danbri_> whatever, you got the data out, right?

17:07:11 * DanC noodles on the relationship between http://www.w3.org/2002/12/cal/ and http://esw.w3.org/t/view/ESW/RdfCalendar

17:07:13 <libby> yep

17:07:24 <danbri_> can we see?

17:07:33 <libby> I'll make it public as soon as I've checked it

17:07:39 <DanC> hmm.. DontWorryBeCrappy

17:07:53 <libby> I've got the content but not validated it yet....

17:08:16 <DanC> "This is a Wiki space "... let's not do that, OK?

17:08:23 <DanC> (in RdfCalendar)

17:08:55 <DanC> let's use the wiki genre... start with an OpeningStatement

17:09:06 <DanC> discussion goes below the line (cf PPR:WikiMaster)

17:09:07 <danbri_> i'd like some intro-to-wikiness to link to

17:09:14 * eikeon modified http://eikster.com/2003/events/ to show upcoming events in chronological order... also can take a start query parameter: http://eikster.com/2003/events/?start=2003-03

17:09:26 <danbri_> okay... i'm not too schooled in the art and science of Wiki

17:09:32 <DanC> eikon, is that KB world-writeable?

17:09:38 * danbri_ mostly uses them for personal scrawls

17:09:44 <libby> neat, eikeon

17:10:00 <libby> yeah, I'm very interested in an events crawler

17:10:14 <eikeon> DanC: Yes... by way of submitting URLs to RDF/XML.

17:10:26 <danbri_> eikeon, yes what danc asked... is this the same database as your foaf crawler?

17:10:35 * DanC is torn between BLURB: Events crawler and RdfCalendarEventsCrawler wiki node

17:10:46 <danbri_> ie. if I point to some event descriptions from my foaf doc, via rdfs:seeAlso will it show up on your site?

17:11:09 <eikeon> danbri_: Yep... all my sites are currently comming from the same store.

17:11:11 * DanC suppses BLURB is in order, as record of the meeting

17:11:13 * danbri_ shares DanC's dillema... let's go with IRC for now

17:11:18 <danbri_> yup.

17:11:31 * DanC liked includding the letter in the blurb...

17:11:48 <libby> yeah, bit fiddly but bettter later

17:12:13 <DanC>http://eikster.com/2003/events/

17:12:14 <dc_rdfig> J: http://eikster.com/2003/events/ from DanC

17:12:23 <libby> danbri_, could we have typedseealsos? what type would it be....?

17:12:41 * DanC looks for a <title> at http://eikster.com/2003/events/ , loses

17:13:08 <danbri_> So are we talking about events crawler apps for a while?

17:13:18 <DanC> two events I was thinking about today: ISWC, and the paper deadline, and ExML likewise

17:13:28 * eikeon quick adds a title... upps!

17:13:36 <danbri_> Regarding typed seeAlsos, yes...

17:13:47 <danbri_> BLURB:On the use of rdfs:seeAlso for RDF harvesting

17:13:47 <dc_rdfig> K: On the use of rdfs:seeAlso for RDF harvesting from danbri_

17:13:49 <DanC> evidently so, Danbri... I hope it ends up as a wiki node... EventCrawling?

17:14:18 * DanC decided there was no need to blurb the general concept, when eikon's specific implementation would do. but doesn't mind it either

17:14:45 <danbri_> K:EventCrawling can be seen as a specialisation of more generic RDF crawling. For FOAF documents, we currently have a bunch of crawlers which will traverse rdfs:seeAlso links between RDF documents.

17:14:45 <dc_rdfig> Added comment K1.

17:15:25 <danbri_> K:ie. when they find an RDF doc, they pay special attention to assertions of the form (?x rdfs:seeAlso ?y), typically putting ?y on the queue for further investigation / retrieval / parsing...

17:15:25 <dc_rdfig> Added comment K2.

17:16:02 <danbri_> K:In FOAF documents, we try to be kind to such crawlers, by mentioning the type of the thing that has the rdfs:seeAlso property, and (more experimentally) the type of the thing (document) referenced.

17:16:02 <dc_rdfig> Added comment K3.

17:16:24 <DanC> so what's the opening statement for RdfCalendar? or is the application really ModellingIcalInRDF?

17:16:43 * DanC checks www-rdf-calendar archive cover page...

17:17:05 <danbri_> K:This technique should extend to calendars and event crawling, quite naturally. You might write <util:Meeting><rdfs:seeAlso><rcal:VCalendar rdf:about="http://example.com/.../mycal.rdf"/>...

17:17:05 <dc_rdfig> Added comment K4.

17:17:44 * danbri_ remembers that he found markup hard to auto-escape in the TWiki, but seems OK in IRC/chump

17:17:44 <libby> K:vcalendar may not work as you can have mukltiple vcalendars in a docuiment...

17:17:44 <dc_rdfig> Added comment K5.

17:17:47 <DanC> K:I recently connected my foafwho document to my upcoming travel itineraries

17:17:47 <dc_rdfig> Added comment K6.

17:18:09 <danbri_> K:I wasn't quite sure whether vcalendar was the right thing, that's interesting to know.

17:18:09 <dc_rdfig> Added comment K7.

17:18:46 <DanC> btw... toIcal.py no longer finds Vevents that aren't conected from Vcalendars by ical:component

17:19:41 <danbri_> K:Just found Dan's [http://www.w3.org/People/Connolly/home-smart.rdf|home-smart.rdf] doc. It has a bunch of (untyped) seeAlsos to more info about him.

17:19:41 <dc_rdfig> Added comment K8.

17:19:53 * DanC asserts write lock on RdfCalendar

17:20:04 <danbri_> ack

17:20:06 <DanC> 'untyped'?

17:20:22 <danbri_> you don't mention a class for the thing referenced

17:20:25 <DanC> hmm... 'the RDF Calendar Task Force'

17:21:02 <libby> rdf interest group calendar taskforce?

17:21:28 <danbri_> <rdfs:seeAlso>

17:21:28 <danbri_> <FOAFShopDoc web:about="http://rdfweb.org/people/danbri/rdfweb/danbri-foafshop.rdf">

17:21:28 <danbri_> <wot:assurance web:resource="http://rdfweb.org/people/danbri/rdfweb/danbri-foafshop.rdf.asc"/>

17:21:28 <danbri_> </FOAFShopDoc>

17:21:34 <danbri_> ...in http://rdfweb.org/people/danbri/rdfweb/webwho.xrdf

17:21:51 <danbri_> is an example of giving a bit of fwd info regarding the referenced document.

