Semantic Web Interest Group IRC Chat Logs for 2003-09-17

00:07:12 <tav> tav is now known as tav|offline

00:59:46 <tav|offline> tav|offline is now known as tav

01:04:30 <dmiles_afk> dmiles_afk is now known as dmiles

01:56:52 <bitsko> B:

01:56:53 <dc_rdfig>http://bitsko.slc.ut.us/blog/2003/09/16/avail-20031001

01:56:54 <dc_rdfig> Available: Senior Unix/XML/RDF developer (midwest/MN/IA, US)

01:56:55 <dc_rdfig> (1:bitsko) Contracts welcome -- several years succsessfully delivering projects to customers remotely.

01:56:56 <dc_rdfig> (2:bitsko) Contact Ken MacLeod at ken@bitsko.slc.ut.us

01:57:20 <bitsko> B1:Contracts welcome -- several years successfully delivering projects to customers remotely.

01:57:21 <dc_rdfig> Replaced comment B1.

02:07:03 <adamhill> adamhill is now known as _adam

02:07:13 <_adam> _adam is now known as _adam_work

02:36:07 <[1]adamhill> [1]adamhill is now known as _dreaminofjeanni

02:36:14 <_dreaminofjeanni> _dreaminofjeanni is now known as adamhill

04:26:44 <adamhill> adamhill is now known as _dreaminofjeanni

06:02:59 <mdupont-zzz> mdupont-zzz is now known as mdupont

07:11:31 <mdupont> mdupont is now known as mdupont-work

10:20:40 <darobin_> darobin_ is now known as darobin

11:35:13 <arnarl> hi

11:35:23 <kao-away> hi there

11:35:27 <kao-away> kao-away is now known as kao

13:45:33 <JimJibber> JimJibber is now known as JibberJim

14:32:18 <sbp-> sbp- is now known as sbp

16:09:15 <DanC> hmm... OWL faq in the wiki or not?

16:09:42 <DanC> I'm certainly glad I moved the URI scheme list into the wiki... the synergy with topics like FollowYourNose sure is nice

16:10:16 * DanC wonders if bijan, sandro, et. al. are interested...

16:10:28 * DanC wonders if bijan, sandro wanna talk about when to hold sw-meaning telcon

16:10:30 <bijan> Could be

16:10:42 <bijan> Could be

16:10:50 <bijan> I started a "ground laying" email yesterday

16:10:54 <bijan> But ran into meeting hell :)

16:11:19 <bijan> DanC: I forward my pellet report note to public-webont-comments

16:11:24 <DanC> cool.

16:11:27 <sandro> +1 faqs in wiki

16:11:28 <bijan> So you can freely include all that info in cr report

16:11:37 <DanC> will do presently

16:11:43 <bijan> Great

16:12:21 * DanC looks for owl cookbook entry from debm...

16:12:39 <bijan> I don't know if SWI Prolog got in, but they do have some, somewhat random to my eye, owl support

16:12:54 * DanC likes random OWL support ;-)

16:13:12 * DanC has been known to perpetrate random OWL support himself

16:13:13 <bijan> Heh

16:13:26 <bijan> senseless acts of OWL support?

16:13:56 * DanC hopes sw-meaning can remain largely chairless

16:14:14 <JibberJim> Just lots of comfy bean-bags?

16:14:20 <DanC> exactly!

16:14:32 * bijan suspects that there will be sat on laps

16:14:37 <sandro> lol (/me has two in his MIT office..... but never goes there.)

16:14:54 <bijan> After all, we're allabout meaning-full relations?

16:14:59 <danbri> how often do you go into mit nowadays?

16:15:01 * danbri noses

16:15:03 <sandro> <snort>

16:15:16 <sandro> It's been about a month, I think. Maybe more.

16:15:18 * DanC would be very happy for sw-meaning to spontaneously decide when to have its 1st telcon... or to somehow decide not to have a telcon after all.

16:15:20 <bijan> We may need a chair just for pun control

16:15:44 <sandro> (combination of car trouble, phone trouble, family health trouble, ... and general lack of good reason.)

