Semantic Web Interest Group IRC Chat Logs for 2003-02-11

00:24:35 <danbri>http://www.w3.org/2001/12/rubyrdf/xsltrdf/README.html

00:24:36 <dc_rdfig> A: http://www.w3.org/2001/12/rubyrdf/xsltrdf/README.html from danbri

00:24:51 <danbri> A:|Parsing RDF using XSLT in Ruby, by Max Froumentin

00:24:51 <dc_rdfig> Titled item A.

00:25:01 <danbri> A:"These files support XSLT-based RDF parsing, with the output in (something approximating) the RDFCore WG's NTriples syntax."

00:25:01 <dc_rdfig> Added comment A1.

00:25:17 <danbri> A:"This package includes a full RDF parser in XSLT by Max Froumentin. See below for parser todo and buglist."

00:25:17 <dc_rdfig> Added comment A2.

00:39:44 <tim-lex> C:I've coded this in (checks for >= 2.2.0) but not checked in in yet

00:39:44 <dc_rdfig> Label C not found.

00:40:15 <LTjake> hello.

00:55:11 <danbri> tim-lex> C:I've coded this in (checks for >= 2.2.0) but not checked in in yet

00:55:25 <danbri> you're annotating yesterday's discussion, amybe?

00:56:11 <danbri> hey tim, if I check out Cwm/swap, how do I run the regression tests?

00:57:36 <danbri> cd test; make

00:57:38 <danbri> ...fails

00:57:52 <eikeon> danbri: I believe you do the following from the test directory

00:57:54 <eikeon> PYTHONPATH=..;python2.3 retest.py -n -f regression.n3

00:57:56 <danbri> TypeError: 'intern' is an invalid keyword argument for this function

00:58:15 <danbri> ...may be misconfig'd python expat stuff. I just installed the most apprpriate looking debian module

00:58:23 * danbri gotta run, night

00:58:30 * tim-lex glancing back , notes Euler is backward, cwm forward chainer

00:58:50 <danbri> I tried this: PYTHONPATH=..;python2.2 retest.py -n -f regression.n3

00:58:52 <danbri> same error

00:59:28 <tim-lex> What line # of what file?

01:01:14 <tim-lex> Sounds like an error in the retest.py maybe.

01:01:49 <tim-lex> danbri: ooops i was ... no datestamps on mu irc

01:01:52 <tim-lex> danbri: ooops i was ... no datestamps on my irc

01:02:06 <tim-lex> To test, in swap, make tested

01:02:16 <tim-lex> (make test doesn't work as test is a directory)

01:02:57 <tim-lex> That does more or less cd swap python retest.py -n -f regressison.py OWTTE

01:03:03 <tim-lex> tim-lex is now known as tim-gone

01:49:31 <danbri> tim,

01:49:33 <danbri> PYTHONPATH=..;python2.2 retest.py -n -f regression.n3

01:49:33 <danbri> Loading tests from regression.n3

01:49:33 <danbri> 1) Parse a small RDF file, generate N3

01:49:33 <danbri> Traceback (most recent call last):

01:49:33 <danbri> File "../cwm.py", line 791, in ?

01:49:34 <danbri> doCommand()

01:49:36 <danbri> File "../cwm.py", line 543, in doCommand

01:49:40 <danbri> p.load(_inputURI)

01:49:42 <danbri> File "../sax2rdf.py", line 515, in load

01:49:44 <danbri> return self.loadStream(f)

01:49:46 <danbri> File "../sax2rdf.py", line 522, in loadStream

01:49:48 <danbri> self._p.parse(s)

01:49:50 <danbri> File "/usr/lib/python2.2/site-packages/_xmlplus/sax/expatreader.py", line 107, in parse

01:49:52 <danbri> self.reset()

01:49:54 <danbri> File "/usr/lib/python2.2/site-packages/_xmlplus/sax/expatreader.py", line 232, in reset

01:49:56 <danbri> self._parser = expat.ParserCreate(None, " ", intern = self._interning)

01:49:58 <danbri> TypeError: 'intern' is an invalid keyword argument for this function

02:19:14 <dmiles> hrrm is there an ODBC component that makes XML requests

02:19:23 <dmiles> must find one

02:20:42 <dmiles> xmlhack: ODBC-XML server is the oposite direction

03:11:08 <danbri> OK, I upgraded (by debian packages) to Python 2.3 + assocaited XML support package. The above problem is now fixed. I'll back out the new packages and investigate if anyone cares.

03:13:47 <reagle>http://www.w3.org/2000/10/swap/doc/Shortcuts.html

03:13:48 <dc_rdfig> B: http://www.w3.org/2000/10/swap/doc/Shortcuts.html from reagle

03:15:06 <danbri> B:Aha, suddenly various things make sense. I wondered where log:forAll was dissapearing to...

03:15:06 <dc_rdfig> Added comment B1.

03:15:11 <danbri> thanks Joseph :)

03:15:23 <reagle> yes, i just found these other two secret pages added

03:15:26 <reagle> :)

03:15:31 * danbri now has Cwm working again, wonders about helping with a Debian package for it...

03:15:46 <reagle> every built a package before?

03:16:03 <reagle> deb packages itself are fairly easy, up to debian policy quality is a higher threshold

03:16:04 <danbri> Is there a tar.gz bundle? we could look at that and figure out where in the filesystem stuff might go.

03:16:11 <reagle> (kind of like a w3c TR ;)

03:16:25 <danbri> I'm learning the ropes, making one for RubyRDF stuff.

03:16:36 <danbri> completely agree re simple/crap vs full horror...

03:17:08 <danbri> With Ruby it's difficult, I hate hardcoding the target directories but Debian seems to prefer that over trusting Ruby installers to put things in their preferred place at install time.

03:17:13 <reagle> well i do appreciate a proper debian policy package, *if* someone else builds it

03:17:16 <danbri> Fink on Macs seems more the latter...

03:18:02 <danbri> current scribbles and ruby deb thing are at: http://www.w3.org/2001/12/rubyrdf/debian/packager/

03:18:52 <danbri> (thinks ..ooOO "why am i not sleepy?")

03:19:09 <reagle> yea, shouldn't it be like 3am where you are?

03:19:31 * danbri glances around... yeah, that'd be about right

03:20:28 <dmiles> where log:forAll was dissapearing to?

03:20:30 <danbri> I like the idea of packaging rubyrdf and cwm with the same conventions, ie figuring out where docs, tests, scripts live and doing it consistenty. Currently I only install the basic .rb library files, which is easy enough...

03:20:40 <danbri> [[

03:20:40 <danbri> So instead of

03:20:40 <danbri> this log:forAll :x, :y.

03:20:40 <danbri> { :x :parent :y } => { :y :child :x }.

03:20:40 <danbri> you can write

03:20:41 <danbri> { ?x parent ?y } => { ?y :child ?x }.

03:20:43 <danbri> ]]

03:20:53 <reagle> i figure once you have the hang of debian policy, then your good to go (again, just like TR)

03:20:55 <danbri> dmiles, you can omit it if you use '?' with variables

03:21:10 <dmiles> ok i like that

03:21:16 <danbri> have you got the hang of debian policy? :)

03:21:19 <dmiles> :var is existentially quants?

03:21:21 <reagle> no

03:21:31 * eikeon has heard danbri say g'nite about 10 times... /me now thinks danbri is a machine :)

03:21:42 <dmiles> forAll :var = for each?

03:21:52 <dmiles> :)

03:22:08 <danbri> yeah, i'm going through a not tired phase. i figure i'll run myself into the ground in a 24 hour period then spend the rest of the week recovering ;)

03:22:27 <dmiles> forAll ?X = syntactily forAll (as well)

03:23:07 <reagle> is max's rfc14n.xsl produce real c14n xml?

03:23:28 <reagle> by that i mean the REC? or just *a* canonical form

03:34:12 <danbri_> 03:23:07 <reagle> is max's rfc14n.xsl produce real c14n xml?

03:34:12 <danbri_> 03:23:28 <reagle> by that i mean the REC? or just *a* canonical form

03:34:12 <danbri_>

03:34:15 <danbri_> the latter

03:34:31 * reagle is gonna have to trademark that term

03:34:38 <reagle> <smile/>

03:34:43 <danbri_> eikeon, have you looked at debian packaging for rdflib?