17:22:14 <DanC> what's a task force?

17:22:19 <danbri_> I think similar might be useful for calendar citation, though at this stage not critical

17:22:31 * danbri_ can't remember how it came to be called a task force

17:22:33 <DanC> I suppose rdf calendar is an RDF IG task force if the RDF IG chair says it is.

17:22:39 <danbri_> It is.

17:22:48 <DanC> so let it be written, so let it be done.

17:24:14 * libby wonders if anyone else has any questions to add to danc's suggestions

17:24:28 <soccos|away> soccos|away is now known as soccos

17:24:33 <libby> syncing with palms seems to be a big issue for people

17:24:45 <libby> though not strictly an RDF issue

17:25:04 <danbri_> re crawling, seeAlso alone lets us harvest up the data, but (as per Eikeon's comments re lack of an rdfs:Schema class) it can be hard to figure out which docs are/represent/describe calendars without some explicit class.

17:25:37 <danbri_> libby, re K:vcalendar, what exactly is one of those?

17:25:45 <danbri_> <libby> K:vcalendar may not work as you can have mukltiple vcalendars in a docuiment...

17:25:58 <danbri_> a grouping construct that collects up some events?

17:26:02 <libby> hm

17:26:08 <libby> we did have a discussion about this

17:26:18 * libby returns to rfc2245

17:26:22 <soccos> palmsync++

17:26:28 <libby> this is why we went with the 'component' in the end

17:26:31 <danbri_> I'd like a term for the class of RDF documents that (typically, hmm) describe k:vcalendars

17:26:39 <libby> tricky

17:27:05 <danbri_> motivation: to build directory-like views on top of an EventCrawler like Eikeon's, ie something similar to icalshare.com

17:27:13 <danbri_> ...I want to write claims like:

17:28:07 <danbri_> <foo:CalendarDoc dc:title="Phases of the moon" dc:description="A calendar for phases of the moon."/>

17:28:21 <DanC> libby, you wrote [[RdfCalendarSchema][document describing how the RDF calendar schema is generated]]. Why not just RdfCalendarSchema?

17:28:24 * soccos lurks, makes coffee, eats breakfast, waits for conflicting meeting to start

17:28:30 <DanC> or better: MaintainingTheRdfCalendarSchema

17:28:51 <DanC> or CalendarSchemaMaintenance

17:28:53 <libby> rfc 2445: The Calendaring and Scheduling Core Object is a collection of

17:28:53 <libby> calendaring and scheduling information. Typically, this information

17:28:53 <libby> will consist of a single iCalendar object. However, multiple

17:28:53 <libby> iCalendar objects can be sequentially grouped together.

17:29:27 <libby> danc, if you like....

17:29:33 * danbri_ re-reads that a couple times

17:29:48 <libby> 'how generated is maybe differenet from generic about the schema. not sure

17:30:42 <DanC> hmm... RdfCalendarTimeZoneNotes ... Notes seems superfluous.

17:30:47 <danbri_> So it would be a mistake to use k:vcalendar as a class whose instances were RDF/XML documents using the rdf cal vocabulary?

17:30:50 <DanC> (luckily, renaming in Wiki space is cheap)

17:31:15 * danbri_ nods re 'Notes'

17:31:21 <DanC> it's a mistake if the instances are *only* allowed to be documents.

17:31:33 <DanC> but I do write { <> a ical:Vcalendar } still sometimes.

17:32:10 <danbri_> It seems somewhat analagous to the (confusion around) the rss:channel class

17:32:34 <danbri_> rss folk don't agree on whether the channel is the thing that has the RSS feed's URI, or something else.

17:33:12 <DanC> ok, pls review http://esw.w3.org/t/view/ESW/RdfCalendar Revision r1.7 - 26 Mar 2003 - 17:32 GMT

17:33:18 <libby> I dont see anything else about it

17:33:22 <libby> about vcalendar

17:33:31 <libby> in rfc2445

17:33:31 <danbri_> I want to work on guidelines for folk to put RDF calendar descriptions on their own websites, so would appreciate help on this thing.

17:33:49 * DanC sees a SemanticWebModellingIssue ...

17:33:52 <libby> that would be neat danbri_

17:34:22 <danbri_> have a look at http://icalshare.com/ -- there are basically short blurbs for each 'calendar', which to do an RDF version would need to live in the RDF model somewhere.

17:35:19 <DanC> I:DanC attending

17:35:19 <dc_rdfig> Added comment I4.

17:35:21 * danbri_ looks at http://esw.w3.org/t/view/ESW/RdfCalendar

17:35:40 <danbri_> I:[http://www.w3.org/People/DanBri/|DanBri] attending

17:35:41 <dc_rdfig> Added comment I5.

17:35:41 <soccos> soccos is now known as soccos|away

17:35:45 * DanC encourages folks to make a handy list of attendees in I:

17:36:49 * DanC wonders if anybody has been nominated as chair or has accepted; I'd like the floor to call for review of RdfCalendar Revision r1.7 - 26 Mar 2003 - 17:32 GMT

17:36:56 <libby> re http://esw.w3.org/t/view/ESW/RdfCalendar, should there be a bit about when you can add stuff, or is that covered by generic wikiness

17:37:20 <libby> apologues danc, I'm a bit fllustered after an open day here

17:37:36 <DanC> it's covered by generic wikiness to my satisfaction, though your question is evidence that more explictness is in order

17:37:50 * danbri_ nominates BitFlusteredLibby as rdfcal chair

17:37:50 * FloK is away: I'm busy

17:37:51 <libby> I sort of assumed you'd have the floor since the topic suggestion was yours.

17:37:54 <DanC> no need to apologize.

17:38:05 <libby> anyway, do have the floor

17:38:07 <libby> :)

17:38:11 <DanC> thx.

17:38:18 <eikeon> I:[http://eikeon.com/|eikeon]

17:38:18 <dc_rdfig> Added comment I6.

17:38:36 <DanC> ok... so that's 2 requsts for a bit of wiki docs... or training...

17:38:47 <danbri_> I read the 1.7 RdfCalendar. It misses some of the earlier context, which I could add

17:38:59 <danbri_> Are you still writing? Or can I add a few things?

17:39:01 <DanC> earlier in the sense of text that was there?

