Semantic Web Interest Group IRC Chat Logs for 2004-03-16

00:33:49 <earle> earle is now known as grault

03:11:37 <IsoosI> :)

03:12:02 <DanC> what's so funny?

03:13:25 <IsoosI> just having a good day :)

03:14:02 <IsoosI> how are things going with you? :)

03:19:08 <DanC> good... found cheap zope hosting. http://www.nidelven-it.no/hosting

03:19:20 <DanC> moving my chuch web site (http://www.fellowshipofgrace.org/) there

03:19:35 <DanC> ah... 9pm has come and gone... time to put the kids to bed

03:26:49 <IsoosI> cool

03:40:55 <eaon> eaon is now known as eaon|zZz

04:37:41 <jsled_laptop> Is the tarball at http://www.w3.org/2000/10/swap/cwm.tgz relatively current?

04:53:23 <DanC> doubt it.

04:53:59 <DanC> if you're feeling generous, you could try it and report any problems to public-cwm-bugs@w3.org

04:54:12 * DanC wanders off for the night...

05:35:36 <ericP> does anyone know if OilEd has migrated to OWL?

09:22:07 <arnarl|away> arnarl|away is now known as arnarl

10:05:33 <swh> anyone know if the DAWG working group is still open for people to join? Our local guy still hasnt put my name forward :(

10:11:42 <ericP> yup

10:11:44 <ericP>http://www.w3.org/2003/12/swa/dawg-charter

10:11:45 <dc_rdfig> A: http://www.w3.org/2003/12/swa/dawg-charter from ericP

10:11:45 <sh1m> I haven't heard that it isn't.

10:11:56 <sh1m> ericP, isn't it some insane time in .jp?

10:11:57 <ericP> A:|RDF DAWG Charter

10:11:57 <dc_rdfig> Titled item A.

10:12:12 <ericP>http://www.w3.org/2002/09/wbs/33280/2004-02-RDFDataAccess-nom/

10:12:12 <dc_rdfig> B: http://www.w3.org/2002/09/wbs/33280/2004-02-RDFDataAccess-nom/ from ericP

10:12:26 <ericP> B:|RDF DAWG Signup

10:12:30 <dc_rdfig> Titled item B.

10:12:50 <ericP> actually, it's an unusually sane time

10:13:01 <ericP> even close to business hours: 19:12

10:13:11 <sh1m> wow. Impressive.

10:13:25 <ericP> however, i am frequently on at less sane hours

10:13:36 * sh1m busy lounging around on irc while his boss is in a meeting.

10:13:40 <sh1m> aren't we all

10:13:48 <ericP> gotta run and finish a converstaion

10:14:06 <ericP> you're not supposed to lounge on irc anymore

10:14:10 <ericP> that's what the web is for

10:14:18 <swh> thanks

10:14:22 <ericP> ok, really gotta run. bbiab.

10:42:54 <chaalsNCE_> chaalsNCE_ is now known as chaalsNCE

10:45:39 <sh1m> hey Charlie

11:09:27 <eaon|zZz> eaon|zZz is now known as eaon

11:20:25 <sh1m> ericP, you still busy?

11:20:58 <sh1m> libby, hellow btw

11:21:15 <libby> heya sh1m

11:22:03 <GNUPredator> GNUPredator is now known as M7U90nt

11:29:25 <dajobe> hi nick

11:34:34 <nmg> morning dajobe

11:35:18 <nmg> is there an SWBP telecon this week?

11:35:29 <dajobe> I think so, thursday 3pm uk time

11:35:37 <dajobe> libby's attending

11:37:16 * nmg fails to see an agenda on the mailing list

11:37:40 * nmg will be attending as well, providing dder manages to put the nomination in

11:38:01 <dajobe> agenda should be due today - isn't it 48 hours notice?

11:38:08 <dajobe> maybe 24

11:39:29 <dajobe> 24 http://www.w3.org/2004/02/Process-20040205/policies.html#GeneralMeetings

11:41:11 <aharth> what are the current topics for the SWBP wg?