03:34:49 * danbri_ grins

03:34:54 <eikeon> danbri_: no

03:35:15 <danbri_> pity, I was wondering about freeloading off your investigations. I don't know much about python libraries...

03:35:34 * eikeon runs RedHat

03:36:10 <eikeon> Took me long enough before I started creating a setup.py for rdflib

03:36:37 <danbri_> does that install it system-wide?

03:38:31 <eikeon> Yes.

03:39:13 <eikeon> Python has a distutil library from which setup.py scripts are usually built.

03:39:46 <eikeon> .google python distutils

03:39:47 <datum> python distutils: http://www.python.org/sigs/distutils-sig/

03:40:01 <eikeon> Hum... can find a better link than that.

03:40:59 <eikeon> http://www.python.org/doc/current/lib/module-distutils.html

03:44:22 <eikeon> For a python app/lib I think you would probably want to build a setup.py and then a debian package around that... am guessing though.

03:44:27 <danbri_> logger_1, pointer?

03:44:27 <danbri_> See http://ilrt.org/discovery/chatlogs/rdfig/2003-02-11#T03-44-27

03:45:02 * reagle isn't sure if they are compatible really... but imgseek but be a good example

03:45:13 <reagle> (a nice pthon-qt package)

03:46:12 <reagle> hrmm... nevermind

03:46:17 <reagle> maybe i just used distutils...

03:48:16 * danbri_ stashes the links in http://esw.w3.org/t/view/ESW/SoftwarePackaging for future pondering

03:50:22 * danbri_ heads off clutching todays new reading... http://www.amazon.com/exec/obidos/ASIN/0743247728/qid=1044935360/sr=2-2/ref=sr_2_2/002-7869949-7558421

03:50:24 <danbri_> nn

03:57:05 * tim-gone suspects that this time danbri really has left

03:59:49 <eikeon> :)

04:00:10 * DanCon browses tag minutes... http://www.w3.org/2003/02/06-tag-summary

04:02:04 <DanCon> all that eulersharp stuff looked tasty...

04:02:31 <DanCon> not to mention eikon's work on a nicer RDF/XML parser for cwm...

04:02:41 <DanCon> that reminds me...

04:03:04 <DanCon> eikon, namespace prefixes shouldn't matter for most of the tests...

04:03:10 <tim-gone> Yes, i have added code so taht if you have rdflib uyou can use it by cwm --rdf=l

04:03:39 <DanCon> but our test harness isn't sophisticated enough to know that prefixes don't matter

04:03:47 * tim-gone spellchecks http://www.w3.org/2000/10/swap/doc/Rules.html

04:04:19 * tim-gone had not mentioned on account of it being a bit ragged esp. spelling, missing links etc

04:04:34 <DanCon> you have rdflib integration working, timbl? maybe I should put RDFCore testing with cwm back on the front burner...

04:05:11 <tim-gone> No, not working or testing just coded. I think I should allow for an environ variable too.

04:05:29 <DanCon> I was doing the whole RDFCore test manifest in one --think run, but I'm considering 2 passes:

04:06:17 <DanCon> (1) convet manifest to cwm test cases, with args and all, (2) run tests with retest.py

04:07:00 <DanCon> chuckle... [[

04:07:01 <DanCon> # This mostly finds bugs in cwm. It doesn't

04:07:01 <DanCon> # really work.

04:07:12 <DanCon> ]] -- swap/test/rdfcore.n3

04:09:26 <tim-gone> There are two attitudes to testing - one to do it all inside cwm because it keeps you in N3, two to make the test call cwm etc on the command line to separate the tester from the testee.

04:10:14 <DanCon> , two to get *some* info about tests that pass before cwm crashes.

04:10:46 <DanCon> it looked so promising... [[

04:10:47 <DanCon> Total 0 new statements from rules in 0_work

04:10:48 <DanCon> Grand total of 643 new statements in 3 iterations.

04:10:48 <DanCon> Begining output.

04:10:48 <DanCon> ]]

04:10:52 <DanCon> but then:

04:11:04 <DanCon> [[

04:11:06 <DanCon> File "/home/connolly/w3ccvs/WWW/2000/10/swap/notation3.py", line 1278, in representationOf

04:11:06 <DanCon> if prefix != None : return prefix + ":" + value[j+1:]

04:11:06 <DanCon> UnicodeError: ASCII decoding error: ordinal not in range(128)

04:11:06 <DanCon> ]]

04:11:48 <tim-gone> ooo . are you testing unicode?

04:12:08 <DanCon> RDFCore tests include unicode stuff, yes.

04:12:09 <tim-gone> How would I immport from a module identified in a string?

04:12:13 * eikeon back... reading log now.

04:12:34 <tim-gone> Well, it is about time we tested it all then.

04:12:59 <DanCon> import... didn't we discuss this? ah, no, it was ericp I discussed this with. There's an import module, as I recall.

04:13:40 * tim-gone is really happy that not only could he usethe US state data in a test of new cwm --mode=s, but also could using it for checking for state tax due from credit card statements

04:14:24 <DanCon> --mode=s sounds cool; have you announced it? has sandro or anybody reviewed i?

04:14:41 <DanCon> though the one-letter flag is a bit retro-cryptic

04:15:23 <DanCon> --read-schemas would be more obvious

04:15:47 <DanCon> hmm... I am getting some output...

04:15:48 <DanCon> [[

04:15:49 <DanCon> :cwm :failsIf <http://www.w3.org/2000/10/rdf-tests/rdfcore/rdf-charmod-literals/Manifest.rdf#test001>

04:15:50 <DanCon> ]]

04:16:03 <DanCon> [[

04:16:04 <DanCon> ...

04:16:06 <DanCon> :failsOnlyIf <http://www.w3.org/2000/10/rdf-tests/rdfcore/rdf-charmod-literals/Manifest.rdf#test001>

04:16:06 <DanCon> ]]

04:16:54 <tim-gone> what does failsIf x mean? Fails on test of x ?

04:17:14 <DanCon> { :T :inF :F; :expectedF :G. :F log:notIncludes :G.} log:implies {

04:17:14 <DanCon> :cwm :failsIf :T.}.

04:17:25 <DanCon> -- swap/test/rdfcore.n3

04:21:07 * eikeon compares rdfcore test cases with expected results at store level (for the positive test cases).

04:21:48 * reagle sends stupid/simple example of rng (rdf) annotated relaxng schema

04:21:53 <eikeon> By loading input document into one store... and expected document (the ones in n3) into another store.

04:23:09 <tim-gone> I have a test which runs down a list of files converting them from n3 into xml and back again. I should put it back into the regression test.

04:23:24 <DanCon> aha... prefix-confusion#2 regards bagid, which sax2rdf doesn't grok.

04:23:49 * tim-gone considred bagid harmful

04:24:45 <DanCon> I think bagid is worthless, but I couldn't talk the WG out of it. Do we add support to cwm or ask for it to be taken out, tim?

04:25:06 <DanCon> or document the omission?

04:25:20 * eikeon agrees... as I never knew what it was for until seeing sax2rdf

04:25:41 * reagle enjoyed the rant considered harmful, considered harmful

04:25:43 <DanCon> but sax2rdf got it wrong.

04:26:40 <eikeon> DanCon: I was not suggesting it be kept... just that I had no idea what it was for before.

04:27:06 <reagle>http://www.meyerweb.com/eric/comment/chech.html

04:27:07 <dc_rdfig> C: http://www.meyerweb.com/eric/comment/chech.html from reagle

04:27:09 <DanCon> umm.. but you do now, after looking at sax2rdf?

04:27:58 <reagle> C|"Considered Harmful" Essays Considered Harmful

04:28:07 <eikeon> No... but have at least a vague idea of what it might have been for... and it does not seem to be that anymore.

04:28:08 <reagle> C:|"Considered Harmful" Essays Considered Harmful

04:28:08 <dc_rdfig> Titled item C.

04:28:46 * reagle thinks rdf chump should try to find the title of a URI itself, my command line chumper does with very good results.