16:15:50 * danbri wanders into ilrt from time to time for change of scene

16:16:25 * DanC files comments on MIT ambiance under "if you don't have anything nice to say..."

16:16:34 * bijan moved, now has 5 mile bike ride rather that two block walk commute

16:16:48 <bijan> But this is the first day I'm working from home sinc ethe move :)

16:16:52 <bijan> Just feeling crappy

16:16:57 * JibberJim thought that ended with "... write about it in your blog" DanC...

16:17:15 <DanC> JibberJim, you're really on today!

16:17:42 <JibberJim> It's a nice sunny day, life is good...

16:20:10 * DanC finds faq foo from debm under unhelpful subject... http://lists.w3.org/Archives/Public/www-webont-wg/2003Sep/0093.html

16:21:02 <DanC> do we have a ClosedWorldAssumption wiki topic yet?

16:21:54 <DanC> nope... hmm...

16:22:43 <DanC> noodling on a name for this pattern... http://www.ksl.stanford.edu/people/dlm/webont/HowToDoIt/closingRoles.html

16:23:00 <DanC> hm... peeking into the URI, I see debm chose ClosingRoles

16:23:14 <DanC> is this pattern already known by that name?

16:24:09 <bijan> Eek, I don't find this characterization helpful!

16:24:21 <DanC> "(again, the order of closing roles that cannot have derived fillers isn't significant)" -- http://www.cs.vassar.edu/faculty/welty/papers/phd/HTML/dissertation-22.html

16:24:25 <bijan> ""A specific Specifically Many applications find some use for knowing that there will be no added information added to some aspect of the knowledge base after a certain point. OWL and DAML+OIL are said to make the "open world assumption", thereby assuming that just because some piece of information is not known now, there is no guarantee that it will not be known later.""

16:24:42 <bijan> CWA isn't about *epistemics* in this way

16:25:10 <bijan> It's not that "may be known later", it's going from "not known" to *false*.

16:25:59 * bijan also getting some format issues with that page

16:26:18 <DanC> this seems to confirm the meta-point, which is that a wiki is likely a good mechanism to extract community wisdom

16:27:06 <bijan> Yep

16:27:18 <DanC> anyway... I don't know any better than ClosingRoles, so I'll try that... if there's a better name, the wiki will find it.

16:27:57 <bijan> Acutally wiki's confirm the kantian/sellarian point that knowledge/wisdom and community (and world!) are mutually consitutive

16:28:31 * bijan notes speaking of wiki; goes back to MoinMoin hacking

16:29:18 * DanC bumps into the problem of plurals... "carefully relaxing ClosedWorldAssumptions." -- http://esw.w3.org/topic/SemanticWebArchitecture

16:30:28 <bijan> Hmm. There's some mointrick for this, I believe

16:30:44 <bijan> and, of course, always editorial

16:30:52 <bijan> 'carefully relaxe the CWA"

16:31:20 <bijan> Even, 'carefully relax the CWA it all its many guises and varients, however they should surface"

16:35:36 <DanC> hmm... I think there isn't one master ClosedWorldAssumption, so I'm leaving it plural for now.

16:36:38 <reagleBRKLN> reagleBRKLN is now known as reagleLUNCH

16:37:36 <bijan> Hmm.

16:37:54 <bijan> I always say, and have heard, The ClosedWorldAssumption

16:38:16 <bijan> Regardless, my rewrites preserved diversity in that nominal unity :)

16:38:26 * DanC offers http://esw.w3.org/topic/ClosedWorldAssumptions last modified 2003-09-17 11:38:03

16:40:09 <DanC> "CWA in guises..." sounds good

16:50:31 * DanC puzzles over the relationship between the bestFriend/spouse situation and the closed world assumption

16:53:19 <_joshua> specialFriend

16:53:24 <_joshua> specialFriendWithBenefits

16:53:26 <_joshua> secretFriend

16:55:50 * DanC saves http://esw.w3.org/topic/ClosingRoles 2003-09-17 11:55:22 in hopes of getting help

17:00:02 <reagleLUNCH> reagleLUNCH is now known as reagleHOME

17:01:52 <ndw> DanC, simple typo: rdfs/s/:subPropertyOf ...