11:41:21 * aharth may join as well

11:43:36 <libby> there's some info here aharth: http://www.w3.org/2004/03/04-SWBPD

11:43:48 <libby> (near the bottom)

11:44:23 <aharth> cool, thanks

11:44:48 * nmg is still a little narked that he couldn't get to the tech plenary

11:45:16 <libby> boo

11:45:19 <libby> shame

12:29:44 <nmg> nmg is now known as nmg_lunching

14:07:41 <nmg_lunching> nmg_lunching is now known as nmg

14:15:27 <libby> oh cool, the mozilla calendar people hang out on irc: #calendar on irc.mozilla.org

14:16:46 <sh1m> heh I bet you are a happy chappy now

14:18:24 <libby> in some sense of 'chappy' yeah ;)

14:19:33 <sh1m> hehe I didn't mean it like that. And now I am going to leave before I say anything "but you are all woman" bleh!

14:20:06 <libby> hehheeh

14:35:10 * chaalsNCE waves to libby-who-is-all-woman

14:36:05 <libby> hey chaalsNCE

14:36:13 <sh1m> I have no qualms in pointing out what a rugged beast of a man chuck is.

14:36:54 <sh1m> But I reckon its the crocadile dundee hat that gives it away

14:37:00 <libby> hehehhe

14:37:09 <libby> stop making me laugh

14:37:13 * libby very serious

14:37:20 * chaalsNCE imagines :-)

14:37:45 * sh1m points

14:37:52 <sh1m> look libby is working! *gasp*

14:37:56 <sh1m> er isnt...

14:40:31 <sh1m> libby, is all your iCalander stuff in python?

14:41:04 <sh1m> or maybe Dan's

14:41:19 <libby> DanC's stuff is in pythpon and before that perl - I had a converter in java but it's not up to date

14:43:12 <sh1m> hmm, I am talking to someone who wants a calander type product in the python based cms i muck about with

14:43:20 <sh1m> just thinking about portability

14:44:04 <dajobe> which cms?

14:44:25 <chaalsNCE> plone, right?

14:48:14 <sh1m> Plone

14:48:15 <sh1m> #plone

14:48:58 <sh1m> It is built on Zope's CMF so decent metadata is just starting to come through, and there are a couple of RDF powered proposals for things comming out now.

14:50:24 <M7U90nt> M7U90nt is now known as n000b

14:50:30 <chaalsNCE> sh1m, I have been doing a bit of [calendar manipulation in RDF using cwm|http://www.w3.org/2001/sw/Europe/200401/actions] to deal with stuff in a local format and getting it to be iCal/RDF happy.

14:52:11 * sh1m looks

14:54:05 <dajobe> I heard good things about plone. We use it here at ILRT.

14:54:28 <sh1m> woo!

14:54:34 * sh1m is so proud

15:08:20 <dajobe> of course I also hear "zope crashed/spinning out of control again, reboot it" :)

15:08:43 <sh1m> ah well, zope 2.6 sucks

15:08:59 * eaon has a zope 2.5.something instance which transfers sometimes 10gb a day

15:09:18 <sh1m> it also eats ram for breakfast, elevens, tweveles, lunch, afternoon tea, dinner, supper and a bedtime snack.

15:09:20 <eaon> i haven't had much problems with zope yet - only with bad zope apps :P

15:09:30 <eaon> hehe yeah

15:20:14 <n000b> n000b is now known as M7U90nt

15:22:43 <CaptSolo_> CaptSolo_ is now known as CaptSolo

15:52:22 <ericP> sh1m, tx for the comprehensive list of meals

15:54:26 <chaalsNCE> sh1m, you forgot second lunch, second dinner, apero, and midnight fridge raid.

15:54:59 <LotR> apero?

16:01:50 <swh> swh is now known as swh_afk

16:05:34 <chaalsNCE> aperitivo / aperitif / little snacky before dinner.

16:10:52 <sh1m> that would be the aussie thing of cute short versions of perfectly good words ;)

16:12:26 <sh1m> ericP, what are the requirements of someone wanting to join DAWG?

16:17:58 <sh1m> danbri you handy for a question?

16:18:43 <danbri> su...

16:18:45 <danbri_> ...re!

16:18:52 <danbri_> what's up?

16:19:16 <ericP> sh1m, loosely, you should respresent a member of W3C

16:19:43 <sh1m> ericP, thats what I wanted to know. And if one didn't then it would be a no no. right?

16:19:55 <ericP> pretty much

16:20:12 <sh1m> ok

16:20:14 * danbri_ wonders whether sh1m's employers might be persaudable to join

16:20:20 <sh1m> not for me boss

16:20:23 <ericP> there's an invited expert mechanism, but you have to demonstrate some expertise that we wouldn't get from someone in the WG

16:20:29 <sh1m> i have a _heck_ of a lot to fo

16:20:54 <sh1m> ericP, yeah aparently WCAG had to change the rules so I could join the party.