04:29:00 <reagle> (i'm surprised to the degree to which people are using decent titles now-a-days)

04:29:08 * tim-gone consideres '"condidered harmful" considered harmful" harmful

04:29:25 <reagle> yes, that's the recursive joke of the essay :)

04:30:14 <reagle> DanC: related to my earlier griping about URI's with frag_id's as second class citizen's, one can't easily get the title of a frag_id from within a page automatically...

04:30:49 <tim-gone> Whatever you say, I can always make a meta joke out of it -- IJacobs

04:31:20 <reagle> i proposed a convention whereby people associate the element content of an element with an ID as the "title" of that URI, but then folks said since you can't count on a title on a web page itself: fat chance.

04:31:24 <tim-gone> re Bagid - if mo one on #rdfig thinks they are useful, then the normal procedure would be top take them out, no?

04:31:28 <DanCon> so, back to bagid, tim... spec is in last call. test case is clear. SNOFHYP.

04:32:05 <tim-gone> I can't even remember properly what it was kinda supposed to be for.

04:32:21 <tim-gone> It is for reification? Or you don't see it unless you reify?

04:32:27 <DanCon> it was a primitive form of quoting.

04:32:48 <tim-gone> Well, far be it for me to complain about quoting ....

04:33:25 <tim-gone> Yes, testing cwm regression with rdflib2rdf falls over a lot on prefixes.

04:33:54 <DanCon> bagid is a handle on the reified form of the triples from an rdf:Description. But the unreified form is asserted as well. And reification is one big use/mention bug.

04:34:03 <tim-gone> Could change test to put it through cwm an extra time or to add a file of prefix es like slat on the end or make cwm abl eto invent prefixes.

04:34:27 <tim-gone> bagid: I propose we suggest that it be removed

04:34:35 <tim-gone> Does anyone use it?

04:34:59 * eikeon re-reads: http://www.w3.org/TR/rdf-syntax-grammar/#section-Syntax-reifying

04:35:09 <tim-gone> One argument is: no one uses it enough to leave it in .

04:35:14 <DanCon> bagid: yes, it was claimed, in the WG, that folks use it.

04:35:35 <DanCon> uses it enough: i tried that argument. didn't carry.

04:35:42 <tim-gone> Anotehr argument is: by the time you've fixed theuse-mention bug you'll have been another 6 months in last call.

04:35:54 <tim-gone> Anotehr argument is: keep it simple at this level please.

04:36:20 <DanCon> The WG was pretty conservative about taking stuff that had REC status out, if anybody used or suported it at all.

04:36:38 <tim-gone> Deprecate it?

04:37:06 <DanCon> deprecate: we thought about that... I wasn't crazy about it.

04:37:13 * tim-gone thinks: Boy, we had less of a quorum on this chaneel to rip the --> out of n3 ;-)

04:38:02 <DanCon> I believe the "does anybody use it?" challenge resulted in primer editor action to show how... let's see...

04:38:49 <tim-gone> I'm not sure that a test case is enough to show interoperability required for Candidate Recommendation.

04:38:55 <DanCon> ooh... good news: no bagid in http://www.w3.org/TR/rdf-primer/ of 23 Jan

04:39:17 * DanCon reads http://www.w3.org/TR/rdf-primer/#reification

04:40:24 <DanCon> not a single RDF/XML example in 4.3 RDF Reification. bzzzt! reificiation is all about RDF/XML syntactic sugar!

04:41:16 * DanCon files comment...

04:43:15 <DanCon> logger, pointer?

04:43:21 <DanCon> logger_1, pointer?

04:43:21 <DanCon> See http://ilrt.org/discovery/chatlogs/rdfig/2003-02-11#T04-43-21

04:44:04 <tim-gone> cvs commit -m "CWM_RDF_PARSER environment variable can now be set to rdflib2rdf" cwm.py

04:44:47 <tim-gone> Danc, was your comment to suggest removal of bagid, or just about the state of the doc?

04:45:26 <DanCon> just about the primer, for that one.

04:45:32 * DanCon reviews http://www.w3.org/TR/rdf-syntax-grammar/#section-Syntax-reifying

04:46:20 * eikeon sees bagid as syntactic sugar... what else can it be... if it goes away when I serialize.

04:46:48 <tim-gone> (cwm.py new revision: 1.123; previous revision: 1.122)

04:47:14 <tim-gone> Eikeon, I have taken the liberty of adding your adapter code rdflib2rdf.py

04:47:36 <eikeon> Saw that... cool! :)

04:48:00 <tim-gone> thanks!

04:49:02 * eikeon cvs updates

04:49:26 <DanCon> timbl, can you say more about a test not being enough? the process is to ensure the spec can be implemented interoperably, no?

04:49:51 <DanCon> seems to me a test and >=2 implementations establishes that pretty well.

04:50:09 <tim-gone> Yes, certainly.

04:50:10 <DanCon> you seem to be talking about requirements traceability.

04:50:25 <tim-gone> But the object of the game is to get useful and used features generally deployed.

04:51:12 <eikeon> Is bagid useful as an abreviation maybe?

04:51:54 * eikeon runs document with bagid in it through parser->serialize to see what come out the other end ;)

04:52:02 <DanCon> WGs have to show each feature is useful and used? I'd like features to be useful and used, but I don't think any such criterion is an established part of W3C process.

04:52:40 <tim-gone> We are generally sensitive to the danger of having only working group or academic use of a feature .... if people jump up and down and insit it is useful that's one thing, but if people all have to look up the spec to find out what it is we are talking about here and no one knows really what its for... then it sdoes bear a second thought maybe.

04:53:26 <tim-gone> As part of the process, no. But the CR exit criteria are established on entry into CR in order to make sure the community's needs of the spec are met during CR>

04:53:45 <DanCon> hmm... putting the general case aside, I think the RDF Core WG assumed anything that got in the '98 REC should stay, unless they/we just couldn't make heads nor tails of it.

04:53:47 <tim-gone> They are not hard and fast.

04:54:10 <tim-gone> should stay: reasonable.

04:54:27 <tim-gone> I wonder whether google could find uses of bagid.

04:54:32 <DanCon> (or was in '99? anyway...)

04:54:43 <eikeon> lol re: google

04:55:08 <tim-gone> Should stay: On teh otehr hand, if its been a rec and *still* no one has used it, that is if anything more damning.

04:55:23 <DanCon> hmm... there's a way to search google for *.rdf, no? the foaf folks do it somehow.

04:56:12 <DanCon> I'd feel more comfortable if you sent the "pls take bagid out" comment. I was there when the WG decided to keep it, and I didn't object.

04:57:15 <tim-gone> Google find about 1430 occurrences of bagid, with the first 100 being all discussions of RDF syntax.

04:57:33 <DanCon> I *did* argue that the primer shouldn't have any sections with no RDF/XML examples, so I feel justified sending that comment.

04:58:01 <mdupont> morning

04:58:19 <mdupont> or evening DanCon

04:59:32 <DanCon> hi

05:00:46 <DanCon> bagid was originally motivated by PICS requirements, as I recall

05:01:27 <DanCon> phpht. no bagid in http://www.w3.org/TR/rdf-pics

05:02:43 <DanCon> ah... there are no actual examples of labels in /TR/rdf-pics

05:04:15 * tim-gone failed to find example usied in earnest google inurl:rdf bagid

05:04:44 <tim-gone> Ok, I am happy to send the comment to the list, but probably should not do it now.

05:05:55 <DanCon> no?

05:05:57 <mdupont>http://news.com.com/2100-1001-984052.html?tag=fd_top

05:05:58 <dc_rdfig> D: http://news.com.com/2100-1001-984052.html?tag=fd_top from mdupont

05:06:04 <mdupont> D:|.Net patent could stifle standards effort

05:06:05 <dc_rdfig> Titled item D.

05:06:13 <mdupont> D:Microsoft is in the process of applying for a wide-ranging patent that covers a variety of functions related to its .Net initiative.

05:06:13 <dc_rdfig> Added comment D1.

