Semantic Web Interest Group IRC Chat Logs for 2003-12-11

00:38:50 <DanC_jam> .time

00:38:51 <datum> Thu, 11 Dec 2003 00:38:51 GMT

00:38:57 <DanC_jam> BLURB: XML Query and RDF Query and life, the universe, and...

00:38:57 <dc_rdfig> A: XML Query and RDF Query and life, the universe, and... from DanC_jam

00:39:27 <DanC_jam> A:attendance list in paper

00:39:28 <dc_rdfig> Added comment A1.

00:39:38 <DanC_jam> logger, chump A

00:39:38 <DanC_jam> A:See [http://ilrt.org/discovery/chatlogs/rdfig/2003-12-11#T00-39-38|discussion]

00:39:39 <dc_rdfig> Added comment A2.

00:39:44 <DanC_jam> . DanBri

00:39:47 <DanC_jam> . Jerome Simeon

00:40:07 <DanC_jam> . Peter Funkhauser (sp?)

00:40:26 <DanC_jam> . Mikah Dubinko (sp?)

00:42:13 <DanC_jam> . Eric Prud... (sp?)

00:42:13 <DanC_jam> . Dan Connolly

00:42:13 <DanC_jam> . [missed] @ Adobe

00:42:13 <DanC_jam> . Nabil Alshama (sp?) @ Adobe

00:42:13 <DanC_jam> . Michael Rys @ Microsoft

00:42:13 <DanC_jam> . John Mostrum (sp?) @ Adobe

00:42:29 <DanC_jam> . Howard Kats @ fatdog

00:42:57 <DanC_jam> . Paul Turn (sp?) @ Network Inference

00:43:11 <DanC_jam> . John Cowan @ Reuters

00:43:29 <DanC_jam> . Uche O[sp?]

00:43:34 <DanC_jam> . [missed?]

00:44:03 <DanC_jam> . [missed?] @ Adobe

00:44:08 <DanC_jam> . Jonathan Robie

00:44:20 <shellac> Uche Ogbuji

00:45:00 <DanC_jam> . Martin Duerst

00:45:26 <DanC_jam> Paul C gives status of XQuery:

00:45:36 <DanC_jam> . long lifetime of WG, many pages

00:45:57 <DanC_jam> . went to last call 12 Nov 2003, incl. XPath 2.0, XSLT 2.0

00:46:16 <DanC_jam> (Paul C clarifies that everything's public)

00:46:34 <DanC_jam> . except formal semantics doc; still has 2 open issues

00:47:13 <DanC_jam> . last call duration 3 months

00:48:07 * DanC_jam notes a blurb on this meeting called A

00:49:40 * danbri wondres if folk have been taking notes here

00:49:46 <danbri> I was scribbling into a text editor

00:50:17 * DanC_jam has been taking some notes, encourages danbri to mail his editor buffer to www-archive and link it from A

00:50:29 <shellac> what's going on there?

00:51:19 <DanC_jam> A:[http://lists.w3.org/Archives/Public/www-rdf-rules/2003Dec/0030.html|BOF notice]

00:51:19 <dc_rdfig> Added comment A3.

00:52:44 <shellac> cheers DanC_jam

00:56:27 <danbri> done, waiting for it to arrive in archive.

00:56:28 <DanC_jam> PaulC: there's also fulltext work... wasn't making a lot of progress until recently

00:56:28 <DanC_jam> PaulC: and we're looking at an update language

00:56:28 <danbri> A:See also some [http://lists.w3.org/Archives/Public/www-archive/2003Dec/0029.html|notes I took] on PaulC's intro. Moving to IRC note-taking now.

00:56:29 <dc_rdfig> Added comment A4.

00:56:29 <danbri> Michael Rys (to J Robie): while as you say it is true many impls map xqueries to sql query... many of them haven't addressed all the subtleties of different semantics of these 2.

00:56:40 <danbri> rys: its easier to model the relational model into xml than rdf's graphs

00:57:03 <danbri> ...even in case of relational mappings. at the 50,000 ft level v easy, but in practice at ground level, some difficulties

00:57:14 <danbri> ...am not advocating you use xquery for a purely relational store

00:58:12 <danbri> ...so some caveats there.

00:58:12 <danbri> Paul: (invites danc to give rdf intro)

00:58:12 <danbri> DanC: an rdf REC from '99. RDFS was PR then too, parked at CR until moved into IG mode, then convened a WG to address those issues, now done, LC in Jan, 2nd LC, now claim we're pretty much done, anticipating PR soon.

00:58:50 <danbri> ...meanwhile, people playing w/ rdf since '99, many folks built query engines, stored in sql, flatfile, several QLs

00:58:58 <danbri> ...people starting doing interop testing, collecting test cases

00:59:09 <danbri> ...getting WGish, so we sketched a charter for discussion

00:59:27 <danbri> paulc: I forgot to mention XML Schema

00:59:37 <danbri> ...one thing is we're in lockstep with XML Schema

01:00:00 <danbri> ...validation semantics after construction based on XML Schema. Also did the F+O library.

01:00:10 <danbri> ...which fleshed out what XML Schema began.

01:00:28 <danbri> jerome: another big thing...

01:00:33 <danbri> xquery worked on use cases early on.

01:00:43 <danbri> design of language was really driven by that. a good experience.

01:00:52 <danbri> ...in rdf query context, rdf folks might want to try that.

01:02:59 <danbri> paul: historically, doing datamodel 1st, and being usecase driven, those were key.

01:02:59 <danbri> john cowan: is xquery tightly bound to complex type validation

01:02:59 <danbri> DanC: in OWL we did requirements. In RDF and OWL we did testcases

01:02:59 <danbri> danbri: do what extent do folks using relax-ng 'not get to play'?

01:02:59 <DanC_jam> re use cases... lemme look at what we suggested way back in June 2001 http://lists.w3.org/Archives/Public/www-xml-query-comments/2001Jun/0007

01:02:59 <danbri> jonathon: xquery works perfectly well w/ just wf xml, dtd-based xml, etc

01:03:02 <danbri> ...how it works with relax-ng being worked out.

01:03:08 <danbri> ..since it lacks type validation

01:03:22 <DanC_jam> phpht. 404 @ http://swordfish.rdfweb.org/discovery/2001/06/squishtests/

01:03:44 <danbri> cowan: some relax-ng extensions do that, for those relax-ng schemas that are unambig

01:03:48 <danbri> .google squishtests

01:03:50 <datum> squishtests: http://lists.w3.org/Archives/Public/www-xml-query-comments/2001Jun/0007.html

01:04:09 <danbri> paulc: we had requirement to use xml schema

01:04:59 <danbri> ..

01:05:19 <danbri> paulc: user defined types, functions... defined in terms of xml schema datatypes

01:05:29 <danbri> ...so cant do what relax-ng allows, which is plugging in arbitrary datatype systems

01:05:40 <danbri> cowan: to clarify... does this mean that you can implement these functions in xquery?

01:05:53 <danbri> paulc: we give you a syntax for defining modules, functions etc. turing complete

01:06:17 <danbri> martind: rdf can use xml schema datatypes, but also allows others

01:06:24 <danbri> paulc: there are a couple hooks in there

01:06:29 <danbri> ...but not much

01:06:43 <Davey> Davey is now known as D[a]vey

01:07:02 <DanC_jam> "the standard sql xml..." hmm... I'll have to learn what that is.

01:09:47 <danbri> danbri: last thing under (2) in agenda... that OWL has RDF query-esque features: you can give membership criteria for classes

01:15:13 <danbri> last thing the logger got: http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/2003-12-11.html

01:15:21 <danbri> logger, pointer?

01:15:21 <danbri> See http://ilrt.org/discovery/chatlogs/rdfig/2003-12-11#T01-15-21

01:15:44 * danbri blahs at ilrt.org -> www.ilrt.bris.ac.uk redirects

01:16:01 * danbri not trying to capture all discussion, wants to listen.

01:17:36 <danbri> jerome: old xquery (1yr ago) had an arrow operator

01:17:45 <danbri> semantic is just like rdf-navigate function shown here

01:18:02 <danbri> paulc: in our old requirements doc, we had a requirement to be able to follow pointers

01:18:19 <danbri> ...originally we wanted to do it, but hard, so we cut it.

01:18:36 <danbri> robie: disagree w/ history. it was that xpointer wasn't done

01:18:49 <danbri> paulc: yup, agree. if xpointer existed, we'd likely have done it.

01:19:02 <danbri> cowan: easy for xqueries w/ unique resolution

01:19:44 <danbri> paulc: We didn't say it'd be easy to implement! that's never stopped us before ;)

01:19:54 <danbri> jerome: issue number is 0134

01:20:14 <danbri> .google xquery issue 0134

01:20:16 <datum> xquery issue 0134: http://lists.ibiblio.org/pipermail/xquery/2001-February/000131.html

01:20:27 <danbri> 'dereference operator and links'

01:20:45 <danbri> robie: re syntactic web, it was important to be able to query rdf and xml together

01:21:01 <danbri> ...also, there are at least two natural ways to normalise rdf data

01:21:17 <danbri> (i) turn it all into triples

01:21:23 <danbri> (ii) merge everything by subject

01:21:27 <D[a]vey> that has to be the nicest .ac.uk website I've seen

01:21:54 <danbri> cowan: that's close to the xml serialization

01:23:30 <danbri> paulc: you made a key point... something that motivated the time i spent on xquery...

01:23:47 <danbri> ...i became v convinced that sql wasn't going to be the ql we use for everything

01:23:57 <danbri> ....i spent a long time doing object extensions for sql etc

01:24:01 <DanC_jam> (which use case?)