16:21:10 <ericP> was it a fun party?

16:21:57 <OliB> hi!

16:22:08 <OliB> while i seem someone talking...

16:22:37 <sh1m> ericP, its a long long party I'll tell you that ;)

16:22:40 <OliB> where should i send my suggestion, if i have use cases of my own that i would like to propose regarding dawg ?

16:23:25 <OliB> www-rdf-rules ?

16:27:12 <ericP> OliB, that would probably be best

16:27:35 <ericP> DanC and I can discuss this mechanism, but I'm sure that whatever you send there will be noticed.

16:33:52 * chaalsNCE has a slight delay in looking for Libby

16:33:55 <chaalsNCE> Libby?

16:38:46 <libby> hello

16:42:04 <chaalsNCE> Does stuff other than the weblog geet into the swad-e newsletter?

16:42:11 * chaalsNCE thinks not... just checking

16:45:42 <swh_afk> swh_afk is now known as swh

16:46:00 <libby> I can do if you want to

16:46:22 <libby> I have put stuff from steve cayzer's weblog in the newsletter on his request

16:52:42 * chaalsNCE ponders. Having to blog stuff is yet another new game that I didn't feel a need to learn. On the other hand, it seems to be where the world wants to go - at least the bloggers' world.

16:53:32 <libby> it's up to you charles. I think blogging is a useful dissemination tool. but I'm just after content really

16:53:49 <libby> it's nice to see what people on the project are doing in one place too

17:02:12 * bryce sends message to mailing list re: image manipulation schema

17:02:37 <DanC> libby, you're thinking of attending the DAWG telcon this week?

17:02:49 <libby> yeah ericp suggested I come along

17:02:59 <libby> if you think that's ok

17:03:02 <DanC> er... are you joining the WG?

17:03:24 <libby> I don;t think so no

17:03:27 * DanC is confused

17:03:32 <libby> I can;t commit that much time

17:04:11 <DanC> I think ericp got confused between BPDWG and DAWG

17:04:12 <libby> if it's not ok, andyS and alberto and I were all workign together - they can talk about it (and are planning on/have joined)

17:04:26 <DanC> andyS and Alberto have joined, yes.

17:04:39 <libby> it was just to talk abotu the work we did on rdf query tests in swad-e

17:05:21 <libby> - http://www.w3.org/2003/03/rdfqr-tests/

17:05:59 <DanC> I'm sure we'll get to that eventually. I think alberto mentioned it in his intro. but I don't expect to cover that in this week's telcon

17:06:25 <libby> ok, it's prompted me to revisit anyway.

17:07:20 <DanC> ok, so we'll expect you at the BPDWG telcon but not at DAWG, ok libby?

17:07:41 <DanC> indeed, alberto's intro cites...

17:07:41 <DanC> [4] http://www.w3.org/2003/03/rdfqr-tests/[5] http://www.w3.org/2003/03/rdfqr-tests/rdf-query-testcases.html[6] http://rdfstore.sourceforge.net/2002/06/24/rdf-query/

17:08:45 <libby> I probably misled ericp; initially I said I would join DAWG at least for a while;

17:09:49 <libby> I'd like to help him out if it's useful to him

17:09:57 <libby> ...and the group :)

17:13:02 <DanC> it's completely up to you and the folks at bristol whether you join or not. But I'd like to keep it fairly clear who's in the WG.

17:13:15 <DanC> you're more than welcome to join.

17:13:36 <libby> thanks :)

17:13:49 <sbp>http://www.tbray.org/ongoing/When/200x/2004/03/15/EndTAG

17:13:50 <dc_rdfig> C: http://www.tbray.org/ongoing/When/200x/2004/03/15/EndTAG from sbp

17:14:09 <sbp> C:|Eddies in the TAG-Time Continuum

17:14:09 <dc_rdfig> Titled item C.

17:14:26 <libby> my issue is that after Oct I can't commit to anything; thought it would be more useful to join BP only because of that...

17:15:09 <sbp> C:"Is he? What's he doing there?"--A. Dent. Farewells to TBray and thanks for all his good work will no doubt be piling in soon, and, to start the ball rolling, farewell TBray, and thanks for all the good work!

17:15:09 <dc_rdfig> Added comment C1.

17:45:25 * DanC doesn't grok the title sbp gave to C

17:46:04 <ndw> "Eddie's in the space time continue? Is he? What's he doing here?"