05:06:31 <mdupont> http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PG01&p=1&u=/netahtml/PTO/srchnum.html&r=1&f=G&l=50&s1='20030028685'.PGNR.&OS=DN/20030028685&RS=DN/20030028685

05:06:32 <dc_rdfig> E: http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PG01&p=1&u=/netahtml/PTO/srchnum.html&r=1&f=G&l=50&s1='20030028685'.PGNR.&OS=DN/20030028685&RS=DN/20030028685 from mdupont

05:06:43 * DanCon covers ears to news about p* thingies

05:07:06 <DanCon> logger_1, pointer?

05:07:06 <DanCon> See http://ilrt.org/discovery/chatlogs/rdfig/2003-02-11#T05-07-06

05:07:13 <mdupont> E:|PATENT20030028685 Application program interface for network software platform

05:07:14 <dc_rdfig> Titled item E.

05:07:23 <mdupont> E:An application program interface (API) provides a set of functions for application developers who build Web applications on Microsoft Corporation's .NET.TM. platform.

05:07:23 <dc_rdfig> Added comment E1.

05:08:12 <mdupont> looks like half of the state of washington is on the authors list ;)

05:08:40 <mdupont> DanCon: sorry to just blurt out

05:13:04 <mdupont> E: [0030] In the described implementation, the .NET.TM. platform utilizes XML (extensible markup language), an open standard for describing data. XML is managed by the World Wide Web Consortium (W3C). XML is used for defining data elements on a Web page and business-to-business documents. XML uses a similar tag structure as HTML; however, whereas HTML defines how elements are displayed, XML defines what those elements contain. HTML uses predef

05:13:04 <dc_rdfig> Added comment E2.

05:13:05 <mdupont> d tags, but XML allows tags to be defined by the developer of the page. Thus, virtually any data items can be identified, allowing Web pages to function like database records. Through the use of XML and other open protocols, such as Simple Object Access Protocol (SOAP), the .NET.TM. platform allows integration of a wide range of services that can be tailored to the needs of the user. Although the embodiments described herein are described in c

05:13:05 <mdupont> unction with XML and other open standards, such are not required for the operation of the claimed invention. Other equally viable technologies will suffice to implement the inventions described herein.

05:13:19 <mdupont> E2:

05:13:20 <dc_rdfig> (mdupont) [0030] In the described implementation, the .NET.TM. platform utilizes XML (extensible markup language), an open standard for describing data. XML is managed by the World Wide Web Consortium (W3C). XML is used for defining data elements on a Web page and business-to-business documents. XML uses a similar tag structure as HTML; however, whereas HTML defines how elements are displayed, XML defines what those elements contain. HTML uses pr

05:13:26 <mdupont> E2:

05:13:26 <dc_rdfig> (mdupont) [0030] In the described implementation, the .NET.TM. platform utilizes XML (extensible markup language), an open standard for describing data. XML is managed by the World Wide Web Consortium (W3C). XML is used for defining data elements on a Web page and business-to-business documents. XML uses a similar tag structure as HTML; however, whereas HTML defines how elements are displayed, XML defines what those elements contain. HTML uses pr

05:13:29 <mdupont> E2: ""

05:13:29 <dc_rdfig> Deleted comment E2.

05:18:12 <tim-gone> tim-gone is now known as tim-away

05:20:17 <mdupont> D: QUOTE " One person affiliated with the World Wide Web Consortium (W3C), another major standards body, said it's difficult to comment on the .Net patents without knowing Microsoft's specific plans. The W3C is in the process of developing a policy that would let the organization include patented technology in its standards as long as companies agree to provide the technology royalty-free. The person, who asked not to be identified, said Micro

05:20:17 <dc_rdfig> Added comment D2.

05:20:17 <mdupont> t has agreed to such terms in the past."

05:20:27 <mdupont> D:t has agreed to such terms in the past."

05:20:27 <dc_rdfig> Added comment D3.

05:27:55 * DanCon thought it was against W3C communications policy to be quoted anonymously.

05:28:53 <mdupont> it looks like they want to patent the specific .NET api

05:29:36 * DanCon la la la not listening about p* la la la

05:32:23 * DanCon is fading... night, all

05:32:45 <eikeon> g'nite

05:33:58 <mdupont> E: QUOTE "[0030] In the described implementation, the .NET.TM. platform utilizes XML (extensible markup language), an open standard for describing data. XML is managed by the World Wide Web Consortium (W3C). XML is used for defining data elements on a Web page and business-to-business documents. XML uses a similar tag structure as HTML"

05:33:59 <dc_rdfig> Added comment E2.

05:34:58 <mdupont> E: QUOTE "The services namespace 300 defines additional namespaces, including a description namespace 302 ("System.Web.Services.Description") containing classes that enable a developer to publicly describe a Web service via a service description language (such as WSDL, a specification available from the W3C),"

05:34:59 <dc_rdfig> Added comment E3.

05:35:25 <mdupont> E: QUOTE: [0054] A services namespace 300 ("System.Web.Services") containing classes that enable a developer to build and use Web services.

05:35:25 <dc_rdfig> Added comment E4.

05:35:58 * MarkB is happy as long as they don't try to snarf GET, PUT, and POST

05:36:06 <mdupont> E: QUOTE: a discovery namespace 304 ("System.Web.Services.Discovery") containing classes that allow Web service consumers to locate available Web Services on a Web server, and a protocols namespace 306 ("System.Web.Services.Protocols") containing classes that define the protocols used to transmit data across a network during communication between Web service clients and the Web service itself.

05:36:06 <dc_rdfig> Added comment E5.

05:36:32 <mdupont> E: (Comment : this looks like the Soap UDDI stuff)

05:36:32 <dc_rdfig> Added comment E6.

05:39:43 <mdupont> E:QUOTE "W3cXsd2001.txt"

05:39:44 <dc_rdfig> Added comment E7.

05:40:05 <mdupont> E:Looks like the contain the w3c schemas in there verbatim as part of the subimission

05:40:05 <dc_rdfig> Added comment E8.

05:40:19 <mdupont> E8:Looks like the contain the w3c schema in there, verbatim, as part of the submission

05:40:20 <dc_rdfig> Replaced comment E8.

05:52:05 <mdupont> hi golbeck GabeW

05:52:10 <GabeW> hi'

05:52:58 <mdupont> E: CONCLUSION [0101] Although the invention has been described in language specific to structural features and/or methodological acts, it is to be understood that the invention defined in the appended claims is not necessarily limited to the specific features or acts described. Rather, the specific features and acts are disclosed as exemplary forms of implementing the claimed invention.

05:52:59 <dc_rdfig> Added comment E9.

05:53:23 <mdupont> it looks like microsoft invented a web api

05:53:27 <mdupont> and soap

05:59:01 <golbeck> hi

06:00:09 <mdupont> E:it looks like ms is covering all the non core asp.net platform in this patent

06:00:09 <dc_rdfig> Added comment E10.

06:00:26 <mdupont> E:specifically the apis for the client/server interaction

06:00:27 <dc_rdfig> Added comment E11.

06:00:35 <mdupont> E:but also the winforms

06:00:35 <dc_rdfig> Added comment E12.

06:01:05 <mdupont> E:this could lock in the users of non-standard apis, but still leave the core open.

06:01:05 <dc_rdfig> Added comment E13.

06:37:04 <jordan> jordan is now known as graduate

09:20:12 <arnarl> hi

10:47:37 <burtonator> has anyone thought about applying the XSLT processing model to RDF graphs so one can do transformations on the data?

10:54:38 <dajobe> that would be a bit insane without an XML format for the triples

10:54:58 <dajobe> since the rdf/xml syntax hides the triples somewhat

10:55:29 <burtonator> hm... what is wrong with accessing them like they way they are done with jena SQL select query sytax

10:55:35 <burtonator> AKA ?subject

10:56:47 <dajobe> I don't get that

10:57:03 <dajobe> if you mean an xml lang to transform the rdf graph, that has been considered a couple of times

10:57:08 <dajobe> more of an rdf graph/path language

10:57:17 <burtonator> well... I guess first there would need to be a standardized RDF pattern mathing mechanism

10:57:19 <dajobe> I think, see Versa lang in 4suite

10:57:20 <burtonator> WHERE clauses

10:57:28 <dajobe> yeah, see above

10:57:36 <burtonator> ah...