17:02:05 * ndw can't really offer more help :-)

17:02:21 * bijan was going to make an editorial comment, but the wiki cry of "FIX IT YOURSELF" came uip :)

17:02:24 <bijan> TO my fleshy lips

17:02:33 <bijan> And even to my not-too-fleshy lips

17:02:43 * bijan unsure where the lip stuff is coming from...

17:03:03 * ndw knew that was the right answer but was overtaken by the evil of laziness

17:03:56 <bijan> EvilOfLaziness...

17:04:13 <bijan> The key point of the EvilOfLaziness is that *MY* laziness is ALWAYS good

17:04:21 <bijan> Everyone else's is evil :)

17:04:52 <DanC> "The three principal virtues of a programmer are Laziness, Impatience,

17:04:52 <DanC> and Hubris. See the Camel Book for why.

17:04:52 <DanC> " -- man perl(1)

17:05:32 <DanC> in this case, my laziness trump's norms. Since I know he can fix it himself, I'm not gonna fix it for him :)

17:06:01 <DanC> indeed, it's my laziness that prompted me to invest in this esw wiki.

17:06:05 <ndw> Damn. Out lazied. Does that mean I trumped bijan since he did fix it?

17:06:33 <DanC> it means the wik worked.

17:06:35 <_joshua> By the same token, industrious, patience, and humility are also the virtues required for a community

17:07:07 <bijan> Hmm.

17:07:11 * DanC waves to Stuart, noting the better part of a quorum of sw-meaning folk

17:07:13 * ndw waves to Stuart

17:07:42 * Stuart was thinking Hmmm.... no Libby.... but I'll see her on Friday!

17:07:50 <bijan> Well, there must be a third

17:08:12 * bijan 's principle programming virtue are Money, Literacy, and

17:08:18 <DanC> shall we play tiddlywinks or whatever for sw-meaning telcon time?

17:08:23 <bijan> Of course, it would be good to successful post the first two

17:08:36 <Stuart> Bijan.... in isolation that as little existential for me :-)

17:08:37 <bijan> Which gives me my third "typing accuracy"

17:08:44 <bijan> Hiya Stuart

17:08:47 <Stuart> Hello

17:08:57 <bijan> Feel the isolation angst!

17:09:09 <Stuart> Anyone prepared to make an opening bid>?

17:09:18 <bijan> If we're going to do any swmeaning planning it'll have to be quick. I have to run in about 10 mintues

17:09:41 <bijan> I did add a comment to your page, dan, but I've not integrated them into your comments yet

17:09:58 <Stuart> Ditto... I suppose its a bit rude to check in and then rush back out.

17:10:08 * sandro wonders if anyone tried to match up times people have given.

17:10:12 <ndw> What's our geographic distribution? USA both coasts and Europe?

17:10:28 <DanC> sw-meaning telcon Wed, 24 Sep at 9am US/Chicago

17:10:45 <bijan> Stuart: not at all!

17:10:48 <sandro> for 1 hr?

17:10:58 * ndw will see DanC's bid

17:11:09 * DanC fidgets with http://www.timeanddate.com/worldclock/meeting.html

17:11:22 <DanC> yes, 45 to 90 minutes.

17:11:36 <bijan> Ouch. WSD F2F

17:11:40 <DanC> bzzt.

17:11:40 <bijan> For me

17:12:02 * DanC hopes for a suggestion from bijan

17:12:31 <DanC> not, it's not rude to pop in/out of IRC. it's the norm.

17:12:31 <bijan> Fri, that time

17:12:47 <Stuart> Weds was out for me.

17:13:19 <DanC> Fri 9am Chicago is 10am Boston, which is RDFCore telcon time. OK for me, and I think DanBri wouldn't mind missing... but it's a real conflict for Pat Hayes (I hope).

17:13:21 <ndw> Friday 26 Sep at 9:00 CDT is bid by bijan

17:13:33 <ndw> I can do Friday

17:13:39 <Stuart> Friday 27th 9am East Coast/2pm UK would work for me

17:13:56 <ndw> No, 10am EDT, Stuart. So 3pm UK if you're right about 9/2.