17:46:04 <Talliesin> Talliesin is now known as Tallie_pint

17:46:15 <sbp> HHG reference

17:46:39 * ndw adds a reference to "DanC" to direct his attention my preceding thought

17:47:14 * sbp fully expects a Chesterfield sofa to pop out into the middle of www-tag given TBray's departure

17:47:23 <DanC> yeah, I recognize the HHG reference; I don't see why it makes a good title for tbray's article

17:47:45 <sbp> ah, because his leaving will certainly affect the TAG in various subtle ways

17:48:14 <DanC> that's an interesting comment/observation on his essay. I still don't see how it's a good title.

17:48:45 * DanC tries to remember what font packages to install to get documents like http://kanzaki.com/docs/sw/xh2rdf.html to look right in galeon

17:48:57 <sbp> I would've used his choice of "</TAG>", but I hate using end tags (ha) with no opening tag in sight

17:49:12 <danbri_> C:|Eddies in the TAG-Time Continuum - Tim Bray leaves the TAG

17:49:17 <danbri_> C:|Eddies in the TAG-Time Continuum - Tim Bray leaves the TAG

17:49:17 <dc_rdfig> Titled item C.

17:49:21 <danbri_> ok compromise?

17:49:26 <sbp> yup, v. good

17:49:32 <DanC> better, yes

17:51:02 <sbp> I'm recognizing a few patterns here. DanC gives a subtle hint to sbp (sbp is an existentially quantified variable, though), sbp pretends to miss it, danbri mediates, hillarity results...

17:51:38 <sbp> if only we had an Rdfig/SwigSocialWiki

17:51:48 <DanC> ?

17:52:10 <DanC> collaboration patterns are welcome in the ESW Wiki

17:52:18 <DanC> witness InstantGig

17:52:25 * sbp looks

17:53:43 <sbp> interesting. no link to GarageBand? not pertinent enough?

17:53:56 <DanC> lassitude

17:54:13 * sbp nods

17:54:17 <DanC> assumed it was known to the readers at the moment

17:54:59 * sbp would add his stereo-as-multitracking trick to some page in this area but for lassitude

17:56:56 <DanC> 2004-03-16 11:56:17 version of http://esw.w3.org/topic/InstantGig now has a link

17:57:55 <sbp> thanks

17:58:20 <DanC> chuckle... "A bookmarklet accident, sorry" -- http://esw.w3.org/topic/KNOWLEDGE_20MANAGEMENT_3a_20SEMANTIC_20DRIFT_20OR_20CONCEPTUAL_20SHIFT

17:59:45 * sbp raises an eyebrow as he observes the first known case of bookmarklet vandalism

18:00:16 * DanC puzzles over ericp's changes to http://esw.w3.org/topic/CurriedFunction

18:00:39 <DanC> "These properties have no useful semantics." yes, they do.

18:00:45 <sbp> is there a styleguide for the ESW wiki at all?

18:01:11 <DanC> well, yes, the FrontPage

18:01:19 <sbp> hmm. title searches use GET? ew

18:01:33 <DanC> why should they not use GET?

18:01:37 <sbp> er, POST rather

18:01:44 <sbp> they should use GET, they don't

18:01:48 <DanC> ah. bug.

18:02:05 * DanC makes a note in http://esw.w3.org/topic/AboutThisService

18:02:10 <DanC> logger, pointer?

18:02:10 <DanC> See http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/2004-03-16#T18-02-10

18:02:46 <DanC> "Eeek! the ?action=subscribe thingy is an unsafe GET!" -- http://esw.w3.org/topic/MoinMoinToDo

18:04:37 * sbp reads "WikiNames vs. Normal Titles"

18:04:51 <sbp> WikiNames aren't very i18n friendly, but that's the whole IRI debate

18:05:32 <DanC> hmm

18:06:16 <DanC> I don't think WikiNames are among the larger barriers for participation by non-english-speakers in the ESW community

18:06:17 <sbp> most non-Latin scripts don't have a case distinction (except for Cyrillic, Greek, and Armenian, which are close to/derived from the same roots as Latin script)

18:06:54 <DanC> wikipedia attacked the multilingual problem space head on; bless them!

18:06:59 <sbp> me neither, particularly. would be more of an annoyance in places such as Wikipedia, but they don't follow a com... heh, right

18:07:31 <golbeck_> golbeck_ is now known as golbeck

18:07:36 <DanC> there is definitely some room for more of a styleguide.