17:39:07 <DanC> you have the write lock, danbri

17:39:28 <libby> I:[http://ilrt.org/people/libby/|libby] attending

17:39:28 <dc_rdfig> Added comment I7.

17:40:02 <DanC> below the line, if you like, add: Remember, this is world-writeable. That's PPR:WhyWikiWorks!

17:40:40 <DanC> eventually I expect/hope that to be supurfluous (sp?). But it seems in order at this stage of the ESW community

17:41:18 <DanC> I have imported a certain amount of wisdom from PPR into the ESW wiki; e.g. http://esw.w3.org/t/view/ESW/?topic=WikiConsensus

17:42:19 * DanC notes the time... wow! 41 minutes gone by!

17:43:02 <DanC> libby, do you see that WikiConsensus bit?

17:43:29 <libby> looking now...

17:44:07 <eikeon> J:|Eikster: Upcoming Events discovered on the Semantic Web

17:44:08 <DanC> 'PPR:WikWikiWors' uses the InterWiki doodad; it's like a qname. PPR is the Portland Pattern Repository; the original WikiWikiWeb

17:44:08 <dc_rdfig> Titled item J.

17:44:23 <libby> PP?

17:44:28 <libby> anyweay, m,akes sense

17:44:38 <libby> wikiconsensus I mean

17:44:53 * DanC awaits danbri's save

17:45:16 <DanC> missing a ], danbri

17:45:17 <libby> ah right

17:45:41 * danbri_ resaves http://esw.w3.org/t/view/ESW/RdfCalendar

17:45:50 * danbri_ couldn't find timbl's business case doc

17:46:12 <DanC> hmm... using _created_ as link text rubs me the wrong way... I gather it's normal in blog-land, but I'm kinda old-school... I like noun phrases as anchor text.

17:46:47 * DanC found it... http://www.w3.org/DesignIssues/Business

17:46:49 * otingocni wonders business case fro what

17:46:50 * danbri_ couldn't think of a rephrase, help y'self if you want to noun it

17:46:59 <DanC> <- http://www.w3.org/2000/01/sw/

17:47:01 * danbri_ renounces writelock

17:47:12 * DanC asserts the write lock on RdfCalendar

17:47:15 * eikeon works on an rdf/xml view of events page -- I should really start doing the rdf/xml versions first then the xhtml ones.

17:50:09 * libby wonders if we need somethign pointing ot usecases on http://esw.w3.org/t/view/ESW/RdfCalendar. only because you can get the interop or some of it from icalendar

17:50:16 <danbri_> Libby, did the 'opening hours' use case -> test case get into the repository?

17:50:21 <libby> yep

17:50:30 <libby> I adedd it yetsrday

17:51:03 <danbri_> which one is it?

17:51:05 <libby> also linking report of workshop would be good: dev_workshop_report_2

17:51:07 <libby> oops

17:51:12 <libby>http://www.w3.org/2001/sw/Europe/reports/dev_workshop_report_2/

17:51:13 <dc_rdfig> L: http://www.w3.org/2001/sw/Europe/reports/dev_workshop_report_2/ from libby

17:51:18 <libby> aw crud

17:51:29 <danbri_> yeah, we have some things on front page of mailing list, some things in wiki, somethings in testcases dir

17:51:41 <libby> some usecases here: http://www.w3.org/2001/sw/Europe/reports/dev_workshop_report_2/#usecases

17:51:50 <libby> yeah, it's tricky

17:51:51 * DanC releases write lock, unsure of what to use as an ETag/version number

17:52:13 <danbri_> I click the little 'unlock' thing whenever I save, fwiw

17:52:33 <danbri_> ah http://www.w3.org/2002/12/cal/test/openingHours.rdf

17:52:44 <libby> yep, that's it

17:53:06 <otingocni> Use cases for sem web - i put 2 slides on each of Dan's trip, Sun's SworRDFish, and something else into the pcforum talk

17:54:06 * DanC hopes to submit Semantic Web Travel tools to ExML; deadline looms

17:54:07 <danbri_> hey otingocni, aren't you...? didn't you invent? no, it can't be...

17:55:11 <libby> so for - how to convert .ics to .rdf

17:55:11 <libby> <danbri_> - how to convert .rdf to .ics

17:55:11 <libby> <danbri_> - how to use rules to convert itineraries to calendars

17:55:11 <libby> <danbri_> - how to scrape calendars from HTML documents

17:55:20 <libby> can we start new wiki pages?

17:55:29 <libby> what's the best thing to do?

17:55:37 <DanC> hmm... "The Personal Information disaster" might have worked better as link text than [Business Model for the Semantic Web

17:55:54 <DanC> maybe a better TopicName than DoingThingsTheComputerCouldDoForMe

17:56:02 <danbri_> I:In other news, I've an [http://www.apple.com/powerbook/index12.html|12in Apple Powerbook G4] on order. So I'll hopefully get to play with Apple iCal app soon :)

17:56:02 <dc_rdfig> Added comment I8.

17:56:09 * DanC asserts write lock, refactors...

17:59:26 * danbri_ works on openingHours.rdf test

17:59:33 <danbri_> ...hits a parser bug

18:00:45 * DanC done refactoring http://esw.w3.org/t/view/ESW/RdfCalendar

18:01:28 <DanC> hmm... 60min gone by, and we haven't really parallelized the howto writing tasks.

18:01:50 <libby> I'd be happy to take one

18:01:54 <DanC> which one?

18:01:55 <eikeon> J:Added [rdf/xml version of events page|http://eikster.com/2003/events/rdf]

18:01:55 <dc_rdfig> Added comment J1.

18:02:17 <danbri_> Is there a wiki entry for EventsHarvester or EventDiscovery ? I'd take that, fold in stuff from chump...

18:02:41 <libby> I could start how to convert .ics to .rdf

18:02:46 <ryanlee> i wrote something in w3c space on scraping calendars: http://www.w3.org/2003/03/to-ical.html

18:02:57 <libby> or write up what I did with the scraping, although it's not great, as I say

18:03:16 <libby> ryan, maybe you'd liek to take that one then?

18:04:04 <DanC> ok, I added: [[

18:04:06 <DanC> Here are some goals or use cases we're working toward:

18:04:06 <DanC> * EventDiscovery? or EventHarvesting?

18:04:06 <DanC> * CalendarScraping?