10:57:40 <burtonator> I am just thinking out loud btw

10:57:53 <dajobe> there was an rdf path mailing list, no mails for a year or so

10:58:00 <burtonator> ha

10:58:01 <burtonator> yeah

10:58:21 <burtonator> it might pick up when RDF things get going a little more

10:58:25 <dajobe> yup

10:58:44 <dajobe> if you google for "rdf path" and "rdfpath", there are lots of the pointers

10:58:54 <burtonator> ah... great

10:59:04 <burtonator> I am using Mozilla RDF templates for this

10:59:15 <burtonator> they are similar to what I want but their RDF support is SO old it isn't funny

11:02:14 <dajobe> :)

11:02:41 <dajobe> i've never used XUL stuff. danbri has or knows who to ask

11:02:56 <dajobe> or not XUL, I don't even know what the things are called

11:03:43 <burtonator> that is probably me

11:03:53 <burtonator> danbri and I have been talking about XUL/RDF lately

11:04:24 <dajobe> well, danbri would probably now ask you to record your efforts somewhere such as the ESW wiki :)

11:04:32 <dajobe> but I think you have your own, mebbe?

11:04:49 <burtonator> hm...

11:04:54 <burtonator> I should blog about them I guess

11:05:14 <burtonator> and RDF/RSS developers experiences with Mozilla/XUL/RDF/RSS

11:05:43 <dajobe> that would be good

11:06:37 <dajobe> is that on http://relativity.yi.org/ ?

11:06:47 <burtonator> peerfear.org

11:06:58 <burtonator> where did you get relativity... that is deprecated btw

11:07:05 <dajobe> google

11:07:20 <dajobe> you need a <h1><blink>deprecated</blink></h1> there then

11:07:29 <burtonator> ha

11:07:43 <burtonator> google never forgets

11:36:34 <danbri> dc_rdfig:view

11:36:35 <dc_rdfig> A: Parsing RDF using XSLT in Ruby, by Max Froumentin (http://www.w3.org/2001/12/rubyrdf/xsltrdf/README.html)

11:36:36 <dc_rdfig> B: http://www.w3.org/2000/10/swap/doc/Shortcuts.html

11:36:37 <dc_rdfig> C: "Considered Harmful" Essays Considered Harmful (http://www.meyerweb.com/eric/comment/chech.html)

11:36:38 <dc_rdfig> D: .Net patent could stifle standards effort (http://news.com.com/2100-1001-984052.html?tag=fd_top)

11:36:39 <dc_rdfig> E: PATENT20030028685 Application program interface for network software platform (http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PG01&p=1&u=/netahtml/PTO/srchnum.html&r=1&f=G&l=50&s1='20030028685'.PGNR.&OS=DN/20030028685&RS=DN/20030028685)

11:37:40 <danbri> E:Please don't post details of patents in RDF IG weblog! Many developers like to steer completely clear of any such knowledge. Thanks.

11:37:41 <dc_rdfig> Added comment E14.

11:39:27 <dajobe> I can remove?

11:39:35 * danbri hmms

11:39:49 * danbri would prefer mdupont being here...

11:40:05 <danbri> yeah, zap it. Sorry Mike.

11:40:25 <dajobe> hold on

11:43:18 <dajobe> gone

11:43:57 <dajobe> you want D: to go to?

11:44:10 <dajobe> nah, that's just news

11:44:40 <danbri> just chop the details. you could leave the raw url and a health warning, perhaps?

11:47:16 <burtonator> ignorance is bliss

11:47:19 <dajobe> gah

11:48:27 <burtonator> sweet... basic FOAF support in NewsMonster

11:48:31 <burtonator> just XSLT though

11:50:32 <danbri> danbri has changed the topic to: #rdfig RDF Interest Group 24x7 hack'n'chat, see http://rdfig.xmlhack.com/

11:50:39 <burtonator> gotta blog about this now...

11:50:45 <danbri> cool"

11:50:46 <danbri> !

11:50:54 <danbri> is newsmonster released / public?

11:51:02 <danbri> what does the foaf support do?

11:51:04 <burtonator> not public

11:51:08 <danbri>http://www-uk.hpl.hp.com/people/marbut/someQuestionsOnCCPP.htm

11:51:12 <burtonator> I keep running into problems

11:51:24 <burtonator> I might just say "fuck it... I tried my best"

11:51:36 <danbri>http://www-uk.hpl.hp.com/people/marbut/someQuestionsOnCCPP.htm

11:51:42 <danbri> oops chump dead

11:51:43 <dajobe> wait a sec

11:51:51 * burtonator kicks chump

11:52:48 <dajobe> go ahead

11:52:49 <burtonator> are there any other RSS feeds that push the RDF edge?

11:52:58 <dajobe> you mean featureful?

11:53:13 <dajobe> aaronsw's uses mod_content and CDATA stuff

11:53:35 <danbri>http://www-uk.hpl.hp.com/people/marbut/someQuestionsOnCCPP.htm

11:53:35 <dc_rdfig> E: http://www-uk.hpl.hp.com/people/marbut/someQuestionsOnCCPP.htm from danbri

11:53:51 <burtonator> right... I handle aarons

11:53:59 <danbri> E:|Some Questions and Answers On CC/PP and UAProf, by Mark Butler 2002-03-20

11:53:59 <dc_rdfig> Titled item E.

11:54:12 <burtonator> Ben is doing Foaf in dc:contributor

11:54:15 <danbri> burtonator, you might try a japanese feed for char encoding issues

11:54:17 <burtonator> so that is what I am doing.

11:54:18 <mortenf_> you could check out mattb's picture logs.

11:54:21 <danbri> nice

11:54:46 <dajobe> most of mattb's site is rss-based

11:54:47 <burtonator> mortenf: URL?

11:54:57 <danbri> picdiary.com

11:55:03 <mortenf_> :-)

11:55:57 <dajobe> here's a japanese rss1.0 http://www.kanzaki.com/info/rss.rdf that has a nice mix of fonts

11:56:08 <dajobe> hmm, bad XML. doh

11:56:10 * mortenf_ is off for lunch, but hopefully will have an even more heavily annotated photo rss feed by tonight...

11:56:16 * danbri just found http://www.kanzaki.com/info/rss.rdf too

11:57:38 <dajobe>http://rdflib.net/2003/02/07/rdflib-1.2.2/

11:57:38 <dc_rdfig> F: http://rdflib.net/2003/02/07/rdflib-1.2.2/ from dajobe

11:57:42 <dajobe> F:|RDFLib 1.2.2

11:57:42 <dc_rdfig> Titled item F.

11:57:51 <dajobe> F:"As of release 1.2.1 RDFLib now passes all of the positive and negative rdfcore [parser] test cases."

11:57:51 <dc_rdfig> Added comment F1.

11:57:56 <dajobe> F:well done eikeon

11:57:57 <dc_rdfig> Added comment F2.

11:58:16 <burtonator>http://www.peerfear.org/download/newsmonster-foaf.png

11:58:16 <dc_rdfig> G: http://www.peerfear.org/download/newsmonster-foaf.png from burtonator

11:58:34 <burtonator> G:|NewsMonster showing FOAF dc:contributor from Content Syndication with RSS

11:58:34 <dc_rdfig> Titled item G.

11:59:12 * dajobe lol at "chumps" bookmark

12:00:15 <danbri> burtonator, do you have any grouping mechanism in newsmonster?

12:00:53 <danbri> eg. if I had a bunch of RSS feeds, like 100s, and they were categorised, could I have them be organised in the moz ui, rather than a massive huge list?

12:01:06 <danbri> tabs or trees or whatever? maybe when you do more xul stuff?

12:01:41 <burtonator> you mean with dc:category?

12:02:05 <burtonator> or so you can say that this is the "Linux" category... and organize your feeds like that

12:02:06 <burtonator> ...

12:02:30 <danbri> that would be one way (dc:subject), but I was thinking more 'feeds from my collaborators', 'feeds from photo synidcation sites' etc., using richer RDF to categorise them...

12:02:48 <danbri> so maybe i'd use external rdf query tools to create the categories from other data

12:02:55 <danbri> feeds about the War...