18:08:06 <DanC> I see lots of people "doing it wrong" in the ESW wiki. IslandTopics, redundnant headings, topics based on issues rather than patters, etc.

18:08:33 <sbp> going through http://spacecrib.sytes.net/friki/ I observed an intriguing, but humourous, pattern based partly on experience with ESW: <sbp> "Named for an obscure '80s alternative music hit" - 80s hits seem to be a good source of WittyMemedTitles (cf. ILoveNuclearPower, and esw:DontWorrybeCrappy)

18:09:03 <sbp> yes. I've also noticed you taking people up on it fairly often, which is why I naturally asked if you'd put all such occurances into a style guide

18:09:06 <DanC> I love nuclear power? what 80's hit does that hark to?

18:09:39 <sbp> not sure. the quote's from http://spacecrib.sytes.net/friki/view?ILoveNuclearPower but it doesn't state the actual source

18:11:23 <DanC> perhaps "this page is about..." is an anti-pattern. e.g. "This page serve as a dashboard..." http://esw.w3.org/topic/QaFramework

18:11:36 <DanC> if you have to say what the page is about, you're doing something wrong. just say it!

18:12:01 <sbp> yes. something I've been meaning to write about recently is abstract/introduction repetition that I see occuring, for example, in a lot of W3C specs, and IETF RFCs

18:12:38 <DanC> repeating the abstract in the introduction is normal/correct in the tech report genre.

18:13:29 <sbp> it's silly though. I can't imagine any other genre of writing where it would be considered acceptable. it's really very silly--I've not seen any other genre of writing where it would be considered acceptable

18:14:00 <dajobe> haha

18:14:09 * DanC doesn't see why it's silly in tech reports

18:15:17 <sbp> because you're giving people the same information twice; it's most pernicious when the text is actually copied from the abstract into the introduction. it can be skillfully subverted by a writer who takes the abstract text and expands upon it in the introduction adding more information as they go along

18:16:18 <jsled> Speaking of DontWorryBeCrappy ... here's something I've been thinking about recently, and mostly-wrote-up last night. Feedback appreciated. http://www.asynchronous.org/twiki/bin/view.pl/Main/AppDataExchange

18:27:47 <ericP> 2004-03-16T18:00:39Z <DanC> "These properties have no useful semantics." yes, they do.

18:28:06 * DanC edited http://esw.w3.org/topic/CurriedFunction

18:28:12 <ericP> i think i make a pretty strong case. i'd be interested in hearing support for "yes, the do."

18:28:36 <DanC> pls read updated http://esw.w3.org/topic/CurriedFunction

18:29:43 <ericP> red

18:30:18 <ericP> what practical use can you make of the :chicagoBetween property when looking for cities between NCY and LA?

18:31:02 <DanC> you answered that yourself: ?city :betweenProperty ?p.

18:31:02 <DanC> :NewYork ?p :LosAngeles.

18:31:25 <ericP> right, i haven't done any sleceting there, i've asked for *any* relation

18:31:34 <DanC> yes. quite.

18:31:44 <ericP> this means that the property itself is adding nothing but syntax

18:31:53 <ericP> at least to that query

18:32:04 <DanC> yes, the CurriedFunction pattern is all about syntax.

18:32:23 <DanC> it's a pattern that you apply when the syntax has only 2-ary relationships.

18:32:26 <ericP> which is why i say that the addtional property adds no sematncis

18:32:41 <DanC> but you're wrong. it does add semantics.

18:33:11 <ericP> what use can i make out of it?

18:33:29 <DanC> you can express a 3-ary relationship in RDF.

18:34:06 <ericP> but it can't be used to recover them so it doesn't *mean* anything beyond a syntactic trick

18:34:31 <DanC> right, it's nothing beyond a syntactic trick. but that syntactic trick is not nothing.

18:34:57 <DanC> that syntactic trick is the CurriedFunction pattern.

18:34:59 <ericP> but it is practically devoid of semantics. all the sematnics are in the :between property

18:35:44 <DanC> "practically devoid of semantics"? I have no idea what you mean by that.

18:36:07 <ericP> if i invent words everytime i want to write something down, i can't communicate with anyone

18:36:38 <DanC> you can if you relate those new words to other words that they know.

18:36:39 <ericP> if i give people a book relating thw words i'm uttering to words they already know, the semantics are all in that book

18:37:04 <DanC> and...? what's your point?