01:24:22 <danbri> ...concerned about msg we send if we have one language for xml , one for rdf

01:24:26 <danbri> ...how to integrate?

01:24:48 * danbri stops scribing a bit

01:25:58 <DanC_jam> I like that idea... an xquery "standard library" for querying RDF

01:26:09 <DanC_jam> (Henry Thompson walked in at some point)

01:28:37 <DanC_jam> "skolem functions" come up... I grok them now, but I remember distinctly in the QL '98 workshop when everybody was using the term and I didn't grok.

01:30:49 <danbri> +1 re std lib

01:31:18 <danbri> (Norm Walsh walked in too)

01:31:29 <shellac> the standard lib is a syntax normaliser?

01:31:46 <DanC_jam> hmm... did Peter say skolem functions are an open problem? looks solved to me. cwm implements it, just cuz that was the obvious thing to timbl

01:31:48 <danbri> it could do that, plus fancier things (didn't discuss detail much)

01:32:23 <shellac> fancier things ... mumble detail :-)

01:33:11 <DanC_jam> heh... the infamous "clarification question", cloaking an argument

01:34:42 <shellac> a fine and noble tradition

01:36:12 * danbri tries to find out if a user-defined function that mapped rdf's xml syntax to triples could have a widely agreed name

01:36:29 <danbri> "could we have a module that had implementions in either c, or xquery, but had same name for the function"

01:37:16 <DanC_jam> Rys explains that xquery is functional except for node construction. I commented on that "except for node construction" (in www-xquery-comments) and I'm not really satisfied by the explanation I've got so far, so I intend to re-raise it now that they're in last call.

01:38:29 <danbri> [action] danbri to study this part of the xquery spec (names for external functions, dep on impl or not?)

01:38:47 <danbri> henry: both I and Norm have some experience using xpaths to navigate graphs

01:38:51 <danbri> ...it works

01:38:55 <DanC_jam> on node construction, 12 Apr 2001 http://lists.w3.org/Archives/Public/www-xml-query-comments/2001Apr/0014.html

01:39:01 <danbri> ...should be able to transfer this experience to rdf graph nav

01:39:14 <danbri> ...will leave to email issues re circularity

01:39:26 <danbri> norm: I have a practical rdf session tommorrow

01:39:34 <danbri> ....but more dog/pony show than detailed tech

01:39:41 <danbri> cowan: is that full xpath?

01:39:50 <danbri> henry: no, multiple parents etc., meaning would be unclear

01:40:05 <danbri> [...]

01:41:30 <DanC_jam> A:speaking of PaulC's exhortation to study xquery, my comments that I want to double-check: [http://lists.w3.org/Archives/Public/www-xml-query-comments/2001Apr/0014.html|12Apr on xquery constructors not being functional]

01:41:30 <dc_rdfig> Added comment A5.

01:44:21 <danbri> ...

01:44:28 <danbri> rys: couple of statements

01:44:33 <danbri> i) tech comment

01:44:47 <danbri> agree w/ peter and henry's comments... basically that yes, we can define an xml view over rdf

01:45:08 <danbri> but since rdf is a graph model, and subsumes trees, maybe forests too,... it is a more elab model

01:45:21 <danbri> ie extended ER model w/ Description Logic capabilities

01:45:29 <danbri> ...if you go this route, what are you querying?

01:45:34 <danbri> abstract notion of relationships?

01:45:42 <danbri> probably don't want to query physical relations

01:46:10 <danbri> eg in rdbms, it eventually bottoms out in page-fetches, but we don't have Page-FetchQL

01:46:21 <danbri> ...rdf systems mix instance data w/ schema stuff

01:46:26 <DanC_jam> A:... and [http://lists.w3.org/Archives/Public/www-xml-query-comments/2001Apr/0014.html|14 Mar 2003 on fn:escape-uri]

01:46:26 <dc_rdfig> Added comment A6.

01:46:49 <danbri> ...rdf systems mix instance data w/ schema stuff

01:46:57 <danbri> ...other prob: if you map rdf into triples

01:47:02 <danbri> ...triples are very structured data

01:47:08 <danbri> ...in that case, xquery is overkill

01:47:16 <danbri> ...what does it give you to work w/ triples? An iterator

01:47:40 <danbri> ...similar stuff to what sql offers you

01:47:45 <danbri> ...except a nice xml serialization

01:47:49 <danbri> if you go down to triples

01:48:02 <danbri> ...but also you have this optimization issue, of all the semantics in the data, not in the structure

01:48:09 <danbri> ...we used OEM as a graph model

01:48:16 <danbri> ...for data integration, we had an operator called fusion

01:48:27 <danbri> ...a full left outer join based on a skolem function

01:48:51 <danbri> ...Given prop identifier, eg resourceID, you can recreate objects out of disperate parts of object descriptions

01:49:02 <danbri> (but this wasn't testsed/deployed outside research prototype)

01:49:15 <danbri> (ii) non-tech part

01:49:24 <danbri> ...how much are we talking research? vs products

01:49:30 <danbri> ...i represent a company normally, perhaps not here

01:49:35 <danbri> (Microsoft)

01:50:01 <danbri> [michael notes that in ansi context, you stand up to say things with yr corporate hat on]

01:50:13 <danbri> rys: i see rdf as a resrch proejct that hasn't moved outside research area

01:50:25 <danbri> ...some apps, but nowhere near where rdf folks want it to go

01:50:33 <danbri> ...ppl seem not to be that enthused about it yet.

01:50:49 <danbri> ...so Q: how importqant is it to have interop across rdf at this stage

01:50:58 <danbri> ...how much would this be research?

01:51:18 <danbri> ...if we can answer this q... worthwhile to do standardisation?

01:51:30 <danbri> ...would make sense as part of general RDF research effort, yeah fine go ahead

01:51:38 <danbri> ..but lots of tech issues that are research

01:51:40 <danbri> jerome:

01:51:50 <danbri> 2 things

01:51:53 <danbri> 1. techical

01:52:02 <danbri> ...i used to take j robie's view

01:52:11 <danbri> ...that rdf was close enough to xml, could use xquery

01:52:21 <danbri> ...but then i started working w/ peter patel schneider

01:52:28 <danbri> ...we realised that more differences than we thought

01:52:32 <danbri> not just tree v graph

01:52:39 <danbri> also ordering is a big diff

01:52:49 <danbri> also rdf strong distinction between classes and relationships

01:52:56 <danbri> while at xml level just elements

01:53:04 <danbri> ...doing rdf in xml, you'll do both as elements

01:53:18 <danbri> ...querying relationships vs classes matters in rdf world

01:53:36 <danbri> ...there might be enough diff in data model

01:53:52 <danbri> ...to justify an rdf ql

01:54:09 <danbri> 2. w/ xquery hat, i'd like to resist urge to add stuff to this already complex language

01:54:25 <danbri> took a long time to design and implement it already, so new rdf-specific stuff would be costly.

01:54:33 <danbri> similarly, why not at ebxml extensions

01:55:01 <danbri> also...

01:55:07 <danbri> if you come up w/ a good rdf query language...

01:55:10 * DanC_jam recalls chairing HTML when everybody wanted "just a few attributes, and this one tag" for their thing

01:55:22 <danbri> ...you can compile them into this sort of xquery stuff

01:55:58 <danbri> [missed some]

01:56:20 * DanC_jam would like to get outa here

01:56:32 <danbri> ericP: responding to rys

01:56:43 <danbri> ...re is it appropriate to use xquery for querying rdf

01:56:54 <danbri> ...we have ways to use xquery to query stuff we'd normally access w/ sql

01:57:04 <danbri> ...we say xquery good for this, but also we don't tell them to scrap sql

01:57:14 <danbri> ...so 'should there be a separate rdf ql'

01:57:23 <danbri> ...would you tell them to scrap sql?

01:57:42 <danbri> current rdfq strawman charter says, if we don't explicitly use xquery, come up w/ a binding to it

01:57:53 <danbri> ...so it gives xquery access to rdf data, but not necc v-versa

01:58:24 <danbri> ...probly both rdf and xml folks feel their representation is the fundamental conduit

01:58:40 <danbri> paulc: I didn't see the charter

01:58:48 <danbri> robie: re yr charter...

01:58:55 <danbri> ...good to have use cases and requirements way up front

01:59:14 <danbri> v easy to solve problems that aren't real, and give theoretical justifications

01:59:25 <danbri> ...when we did xquery we had 7 diff languages as starting pt

01:59:46 <danbri> ...we did use cases to figure out what prob we were solving

02:00:08 <danbri> ...i'm still not sure what prob rdf solving

02:00:16 <danbri> ...use cases are yr concrete reqs

02:00:38 <danbri> 2nd thing

02:00:44 <danbri> ...important to think of strawman architectures

02:00:53 <danbri> ...practical ways to implement

02:00:59 <danbri> ...also 1 other model for integration

02:01:02 <danbri> ...text search

02:01:27 <danbri> ...completely different datamodel, words/sentences/paragraphs/synonyms/stemming [soundex/...]

02:03:29 <danbri> ...should we explore these together?

02:03:56 * danbri listens not scribes

02:04:17 <danbri> ...

02:04:29 <danbri> paul (net infererence): my company comes at this from an OWL perspective

02:04:39 <danbri> ...what does all this mean in terms of querying OWL fits in?

02:04:59 <danbri> ...i would say we are sensitive to (i) too many standards (ii) appreciate possible kitchensink problem.

02:05:18 <danbri> ...having said all that, what we're currently doing is implementing our OWL qery impl in xquery

02:05:25 <danbri> ...or a strange subset of it ;)

02:05:44 <danbri> ...had good discussion w/ ericp, danb and others... re ways of doing this

02:05:53 <danbri> cowan: is your rdfq against the rdf serialization?