17:13:59 <bijan> DanC: I could go later or earlier than the RDF Core telcon time

17:14:22 <DanC> earlier conflicts with family foo for me, but I could perhaps do it once.

17:14:31 <Stuart> 3pm UK is still fine (unless we bleather on for hours and hours....)

17:15:11 * Stuart failed to parse 9/2 norm

17:15:42 * DanC reviews friday times in chicago, Boson, and London http://www.timeanddate.com/worldclock/meetingtime.html?day=26&month=9&year=2003&p1=136&p2=64&p3=43&p4=-1

17:16:07 <ndw> You said 9am was 2pm in UK, so I inferred that 10am was 3pm without checking your math, Stuart :-)

17:16:43 * sandro wonders if the guy from China is thinking of joining....

17:17:20 <Stuart> ....a there was me looking at a meeting in early February or September

17:17:24 <DanC> I bid Friday, September 26, 2003, at 16:30:00 UTC time http://www.timeanddate.com/worldclock/fixedtime.html?year=2003&month=9&day=26&hour=16&min=30&sec=0

17:18:14 <Stuart> Na... that's 17:30 UK...on a Friday... really bad karma

17:18:31 * DanC passes the tiddlywink to stuart, hoping for a counter-proposal

17:19:33 <DanC> I bid Thu 25 Sep 2p Chigago

17:19:53 <Stuart> Ok... I'll see your 16:30 UTC and raise you 15:30 UTC (hoping we haven't been there already).

17:19:58 <sandro> That's 30-minutes post WebOnt, yeah?

17:20:14 <sandro> No, it's more like -30

17:20:20 * DanC notes the futility of pronouns

17:20:36 <sandro> Number your bids?

17:20:37 <Stuart> [race condition... which bid]

17:21:12 * ndw seconds 25 Sep 2p Chicago, just to put momentum somewhere

17:21:16 * DanC compares Fri 15:30 UTC with RDFCore telcon time... seems ok

17:22:07 * DanC seconds 16:30 UTC Fri http://www.timeanddate.com/worldclock/fixedtime.html?year=2003&month=9&day=26&hour=15&min=30&sec=0

17:22:42 <Stuart> meeting duration?

17:22:45 <bijan> I can make that work, once at least

17:22:47 <ndw> DanC meant 15:30 even though he said 16:30, yes?

17:22:54 <DanC> sorry...

17:23:01 <DanC> I second 15:30 UTC http://www.timeanddate.com/worldclock/fixedtime.html?year=2003&month=9&day=26&hour=15&min=30&sec=0

17:23:11 <bijan> Especially if we hold to an hour

17:23:12 * Stuart 's heart starts beating again

17:23:13 <DanC> for 45 to 90 minutes

17:23:13 * ndw thirds 15:30 UTC on 26 Sep

17:23:27 <DanC> we need it to stop at 16:30 sharp? I can live with that.

17:23:41 <bijan> It would work best for me, yeah

17:24:03 <bijan> Ok. must run. I endorse this and committ, all things willing enough, to being there.

17:24:08 * Stuart will do 45-90 mins starting 15:30 UTC Friday 26 Sep

17:24:36 <DanC> ok, 15:30 UTC Friday 26 Sep for 45-60 seems to meet a lot of relevant constraints.

17:24:45 <DanC> thanks, all, for playing.

17:24:57 * Stuart guesses we need to ask the folks that aren't here.

17:25:10 * ndw suggests DanC grab a bridge

17:25:18 <DanC> indeed, I need to announce it by email 7 days in advance;

17:25:45 <DanC> the next step in the transaction is to send an agenda 24 hours in advance, or to figure out something else in the mean time.

17:26:09 <Stuart> Ok... now I'll toddle out the room and head for home...

17:26:24 * DanC looks, with hope, at sandro

17:26:47 * DanC realizes he already wrote "I need to..."

17:26:49 <Stuart> Dan... do you have enough input from the intro's to structure an agenda (*not* volunteering).

17:26:53 <sandro> <looks around> what? what? :-)

17:27:00 <DanC> no, I don't have enough for an agenda.

17:27:02 <DanC> only a time.