18:04:07 <DanC> ]]

18:04:10 <libby> ryanlee, that's useful

18:04:12 <danbri_> Any objections to my doing EventDiscovery 'How do I share and find RDF calendar documents?'

18:04:20 <libby> not form me

18:04:32 <DanC> you have the write lock on EventDiscovery, danbri

18:04:33 <danbri_> OK, that'll be my TODO to get done by next meeting

18:04:39 * danbri_ makes a star

18:04:39 <danbri_> t

18:04:51 * DanC looks forward to when the ESW wiki groks PPR:EditConflictResolution.

18:04:52 <ryanlee> i'm something of a newbie when it comes to wikis, but i'd be willing to toss something in

18:05:15 <danbri_> please do!

18:05:16 <DanC> I noted the lack of PPR:EditConflictResolution in AboutThisService, linked from WebHome and WikiChoice

18:05:39 <libby> ryanlee, just do this: http://esw.w3.org/t/edit/ESW/CalendarScraping?topicparent=ESW.RdfCalendar and go ahead

18:05:46 * eikeon would like to add sharing in addition to harvesting and discovering -- there may be some issues worth discussing on how to best export things once gathered.

18:06:02 <DanC> ok, ryanlee has the wite lock on RdfCalendar

18:06:16 <libby> sounds interesting eikeon, like, as repeatign events?

18:06:22 <DanC> libby, you're aware that twiki doesn't prevent lost updates, yes?

18:06:40 <libby> danc, can only one edit on any page go on at once?

18:06:43 <DanC> (or at least: not the way the ESW Twiki is currently running)

18:06:56 <ryanlee> i'll bookmark that and work on it later then...cancel to release lock, i presume

18:07:03 <libby> yep, I've lost stuff on theer already....(not today)

18:07:28 <DanC> libby, if you and I start editing the same page, the guy who saves last wins. It should be the guy who saves first wins and the other guy gets an error message. Twiki just throws the first edit away. lost update.

18:07:35 <libby> so, can you have several people editing different pages on twiki?

18:07:43 <libby> right, ok

18:07:50 <DanC> yes, editing different pages is safe, I believe

18:07:52 <libby> I'm not editing anything....

18:08:08 <libby> that's useful to know :)

18:08:23 * DanC starts on CalendarScraping

18:09:26 <danbri_> danc, do you have a Wiki node for concept of 'turn RDF back into app specific formats to make it useful'

18:09:41 <libby> so I could start on icsToRdf or somesuch?

18:09:42 <danbri_> ie rdf2ical, rdf2ldif, rdf2html, rdf2wap...

18:09:51 * sandro plays with RACER (http://www.fh-wedel.de/~mo/racer/) and looks at the DIG protocol (http://potato.cs.man.ac.uk/dig/interface1.0.pdf), wonders if Sean Bechhofer ever hangs out here.

18:10:14 <danbri_> eg. one might merge data in RDF, then generate LDIF person records so an off-the-shelf LDAP server could eat the data

18:10:21 * DanC saved http://esw.w3.org/t/view/ESW/CalendarScraping

18:10:46 <eikeon> libby: Both, I think. Am interested in sharing the info about events that Eikster is able to discover... not try and become a central source, but just a broker of sorts.

18:11:04 * libby was also wondering about that

18:11:27 <DanC> hm... for going the other way, danbri, I tend to call my script grokTravItin.pl or grokThis or grokThat

18:11:40 <DanC> which suggests GrokkingLegacyFormats

18:11:48 <libby> I've got a tool a bit like mattb's, which on the commandline allows you to create RDF by searching a database. strikes me, that might be useful for events, not just codepictions

18:11:50 <DanC> what did timbl call it in his slide...

18:12:02 <eikeon> First pass I have is listing all the vevents plus depth of the graph beyond that... not sure if there is a better way to pull out the data.

18:12:16 <DanC> Application integration: Export http://www.w3.org/2000/10/swap/talk/slide5-0.html

18:12:33 * otingocni wonders promted by sandro's off-cal note re RACER: Is *anyone* keeping a product registry of RDF products?

18:12:40 <sandro> how about DomainSpecificFormats

18:12:58 <DanC> the daml tools repository is a registry of RDF products

18:13:16 <sandro> off-cal -- I missed the fact that it was > an hour-long, and talk seemed to have died down here.

18:13:27 <DanC> DomainSpecificFormats is as good as anything I can think of.

18:14:09 <DanC> remember: renaming and refactoring in Wiki-space is cheap. Don't feel you need a great name before you can start writing.

18:14:30 <DanC> And don't feel you can't just scribble tangentially related stuff on an existing page.

18:14:41 <DanC> DontWorryBeCrappy

18:14:51 <DanC> (tm Tim Finin)

18:15:13 <eikeon> otingocni: I did a tiny bit of work trying to come up with a vocabulary for projects and releases: http://eikster.com/2003/releases/

18:15:48 <DanC> agenda + next meeting?

18:16:10 <libby> for what danc?

18:16:15 <libby> talke about it now?

18:16:22 <libby> the agenda I mean?

18:16:31 <DanC> agenda + is zakim-notatoin for "please add to the agenda..."

18:16:40 <libby> ah....

18:18:22 <DanC> I wish the record of this meeting would magically get linked from http://www.w3.org/2002/12/cal/#mtg

18:18:45 * danbri_ stops scribbling in http://esw.w3.org/t/view/ESW/EventDiscovery

18:19:22 <DanC> I'm considering moving the meeting index into wiki space to facilitate wishes like that being granted by other folks.

18:19:40 <danbri_> 2nd'd

18:20:13 <libby> go for it. it's a pain updating it...

18:20:34 <DanC> I:RESOLVED: to use the wiki for meeting index. ACTION DanC

18:20:34 <dc_rdfig> Added comment I9.

18:21:19 <danbri_> I:New doc, [http://esw.w3.org/t/view/ESW/EventDiscovery|EventDiscovery], 'How do I share and find RDF calendar documents?'

18:21:19 <dc_rdfig> Added comment I10.

18:21:33 * danbri_ not quite happy with that doc yet, but it gives a focus for attention

18:21:40 <DanC> well played

18:21:55 <DanC> we're not likely to achieve perfection during synchronous chat

18:23:20 * danbri_ postpones perfection temporarily ;)

18:23:49 <danbri_> OK, time to arrange next meeting?

18:23:57 <DanC> can I squeeze another agendum in? timezones and ModellingQuantities, among SemanticWebModellingIssues

18:24:11 <danbri_> are we on a fortnightly or weekly thing at moment? fortnight seems good to me.