02:06:02 <danbri> paul: it is all custom functions

02:06:11 <danbri> ...interested to see folks might find custom libraries acceptable

02:06:17 <danbri> ...thats the path we're on right now

02:06:39 <danbri> cowan: no doubt it would work, but is it best/cleanest

02:09:01 <DanC_jam> A:hmm... took a look at [http://www.w3.org/TR/xquery-use-cases/|XQuery use cases]. the TOC looks like a lot of greek letters, but if you read you get some stories.

02:09:01 <dc_rdfig> Added comment A7.

02:09:57 <DanC_jam> A:the [http://www-db.research.bell-labs.com/user/pfps/papers/yin-yang.pdf|yin/yang paper] was nominated by Robie

02:09:57 <dc_rdfig> Added comment A8.

02:10:55 <DanC_jam> A:follow-up discussion is vaguely directed at [http://lists.w3.org/Archives/Public/www-rdf-rules/|www-rdf-rules]

02:10:55 <dc_rdfig> Added comment A9.

02:11:23 <deltab> is there a use case that includes URI encoding?

02:11:35 <danbri> adjourned...

02:14:09 <deltab> DanC_jam: A5 and A6 have the same URI

02:14:35 <shellac> thanks for the scribing dan 'n dan

10:09:52 <md-zzzz> md-zzzz is now known as mdupont

12:25:57 <karlcow> karlcow is now known as karl_

12:26:09 <karl_> karl_ is now known as karlcow

12:26:53 <ircleuser> ircleuser is now known as teefal

13:43:14 <larsbot> does anyone know of an online list of commercial projects that have used RDF?

13:44:07 <danja> larsbot - maybe trawl Dave Beckett's big list?

13:44:29 <larsbot> do you have a URI for it?

13:44:45 <larsbot> found it. thanks!

13:45:36 <danja> not sure how up-to-date - has it got SemaView stuff?

13:45:47 <larsbot> yes

13:46:06 <danja> good man that dajobe

13:46:55 <larsbot> yep :-)

13:47:46 <danja>http://www.ilrt.bris.ac.uk/discovery/rdf/resources/

13:47:46 <dc_rdfig> B: http://www.ilrt.bris.ac.uk/discovery/rdf/resources/ from danja

13:48:27 <danja> B|Dave Beckett's Resource Description Framework (RDF) Resource Guide

13:48:34 <danja> oops

13:48:43 <danja> B| Dave Beckett's Resource Description Framework (RDF) Resource Guide

13:49:05 <danja> B:| Dave Beckett's Resource Description Framework (RDF) Resource Guide

13:49:05 <dc_rdfig> Titled item B.

13:49:10 <danja> ah

13:49:36 <danja> B: The HUGE list

13:49:37 <dc_rdfig> Added comment B1.

13:49:48 <danja> B: still going strong

13:49:48 <dc_rdfig> Added comment B2.

13:51:04 <danja> since? ages ago...

13:55:42 * Ankh just found a beermat with Libby's cell phone number on it

13:55:55 <Ankh> not entirely sure why

14:01:33 <Ankh> morning Dan

14:26:48 <DanC_jam> morning.

14:47:52 <libby> oh hello liam, didn;t realize it was you!

14:48:10 <libby> !pic liam

14:48:20 <whwhwhwh> Gregory Todd Williams http://kasei.us/images/greg.png

14:48:21 <whwhwhwh> William Loughborough http://swordfish.rdfweb.org/photos/2001/05/dms/wai2.jpg

14:48:22 <whwhwhwh> Stuart Williams http://www.w3.org/2002/05/tagboard.jpg

14:48:23 <whwhwhwh> William Kearney http://www.ideaspace.net/users/wkearney/images/109-0913_IMG-SM.JPG

14:48:24 <whwhwhwh> Gregory Todd Williams http://kasei.us/images/greg.png

14:49:23 <libby> !pic liam quin

14:49:26 <whwhwhwh> Liam Quin http://swordfish.rdfweb.org/photos/2003/05/08/2003-05-08-Images/7.jpg

15:01:29 * timbl impressed with wh^4

15:02:48 <libby> oh thanks :)

15:04:14 <timbl> !pic bush

15:04:18 <whwhwhwh> no results found, sorry

15:05:58 * libby was very chuffed with this:

15:06:05 <libby> !picpath libby, neil

15:06:27 <jordan-> wh^4?

15:06:56 <libby> hm

15:07:27 <whwhwhwh> Libby Miller to Neil Tennant via Dan Connolly, Tim Berners-Lee, Bill Clinton, Bono, Brian Eno

15:07:28 <whwhwhwh> in 6 steps

15:07:29 <whwhwhwh> see http://swordfish.rdfweb.org/discovery/2002/02/paths/byname.jsp?name=libby&name2=neil

15:08:21 <Jhendler> Libby, neat -- I think photomarkup/process a big win for SemWeb -- if owl.mindswap.org ever comes up, see our image demo -- we're storing svg

15:08:44 <Jhendler> overlays separate from the photos, so we keep the metadata in RDF w/pointer back to image and put it together at display time

15:08:54 <Jhendler> can highlight who is who in the images and the like

15:09:08 <Jhendler> unfortunately, construction in our lab killed a disk last night -- sigh

15:09:11 <libby> right, neato, kinda like jibberjim's stuff

15:09:14 <libby> :(

15:09:31 * libby used to be a big pet shop boys fan

15:10:27 <Jhendler> yeah, we interacted with jibberjim on this - ours a bit different - expands on his ideas

15:10:34 <libby> cool

15:11:26 <Jhendler> we have a separate tool for marking up images that can export to our website through an RDF API ... so we can do arbitrary markup (not just foaf) and have it flow to our site stuff

15:11:55 <Davey> Davey is now known as D[a]vey

15:37:00 <libby> hello eikeon

15:45:01 * DanC_jam would like to talk to somebody from network inference...

15:47:16 <DanC_jam> Paul Turner maybe... anybody see him? danbri?

15:47:25 <DanC_jam> ericp?

15:49:28 <danbri> gone home, i believe.

15:49:48 <eric>http://www.xmlconference.org/xmlusa/2003/thursday.asp#3

15:49:48 <dc_rdfig> C: http://www.xmlconference.org/xmlusa/2003/thursday.asp#3 from eric

15:49:52 <DanC_jam> got his email address handy? has the attendance sheet from last night gone out?

15:50:31 <eric> C:|Russian Dolls and XML: Designing Multi-Version XML Documents - Robin La Fontaine & Thomas Nichols

15:50:31 <dc_rdfig> Titled item C.

15:51:05 <danbri> try paul.turner@networkinference.com (per http://groups.yahoo.com/group/jena-dev/message/4124)

15:51:13 <DanC_jam> thx.

15:52:10 <Ankh> how was the query bof?

15:52:36 * libby was wondering that

15:53:15 <danbri> I found it pretty useful

15:53:32 <danbri> Not polarised or anything, just ppl talking around aspects of the problem(s)

15:53:35 <Ankh> cool

15:53:41 * Ankh is an XML Queerie

15:53:48 <Ankh> er wait, I didn't say that :)

15:53:51 <libby> heh

15:53:52 <danbri> tell me something i don't know

15:54:14 <Ankh> danbri: :-)

15:56:05 <LotR> xml queery. heh.

15:57:22 <Jhendler> libby - dawns on me - you have a foaf DB w/a bunch of photos and people, we have a foaf-based DB with a bunch of people and photos

15:57:46 <libby> :)

15:57:52 <Jhendler> we should get a couple of our folks to try out RDFApi or the like by linking these -- be a good "hands across the ocean" demo

15:58:07 <libby> have you got a netapi-type access to it?

15:58:08 <Jhendler> and that sort of thing. We don't count for SWAD-E, but it wuld still be cool to do

15:58:17 <D[a]vey> D[a]vey is now known as Davey

15:58:18 <Jhendler> working on it -- what about you folks?

15:58:20 <Ankh> I nevr did put the foaf gallery up for the Gnome project, looks like it doesn't make email addresses public any more?

15:58:23 * libby has but not exactly well doicumented or stanards-compliant etc

15:58:48 <libby> all wh4's data goes via varioue badly-documented netapis :)

15:59:05 <Jhendler> eikeon and kendall are working on doing this for all out stuff - would be interesting to get some communication going on it

15:59:11 <Jhendler> s/out/sour

15:59:18 <Jhendler> s/sour/our (Argh)

16:00:19 <libby> Ankh: yep, we tend to use foaf:mbox_sha1sums these days, you were quite right :)

16:00:30 * Davey gets lost in FOAF recursion

16:00:37 <libby> cool Jhendler. I don;t know how to do it right though

16:01:52 <Jhendler> libby -- seems the easiest is to get the appropriate people in a chat or email discussion - daniel (Eikeon) could talk to someone from your place and we could see if it is a SMOP or real work

16:01:52 <libby> I want to be able to tell wh4 abotu a new service, i.e. how to access it, what questions you can ask it, and how to interface with a human. not worked out how to do that yet :)

16:02:01 <libby> SMOP?

16:02:16 <Jhendler> sorry, old's hacker's term - Small Matter of Programming

16:02:52 <libby> ah, gotya. yeah, be nice to chat with daniel

16:02:52 <Jhendler> (usually used to mean something that is straightforward, but a lot of work -- sort of like "Repatch the compiler to output the new codes, no problem, just a SMOP"

16:03:29 <Ankh> libby cool (re. md5), that woud make it useable for the project Id wanted, although I don't have time any more :(

16:03:56 <libby> aw boo :(

16:04:29 <eric> C:Russian doll because when you look at one of their documents, there are many different XML documents hidden here.