17:27:15 <DanC> this proposal for time will also be a call for agenda input

17:27:32 <DanC> and call for chair nominations, probably.

17:27:52 <bitsko> bitsko is now known as bitsko|lunch

17:27:58 <Stuart> sounds good....

17:28:06 <DanC> ciao, Stuart

17:28:11 <Stuart> bye

17:29:36 <sandro> Sounds good, Dan.

17:30:05 * sandro flips through the intro's to see if this time is brazenly dis'ing anyone.

17:37:23 <DanC> logger, pointer?

17:37:23 <DanC> See http://ilrt.org/discovery/chatlogs/rdfig/2003-09-17#T17-37-23

17:43:48 <sandro> Check complete: no clear conflict between that time and stated constraints. (Peter committed "mid-morning" on fridays.)

18:18:08 * danbri catches up

18:18:11 <danbri> what was decided?

18:19:06 <DanC> more or less.

18:19:36 <DanC> London Fri 4:30 PM Fri 26Sep

18:19:42 * danbri reading yr mail

18:19:46 <danbri> ok that's after RDFCore. I'm in.

18:22:32 * DanC noodles on connections between addressbook used for auto-completion in my mailer and my list of people I've met in my travels

18:22:45 <DanC> ... and the one in my WearableGizmo

18:23:07 <DanC> i.e. the one used to recognize incoming calls and to place outgoing calls. (and to compose mail on the gizmo)

18:24:10 <DanC> ... and my email inbox whitelist...

18:24:55 <DanC> and ConfidentialityVersusAvailability

18:32:19 <_adam_work> _adam_work is now known as adamhill

20:08:28 <mdupont-work> mdupont-work is now known as mdupont

20:17:03 <bitsko|lunch> bitsko|lunch is now known as bitsko

20:40:35 <dajobe>http://search.cpan.org/author/ZOOLEIKA/RDF-Simple-0.04/

20:40:36 <dc_rdfig> A: http://search.cpan.org/author/ZOOLEIKA/RDF-Simple-0.04/ from dajobe

20:40:51 <dajobe> A:|Perl RDF::Simple 0.04, 15 Sep 2003 by Jo Walsh

20:40:51 <dc_rdfig> Titled item A.

20:48:46 <DanC>http://www.w3.org/2003/08/binary-interchange-workshop/

20:48:46 <dc_rdfig> B: http://www.w3.org/2003/08/binary-interchange-workshop/ from DanC

20:49:00 <DanC> B:|W3C Workshop on Binary Interchange of XML

20:49:00 <dc_rdfig> Titled item B.

20:49:08 <DanC> B:24th, 25th and 26th September, 2003, Santa Clara, California, USA

20:49:08 <dc_rdfig> Added comment B1.

20:49:36 <DanC> B:browsing position papers...

20:49:36 <dc_rdfig> Added comment B2.

20:51:00 <DanC> B:"While we expect to see non-standard binary forms used internally within certain vendors? implementations, including perhaps our own, we are not yet convinced that there is justification to standardize an interchange format other than XML 1.x." says [Conner and Mendelsohn of IBM|http://www.w3.org/2003/08/binary-interchange-workshop/19-IBM-IBMPositionPaperBinaryXMLWorkshop-updated.html]

20:51:00 <dc_rdfig> Added comment B3.

20:52:51 <dajobe> B:the ExpWay [http://www.w3.org/2003/08/binary-interchange-workshop/15-Expway-PositionPaper-20030816.zip|position], html in a zip (sigh! they love their binary) ponders future work on binary formats for RDF "that is fast, easily fragmentable, and easy to index (preferably in ways that are harder with such storage methods as RDBMSs). Current ongoing work from various parties on addressing the mismatch between trees and graphs ma

20:52:51 <dajobe> y unlock possibilities for integration in that domain, and we will be glad to explore solutions in that space again."

20:52:51 <dc_rdfig> Added comment B4.

20:52:53 <dajobe> gah

20:53:08 <dajobe> B:... may unlock possibilities for integration in that domain, and we will be glad to explore solutions in that space again."

20:53:09 <dc_rdfig> Added comment B5.