18:24:15 <danbri_> fine by me

18:24:30 <DanC> uk:fortnight = us:every-other-week?

18:24:32 <danbri_> shall we do next meeting, then chat about that after?

18:24:38 <libby> fortnight ly recenntly

18:24:40 <danbri_> forteen nights, yup

18:24:56 * danbri_ still can't remember what bi-annual means

18:25:00 <libby> sounds good danbri

18:25:28 <DanC> I kinda need to think about an end to the recurrence before I set expectations that I'll be here regularly

18:26:07 <DanC> every other week thru WWW2003?

18:26:30 <deltab> danbri_: every two years, and twice a year

18:26:43 <danbri_> How about tagging it to the end of the SW Activity current phase (or its next minor extension). Then we can rethink goals as part of RDFIG rethink

18:26:50 <danbri_> yep, thru WWW2003 works for me.

18:26:57 <DanC> I believe biannual means twice per year; every other year is semiannual.

18:27:09 <libby> this is kinda why I wanted to discuss when are we donr a couple of weeks ago

18:27:21 * danbri_ nods

18:27:23 <DanC> aha! when are we done as measured on the calendar!

18:27:32 <libby> well, both :)

18:27:34 <DanC> I thought you meant as measured by what we get done.

18:27:38 <deltab> both meanings are common, as far as I know

18:27:46 <libby> I saw the two as being closely connected

18:27:52 <ryanlee> also try biennial (every two years)

18:27:56 <DanC> connecting them closely is Very Hard Work.

18:28:10 <deltab> and twice-yearly

18:28:12 <DanC> it's commonly called Management

18:28:22 <libby> well, actually, I got an answer that a lot of peopel tyhought was the minimum, which was round-tripping to icalendar

18:28:34 <libby> but coudl do that fast or slow

18:28:44 <libby> I'm happy with daedlines.

18:29:22 <DanC> 1630Z looms. pls call the question on every-other-week 'till WWW2003

18:29:29 <DanC> i.e. any objection?

18:29:39 <libby> not form me

18:29:43 <danbri_> Yeah, there are at least two ways to think of these 'lite' activities. As 'do it fast and cheap' versus 'a way to track what we're doing in an area of ongoing interest'.

18:29:47 <danbri_> nor me.

18:29:59 <DanC> I:RESOLVED: to meet every other weds at 1500Z for up to 90 min thru WWW2003.

18:29:59 <dc_rdfig> Added comment I11.

18:30:00 <deltab> itym 1830Z

18:30:02 <libby> hm , now we can write a testcase

18:30:34 <libby> ok, danc, want to talk about you extra agenda item for a few minutes?

18:30:37 <deltab> heh: 183000Z #rdfig <deltab> itym 1830Z

18:30:42 <DanC> hearing no motion to extend the meeting, I gather we are ADJOURNED.

18:30:48 <danbri_> ADJOURNED.

18:31:10 <DanC> yes, I still wanna talk about quantities and timezones.

18:31:41 <danbri_> danbri_ has changed the topic to: RDF/SemWeb 24x7 calendar-flavoured hack'n'chat http://rdfig.xmlhack.com/

18:31:51 * danbri_ listens

18:31:51 <DanC> are you familiar with the quantities problem, libby? i.e. [ :magnitute "20"; :units :meter] is a train wreck waiting to happen.

18:32:28 <libby> could you explain a bit?

18:32:33 <DanC> because we'd like [ :magnitute "20"; :units :meter] = [ :magnitute "2000"; :units :centimeter]

18:32:51 <libby> ok, yep

18:32:55 <danbri_> but then you lose the paired associations... yup

18:32:55 <DanC> but then we get [ :magnitute "20", "2000"; :units :meter, :centimeter]

18:33:15 <DanC> well, that's exactly the way we've modelled timezones.

18:33:28 <libby> crud, really?

18:33:32 <eikeon> Re: next meeting --> http://eikster.com/2003/events/?start=2003-04-09

18:34:07 <DanC> we can't say [ :dateTime "...10:00"; :tzId ".../Boston"] = [ :dateTime "...11:00"; :tzId ".../Chicago"]

18:34:32 <deltab> what about [ :meters "20"; :centimeters "2000" ] ?

18:34:45 <DanC> exactly, deltab, InterpretationProperties to the rescue.

18:35:04 <DanC> [ :timInBoston "10:00"; :timeInChicago "11:00"] is just fine.

18:35:27 <danbri_> binary relations mean we're always trying to smooth out bumps in the carpet like this. I like the interpretation property idiom...

18:36:11 <DanC> I haven't worked out a way to "fix" icalendar in this way... but for now, keep in mind that you *cannot* equate ical datetimes from different time zones.

18:36:34 <danbri_> Is that noted in the wiki or website somewhere?

18:36:47 <DanC> have you seen my start on NaryRelations, danbri?

18:37:19 <DanC> noted: not yet, danbri. I was trying to write it up before today's meeting, but I didn't get to it.

18:38:36 <DanC> I believe the icalendar community intends that 10am in Boston is *not* identical to 11am in Chicago. i.e. timezones really are part of the identity of the value of the dtstart property

18:39:00 * danbri_ reads http://esw.w3.org/t/view/ESW/NaryRelations ...useful

18:39:13 <sandro> Hrm. I cant register on TWiki because I already created a page for myself.

18:39:14 <DanC> for recurring events, it matters a lot.

18:39:31 <DanC> I did the same thing, sandro. pls make a note in AboutThisService

18:39:56 * DanC should link the discussion danbri and I had about attribution and such

18:40:00 <sandro> What's the work around?

18:40:10 <DanC> don't bother trying to login or anything

18:40:31 <DanC> i.e. don't expect it to treat you differently from other people automatically

18:41:02 * libby puzzles over: <DanC>I believe the icalendar community intends that 10am in Boston is *not* identical to 11am in Chicago. i.e. timezones really are part of the identity of the value of the dtstart property

18:41:02 <sandro> I mean -- should I somehow delete my page, or just forget abour registering?

18:41:37 <otingocni> Hmmm... have you seen UseRealURIs?

18:41:40 <DanC> well, you could fininsh installing MoinMon on swada, sandro. 1/2 ;-)

18:42:01 <DanC> libby, consider round-tripping.

18:42:03 <sandro> Sounds like a pair-programming commitment, Dan.