16:04:29 <dc_rdfig> Added comment C1.

16:05:33 <eric> C:This stuff is called Unified Delta.

16:05:34 <dc_rdfig> Added comment C2.

16:08:52 <eric> C:Robin La Fontaine giving a list of the benefits of expressing differences between XML documents in XML.

16:08:52 <dc_rdfig> Added comment C3.

16:10:20 <eric> C:Unified Delta provides "n XML files in one". The source data and the changes are integrated in a single document. All the versions of the document must have the same root element.

16:10:22 <dc_rdfig> Added comment C4.

16:11:09 <eric> C:A particular source document can be extracted with 60 lines of XSLT.

16:11:09 <dc_rdfig> Added comment C5.

16:11:23 <eric> C:Minor versions can also be easily deleted.

16:11:24 <dc_rdfig> Added comment C6.

16:11:46 <eric> C:All this is done with minimal duplication.

16:11:46 <dc_rdfig> Added comment C7.

16:13:12 <eric> C:Applications include content management, i18n, variant management, collaborative authoring, ...

16:13:13 <dc_rdfig> Added comment C8.

16:13:17 <MDubinko> hi kendall

16:13:27 <kendallclark> hi micah

16:13:41 <kendallclark> BLURB:Curing the Web's Identity Crisis: Subject Indicators for RDF

16:13:41 <dc_rdfig> D: Curing the Web's Identity Crisis: Subject Indicators for RDF from kendallclark

16:14:02 <kendallclark> D:(I think the "Identity Crisis" here came from an article of mine. Which is nice.)

16:14:02 <dc_rdfig> Added comment D1.

16:14:17 <eric> C:Robin La Fontaine giving markup examples.

16:14:17 <dc_rdfig> Added comment C9.

16:14:20 <kendallclark> D:Basically, a topic maps meets RDF talk (well, so far, I was a few minutes late.)

16:14:20 <dc_rdfig> Added comment D2.

16:14:44 <kendallclark> D:(Drops Hendler and Lassila from SciAm article, alas.)

16:14:44 <dc_rdfig> Added comment D3.

16:15:13 <kendallclark> D:I wish I grokked Topic Maps better; I've been saying that for 4 years, -sigh-

16:15:13 <dc_rdfig> Added comment D4.

16:16:35 * eikeon waves to libby, Jhendler, et al from a meeting -- will chat later.

16:16:46 <kendallclark> D:Shows the layer cake, with the comment: "There's no Topic Maps in here because the W3C didn't invent them. You can't expect them to..." -- Uh, we can't? Actually, I think that's something we should expect.

16:16:46 <dc_rdfig> Added comment D5.

16:16:46 <eric> C:The first version of the archive is built out of the original by adding a "dxu:vset" attribute to the root element.

16:16:47 <dc_rdfig> Added comment C10.

16:16:48 <libby> heya eikeon

16:17:20 <kendallclark> D:Ah, Pepper gives me mad props! "Title of an important article by..." me. Heh.

16:17:20 <dc_rdfig> Added comment D6.

16:17:41 <MDubinko> there are 26 W3C documents currently listed as in Last Call

16:19:10 <evlist> C:Modified text goes into "dxu:PCDATA" elements.

16:19:10 <dc_rdfig> Added comment C11.

16:19:41 <danbri> D:Unicode's in the layercake...

16:19:41 <dc_rdfig> Added comment D7.

16:19:46 <danbri> isn't it?

16:20:09 <kendallclark> danbri: yes, it's a silly comment, but has some measure of truth

16:20:23 <kendallclark> at least, i think we should work hard to make sure it's *not* true

16:20:25 <evlist> C:Added/removed elements identified through "dxu:vset" attributes.

16:20:25 <dc_rdfig> Added comment C12.

16:20:41 <kendallclark> well, "we" in the loose sense, i.e., you all!

16:20:51 <danbri> DAML+OIL wasn't created by a w3c wg.; but it fitted nicely(ish) into the stack...

16:21:24 <MarkB> SOAP too

16:21:42 <kendallclark> (i'm just reporting!)

16:21:54 <kendallclark> but i shall bring up these counterexamples in the q&a

16:22:12 <evlist> C:Updated attributes are transfomed into "dxu:attribute" elements.

16:22:12 <dc_rdfig> Added comment C13.

16:22:15 <MarkB> cool

16:22:16 <danbri> what room is it in? i'm not getting enough done here to justify missing talks...

16:22:39 <kendallclark> 109

16:22:54 <evlist> C is in room 111.

16:24:57 <evlist> C:That scales pretty well to mixed content.

16:24:57 <dc_rdfig> Added comment C14.

16:25:47 <evlist> C:The whole stuff is quite eleguant and readable.

16:25:47 <dc_rdfig> Added comment C15.

16:28:03 <evlist> C:The time to extract any version is the same and increases linearly with the size of the archive.

16:28:03 <dc_rdfig> Added comment C16.

16:28:51 <kendallclark> D:URIs identified for 2 distinct purposes: identify info resources; identify things that info resources describe

16:28:51 <dc_rdfig> Added comment D8.

16:29:14 <kendallclark> D:(This has implications for the social meaning cluster, as well.)

16:29:14 <dc_rdfig> Added comment D9.

16:30:02 <kendallclark> D:(And this isn't being offered as a new conclusion; as Pepper points out, the APW, in 2.2.5, talks about this distinction.)

16:30:02 <dc_rdfig> Added comment D10.

16:30:52 <evlist> C:Can give you an interlinear presentation of the same document in different languages.

16:30:52 <dc_rdfig> Added comment C17.

16:32:24 <kendallclark> D:(I've been working on a paper with Bijan P. about using David Lewis's notion of convention (as a 2-party cooperative coordination game) to address some of these issues. Wanted to submit to Phil of CS conference in Italy, but may hold it for next year's ISWC...)

16:32:24 <dc_rdfig> Added comment D11.

16:32:36 <evlist> C:Lets also you produce the list of changes to translated.

16:32:36 <dc_rdfig> Added comment C18.

16:34:43 <evlist> C:Nice concept of variant management applied to SVG documents.

16:34:43 <dc_rdfig> Added comment C19.

16:36:46 <kendallclark> D:Ah, we finally come to the point: "This fundamental ontological fact (gag!) -- that interesting things in teh world don't have addresses -- isn't addressed either by RDF or by the APW."

16:36:46 <dc_rdfig> Added comment D12.

16:37:33 <MDubinko> Paul Cotton: XML Query test suite to have 10,000 - 20,000 cases

16:37:45 <bijan> Wow.

16:37:51 <larsbot> bloody hell

16:38:00 * bijan totally doens't believe that that will be reliable

16:38:04 <MDubinko> they don't do anything small! :-)

16:38:09 <bijan> I mean, who's verifying these cases?

16:38:26 <kendallclark> hmm, after going to xquery for rdf f2f last night, i have to agree w/ micah

16:38:55 <bijan> So, I had a brainsomething last night, re: XQuery and RDF Rules

16:38:55 <evlist> C:See http://www.deltaxml.com/unified/

16:38:55 <dc_rdfig> Added comment C20.

16:39:08 <bijan> Instead of doing "Rules for the Web" in some freestanding way

16:39:18 <kendallclark> bijan: wish you'd been at the f2f (which was sort of bad, IMO)

16:39:49 <bijan> One might take XQuery, which is, after all, a functional programming language with native support for XML, and turn it into a functional/logic programming hybrid with additional support for RDF

16:39:55 <bijan> Somethin glike Mercury, in fact

16:40:21 <kendallclark> D:The come-to-jesus point: Published Subject Indicators. /me stifles-yawn

16:40:22 <dc_rdfig> Added comment D13.

16:40:22 <bijan> One might be able to do this rather cleanly, in fact.

16:40:48 <larsbot> kendallclark: why so negative?

16:41:07 <shellac> why use XQuery?

16:41:12 <kendallclark> larsbot: in general or...?

16:41:39 <evlist> C:No specific provision for metadata, but any metadata in the document is "versionized".

16:41:39 <dc_rdfig> Added comment C21.

16:41:43 <timbl> D: Of course, an RDF document may describe many resources, and resources may be described by many RDF documents. There isn't always a clear "subject" for a web page.

16:41:43 <dc_rdfig> Added comment D14.

16:42:05 <larsbot> kendallclark: I was thinking about D13

16:42:11 <kendallclark> fwiw, not negative, just didn't get enough PSI meat, for my tastes. no biggie.

16:42:14 <timbl> D: /me wonders what can't be names by an RDF document

16:42:14 <dc_rdfig> Added comment D15.

16:43:08 <MDubinko> (still boggling: that's 1 test case a minute for 4-8 working WEEKS)

16:43:10 <kendallclark> larsbot: not negative, just wanted something a bit less superficial.

16:43:25 <bijan> shellac: Cause it's already there, has deep XML support, etc. etc.?

16:43:31 <kendallclark> but it's fine, i can go read docs

16:43:33 <larsbot> kendallclark: I can understand that. difficult to fit that into 45 mins at XML USA, though

16:43:44 <bijan> shellac: Mindshare, avoiding NIH syndrome, etc. etc.

16:43:49 <kendallclark> larsbot: yes, i didn't really mean it as criticism. i'm just a habitual whiner. :>

16:43:56 <larsbot> D:clear subject: true, which is why PSI documents usually have to be written specifically to be PSIs

16:43:56 <dc_rdfig> Added comment D16.