18:37:21 <DanC> (I don't agree the semantics are all in the book, btw)

18:37:45 <ericP> yeah, that was a bad and inaccurate analogy

18:38:21 <ericP> but who is going to use the :chicagoBetween property to recover data?

18:38:39 <sandro> Ewwww, reading http://esw.w3.org/topic/CurriedFunction is nearly impossible, the way it mixed thread-mode and consensus-mode.

18:38:58 <DanC> which version, sandro?

18:39:14 <ericP> basically, it means ','

18:39:18 <sandro> People will query for that property without knowing its name, EricP.

18:39:30 <sandro> (last modified 2004-03-16 13:27:47) , DanC.

18:40:23 <sandro> ericP, are you familariar with Currying, from contexts where its syntax is natural and convenient? (eg Haskell)

18:40:55 <ericP> familiar, but i don't think it's a very good idea except in corner cases

18:41:50 <DanC> you don't think it's a good idea to apply the pattern to RDF? or you don't think haskell is a good idea?

18:42:10 <sandro> Why? The plus is it uses fewer triples. The minus is it confuses some people, and it requires higher-order functions (which RDF doesn't prevent, but some folks might avoid implementing).

18:42:17 <DanC> (haskell borrowed it from ML, I think. I think ML is the first place it appeared in a programming langauge)

18:42:27 <ericP> apply to RDF, or most other scenarios

18:43:26 <ericP> it's like uttering a nonsense word because you have to utter a word

18:43:27 <DanC> I don't think there's any intrinsic reason to prefer ArgumentList, RecordDescription or CurriedFunction. each sucks in its own way.

18:43:42 <sandro> I happen to agree Currying isn't an RDF Best Practice, mostly on the confusion front. But I also find it very appealing for elegance and efficiency.

18:43:56 <DanC> CurriedFunction is nice cuz it's regular. it works the same for 2-ary, 3-ary, or 10-ary functions.

18:44:27 <DanC> RecordDescription and ArgumentList have a sharp disconnect between 2-ary and 3-ary relationships.

18:45:23 <sandro> How does ArgumentList have a sharp disconnect there?

18:45:42 <ericP> i think they all do.

18:46:00 <sandro> Oh, sorry, I thought you meant between 3 and 4. Never mind.,

18:46:20 <sandro> RDF has a sharp one between 1 and 2, too.

18:46:36 <ericP> how is that different from the disconnect in Currying?

18:46:56 * DanC struggles to explain

18:47:22 <sandro> Prolog while naturally supporting n-ary predicates sometimes works better with ArgumentList, eg for VarArgs situations.

18:47:46 <sandro> (the other two would be very problematic in prolog.)

18:48:49 <sandro> (It's common to put optionally arguments tagged into a list argument. But access to args is much more efficient than access to the database, I believe.)

18:49:56 <DanC> I can't seem to explain how 3-ary relations are more regular. But there's a whole school of thought that goes wild with the idea. they call them "monads". well, monads are functions, not arbitrary relations. and the ML programming language is born from the idea.

18:50:02 <ericP> (read another way, currying introduces an infinite number of terms into the field where we look for sematnics

18:50:19 * DanC can't make sense of ericp's use of "semantics"

18:50:20 <ericP> )

18:50:31 <ericP> practicaly use of the term

18:50:48 <ericP> building a query, writing a rule, to solve a real problem

18:51:23 <DanC> I use all three (subj, pred, obj) for building queries, writing rules, solving real problems.

18:52:17 <ericP> predicate is not promoted to a higher level?

18:52:33 <DanC> higher in what sense?

18:52:46 <ericP> don't you look up the predicate in a browser?

18:52:51 <ericP> a foaf:knows b.

18:53:11 <ericP> where do you look to see what it means (assuming a and b are urls)

18:53:12 <ericP> ?

18:53:20 <DanC> at all three.

18:53:30 <DanC> and lots of other places.

18:56:14 <ericP> i'm surprised to say that you don't look for sematnics in the predicate and authority in the subject and object

18:56:33 <ericP> s/in the/from the/g

18:57:45 <ericP> anyways, track what happens when you extend such a relation ni a way that shouldn't contradict the orig meaning, only extensds it.

18:58:08 <ericP> http://esw.w3.org/topic/RecordDescription shows such a case

18:58:09 <dc_rdfig> D: http://esw.w3.org/topic/RecordDescription from ericP

18:59:11 <DanC> I don't see the point you were making in RecordDescription either. I edited it as such.

23:36:27 <eaon> eaon is now known as eaon|zZz