18:42:17 <eikeon> How about having the range of dateTime be a UTC time?

18:42:52 <DanC> no, that's loss of information, eikon. at least in the recurring case. zakim telcons are scheduled on Boston time, not UTC. When sumer time comes, the UTC offset changes.

18:43:02 <eikeon> And require consumers and producers of the rdf to convert between timezones for display.

18:43:10 * danbri_ adds a 'don't bother registering' note to http://esw.w3.org/t/view/TWiki/TWikiRegistration

18:43:10 <DanC> and for a week, the folks in europe get screwed ;-)

18:43:56 <DanC> fininsh..... fininish... to be like tim finin...

18:44:19 * DanC chuckles at his own joke... ahem

18:44:20 * danbri_ has a new test

18:44:23 <sandro> DanC, did you start a MoinMoin install on swada?

18:44:26 <danbri_> 1) Failure!!!

18:44:26 <danbri_> test_rdfcal(TC_FetchRDF) [./tc_fetchrdf.rb:25]:

18:44:26 <danbri_> There should be 23 triples in this calendar, not 23

18:44:31 <DanC> yes, I started, sandro.

18:44:32 <danbri_> hmm :)

18:44:36 <libby> heh

18:44:37 * danbri_ hasn't got hang of this yet

18:44:47 <sandro> Pointers?

18:44:50 <DanC> umm...

18:45:11 <danbri_> I think I remember one reason I didn't investigate MoinMoin heavily: PythonCowardice...

18:45:14 <DanC> connolly@swada ../moin-1.0[531] pwd [newline]/usr/local/src/moin-1.0

18:45:17 <danbri_> ...wanted something hackable

18:45:23 <danbri_> ....didn't know python

18:45:37 <danbri_> ...not a relevant consideration now others are more interested

18:45:52 * danbri_ picked up a bit of python since then too

18:45:58 * sandro cringes at http://daml.olin.edu/wiki/wikidaml/moin.cgi/MileStones

18:46:32 * DanC did some PPR:WikiEngine surfing last night, recorded motivation for choosing MoinMoin in AboutThisService

18:46:39 <danbri_> why? cos they're in the past?

18:47:04 * DanC happened by wikidaml/moin.cgi/MileStones last night too

18:47:25 * libby gotta go home, last one at work. bye!

18:47:44 <danbri_> moinmoin sounds good to me

18:47:47 <sandro> "PPR" ?

18:47:53 * danbri_ wondersr how much hassle migration will be

18:48:13 <DanC> timbl, calendar meeting is over; feel free to chump RACER etc.

18:48:26 * DanC notes #moin is nearby

18:48:44 <DanC> PPR = Portland Pattern Repository, the original wiki

18:48:57 <sandro> AH, I had just learned that from interwiki. :-)

18:50:06 <sandro> Holy! that's a lot of WIkiEngines. http://c2.com/cgi/wiki?WikiEngines

18:50:17 <DanC> some folks just call it Wiki:WikiEngine, but I'd like to support c2.com in hosting a patterns/software discussion, not (too much of a) pissing-and-moaning about virtual communities discussion

18:51:17 <DanC> re UseRealURIs, can I please declare all the InterWiki namespaces?

18:51:28 <DanC> for use in this channel?

18:51:53 * danbri_ joins #moin

18:52:00 <DanC> or... hmm... a handful of them?

18:52:10 <danbri_> DanC, in what syntactic context?

18:52:27 <danbri_> chump has a prior claim on ^\w\w:...

18:52:35 <DanC> ah. ok.

18:53:31 * DanC has another meeting in 8min... sigh...

18:53:55 <DanC> oh crud... hugo is a critical resource in my todo list for today...

18:54:03 <danbri_> doh!

18:54:09 <danbri_> 8pm in france...

18:55:44 <DanC> hmm... an interesting feature for hipAgent connect TODO items to critical resources; prioritize based on availability of critical resources.

18:56:12 <DanC> ala BEEP: Hugo normally leaves the office 40 minutes from now. You should probably talk with him about XYZ.

18:58:11 <DanC>http://rdftwig.sf.net/

18:58:11 <dc_rdfig> M: http://rdftwig.sf.net/ from DanC

18:58:26 <DanC> M:RDF Twig

18:58:26 <dc_rdfig> Added comment M1.

18:58:38 <DanC> M:"a set of XSLT extension functions for Saxon and Xalan to provide access to RDF graphs stored in the Jena repository."

18:58:38 <dc_rdfig> Added comment M2.

18:58:48 <DanC> M: go Norm Walsh!

18:58:49 <dc_rdfig> Added comment M3.

18:59:05 <danbri_> M:Very nice!

18:59:05 <dc_rdfig> Added comment M4.

19:00:08 <DanC> M:|RDF Twig

19:00:08 <dc_rdfig> Titled item M.

19:00:11 <DanC> M1:""

19:00:11 <dc_rdfig> Deleted comment M1.

19:02:50 <bitsko> and here I thought RDF was not XSLTable

19:03:05 <dajobe> wrong

19:03:22 <dajobe> try googling for the many RDF XSLT apps

19:10:11 <bijan> M:4Suite has an XSLT extension that gives access to the Versa RDF query/path language from within XSLT, too.

19:10:11 <dc_rdfig> Added comment M4.

19:10:27 <bitsko> what I can't seem to wrap my mind around is that there seems to be multiple ways to encode RDF that would make it difficult to use with an XSL transform

19:11:20 <bijan> M:I've been wanting to build a pluggable XSLT implementation (perhaps on 4Suite or SXSLT/SXML), i.e., plug in different, domain specific path/query languages.

19:11:21 <dc_rdfig> Added comment M5.

19:11:25 * eikeon wonders if the w3c could add URLs in the body of messages posted to archived lists to the archive page for the message

19:11:41 <bijan> YOu mean in the header?

19:11:46 <bijan> As we discussed?

19:11:48 <bijan> headers

19:12:25 <eikeon> Hum... I did not check the headers... might already be there.

19:13:01 <bijan> Well, what we discussed was the list manager adding archive urls to an x- header in the mail messages.

19:13:01 <eikeon> This was somewhat indepenent thought... was just trying to get to the archive copy of a message I was reading in my mail client.

19:13:31 <bijan> If it were a header, then you could tweak your mail client.

19:13:47 <bijan> If you embedded it in the message, it would lose it's header status, of course, but be more universal.