16:44:15 <kendallclark> D:FWIW, my 'yawn' above wasn't meant as criticism. As Lars points out, it's hard to do a lot of detail in this venue, in 45 minutes.

16:44:15 <dc_rdfig> Added comment D17.

16:44:32 <MDubinko> oo

16:44:38 <bijan> My guess, thought, that it will be necessary to cut down on the semantics, to make that work

16:44:42 <larsbot> kendallclark: I don't mind that. I just wanted to know whether there was some tech point underneath that didn't come out

16:44:52 <kendallclark> larsbot: no. :>

16:44:59 <kendallclark> well, not that i'm ready to talk publicly about, yet.

16:44:59 <shellac> bijan: I know what you mean, but the deep XML support isn't that deep (aside from XSD)

16:45:15 <bijan> Sure it is

16:45:27 <shellac> bijan: and javascript has much more support, and a much bigger mindshare

16:45:27 <bijan> The XQuery type system is some varient of the XSD one

16:46:01 <bijan> Though Javascript 2.0 has a formal semantics, it's not clear how to extend it to a logic programming hybrid

16:46:16 <larsbot> kendallclark: let us know when you are

16:46:21 <bijan> Whereas one might be able to wholesale borrow from the various functional/logic hybrid in the case of XQuery

16:47:51 <shellac> I should add that I've been using XQuery over rdf

16:48:21 <larsbot> bijan: Jonathan Robie did a paper on querying RDF with XQuery, called "The Syntactic Web"

16:48:58 <bijan> Sure

16:49:02 <bijan> There's lots of that floating around :)

16:49:10 <kendallclark> hmm, do i stay for a ruleml talk...ugh.

16:49:12 <bijan> I was suggesting, i think, something slightly different :)

16:49:19 <kendallclark> (no, because i heard the same talk @ iswc)

16:49:19 <larsbot> I hope so... :-)

16:49:53 <bijan> Hence the "turn Xquery into a hybrid functional/logic programming langauge"

16:50:21 <bijan> Instead of "query rdf documents with XQuery"

16:50:25 <shellac> are you thinking of adding rdf extension functions to XQuery, or something deeper

16:50:27 <shellac> ?

16:50:43 <bijan> If you google for the mercury programming langauge

16:50:57 <bijan> Or any of the functional extentions to prolog

16:51:05 <bijan> You should get a sense of waht I meant

16:56:10 <bijan> Well, "want" is strong. I was just going to throw it out as a possibility :)

16:57:26 <shellac> ok, I get the picture

16:58:29 <shellac> I think XQuery would need major surgery :-)

16:58:37 * DanC_jam checks schedule re lunch

16:58:44 <DanC_jam> oops! Missed pepper on URI crisis

16:59:01 <kendallclark> DanC: i loosely churned it in D:

16:59:15 <bijan> Whatever

16:59:21 <bijan> I didn't say it would be trivial

16:59:22 <edd> it is the same talk we have heard for a while now, by the look of it

16:59:43 <kendallclark> from pepper? it's the same as the paper of the same name.

16:59:50 <bijan> I do think that it might be a reasonably clean extension, but that's just a guess

16:59:53 <kendallclark> but he pimped one of my deviants, so that was pleasant.

16:59:56 <shellac> did pepper mention owl lite (or tiny)?

17:00:01 <kendallclark> no

17:00:07 <bijan> Tiny?

17:00:13 <kendallclark> well, i missed first 9 minutes, but i doubt he did.

17:00:50 <shellac> tiny seems to be used for the a subset of OWL lite that's easy to do in rules

17:01:00 <bijan> Oh, owl feather

17:01:06 <shellac> eg jena2, sesame have used the term

17:01:25 <shellac> yeah - feather was another suggestion

17:01:37 <shellac> cuter :-)

17:01:45 <bijan> well, that's what ben grosof used in his talk

17:01:58 <bijan> Descriptoin Logic Programs: DLP

17:09:12 <DanC_jam> I wouldn't expect pepper's talk to be different, but a talk is an invitation to discuss the ideas.

17:09:52 <DanC_jam> discussing that with him was one of my main goals for Extreme Markup, but he didn't make it (weather or some such; I don't recall clearly).

17:10:04 <DanC_jam> Now he and I are in the same building, and I missed it. sigh.

17:10:46 <kendallclark> not much discussion; he took most of his time

17:10:48 * DanC_jam considers UriCrisis for the agenda of the SemWeb Best Practices ftf in March in Cannes

17:13:28 <danbri> shellac, ericp says "thank you kindly for the treehugger example"

17:13:38 <danbri> ..."was v useful"

17:15:08 <kendallclark> i was gonna churn this talk, but some wanker asked me to stop typing because "it distracts me" -- what a dope.

17:15:41 * libby had that once. but was at a librraian's conference

17:16:04 <kendallclark> i'm currently assaulting him w/ Mind Bullets.

17:16:29 * DanC_jam notes lunch in T-14min

17:16:44 * libby thinks the tapping of mac keybraods en mass is like the soothing sound of soft rain falling

17:18:09 <kendallclark> i grant that i type loudly, but come on!

17:22:28 <libby> danbri, ericp, shellac says, "no problem"

17:38:24 <Davey> Davey is now known as D[a]vey

17:43:52 * Jhendler catching up on rdfig-- laughs at the no topic maps in RDF comment - I think Tim and his coauthors certainly knew they existed

17:48:17 <Jhendler> D: the topic maps people seem very confused about naming vs. referencing and all those things

17:48:17 <dc_rdfig> Added comment D18.

17:48:44 <larsbot> Jhendler: what do you mean?

17:50:08 * larsbot curious

17:50:45 <Jhendler> D: Mondeca seems to have been getting it worked out a bit [http://www.mondeca.com/english/faqs.htm|as they describe in their FAQ] and are working on using RDF/OWL to provide the semantics for topic maps

17:50:45 <dc_rdfig> Added comment D19.

17:51:32 <Jhendler> lars - not meant as explicit criticism - I meant that they say things like "RDF cannot do XXX" when RDF can do it

17:51:48 <Jhendler> often because of differences in how these things are described by the two communities

17:51:58 <larsbot> actually, I agree with that

17:52:16 <larsbot> there is a problem with identity in RDF, but it doesn't mean that RDF is left with a big black hole in the middle

17:52:20 <larsbot> just that there is a problem

17:53:03 <Jhendler> right, but RDF assumes (simplifing) that there exists some URI scheme for naming things, and it can use that for pointing - so the TM criticism isn't

17:53:13 <Jhendler> really w/RDF, it's with URIs

17:53:38 <Jhendler> and I wonder why a non-URI naming scheme to documents (simplification of pub sub) is considered to solve this

17:54:10 <Jhendler> that is - if I were to simply say "create URNs for the things you want to point to, use those in RDF" why haven't I met their objections?

17:54:27 <larsbot> first of all: because there are no suitable URN schemes for this yet

17:54:42 <larsbot> secondly: because URNs are not self-documenting

17:55:52 <Jhendler> right -- but notice you're now outside the issue of "RDF" and into - how do we create URIs for what topic mpas can name -- in a sense

17:56:13 <larsbot> in a sense, yes

17:56:23 <Jhendler> I'm not sure I said that real well - but what I mean is that these things are addressing orthogonal issues that can be easily

17:57:00 <Jhendler> put together - but sometimes we say things like "RDF doesn't" (or "Topic Maps don't" when we really should be talking about how to put them together, not how to separate them further

17:57:28 <Jhendler> Mondeca tooks some steps that direction, liked what it saw, and now is doign interesting work in putting OWL and TMs together

17:57:40 <Jhendler> for example

17:58:04 <larsbot> I'm not convinced by what Mondeca is doing

17:58:13 <larsbot> but that's a side issue

17:58:29 <larsbot> also, what Pepper was talking about was how to get PSIs into RDF

17:58:39 <larsbot> which is bringing RDF/TMs together, not separating them

17:59:11 <larsbot> I think it's true to say that RDF doesn't clearly distinguish between RDF nodes that represent information resources and ones that don't

17:59:26 <Jhendler> info resources?

17:59:50 <larsbot> any sequence of bytes

18:00:31 <larsbot> I don't think that's necessarily horribly difficult to fix, but it does require attention

18:00:48 <eaon> eaon is now known as eaon|out

18:00:49 <larsbot> other people within the RDF community have been saying pretty much the same thing for a long time

18:03:42 <Jhendler> don't have time to engage on that now - as you know, I would like to see more TM/RDF stuff happen - athough I am a bit upset that some in the TM community seem deadset on reinventing OWL instead of using it

18:03:56 <Jhendler> but that's another story -- the usual "not invented here"

18:04:05 <larsbot> did you read my take on that?

18:04:30 <kendallclark> Jhendler: hi jim

18:05:10 <Jhendler> howdy Kendall

18:05:32 <larsbot> I've tried to make TM/RDF stuff happen several times, but have met with so little interest from the RDF side that I've pretty much given up. if someone else tries to get something going I'll sign up

18:07:37 <JimJibber> JimJibber is now known as JibberJim

18:27:16 <D[a]vey> D[a]vey is now known as Davey

18:32:24 <kendallclark> hi edd

18:33:31 <edd> hi

18:33:40 * sandro is slightly shocked at larsbot's defn of "information resources", but imagines no one wants to get into it now.

18:34:21 * larsbot does, but may well be alone

18:35:12 * DanC_jam chuckles at "kendallclark: The come-to-jesus point:"

18:35:51 <DanC_jam> I'm sometimes mistaken for a Texan. I'm from the midwest, Kansas City. I think kendallclark is the true Texan here.