12:03:03 <burtonator> there is a way I can do it

12:03:07 * danbri nods

12:03:12 <burtonator> I need to fix some of the RDF support in Moz.

12:03:20 <burtonator> it is kind of schizophrenic in some areas

12:04:04 <danbri> that sounds good. what in particular is up?

12:04:23 <burtonator> a lot of mozilla is over-engineered in some areas and this might be one

12:04:27 <burtonator> what do you mean up?

12:04:36 <danbri> as in 'what's up, doc?'

12:04:41 <danbri> ie what's wrong...

12:04:55 <burtonator> oh... it is not behaving the way with my RDF data

12:05:03 <burtonator> like column sorting

12:05:13 <danbri> I don't think it is used to consuming externally generated RDF

12:05:22 <burtonator> yeah... it isn't

12:05:25 <danbri> they use it more like an internal data API, so bugs w.r.t. RDF specs got missed.

12:05:34 <burtonator> yeah.. tell me about it

12:09:24 <burtonator> danbri: was it you or ed that created this? http://rdfweb.org/foaf/examples/unreadable-thumb.jpeg

12:10:34 <danbri> me, using shellac's rdfauthor visualization of edd's foaf file

12:10:40 <danbri> or edds+mine I forget

12:10:51 <danbri> full vsn: http://rdfweb.org/foaf/examples/edd-foaf.jpeg

12:11:41 <burtonator> ok... that is the one i wanted to use on the semantic page of NewsMonster

12:11:50 <burtonator> images are worth a thousand words for consumers

12:11:55 <danbri> svg one is nice, you can search it with rightclick/'find' in adobe plugin, ihttp://rdfweb.org/foaf/examples/edd-foaf.svg

12:11:58 <burtonator> if it is cool of course

12:11:59 <danbri> fine by me

12:12:19 <burtonator> the svg doesn't work under batik which is my only decent SVG tool

12:12:27 * burtonator wants SVG in Mozilla already!

12:12:29 <danbri> it was a quick hack to make a temprary logo

12:12:33 * danbri too

12:13:40 <burtonator> the hold up is raphs code which he has approved for re-license but Mozilla can't get their act together

12:14:18 <danbri> raph from #infoanarchy etc?

12:15:12 <burtonator> yeah

12:16:25 <danbri> how complete is raph's stuff? does it do .js scripting, interactive stuff?

12:16:36 <burtonator> it is just libart...

12:16:51 <burtonator> but I believe the mozilla SVG impl could/can handle js/dom stuff

12:17:10 * danbri wonders if there's anything the rest of us can do...

12:17:40 <burtonator> hang out in #mozilla and bitch :)

12:17:55 <danbri> yeah, they really like that!

12:18:30 <burtonator> well... they need to be snapped into reality every once and a while

12:18:42 <burtonator> not just them... developers in general... hard working ones particularly

12:18:48 * danbri nods

12:19:19 <burtonator> danbri: did you look at my blognet page? What did you think?

12:19:39 <burtonator> I mean I realize it is just RDF but I think it is a simple meme that could go far

12:19:43 <burtonator> ... like FOAF

12:20:02 <burtonator> I have tentative support from Blogger, Moveable Type, etc

12:20:08 <danbri> you mean http://www.peerfear.org/download/newsmonster-foaf.png ?

12:20:33 <burtonator> ... file:///projects/newsmonster/site/blognet.html

12:20:48 <burtonator> semantic like service discovery with RSS

12:21:20 <danbri> got an http:// uri? i don't cache things yet

12:22:24 <burtonator> HA

12:22:27 <burtonator> sorry

12:22:36 <burtonator> ... http://www.newsmonster.org/blognet.html

12:30:18 <burtonator> danbri: I might be willing to support ical/RDF in Mozilla btw...

12:31:05 <libby> ooh, really?

12:31:24 <danbri> oooh, that'd be nice... are you aware of the mozilla calendar work? they've started investigating the rdf/ical thing too...

12:31:27 <libby> burtonator, are you following the www-rdf-calendar discussions and irc chats?

12:31:50 <burtonator> yeah... I am sure they are are probably using RDF datasources internally

12:32:01 <burtonator> er ... in mozilla

12:32:07 <burtonator> libby: no... just found out about this

12:32:26 <libby> hm , not sure they are in moz for calendar stuff

12:33:30 * burtonator imports his evolution ical into mozilla cal

12:33:47 <burtonator> all the XUL controls support RDF datasoruces

12:33:49 <burtonator> sources

12:34:03 <burtonator> this is how mail and news works I think.. RFC 822 in and then an RDF representation

12:35:29 <libby> tha cal thing is a different project though, orginally donated code.

12:35:59 <burtonator> oh... I know

12:36:13 <libby> ok

12:36:15 <danbri> yeah, using a moz datasource for calendar would be the normal mozilla architecture, but it was a separate component so may not do that yet.

12:36:34 <libby> well, if you fancy coming along to next IRC calendar meetingthat'd be great

12:36:57 <burtonator> I will be... when is it?

12:37:00 <danbri>http://lists.w3.org/Archives/Public/www-rdf-interest/2002Nov/0012.html

12:37:01 <dc_rdfig> H: http://lists.w3.org/Archives/Public/www-rdf-interest/2002Nov/0012.html from danbri

12:37:43 <libby>http://lists.w3.org/Archives/Public/www-rdf-calendar/2003Feb/0005.html

12:37:44 <dc_rdfig> I: http://lists.w3.org/Archives/Public/www-rdf-calendar/2003Feb/0005.html from libby

12:38:13 <burtonator> cool...

12:38:15 <libby> I:|next calendar IRC meeting - wednesday Feb 12th, 17:00 UTC

12:38:15 <dc_rdfig> Titled item I.

12:40:23 <burtonator> ha... 9:00 AM PST tomorrow.

12:40:30 <burtonator> I will probably be sleeping... that is in 5 hours

12:40:53 <libby> I: agenda items to libby.miller@bristol.ac.uk today please

12:40:53 <dc_rdfig> Added comment I1.

12:40:55 <libby> eh?

12:41:06 <mortenf_> erh, it's in appr. 27 hours?

12:41:16 <libby> yeah, that's right :)

12:41:37 <libby> thnks mortenf_, counting on my fingers there

12:41:43 <mortenf_> :)

12:41:48 <libby> so plenty of sleep time

12:41:51 <burtonator> oh... worry... 9:00 wed

12:41:53 <burtonator> gotcha

12:42:04 <libby> well anyway, there's always the logs.

12:42:38 <mortenf_> btw, can anyone point at TGN (Getty Thesaurus) used in RDF?

12:43:13 <mortenf_> i'm seeing no end of "use a controlled vocab, i.e. TGN", but no actual impl./use...

13:04:22 <danbri> anyone know how to reset a forgotten Movable Type password?

13:04:45 <danbri> yeah, TGN isn't publically available except under strict licensing conditions.

13:04:46 <pixel> yeah. hold on.

13:05:24 <pixel> on the login screen, there should be a "forgot your password?" link

13:05:41 <danbri> yeah, but it assumes you got as far as giving it your town of birth or whatever

13:05:49 <pixel> oh. hehehe

13:05:50 * danbri rummaging in the database files

13:05:57 <danbri> not me

13:05:59 <pixel> that's all I know :)

13:06:05 <danbri> ta anyway

13:07:05 <danbri> "Author has not set birthplace; cannot recover password"

13:07:47 <mortenf_> hmm...

13:13:23 <pixel> anyone know the simplest way to use SQL 2000 DTS to import an XML file?

13:14:32 * danbri has no idea, sorry

13:14:42 <pixel> darn.

13:15:39 <pixel> they touted this version as being cozy with XML, but it seems like they just slapped together a couple of new functions, and didn't think about how people might really use it.

13:17:02 <danbri> aha, found MT fix http://www.movabletype.org/cgi-bin/ikonboard/ikonboard.cgi?act=ST;f=10;t=3070;

13:17:49 <pixel> anyone know of a repository (like CPAN) for movable type templates, comment system modifications, etc?