19:13:58 <bijan> Universally accessible

19:14:37 * eikeon sees X-Archived-At: -- so already in header... cool.

19:31:27 * otingocni :-) Just send beers, flowers, chocolate to sysreq@w3.org

19:36:03 <eikeon> otingocni: :)

20:34:07 <taaz> eikeon: have you had a chance to look at that relative link issue in rdflib?

20:36:28 <danbri_> dc_rdfig:view

20:36:28 <dc_rdfig> I: RDF calendar meeting Wednesday March 26th 2003, 17:00 UTC (http://lists.w3.org/Archives/Public/www-rdf-calendar/2003Mar/0038.html)

20:36:29 <dc_rdfig> J: Eikster: Upcoming Events discovered on the Semantic Web (http://eikster.com/2003/events/)

20:36:30 <dc_rdfig> K: On the use of rdfs:seeAlso for RDF harvesting (blurb)

20:36:31 <dc_rdfig> L: http://www.w3.org/2001/sw/Europe/reports/dev_workshop_report_2/

20:36:32 <dc_rdfig> M: RDF Twig (http://rdftwig.sf.net/)

20:40:16 * nwalsh waves to the room

20:40:28 <dajobe> hi norm

20:40:40 <nwalsh> Hi dajobe

20:40:52 <zool> hello

20:42:18 <dajobe> there is a 'localhost' link on the rdftwig site

20:42:30 <nwalsh> Urk.

20:42:41 <nwalsh> Where?

20:42:51 <dajobe> the Extreme Markup Languages one

20:42:53 <zool> mmm old infocom games; i'm playing with a RDF graph based infocom-ish storytelling engine which i keep forgetting about

20:42:55 <nwalsh> Oh, ack!

20:43:15 * nwalsh smacks head for obvious cut and paste error

20:43:24 <danbri_> hi norm

20:43:46 <dajobe> i'll need to start an XSLT section to my rdf guide, at this rate

20:44:31 <nwalsh> Hi Dan

20:44:36 <nwalsh> Hi danbri_, even

20:44:40 <danbri_>http://www.w3.org/2001/12/rubyrdf/pack/tests/coretests.rb

20:44:40 <dc_rdfig> N: http://www.w3.org/2001/12/rubyrdf/pack/tests/coretests.rb from danbri_

20:44:49 <danbri_> N:|DanBri's testing headache, continued

20:44:49 <dc_rdfig> Titled item N.

20:45:33 <danbri_> N:Per [http://rdfig.xmlhack.com/2003/03/26/2003-03-26.html#1048691883.680525|previous discussion] I didn't know whether to use the RDF testcase collection to do code generation, vs just have tests loop through each positive and negative test.

20:45:33 <dc_rdfig> Added comment N1.

20:46:25 <danbri_> N:I chose latter (implemented here), and now remember a previous headache: if an assertion fails, RubyUnit bails on the entire method.

20:46:26 <dc_rdfig> Added comment N2.

20:46:58 <danbri_> N:Which is kinda annoying as I've basically got one big fat test method and make a lot of assertions within it, since I'm driving everything from data files.

20:46:58 <dc_rdfig> Added comment N3.

20:47:13 <danbri_> N:Maybe I should go back to code generation? What do other folk do?

20:47:13 <dc_rdfig> Added comment N4.

20:47:53 <dajobe> N:test my C parser using perl

20:47:54 <dc_rdfig> Added comment N5.

20:48:39 <danbri_> ...no fancy test framework, just write yr own Perl to do it?

20:49:24 <bitsko> N:in that situation, I'd catch the errors and count them, and just raise one failure for the test case

20:49:25 <dc_rdfig> Added comment N6.

20:49:25 <dajobe> fancy test framework in the perl

20:49:30 * nwalsh fixes his rdftwig index.html page

20:50:13 <dajobe> easier to catch crashes (C that is) exceptions etc. via an external program. Writing it in another language meant I didn't need to parse rdf/xml in order to test parsing rdf/xml

20:50:58 <dajobe> what if a parser bug that had a test case caused the description of the that test case to be skipped?

20:51:12 <danbri_> quite!

20:51:21 <danbri_> that previosuly had biased me towards code-generation

20:51:34 <danbri_> I was writing the test files out using an machine that had your parser installed...

20:51:45 <danbri_> ...so that others could run tets without a dependency on raptor/redland

20:52:06 <dajobe> oh, I've just remembered, I depend on 'ntc'

20:52:11 <dajobe> that's in C++

20:52:16 <danbri_> yes, I've been wondering about that...

20:52:33 <danbri_> it's short enough to consider rewriting in ruby

20:52:41 <dajobe> it's fine for the test cases but goes all n! on one bit of nasty daml

20:52:47 <danbri_> eek

20:53:00 <dajobe> thanks to sandro for that one

20:53:15 <dajobe> hmm, maybe there was a bagID in that

20:54:05 <dajobe> no

20:55:57 <otingocni> re test harnesses -- you have to be careful not to test something with itself.

20:56:24 <dajobe> exactly

20:56:27 <dajobe> I test raptor with arp

20:56:40 <otingocni> I was worried about using cwm to test ntriple file equality, when it ignored data types (on N3 input, currently, not on N3 output or XML input)

20:56:55 <dajobe> i put up an ntriples validator, dunno if people notice

20:56:56 <dajobe> d

21:02:05 <otingocni> How about an RDF graph comparer?

21:02:53 <dajobe> hmm

21:03:04 <otingocni> Actually, the most useful thing would be a canonnicalizer. Which Jeremy has. I think canonnicalizingthe ntriples and then doing aunix diff is easy.

21:03:23 <dajobe> that was the idea of ntriples

21:03:26 <otingocni> I could just once cannoncialize each reference ntriples.

21:03:33 <dajobe> but with the bnodes there isn't always a way to do it

21:03:56 <dajobe> you can cut down and match patterns but in the end, you will still have to search

21:04:01 <otingocni> Well, accoring to Mr Carrol (sp?) there is a way to doi it.

21:04:34 <danbri_> N:I'm taking your advice...

21:04:35 <dc_rdfig> Added comment N7.

21:04:37 <dajobe> make a loop of bnodes, now make another one. compare

21:06:08 <dajobe>http://www.hpl.hp.com/techreports/2001/HPL-2001-293.pdf

21:06:09 <dc_rdfig> O: http://www.hpl.hp.com/techreports/2001/HPL-2001-293.pdf from dajobe

21:06:20 <dajobe> O:|Matching RDF Graphs, Jeremy J. Carroll (PDF)

21:06:20 <dc_rdfig> Titled item O.