18:35:59 <kendallclark> an artifact of my v. religious childhood...

18:36:25 <kendallclark> yep, born in houston, lived in dallas till 6 months ago; my first time to live outside TX.

18:36:51 <kendallclark> DanC: you've never heard the 'come to jesus' saying?

18:36:59 <DanC_jam> oh yes, I've heard it.

18:38:06 <kendallclark> ah, good

18:39:02 <kendallclark> hmm, i wonder if i'm the only one who wasn't very cheered by xquery/rdf prospects?

18:39:07 <kendallclark> or maybe i missed the good stuff?

18:39:15 * DanC_jam wonders if kendallclark has seen "Hands on a Hardbody" (hmm... thought I cited it from http://dm93.org/z2001/ShareLifeThruArt and/or http://dm93.org/1993/9303austin/)

18:39:36 <kendallclark> uh... i trust yr not pointing me to p0rn?

18:39:49 <DanC_jam> re xquery/rdf, I had low expectations, and they were met ;-)

18:39:52 <DanC_jam> no, not porn.

18:40:13 <kendallclark> heh, re: expects

18:40:23 <DanC_jam> "Hands On A Hardbody Contest in Longview, Texas"

18:40:31 <kendallclark> ericp gets the bon mot o' the night award, imo: "xquery boots!"

18:41:03 <DanC_jam> . http://www.amazon.com/exec/obidos/tg/detail/-/B00000JYWY/danconnollyA/

18:41:09 <kendallclark> i've seen mystic river movie twice lately. very good.

18:41:39 <DanC_jam> "At first one might think that "Hands on A Hardbody" might be the title of an adult video to get you in the mood...but really it's the true life story of 23 contestants who are vying for the keys to an ordinary Nissan pickup."

18:41:47 <kendallclark> recommend the book?

18:41:52 <kendallclark> oh, oh, i've seen that!

18:42:00 <kendallclark> well, part of it

18:42:03 <DanC_jam> oh yes, Mystic River is a wonderful book.

18:42:22 <kendallclark> i've read lehane stuff

18:42:24 <kendallclark> i liked

18:42:28 * DanC_jam wonders what he did with his notes on Mystic River; my life is an intertwingly mess

18:43:15 <kendallclark> oh, you might like this; the newest NYRB has an excellent essay on Eastwood, focusing on Mystic River. Very worth reading.

18:43:25 <kendallclark> i read it on paper, so don't know if it's on web

18:43:27 <kendallclark> shall look

18:44:15 <kendallclark> ah, yes, excellent

18:44:27 <kendallclark> i probably shldn't churn it here, though...

18:44:41 * DanC_jam finds mystic river notes where they belong: http://dm93.org/z2001/ShareLifeThruArt

18:44:41 <kendallclark> . http://www.nybooks.com/articles/16852

18:44:53 <kendallclark> i recommend that essay

18:45:02 <kendallclark> good plan ride home reading, fwiw

18:45:18 <kendallclark> the nyrb site is generally very excellent; great uri scheme

18:45:36 <kendallclark> (heh, he says self-interestedly, since it's the precise uri scheme i use on monkeyfist.com)

18:46:08 <DanC_jam> logger, pointer?

18:46:08 <DanC_jam> See http://ilrt.org/discovery/chatlogs/rdfig/2003-12-11#T18-46-08

18:49:18 * DanC_jam queues NYRB essay in http://dm93.org/z2001/ShareLifeThruArt

18:50:09 <kendallclark> i like any web pub that doesn't have to do a 'print this' version since the 'regular' representation is v. clean

18:50:24 <kendallclark> that's prolly some kind of design ideal or something

18:52:13 * DanC_jam wishes for "peg this in my cache for the plane" button

18:52:24 <DanC_jam> hmm... javascript bookmarklet would probably work

18:52:48 <kendallclark> how does yr cache/queue thign work?

18:52:59 <sandro> Can't you make your cache big enough to hold your last few month's browsing?

18:53:02 <DanC_jam> my "queue" is just a wiki page

18:53:15 <DanC_jam> dunno, sandro

18:53:23 <DanC_jam> my cache is wwwoffle

18:53:41 <DanC_jam> I expect it'll keep that essay thru Friday's plane trip, but I'd like an explicit promise

18:54:15 <DanC_jam> Norm's presentation of wwwoffle config for "keep this for a year" and such was an eye-opener. I suspected wwwoffle could do that, but wasn't sure.

18:54:20 * DanC_jam wishes for wwwoffle conf in RDF

18:54:21 * kendallclark not tried wwwoffle, should do so

18:55:17 * kendallclark heads off to the ahmed talk

18:55:44 * DanC_jam would like to see the ahmed talk, but (a) has a previous engagement, and (b) saw it at Extreme in Aug

18:55:50 * DanC_jam finds notes fromAug

18:56:35 <DanC_jam> extremem markup notes http://rdfig.xmlhack.com/2003/08/06/2003-08-06.html#1060180738.070295

18:57:02 <DanC_jam> hmm... those notes are a bit raw.

18:57:19 <DanC_jam> 15:17:04 <DanC-AIM> Kaled used uml diagrams for presenting topicmap patterns. Diagrams++

18:57:33 <DanC_jam> -- http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/2003-08-06#T15-17-04

18:59:13 <evlist>http://www.xmlconference.org/xmlusa/2003/thursday.asp#17

18:59:14 <dc_rdfig> E: http://www.xmlconference.org/xmlusa/2003/thursday.asp#17 from evlist

18:59:50 <evlist> E:|Topic Map Design Patterns For Information Architecture - Kal Ahmed

18:59:51 <dc_rdfig> Titled item E.

19:00:09 <evlist> E is in room 109

19:00:17 <kendallclark> well, i hope it's worth it :>

19:00:24 <kendallclark> er, worth attending, i mean

19:00:43 <evlist> yes, I hope so!

19:00:57 <kendallclark> churn away, evlist!

19:01:59 <kendallclark> ah, i have to hear rich s. at 2:45

19:02:33 <evlist> E:The problems with Topic Maps is that there are many options, never one right way, where do I start?

19:02:33 <dc_rdfig> Added comment E1.

19:03:21 <kendallclark> hmm, i have to figure out how to sneak paul ford into conference, w/out him registering

19:03:26 <kendallclark> anyone have a spare badge?

19:05:27 <evlist> E:But library science came to the rescue: lots of studies about structures apply to TM.

19:05:27 <dc_rdfig> Added comment E2.

19:05:28 <evlist> E:But library science came to the rescue: lots of studies about structures apply to TM.

19:05:28 <dc_rdfig> Added comment E3.

19:05:38 <evlist> E3:""

19:05:38 <dc_rdfig> Deleted comment E3.

19:07:24 <evlist> E:The 3 ways to describe TM models: PSI, PSI Metadata, TMCL. A 4th way is missing (prescriptive for human consumption).

19:07:24 <dc_rdfig> Added comment E3.

19:08:32 <evlist> E:This 4th way is Topic Maps design patterns

19:08:32 <dc_rdfig> Added comment E4.

19:10:06 <evlist> E:TP design patterns (TPDP) are smaller than an ontology. May be built on top of themselves.

19:10:06 <dc_rdfig> Added comment E5.

19:11:24 <kendallclark> E:TMDPs are prescriptive & human readable.

19:11:24 <dc_rdfig> Added comment E6.

19:11:46 <evlist> E:Diagramming is an important part of pattern description.

19:11:46 <dc_rdfig> Added comment E7.

19:11:55 <kendallclark> E:(Ooh, he loses me w/ UML diagrams. :<)

19:11:56 <dc_rdfig> Added comment E8.

19:13:33 <kendallclark> E:(A rather crowded session, btw.)

19:13:33 <dc_rdfig> Added comment E9.

19:14:28 <evlist> The churn doesn't seem to catch up!

19:15:06 <kendallclark> ?

19:15:19 <evlist> no, sorry.

19:16:29 <kendallclark> E:Hierarchical Naming Pattern (motive: solve problem of naming hierarchical ordered topics)

19:16:30 <dc_rdfig> Added comment E10.

19:17:06 <kendallclark> E:Requires 2 sorts of names: a long & short name, that is, a context and contextless name.

19:17:06 <dc_rdfig> Added comment E11.

19:17:20 <kendallclark> E:Need to avoid: topic naming constraint

19:17:20 <dc_rdfig> Added comment E12.

19:18:05 <kendallclark> E:Design: create 2 names for each topic, save for the root; scope the short name by the parent topic (i.e., parent is the context for the short name)

19:18:05 <dc_rdfig> Added comment E13.

19:19:12 <evlist> E:Kal Ahmed showing nice UML representations of Topic Maps

19:19:12 <dc_rdfig> Added comment E14.

19:25:55 <kendallclark> wow, this volokh person is a nasty pig.

19:26:10 <evlist> E:Now showing a "topic per concept thesaurus pattern".

19:26:10 <dc_rdfig> Added comment E15.

19:26:51 <evlist> E:And compares this to a "topic per term" pattern.

19:26:51 <dc_rdfig> Added comment E16.

19:30:45 <evlist> E:Now describing the faceted classification pattern.

19:30:46 <dc_rdfig> Added comment E17.

19:36:17 <evlist> E:Kal Ahmed hopes TM navigators will implement some of these design patterns at some point.

19:36:17 <dc_rdfig> Added comment E18.

19:38:09 <evlist> E:Promote your TM patterns on http://www.topicmapcentral.com/wiki/Wiki.jsp !

19:38:09 <dc_rdfig> Added comment E19.