13:18:00 <pixel> (you got me thinking about moveable type)

13:18:06 <danbri> only their website

13:18:45 <pixel> thx

13:21:54 <mortenf_> they are working on a repository, in the mt-dev yahoogroup.

13:22:48 <pixel> excellent.

13:27:54 <dajobe> is it just me or are forums a pain? Can't access via email, news and sometimes not by google. Much prefer a proper web site or mailing list. </rant>

13:28:31 <burtonator> the would be better if they were RDF and FOAF enabled

13:28:56 * pixel has been writing some forum software for the past few years that uses search-engine friendly URLs, and full-text-posts-in-rss but he'll probably never get back to it :)

13:29:39 <pixel> actually, there's more idea in there than there is code.

13:29:59 <dajobe> you need (gack!) lazycodeweb

13:30:10 <pixel> hehe

13:39:21 <danbri>http://esw.w3.org/t/view/ESW/SemWebSpain

13:39:21 <dc_rdfig> J: http://esw.w3.org/t/view/ESW/SemWebSpain from danbri

13:39:33 <danbri> J:|A Wiki page about Semantic Web materials and projects in Spain, or in the Spanish language.

13:39:33 <dc_rdfig> Titled item J.

13:39:55 <danbri> J:Feel free to contribute...

13:39:55 <dc_rdfig> Added comment J1.

15:28:28 <mdupont> yes

15:57:04 <molar> hello

15:57:20 <molar> does anyone know where I can get help with RDFStore

15:57:53 <molar> the website doesn't seem to have been upadated for a while, but I see that in CVS there has been an upadate recently and a new version might be out soon?

15:58:49 <libby> I think thery're doing a lot of work. there is a newish version: http://rdfstore.sourceforge.net/downloads/snap.tgz

15:59:41 <libby> alberto reggiori was doing lots of work on it

16:00:32 <molar> for reasons out of my control i'm using it in windows and cannot get the ppm file to work

16:00:44 <molar> and installing it with nmake throws up errors

16:01:17 <libby> I havn;t tried that...suggest you email alberto...

16:01:24 <molar> ok

16:01:29 <molar> thanks for your advice

16:01:46 <libby> he's very nice, I'm sure will help is possible

16:02:03 <molar> great

16:03:23 <zool> it's worth trying RDF::Core which is pure perl afair.

16:03:53 <molar> i'll have a look

16:03:54 <molar> thanks

16:15:09 <danbri> alberto is still around, i had mail from him recently re rdf thesarus stuff

16:15:42 <molar> I've just emailed him

16:58:21 <eikeon> Is there a single url that lists the documents making up the complete specification of RDF. I had a link to an email listing them called "latest working drafts" that was looking to update.

17:00:01 <eikeon> Looks like they all just cross reference each other. Delinkifying :)

17:00:33 <dajobe> how about http://www.w3.org/2001/sw/RDFCore/#documents

17:00:52 <eikeon> Thanks... relinkifying.

17:01:21 <dajobe> BTW, you said " passes all of the positive and negative rdfcore test cases."

17:01:27 <dajobe> that isn't quite clear enough

17:01:33 <dajobe> you mean all the parser tests

17:01:43 <dajobe> not misc, entailment, datatypes etc.

17:01:54 <dajobe> or maybe you added an inference engine and I didn't notice :)

17:04:34 <eikeon> Yes. Upps! I will fix that.

18:38:53 <danbri> is mit offline again?

18:45:52 <em> danbri, a quick reboot

18:46:05 * em waves

19:50:59 <niq> anyone deal with medical data as RDF?

20:04:11 <dajobe> niq: I think there is some daml, er OWL work going on

20:04:19 <dajobe> and maybe jonathan borden is doing something

20:04:23 <mdupont> hey all

20:04:38 <zool> should i really be switching from daml to OWL?

20:04:39 <mdupont> i just wanted to say that i am sorry if i offended anyone today

20:04:46 <mdupont> with the patent stuff

20:04:53 <mdupont> i did not know it was an issue at all

20:04:58 <mdupont> and wont do it again

20:04:59 <mdupont> :)

20:05:32 <mdupont> I would not have posted it if it did not reference the W3C

20:05:34 <dajobe> s'okay - so-called software patents are a PITA

20:05:45 <mdupont> PITA?

20:05:54 * mdupont looks in the TLA dictionary

20:05:57 <dajobe> pain in the ....

20:06:12 <mdupont> hehe

20:06:16 <mdupont> :)

20:06:40 <dajobe> niq: I found some borden ref - http://lists.xml.org/archives/xml-dev/200204/msg00944.html

20:06:50 <dajobe> snomed seems to be the thing

20:07:11 <niq> oh hi, wasn't watching

20:07:16 * niq reads

20:07:38 <mdupont> oh

20:07:40 <mdupont> :)

20:07:45 <mdupont> i will give you another weel

20:07:47 <mdupont> week

20:07:48 <dajobe> niq and http://www.openhealth.org/ seems to have more links

20:07:52 <mdupont> take your time

20:07:56 <dajobe> eh?

20:08:00 <mdupont> it is not very important

20:08:39 <niq> I have a prospective client who is collecting medical stats and has a view to publishing them; if there's a suitable schema I'll suggest RDF

20:08:54 <dajobe> I'm sure there is some daml/owl one, it's bugging me now

20:08:57 <dajobe> ask bijan?

20:09:14 * niq opens openhealth.org

20:09:25 <mdupont> dajobe: i would like to use your redland as the basis for the dotgnu.rdf c# interface

20:09:32 <dajobe> ah, that

20:09:44 <mdupont> and the dotgnu wants to submit that to the w3c for approval and support

20:09:49 <dajobe> erm?

20:09:53 <mdupont> many of the people here support that idea

20:09:55 <dajobe> not sure what you mean by that?

20:10:09 <dajobe> do you mean design a c# (.NET CLR?) API for RDF?

20:10:12 <mdupont> i mean a RFC, a submission of a API

20:10:13 <mdupont> yes

20:10:14 <mdupont> that

20:10:23 <mdupont> just the interface

20:10:27 <mdupont> not the implementation

20:10:33 <dajobe> ok, we alreayd discussed that a bit

20:10:37 <mdupont> like what ECMA does for the core C#

20:10:46 <dajobe> oh

20:10:47 <mdupont> dajobe: cool

20:10:58 <dajobe> I don't follow the c# stuff

20:11:21 <dajobe> ... so some things that the CLR might provide, would need a modifcation to the api approach

20:11:36 <dajobe> like some langs such as python have seqs, lists, hashes/dicts etc., others not

20:11:55 <dajobe> I have to go with the more general approach for redland, at least to start

20:12:06 <dajobe> and tend to add the detailed language metaphors later

20:12:06 <mdupont> the C# supports collections and Uris

20:12:10 <dajobe> sure

20:12:19 <mdupont> and also with the parrot/dotgnu interop

20:12:31 <dajobe> I was just mentioning on #foaf that I'm toying with aligning the redland py api with eikeon's rdflib one, or at least investigating that

20:12:34 <mdupont> there will the possibilty to implement that api in python/parrot

20:12:46 <mdupont> dajobe: ok cool

20:12:47 <dajobe> parrot is the perl6 thing?

20:12:58 <mdupont> parrot is the python/perl6

20:12:59 * danbri thinking of aligning my ruby api with rdflib and cwm's approach...

20:12:59 <mdupont> thing

20:13:10 <dajobe>http://www.parrotcode.org/

20:13:10 <mdupont> ok, you guys are the experts

20:13:11 <dc_rdfig> K: http://www.parrotcode.org/ from dajobe

20:13:17 <mdupont> i am just a beginner

20:13:31 <dajobe> K:|Parrot - VM for bytecode interpreted languages such as perl6

20:13:32 <dc_rdfig> Titled item K.

20:13:36 <mdupont> and am just interested in getting some dotgnu.rdf stuff worked out

20:13:50 <dajobe> K:or python

20:13:51 <dc_rdfig> Added comment K1.