21:06:42 <dajobe> O:as implemented in Jena, with a *lot* of optimisation

21:06:42 <dc_rdfig> Added comment O1.

21:07:20 * otingocni should get the latest Jena then

21:07:53 <dajobe> it had a scary amount of node colouring, patterns, bitmap encoding in order to make it work fast enough in Java

21:08:31 <otingocni> wow

21:08:47 <otingocni> At least the rdf test cases only have two or three statements!

21:08:58 <DanC> jeremy's c14n-for-sig is non-deterministic

21:09:26 <DanC> so it's not really c14n; but it's close enough for the signature use case

21:09:34 <dajobe> otingocni: yes, that's why a dumb (n!) compare or nearly works mostly

21:10:21 <otingocni> Its not really c14n? I thought it was

21:10:41 <DanC> nope.

21:11:00 <DanC> it assigns arbitrary, not canonical, numbers to bnodes.

21:11:01 <dajobe> O:*It is not necessary to use some of the more sophisticated techniques suggested, due to the large amount of labelling found in RDF graphs.*

21:11:02 <dc_rdfig> Added comment O2.

21:12:54 * danbri_ gets a todo list from tests http://www.w3.org/2001/12/rubyrdf/pack/tests/ts_rdftests-out.txt

21:12:56 <otingocni> Of course. You are right. The signed document is arbitrary but equivalent. Like signing a serialization.

21:13:13 <DanC> right.

21:13:17 <danbri_> its working from the rdf files, and consulting the manifest, so may be finding stuff i could ignore

21:13:42 <dajobe> yes

21:13:53 <otingocni> Check for status "APPROVED"

21:13:53 <dajobe> you should be doing it the other way round

21:14:02 * danbri_ knows

21:14:04 <dajobe> since some test cases might be using multiple 'rdf' files

21:14:08 <danbri_> ok, will do that

21:14:14 <dajobe> might, but don't presently I think

21:14:17 <danbri_> i got the basics there, so am a bit happy

21:14:22 <dajobe> plus some tests have multiple correct answers

21:14:22 <danbri_> also lots todo though

21:14:32 <dajobe> *cough* bagID *cough*

21:14:49 <danbri_> i'm only at the 'did it barf or not' stage, never mind getting the right number of triples...

21:14:56 <danbri_> and certainly never mind getting the _right_ triples ;)

21:14:57 <dajobe> sure

21:20:07 * eikeon is guilty of using rdflib to test itself... upps did I say that out loud.

21:37:32 <dajobe> rdflib parser question

21:37:48 <dajobe> do you allow parsing of content with no given URI

21:42:52 <dajobe> ok, I find lots of cwd & file: detail, so if not you make one

21:43:40 <dajobe> if location[-3:]==".nt":

21:43:40 <dajobe> self.parse_nt_URI(location, None)

21:43:40 <dajobe> else:

21:43:42 <dajobe> ...

21:43:52 <eikeon> Parsing of content with no URI?

21:44:12 <dajobe> yes

21:44:27 <taaz> like a string of rdf data?

21:44:35 <dajobe> could be

21:44:39 <dajobe> or embedded

21:44:54 <eikeon> Yes... can use a StringInputSource

21:44:59 <dajobe> my issue is, the method I use in my parser requires a base URI

21:45:12 <dajobe> so how do you resolve relative urls for that

21:45:20 <dajobe> if isinstance(location, InputSource):

21:45:20 <dajobe> self.parse(location)

21:45:28 <dajobe> so then... ?

21:45:41 <eikeon> Hum... looking... I tried to follow the sax input source thing.

21:46:29 <dajobe> the related Q, should I enforce requiring a base URI for ntriples, which doesn't need it

21:47:54 <eikeon> Let me quick try parsing some rdf/xml with a relative uri from a string and see what happens...

21:48:22 <dajobe> it either means aborting, constructing a file: one, or dieing at the first bit of rdf/xml that needs it "rdf:ID" or relative uri. yuck

21:49:06 <dajobe> dying

21:51:48 <eikeon> Hum... interesting... I keep them relative in that case :)

21:52:02 <eikeon> Wonder if I special cased it... looking in parser code now.

21:52:56 <dajobe> hehe

21:53:06 <dajobe> rdf only has abs uri-refs ;)

21:53:33 <dajobe> I guess I'll stick with giving an error for now

21:53:36 <eikeon> Yep... I should raise an exception.

21:53:38 <dajobe> if they want to fake it, they can pass it in

21:53:55 <eikeon> I think the caller should have to set a base URI.

21:54:06 <dajobe> for rdf/xml

21:54:10 <dajobe> but for ntriples?

21:54:22 <dajobe> I assume you use the same i/f, as I do

21:54:28 <eikeon> For rdf/xml... /me does not really do ntriples.

21:54:41 <dajobe> heh, well done

21:56:32 <soccos|away> soccos|away is now known as soccos

21:57:01 <soccos> soccos is now known as soccos|away

21:57:14 <eikeon> The support rdflib has for parsing ntriples was in support of running the test cases... I did not really intend for rdflib to be able to parse ntriples. Was confused once when I heard someone say it could... then I realized... doh.

21:59:39 <eikeon> Would not have used [if location[-3:]==".nt"] as a test for deciding if something was rdf/xml or ntriples if I meant it for more than just test code ;)

22:00:29 <dajobe> lol :)

22:23:42 <danbri_>http://www.philo.de/rdf/

22:23:43 <dc_rdfig> P: http://www.philo.de/rdf/ from danbri_

22:23:57 <danbri_> P:|Open RDF component kit for Delphi/Kylix

22:23:57 <dc_rdfig> Titled item P.

22:24:51 <danbri_> P:Recently [http://lists.w3.org/Archives/Public/www-rdf-interest/2003Mar/0063.html|announced] to the RDF IG list.

22:24:52 <dc_rdfig> Added comment P1.

22:25:02 <danbri_> oops wrong url

22:25:43 <danbri_> P1:""

22:25:43 <dc_rdfig> Deleted comment P1.

22:25:48 * danbri_ cant find it in the archives

23:49:43 * FloK is back (gone 06:11:55)

23:50:03 * chaalsMEL waves to danbri_