19:41:32 <evlist>http://www.xmlconference.org/xmlusa/2003/thursday.asp#23

19:41:32 <dc_rdfig> F: http://www.xmlconference.org/xmlusa/2003/thursday.asp#23 from evlist

19:42:16 <evlist> F:|Book Builders: Content Repurposing with Topic Maps - Nikita Ogievetsky & Roger Sperberg

19:42:16 <dc_rdfig> Titled item F.

19:48:31 <Davey> Davey is now known as D[a]vey

19:48:32 <evlist> F:Roger Sperberg sets the context (AnswerBooks, Q&A format, tOC generated automatically, index & "end" table in XML, ...)

19:48:32 <dc_rdfig> Added comment F1.

19:53:52 <kendallclark> edd: you here?

19:54:04 <edd> i am

19:54:16 <kendallclark> hey, i've collected The Pefinator, he was asking about you.

19:55:45 <kendallclark> BLURB: Rich Salz, XML Security Standards and Best Practices

19:55:45 <dc_rdfig> G: Rich Salz, XML Security Standards and Best Practices from kendallclark

19:56:29 <kendallclark> G:I missed the first bits, but Rich offers a pretty thorough of the XML security landscape.

19:56:29 <dc_rdfig> Added comment G1.

20:09:33 <evlist> F:Nikita Ogievetsky describing the principle (global external Topic Map common to all the books).

20:09:34 <dc_rdfig> Added comment F2.

20:10:10 <kendallclark> G:"Unlike TPC DoS, one XML req enough to DoS, if it's properly mangled..." -- A bit scary.

20:10:10 <dc_rdfig> Added comment G2.

20:10:31 <kendallclark> G:Suggests a 4x perf slowdown to validate.

20:10:31 <dc_rdfig> Added comment G3.

20:12:14 <kendallclark> G:PKI is 'ridiculously hard'.

20:12:15 <dc_rdfig> Added comment G4.

20:12:31 <evlist> F:The TOC are included in the Topic Map and books can be constructed through their TOC.

20:12:31 <dc_rdfig> Added comment F3.

20:17:49 <kendallclark> G:Best practices: 1. Secure the transport layer; 2. Mask internal resources; 3. Implement XML filtering; 4. Protect against XDoS; 5. Schema-validate all messages;

20:17:49 <dc_rdfig> Added comment G5.

20:21:44 <kendallclark> G:6. Be able to transform all messages; 7. Sign all outgoing messages; 8. Timestamp all messages; 9. Encrypt sensitive message fields; 10. Securely log all sensitive events.

20:21:44 <dc_rdfig> Added comment G6.

20:21:59 <MarkB> D:+1 to Jhendler; I mentioned to Sam Hunting that TopicMaps seem to be doing the equivalent of using java.lang.Object.equals() where "==" is required

20:21:59 <dc_rdfig> Added comment D20.

20:23:12 <kendallclark> G:Best practices are processing intensive.

20:23:12 <dc_rdfig> Added comment G7.

20:24:20 <evlist> F:This systems lets you create new books through drag & drop (but of course that wouldn't be possible for other types of books).

20:24:21 <dc_rdfig> Added comment F4.

20:35:39 Topic now RDF, Semantic Web and XML 2003 Conference Philadelphia day 2. Scratchpad http://rdfig.xmlhack.com/ logs http://www.ilrt.bris.ac.uk/discovery/chatlogs/rdfig/

20:35:39 Users on #rdfig: logger MDubinko wkearney99 reagleBRKLN burtonator_ eikeon kendallclark dajobe ChrisDodo nsrac419 danbri DanC_jam evlist mortenf mike_ hugoh timbl D[a]vey arnarl shellac iwaim mdupont sh1mmer MarkB anselm workbench LotR kasei whwhwhwh roGer^work eaon|out darobin dc_rdfig Wack eikco grault grove mmealling datum GabeW larsbot DanC swh ericP chrisc sandro tansaku karlcow kao kota taaz dngor jql jordan kham CaptSolo sbp xover deltab

20:35:39 <NickServ> Password accepted - you are now recognized

20:35:39 <MemoServ> You have no new memos

20:35:39 <ChanServ> [#rdfig] This channel is logged and blogged: http://logicerror.com/rdfIRCWelcome

20:45:42 * edd rushes here and there, trying to locate kendall

20:51:45 <MDubinko_> MDubinko_ is now known as mdubinko

20:53:21 <evlist>http://www.xmlconference.org/xmlusa/2003/thursday.asp#29

20:53:22 <dc_rdfig> H: http://www.xmlconference.org/xmlusa/2003/thursday.asp#29 from evlist

20:54:03 <evlist> H:|Semantic Integration at the IRS: The Tax Map - Michel Biezunski

20:54:04 <dc_rdfig> Titled item H.

20:54:10 <mdubinko>http://www.xmlconference.org/xmlusa/2003/thursday.asp#27

20:54:10 <dc_rdfig> I: http://www.xmlconference.org/xmlusa/2003/thursday.asp#27 from mdubinko

20:54:32 <mdubinko> I:|Using Fuzzy Logic to Create Links: Resolving References to Cited Court Cases

20:54:32 <dc_rdfig> Titled item I.

20:56:06 <DanC_jam> H:Coolheads Consulting

20:56:07 <dc_rdfig> Added comment H1.

20:59:20 <mdubinko> iwaim: Jack Rugh, Retrieval Systems Corporation and Julia Lennen, Tax Management, Inc./BNA

20:59:33 <mdubinko> d'oh, stupid auto-completion

20:59:53 <mdubinko> I:Jack Rugh, Retrieval Systems Corporation and Julia Lennen, Tax Management, Inc./BNA

20:59:54 <dc_rdfig> Added comment I1.

21:02:28 <evlist> H:"Michel Biezunski is the grand father of Topic Maps" -- Kal Ahmed

21:02:29 <dc_rdfig> Added comment H2.

21:03:17 <mdubinko> I:Challenge: portfoloios with over 500K references

21:03:18 <dc_rdfig> Added comment I2.

21:03:43 <mdubinko> I2:Challenge: portfoloios with over 500K citations

21:03:44 <dc_rdfig> Replaced comment I2.

21:04:46 <evlist> H:Purpose: enhance the productivity of the tax law assistance call centers & server as a model to demonstrate the concept of a central entry point for technical information at IRS;

21:04:46 <dc_rdfig> Added comment H3.

21:05:09 <mdubinko> I:Citations include lots of details (case name, volume, reporter, page, point page, year)

21:05:09 <dc_rdfig> Added comment I3.

21:07:57 <mdubinko> I:more complications: same case can be recorded by multiple recorders, or more than one per printed page

21:07:57 <dc_rdfig> Added comment I4.

21:08:21 <evlist> H:Currently at the 3rd step (all 95 taxpayer information publications, FAQ & tele-tax topics document types).

21:08:21 <dc_rdfig> Added comment H4.

21:09:20 <mdubinko> I:high standard of accuracy: better to have a no-link than a link-to-wrong-case

21:09:20 <dc_rdfig> Added comment I5.

21:10:00 <evlist> H:System supports both XML & SGML.

21:10:01 <dc_rdfig> Added comment H5.

21:11:31 <evlist> H:Topic Map created automatically from information gathered in the document.

21:11:32 <dc_rdfig> Added comment H6.

21:13:59 <mdubinko> I:SGML markup triggers (<cite.parallel ref="TCM\40\99">...) initiate processing

21:13:59 <dc_rdfig> Added comment I6.

21:15:09 <mdubinko> I:Normalization for punctuation, space, etc., then attempt exact match. In the case of multiple matches, uses fuzzy logic to decide

21:15:10 <dc_rdfig> Added comment I7.

21:15:36 <evlist> H:Topics extracted from headers, keywords & tele-tax topics and separated into key topics (chosen by experts), form topics and other topics.

21:15:37 <dc_rdfig> Added comment H7.

21:17:41 <mdubinko> I:word frequency in case names tracked, log weighted

21:17:41 <dc_rdfig> Added comment I8.

21:17:50 <evlist> H:Key topics are available through alphabetic indexes, form topics are accessible though a specialized list & other topics are available through a search engine.

21:17:51 <dc_rdfig> Added comment H8.

21:19:48 <evlist> H:This allows to reduce the number of topics to browse (the total number of topics is ~ 10,000).

21:19:49 <dc_rdfig> Added comment H9.

21:20:46 <DanC_jam> "merge on the basis of names" hmm... if those names are URIs, that coincides with RDF semantics.

21:20:56 <DanC_jam> H:"merge on the basis of names" hmm... if those names are URIs, that coincides with RDF semantics.

21:20:56 <dc_rdfig> Added comment H10.

21:21:13 * DanC_jam wonders about notifying presenters about these notes

21:21:33 <mdubinko> I:each matched word, by weight, contributes to an overall "match number", between 0 and 1

21:21:34 <dc_rdfig> Added comment I9.

21:21:53 <evlist> edd has asked to get these pages mentioned on the blog page for the conference.

21:23:11 <DanC_jam> so did I

21:23:11 <evlist> but it doesn't seem to be there yet (http://www.xmlconference.org/xmlusa/2003/blogroll.asp)

21:23:26 <DanC_jam> hmm... sometimes I wonder what's the difference between topicmaps and, say, PHP

21:23:33 <DanC_jam> I mean: where's the topicmap? I see an HTML document.

21:23:54 <DanC_jam> and they wrote custom code to make these HTML pages from topicmaps, no?

21:24:10 <evlist> yes, it seems so!

21:25:14 <DanC_jam> is the topicmap he's talking about handy in HTTP space? ala the NCI ontology in owl?

21:25:31 <DanC_jam> H:is the topicmap he's talking about handy in HTTP space? ala the NCI ontology in owl?