20:59:33 <dajobe> hmm, the w3c link validator robot attacks the logs again

21:00:17 <danbri> what's the problem? did you report it?

21:00:25 <dajobe> it's not clear

21:00:43 <dajobe> looks like somebody tried to link-validate the irc logs home page

21:01:01 <dajobe> seems like the link validator doesn't think it is a robot, no robots.txt fetches

21:01:19 <dajobe> defacto standards, aren't they great!

21:02:27 <DanC> w3c link validator robot?

21:03:01 <danbri> and why is that a problem? is it doing some kind of monsterous recursive get somehow?

21:03:08 <dajobe> it's stopped now

21:10:41 <JibberJim> - http://www.w3.org/Bugs/Public/show_bug.cgi?id=27 <- it's known.

21:14:37 * DanC hadn't realized W3C offered link validation service to the public somehow

21:15:14 <DanC> so the problem was reported Dec 2002

21:15:44 <DanC> er... bug 27 goes back even further... 2002-10-25

21:15:56 <DanC> sigh... what good does it to to track bugs if you're not gonna fix them?

21:16:51 <DanC> hmm... this ville.skytta accepted an assignment

21:16:55 <DanC> I wonder what that means

21:54:24 <qq[Schuyler]> qq[Schuyler] is now known as Schuyler

21:56:56 <mdupont> mdupont is now known as md-back-08

21:57:05 <md-back-08> md-back-08 is now known as md-back0800CET

22:08:28 <verbosus> Hello everyone. I?m currently playing with the RAP toolkit and am having some problems with some N3 statements, can anyone help?

22:08:49 <deltab> what problems?

22:09:37 <verbosus> deltab, I am writing the following:

22:09:41 <verbosus> @prefix foaf: <http://xmlns.com/foaf/0.1/> .

22:09:41 <verbosus> <#me> a foaf:Person ;

22:09:41 <verbosus> foaf:name "Antonio Cavedoni" ;

22:09:41 <verbosus> foaf:weblog <http://cavedoni.com/blogorroico/> .

22:10:32 <verbosus> I am parsing this N3 and then trying if I can serialize it back onto itself with the RAP toolkit.

22:11:19 <verbosus> But the last URI gets converted to an unknown @prefix. So, is my N3 wrong or is the RAP serializer being clunky?

22:11:52 <verbosus> This is what I get in return:

22:11:57 <verbosus> # Generated by N3Serializer.php from RDF RAP.

22:11:57 <verbosus> # http://www.wiwiss.fu-berlin.de/suhl/bizer/rdfapi/index.html

22:11:57 <verbosus> @prefix foaf: <http://xmlns.com/foaf/0.1/>.

22:11:57 <verbosus> @prefix ns0: <http://cavedoni.com/blogorroico/>.

22:11:57 <verbosus> @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>.

22:11:58 <deltab> I suppose it's allowed to do that

22:12:02 <verbosus> #me a foaf:Person ; foaf:name "Antonio Cavedoni" ; foaf:weblog ns0:.

22:12:32 <verbosus> But that ns0: isn?t a namespace, it is a resource!

22:12:52 <verbosus> Well, at least I think I correctly specified it as such, but am not sure.

22:12:52 <deltab> yes

22:13:38 <mortenf> there's really no diff between a namespace uri and a "regular" uri, but i'm not sure that syntax is allowed?

22:13:39 <deltab> the prefixes aren't part of the model

22:14:29 <verbosus> mortenf: but shouldn?t a proper parser check for at least ?@prefix? to determine whether something is a NS or not?

22:14:44 <mortenf> it doesn't matter, the result is the same.

22:15:11 * verbosus doesn?t quite understand.

22:15:18 <mortenf> once it's parsed (from e.g. n3), it's just triples, you can't tell which namespaces were declared

22:15:36 <timbl_UK> Prefixes are just a shorthand.

22:16:02 <dajobe> my n3 parser is happy with it :)

22:16:09 <timbl_UK> RAP decided to use a shorthand to write that URI. It made up a prefix "ns0", declared it at the top, and used it in the middle.

22:16:09 <mortenf> is :a b:c d: . valid?