20:13:59 <dajobe> that was the april fool that turned real IIRC :)

20:14:08 <dajobe> anyway

20:14:12 <mdupont> i would love if we could work togeather somehow under your guidence

20:14:21 <mdupont> <sp?>

20:14:29 <dajobe> apis are on the track fo things we are working on with SWADE

20:14:40 <_joshua> parrot us supposedly very different from c# java etc

20:14:52 * mdupont remembers "bad spellers of the world untie"

20:15:00 <dajobe> K:An [http://www.python.org/parrot.html|April fool that turned real]

20:15:00 <dc_rdfig> Added comment K2.

20:15:17 <mdupont> _joshua: dotgnu will compile to parrot in the future

20:15:27 <mdupont> the dotgnu ecma core lib will be made available

20:15:30 <mdupont> to the parrot

20:15:41 <mdupont> they are already working on it

20:15:48 <_joshua> that nebulous "will" in the way that all open source may or may not actually ever be completed :)

20:15:52 <zoyd> is WSDL an alternative to RDF ?

20:15:58 <_joshua> no

20:16:02 <mdupont> _joshua: there are already prototypes

20:16:11 <danbri> no, just something that could have been an rdf app...

20:16:20 <mdupont> i gotta run

20:16:26 <mdupont> just wanted to say hi

20:16:31 <danbri> hi :)

20:16:33 <mdupont> :)

20:16:38 <mdupont> and later!

20:16:39 <dajobe> zoyd: try googling for wsdl rdf, there are a few things out there discussing them by Uche Ogbuji and others

20:17:07 <zoyd> ok.

20:17:21 <danbri> .google wsdl rdf uche

20:17:22 <datum> wsdl rdf uche: http://www.w3.org/2002/02/21-WSDL-RDF-mapping/

20:17:41 <dajobe> yeah but there are ibm developerworks articles that expand

20:18:03 <danbri> yeah, ericp's page links to there, see http://www-106.ibm.com/developerworks/webservices/library/ws-soap/code.html

20:19:44 * dajobe closes the rdf helpdesk for a bit ... goes for food

20:19:47 * bijan is never askd

20:20:07 <mdupont> :)

20:20:20 <mdupont> i gotta go to a party

20:20:47 <mdupont> :)

20:20:51 <mdupont> ok, guys

20:21:12 <mdupont> i will hope someone writes back about a possible date

20:21:17 <mdupont> for an kick off chat,

20:21:34 <mdupont> and otherwise, I will just continue work on the c# stuff

20:27:53 * eikeon stops by... sees lots of talk about interop... nice.

20:31:39 <eikeon> Am definitely interested in helping out on that front. And am interested in helping with a native C# one... one that takes advantage of what the language has to offer... have not yet finished seeing how rdflib interface maps to C# yet.

21:36:40 <jhendler> TrustBot, golbeck@cs.umd.edu to timbl@w3.org

21:36:41 <TrustBot> golbeck@cs.umd.edu should trust timbl@w3.org at a level 7.272920203735144 on a scale of 1 - 9 according to my calculations.

21:37:40 <jhendler> TrustBot, explain golbeck@cs.umd.edu to timbl@w3.org

21:37:40 <TrustBot> Trust values are a weighted average. In this case, the maximum trust path from golbeck@cs.umd.edu to timbl@w3.org had value 8.0. The minimum trust path had value 6.0. golbeck@cs.umd.edu has 4 neighbors with paths to timbl@w3.org, and the weighted average was 7.272920203735144.

21:37:41 <TrustBot> The maximum length path is golbeck@cs.umd.edu->7.0-> 7.0-> 8.0-> 9.0 -> timbl@w3.org and the minimum length path is golbeck@cs.umd.edu->8.0-> 9.0 -> timbl@w3.org.

21:38:07 <golbeck> hello AI class

21:40:19 <jhendler> .google golbeck

21:40:21 <datum> golbeck: http://www.cs.umd.edu/~golbeck/

21:40:53 <jhendler> .time

21:40:53 <datum> Tue, 11 Feb 2003 21:40:53 GMT

21:55:26 <dmiles> i am looking for a clear CycL to log:* rdf

21:55:56 <dmiles> actually .. where is the best spec for log: operatators

21:56:33 <dmiles> i wrote a English To CycL translator.. i want to take it a few steps futher into Daml

21:56:37 <dmiles> jllykifsh

21:56:39 <jllykifsh> hi dmiles I will answer you in "#rdfig" until you say "goodbye"

21:56:57 <jhendler> dmiles sounds cool

21:57:44 <dmiles> english2Kif('?I',"Joe likes Susan",X),say(X).

21:57:47 <jllykifsh> (assert

21:57:48 <jllykifsh> (verbSemTrans

21:57:49 <jllykifsh> (#$likes-Generic

21:57:51 <jllykifsh> (named Joe)

21:57:53 <jllykifsh> (named Susan)) true))

21:58:22 <dmiles> yeah.. I wrote it for a special infernce engine (and OpenCyc) but think it would have apeal here

21:59:02 <dmiles> english2Kif('?I',"Joe can see coffee",X),say(X).

21:59:06 <jllykifsh> (assert

21:59:07 <jllykifsh> (modal

21:59:08 <jllykifsh> (can

21:59:09 <jllykifsh> (named Joe)

21:59:11 <jllykifsh> (verbSemTrans

21:59:13 <jllykifsh> (#$and

21:59:15 <jllykifsh> (#$isa ?ACTION3 #$VisualPerception)

21:59:17 <jllykifsh> (#$performedBy ?ACTION3

21:59:19 <jllykifsh> (named Joe))

21:59:21 <jllykifsh> (#$perceivedThings ?ACTION3 ?coffee1))

21:59:23 <jllykifsh> (and true

21:59:25 <jllykifsh> (isaMember ?coffee1

21:59:27 <jllykifsh> (#$Noun N03886725)))))))

21:59:49 <dmiles> N03886725 is the wordnet synset

22:00:17 <dmiles> the #$Noun thing in this case is a Collection denoting function

22:01:07 <dmiles> i would like to have it generate Daml or RDF logic

22:02:07 <dmiles> it also provides a proof why it understands things the way it does so that it can be improved

22:05:13 <dmiles> the goal for now (besides better language coverage) is to hook it up so we can do English to RDF Logic to Query or Assert

22:05:25 <dmiles> So one could "talk" with web pages

22:07:20 <sandro> Very cool.

22:07:34 * sandro runs off

22:07:47 <dmiles> well someone link me to a spec for log:* ;)

22:09:04 <SethR> i saw your write of for a self-aware assistant .. very kewl!

22:09:20 <SethR> is that really happening?

22:09:48 <dmiles> yeah.. it seemed like that grant offering was written for cycorp ..

22:09:54 <dmiles> but applied anyways

22:10:05 <dmiles> no word yet :( .. closing was Dec19th

22:10:23 <SethR> so it's funded already ... or pending?

22:10:29 <dmiles> well i have been working on it anyways.. i just rather get money at the same time

22:10:36 <dmiles> pending

22:11:03 <dmiles> i am living off savings.. but thats giving time to work

22:11:21 <dmiles> but in a couple months maybe will be funded

22:11:29 <jllykifsh> you do not seem quite certain .

22:11:53 <SethR> im trying go define a space where agengt like that could roump around ... see .... uuhh how do you do off here?

22:12:08 <dmiles> off:

22:12:21 <dmiles> off: well for "off" compliant bots ;P

22:12:22 <SethR> off: see http://robustai.net/papers/How_can_we_create_semspace.html

22:12:49 <dmiles> looking

22:12:57 <SethR> off: just dont want that published yet .. its just too flakey

22:13:35 <dmiles> off: "A portal would have a SQL database that remembers the objects and agents present in each semspace along with their states.

22:14:16 <dmiles> reading everything well but that; P

22:14:36 <deltab> eikeon: there's a typo in the examples links at the bottom of http://rdflib.net/2003/02/07/rdflib-1.2.2/doc/installing.html

22:14:36 <dmiles> oh well its just a portal ;P

22:14:48 <dmiles> (the SQL part)

22:15:34 <SethR> i suppose the SQL database could be cyc statements

22:16:34 <dmiles> yeah.. it's all good, ocver the last couple days i been trying to write an ODBC driver for Cyc

22:17:17 <dmiles> its fun trying to clone ANSI 92-SQL