21:25:32 <dc_rdfig> Added comment H11.

21:25:53 <mdubinko> I:logic diagram spans over 5 slides!

21:25:53 <dc_rdfig> Added comment I10.

21:26:13 * DanC_jam would like to see those 5 slides

21:26:36 <mdubinko> they have a few extra handouts, if you come to 111

21:27:17 <mdubinko> I10:logic diagram spans over 6 slides!

21:27:17 <dc_rdfig> Replaced comment I10.

21:28:17 <DanC_jam> H:"Aggregation" slide is cool; sounds familiar, w.r.t. doing RDF stuff in W3C

21:28:17 <dc_rdfig> Added comment H12.

21:28:31 <DanC_jam> H:"Methodology: Incremental process"

21:28:31 <dc_rdfig> Added comment H13.

21:28:45 <DanC_jam> H:"Impact on existing practices: minimal changes in workflow"

21:28:45 <dc_rdfig> Added comment H14.

21:29:01 <DanC_jam> H:"Discovery of new issues: global consistency vs. local consistency"

21:29:01 <dc_rdfig> Added comment H15.

21:32:21 <mdubinko> I:reasons for failed match: not in db, multiple hits, or typographical error in db

21:32:21 <dc_rdfig> Added comment I11.

21:32:26 <evlist> H:Tax Map Maintenance Worshops organized to maintain the consitency of information.

21:32:27 <dc_rdfig> Added comment H16.

21:33:15 <mdubinko> I:solution: editorial interface, allowing human intervention, repair, flagging

21:33:15 <dc_rdfig> Added comment I12.

21:33:45 <DanC_jam> H:"Design decision: not to interfere with pre-existing workflow."

21:33:46 <dc_rdfig> Added comment H17.

21:34:56 <DanC_jam> H:this jives really well with what I'm talking about at the 'practical RDF' town hall tonight: [http://www.w3.org/2003/Talks/simo-semwebapp/|The Semantic Web and its applications at W3C]

21:34:56 <dc_rdfig> Added comment H18.

21:36:07 <mdubinko> fuzzy logic is a natural fit for human exception-handling

21:37:15 <DanC_jam> H:see also [http://lists.w3.org/Archives/Public/www-rdf-interest/2003Nov/0055.html|call for discussion on W3C Semantic Web Best Practices WG charter]

21:37:15 <dc_rdfig> Added comment H19.

21:37:52 <mdubinko> I:results: 98% linking, no false positives (I think that's what she said)

21:37:53 <dc_rdfig> Added comment I13.

21:39:21 <mdubinko> fragment syntax for legal cases anyone?

21:41:04 <evlist> H:Conclusion: "It is possible to benefit from cooperative work while leaving as much freedom as possible to the creators of the information sources."

21:41:04 <dc_rdfig> Added comment H20.

21:45:02 <DanC_jam> like lots of projects, they're struggling with issues around releasing the formal artifact itself; the topicmap

21:46:10 <evlist>http://www.xmlconference.org/xmlusa/2003/thursday.asp#33

21:46:11 <dc_rdfig> J: http://www.xmlconference.org/xmlusa/2003/thursday.asp#33 from evlist

21:46:36 <evlist> J:|Towards Semantic Interoperability of XML Vocabularies - Jacek R. Ambroziak

21:46:36 <dc_rdfig> Titled item J.

21:50:03 <D[a]vey> D[a]vey is now known as Davey

21:51:53 <evlist> J:This work is inspired by [UDEF|http://www.udef.org/].

21:51:54 <dc_rdfig> Added comment J1.

21:53:42 <evlist> J:UDEF is basing their mapping on exact synonymy, which is to narrow to really work.

21:53:42 <dc_rdfig> Added comment J2.

21:55:58 <DanC_jam> I disagree that the wire has "only HTML"; there's a growing amount RDF on the wire: FOAF, RSS, etc.

21:56:51 <evlist> J:Three levels: data, information, knowledge.

21:56:52 <dc_rdfig> Added comment J3.

21:57:21 <evlist> J:XML is an enabler to work at the information layer.

21:57:22 <dc_rdfig> Added comment J4.

21:57:48 <DanC_jam> where's the protocol? has he seen netapi? er... oh yeah; he's a co-submitter, isn't he?

21:58:57 <DanC_jam> "Semantic Web Protocol Use Cases" nifty!

21:59:02 * DanC_jam wonders what letter we're on

21:59:41 * evlist is in room 111 where it's "I"

21:59:48 * evlist is in room 111 where it's "J" (sorry)

22:00:01 <DanC_jam>http://www.xmlconference.org/xmlusa/2003/thursday.asp#35

22:00:01 <dc_rdfig> K: http://www.xmlconference.org/xmlusa/2003/thursday.asp#35 from DanC_jam

22:00:16 <DanC_jam> K:|Semantic Web Servers - Engineering the Semantic Web

22:00:17 <dc_rdfig> Titled item K.

22:00:30 <DanC_jam> K:"Semantic Web Protocol Use Cases" nifty!

22:00:30 <dc_rdfig> Added comment K1.

22:00:34 <evlist> J:Goals: exploit partial interoperability through declarative annotation.

22:00:34 <dc_rdfig> Added comment J5.

22:02:40 <DanC_jam> people laugh at Graham's suggestion of semantic web servers in fridges talking to stoves... I suggest that's like people 10 or 20 years ago laughing at star trek communicators, i.e. cellphones.

22:03:07 <DanC_jam> K:people laugh at Moore's suggestion of semantic web servers in fridges talking to stoves... I suggest that's like people 10 or 20 years ago laughing at star trek communicators, i.e. cellphones.

22:03:07 <dc_rdfig> Added comment K2.

22:04:03 <evlist> J:Conceptual Indexing (work done at Sun labs) is another inspiration.

22:04:04 <dc_rdfig> Added comment J6.

22:07:39 <evlist> J:Concept can be linked by relations (is_a, instance_of, part_of, ...) instead of just synonymy.

22:07:40 <dc_rdfig> Added comment J7.

22:13:07 <DanC_jam> hmm... getStatements is a typical RPC call with a few scalar args; WSDL 2.0 should be able to map this to GET. the argument about the result value should go in the Accept: header

22:13:18 <DanC_jam> K:hmm... getStatements is a typical RPC call with a few scalar args; WSDL 2.0 should be able to map this to GET. the argument about the result value should go in the Accept: header

22:13:18 <dc_rdfig> Added comment K3.

22:13:28 <evlist> J:UDEF relies on the US 2002 [NAICS|http://www.census.gov/epcd/www/naics.html] industry ontology.

22:13:29 <dc_rdfig> Added comment J8.

22:13:32 <DanC_jam> K3:hmm... getStatements is a typical RPC call with a few scalar args; WSDL 2.0 should be able to map this to GET. the argument about the result type should go in the Accept: header

22:13:32 <dc_rdfig> Replaced comment K3.

22:13:58 <DanC_jam> K:hmm.. updateStatements... how about a where clause? how much of SQL can we do?

22:13:58 <dc_rdfig> Added comment K4.

22:15:03 <DanC_jam> K:hm... options()... seems to contradict the goal of interoperability, i.e. going away from "proprietary" protocols

22:15:03 <dc_rdfig> Added comment K5.

22:15:35 <DanC_jam> K:"RDF Data Model in XML Schema". hmm... interesting.

22:15:35 <dc_rdfig> Added comment K6.

22:15:54 <evlist> J:This ontology lacks subsumption transitivity.

22:15:54 <dc_rdfig> Added comment J9.

22:16:11 <DanC_jam> K:soap binding doesn't look straightforward to me at all. cf whenToUseGet-7

22:16:11 <dc_rdfig> Added comment K7.

22:19:51 <evlist> J:In Ambroziak's proposal, a set of rules is defined for each vocabulary. These sets of rules use a common ontology and processors are derived from these rules.

22:19:52 <dc_rdfig> Added comment J10.

22:20:46 <evlist> J:A rule is a triple "(concept, match, extract)".

22:20:47 <dc_rdfig> Added comment J11.

22:22:07 <evlist> J:Concept is a pointer into the ontology, match is an XPath expression and extract a XSLT snippet.

22:22:07 <dc_rdfig> Added comment J12.

22:23:28 <evlist> J:Ambroziak is not sure if XPath is enough for that work.

22:23:28 <dc_rdfig> Added comment J13.

22:25:01 <evlist> J:The generation of XSLT out of these rules isn't implemented yet, but a conceptual model of this mapping has been implemented.

22:25:02 <dc_rdfig> Added comment J14.

22:29:32 <DanC_jam> ahmed's a very polished session chair!

22:30:15 <LotR> yay for bee wax

22:32:11 <verbosus> Hello there.

22:32:32 <verbosus> Does anybody have a copy of sbp’s pyrple? His Web site is not working at the moment.

22:32:45 <mortenf> yeah, hang on

22:34:21 <verbosus> Does that include the docs as well, mortenf?

22:34:27 <mortenf> hmm

22:34:38 <mortenf> yep

22:34:43 <verbosus> Yes, it does.

22:34:48 <verbosus> Thank you very much :-)

22:34:52 <mortenf> sure

22:38:08 <eaon|out> eaon|out is now known as eaon

22:39:18 <Jhendler> J: interesting - the use of OWL for mapping XMLS's to each other has been proposed, and a couple of student level projects have shown the possibility

22:39:19 <dc_rdfig> Added comment J15.

22:39:44 <Jhendler> J: interesting to similar thought in XML community

22:39:44 <dc_rdfig> Added comment J16.

23:35:00 <eaon> eaon is now known as eaon|zZz