22:16:22 <dajobe> if b: and d: prefixes are defined

22:16:37 <verbosus> Right, so there is no way for me to parse that N3 and then have it back as the original?

22:16:37 <mortenf> hmm, ok.

22:16:39 <timbl_UK> and the null prefix is defined

22:16:43 <dajobe> yeah

22:16:44 <mortenf> yep.

22:17:29 * timbl_UK doesn't use that but has seen other people do it. Should put a test into the test suite.

22:17:48 <verbosus> I was thinking to have some sort of application where one could scribble in N3.

22:17:51 <dajobe> is ':' alone allowed?

22:17:59 <deltab> are zero-length localnames allowed?

22:17:59 <verbosus> Then that would be parsed and stored into a DB.

22:17:59 <dajobe> i.e. the default namespace URI

22:18:18 <verbosus> But to edit your scribbles I would need to be able to get my original N3 file back, in some form.

22:18:31 <mortenf> if "some form", then no problem.

22:18:32 <dajobe> deltab: no localname in the "qname" means the namespace URI I think

22:19:20 * verbosus is slowly starting to get it.

22:19:27 <dajobe> other funky cases: <> <#>

22:19:33 <deltab> Notation3.html doesn't allow it

22:19:42 <deltab> it doesn't allow :foo either

22:19:48 <dajobe> it what?

22:19:54 <deltab> the grammar

22:19:56 <verbosus> Right, mortenf. Maybe I could explicitly tell the serializer not to use that shorthand.

22:20:01 <timbl_UK> <> and <#> are definitely allowed

22:20:02 <dajobe> <deltab> Notation3.html doesn't allow it<=what

22:20:29 <deltab> zero-length prefixes or localnames

22:20:37 <verbosus> So that I can get my triples laid out in a more straightforward way, and then edit them easily.

22:20:38 * mortenf finds it hard to read, since not bnf

22:21:02 <timbl_UK> The URI is defined to be a concatenation of the namespace string in the @prefix declation and the local name.

22:21:05 <timbl_UK> Hence, "prefix".

22:21:06 <DanC> .google RAP toolkit

22:21:08 <datum> RAP toolkit: http://www3.wiwiss.fu-berlin.de/rdfapi-php/test/jena-rdql-testcases.php

22:21:32 * timbl_UK woudl like to write it in BNF in RDF

22:21:43 <mortenf> of course :)

22:21:46 <verbosus> Yeah timbl, thanks for the clarification.

22:22:04 <timbl_UK> not BNF?

22:22:24 <timbl_UK> not bnf in ways i didn't notice?

22:23:17 <dajobe> heh: "qname ... localname ??? Allow omit colon when prefix void - keyword clash"

22:23:17 <mortenf> well, i guess it is (close) to bnf, but since it doesn't cover the reality...

22:23:29 <DanC> sbp did an exhaustive study of the syntax of localnames in N3. The quantity of conflicting evidence is remarkable.

22:23:40 <dajobe> sounds exhausting

22:23:57 <DanC> which reality do you subscribe to, mortenf? I subscribe to the test suite.

22:24:48 <timbl_UK> The BNF is currently the proposed syntax which people seemed to be mode happy with here.

22:24:55 <timbl_UK> In the @forAll for example.

22:24:57 <mortenf> well, that's probably the best idea, but :a doesn't seem to be legal according to the bnf

22:25:11 <mortenf> (nor a:)

22:25:19 <timbl_UK> If we had teh BNF in N3 then we could generate test sutes from it.

22:26:10 <timbl_UK> Yes, the BNF does insist on some character.

22:26:21 * DanC notes "the BNF" is ambiguous

22:26:36 <mortenf> i'm refering to http://www.w3.org/DesignIssues/Notation3.html

22:26:42 <timbl_UK> localname :: alpha | localname alphanumeric

22:26:48 <timbl_UK> me too

22:28:14 <mortenf> mind you, i'm not complaining, don't feel like writing a parser anyway (since dajobe is doing it for me...)

22:28:35 <dajobe> it's definitely ntriples+n (n=6 currently) rather than n3--

22:28:48 <mortenf> 'nuf for me

22:29:05 <dajobe> it's committed in the CVS, BTW

22:29:36 <mortenf> i think i'll hold off for a couple of days, just got it all running again (seems a recompile of curl did the trick)...

22:30:07 <dajobe> I'm getting failed locks in the anoncvs for 2000/10/swap

22:30:19 <dajobe> cvs server: failed to create lock directory for `/sources/public/2000/10/swap' (/sources/public/2000/10/swap/#cvs.lock): Permission denied

22:30:49 <timbl_UK> cwm allows: @prefix : <#> . : : : .

22:30:55 * DanC figures out what RAP is... RAP - RDF API for PHP V0.6 http://www.wiwiss.fu-berlin.de/suhl/bizer/rdfapi/

22:31:02 <dajobe> cwm line noise, yay :)

22:31:03 <mortenf> heh.

22:31:19 <timbl_UK> Anyone like to propose arguments for allowing or disallowing zero length local names?

22:31:25 * DanC can see cwm/N3 ascii art coming...

22:31:33 * timbl_UK cwm --morse

22:31:50 <dajobe> well, I read that _-_ encoding thing already - that's crazy

22:31:52 <mortenf> nah, my only angle would be readability

22:31:57 <verbosus> timbl_UK: cwm --braille

22:31:59 <deltab> in general, or in qnames?

22:32:07 <timbl_UK> It is mathematically neat to allow them.

22:32:11 <dajobe> you mean forbidding a: ?

22:32:12 <timbl_UK> in qnames

22:32:12 <DanC> I abstain from local name syntax until we figure out how to keep the code and the grammar documentation in sync, ala yapps.

22:32:27 <timbl_UK> I mean to forbifd or to allow a:

22:32:52 <dajobe> how do you abbreviate that @prefix'd uri?

22:33:12 <timbl_UK> I don't want to use yayayaya. I want to use something which the community will all parse. So RDF.

22:33:38 <timbl_UK> which @prefixed URI?

22:33:59 <dajobe> you do @prefix a: <http://example.org/foo#>

22:34:04 <dajobe> then want to use that

22:34:25 <dajobe> so what could you do instead of: a: a owl:Ontology .

22:34:32 <dajobe> hmm, maybe a was a bad example

22:34:57 <timbl_UK> The rules are strict about concatenation, so you can't slip in out a hash.

22:35:29 <timbl_UK> Maybe we should have ma de it so one is slipped in if it wasn't there, but it is too late now.

22:36:00 <mortenf> nn

22:36:11 * timbl_UK very vaguely remembers stg about "a zero length fragid is the same as no fragid"

22:36:41 <dajobe> well that'll confuse me; won't that meen no diff between n3 <> and <#>

22:36:44 * timbl_UK heading for bed too.

22:36:52 <dajobe> see you Friday :)

22:37:01 <timbl_UK> Yeup!

22:37:11 * DanC looks for a place to file RAP in the wiki so I and others can find it later... chooses WebDataInterfaceDesign

22:37:32 <timbl_UK> dajobe: good point.

22:37:35 * DanC loves that wiki bookmarklet!

22:37:58 <timbl_UK> I don't know where that came to mind from.

22:38:14 * DanC recalls <> vs. <#> discussion in uri@w3.org... pretty sure they're not synonyms, per IETF consensus

22:38:30 * timbl_UK relieved

22:38:44 <DanC> they might be synonyms per RFC1630

22:41:07 * DanC wonders about something more straightforward like RdfSoftwareList or RdfApiList

22:41:46 <verbosus> RdfToolkits?

22:42:38 <DanC> hm... ResourceDescriptionFrameworkSoftwareDevelopmentKits 1/2 ;-)

22:45:57 <DanC> I see daveb filed it under RDF Editors and Tools

22:46:06 <DanC> -> http://www.ilrt.bris.ac.uk/discovery/rdf/resources/#sec-tools

22:46:50 <dajobe> yeah, another bucket

22:48:57 * DanC cites /rdf/resources/ from a few places in the wiki...

23:09:42 <bitsko> bitsko is now known as bitsko|dinner

23:40:20 <adamhill> adamhill is now known as _adam_work