IRC log of dig on 2007-07-03
Timestamps are in UTC.
- 00:26:49 [timbl]
- timbl (n=timbl@146-115-66-146.c3-0.lex-ubr1.sbo-lex.ma.cable.rcn.com) has joined #dig
- 04:19:01 [djweitzner]
- djweitzner has quit ()
- 11:49:57 [yosi_lap]
- yosi_lap (n=syosi@static-71-243-122-114.bos.east.verizon.net) has joined #dig
- 12:04:39 [RalphS]
- RalphS (n=swick@30-6-188.wireless.csail.mit.edu) has joined #dig
- 12:16:58 [timbl]
- timbl has quit ()
- 12:59:32 [djweitzner]
- djweitzner (n=djweitzn@pool-70-108-17-123.res.east.verizon.net) has joined #dig
- 13:01:54 [timbl]
- timbl (n=timbl@30-6-196.wireless.csail.mit.edu) has joined #dig
- 13:15:24 [yosi_lap]
- timbl, I found the turtle tests at http://www.dajobe.org/2004/01/turtle/tests/
- 13:15:38 [yosi_lap]
- I'm working on getting tester.py to use them
- 13:23:40 [timbl]
- Ah.
- 13:23:58 [timbl]
- I have made some progress ... 7 errors now, 31 this morning.
- 13:24:12 [timbl]
- Two of them are js stack overflow!
- 14:31:09 [DanC]
- hmm... svn logs seem kinda klunky... maybe I'll import into hg so I can use `hg view` locally. or maybe there are GUI svn tools that let me navigate between logs and diffs...
- 14:32:57 [DanC]
- apt-get install rapidsvn
- 14:34:37 [yosi_lap]
- I'm making progress
- 14:34:54 [yosi_lap]
- cwm is passing a turtle parse test it should fail --- this bears investigation
- 14:39:26 [DanC]
- rapidsvn doesn't seem to have enough sensible defaults. it keeps complaining about "no diff tool set in preferences" and such
- 14:40:46 [yosi_lap]
- my favorite diff tool is meld
- 14:41:28 [DanC]
- meld... thanks for the tip; it seems to be installed already
- 14:42:39 [DanC]
- hgimportsvn is no speed demon; it's working on 2006-01-29 17:16:47
- 14:43:46 [DanC]
- 2006-04 ...
- 14:48:13 [DanC]
- 2006-06 was a busy month
- 14:50:44 [davidli]
- davidli (i=dli@MACGREGOR-FIVE-FIFTY-NINE.MIT.EDU) has joined #dig
- 14:53:48 [lkagal]
- lkagal has quit ()
- 15:17:52 [yosi_lap]
- timbl, can you cvs update in the pyjs directory and run the turtle parser tests?
- 15:24:10 [timbl]
- hang on
- 15:24:28 [timbl]
- what have you changed?
- 15:25:24 [yosi_lap]
- I changed $SWAP/test/n3/tester.py to be able to understand the manifests listing the turtle tests
- 15:25:40 [yosi_lap]
- that live at http://www.dajobe.org/2004/01/turtle/tests/
- 15:26:01 [yosi_lap]
- I then changed $SWAP/pyjs/Makefile to use those tests
- 15:26:07 [yosi_lap]
- running them on cwm is fascinating
- 15:27:36 [timbl]
- C Makefile
- 15:27:37 [timbl]
- C tester.py
- 15:31:15 [soniaswim]
- soniaswim (i=soniaswi@30-7-172.wireless.csail.mit.edu) has joined #dig
- 15:31:52 [soniaswim]
- soniaswim has quit (Client Quit)
- 15:35:01 [yosi_lap]
- I don't have tabulator here; I can't run those tests
- 15:35:29 [yosi_lap]
- `` js: "../pyjs/testn3.js", line 11: Couldn't open file "/devel/dig/2005/ajar/ajaw/js/log.js". ''
- 15:38:30 [timbl]
- Hmmm.. yes.. you need the dig stiff checked out too
- 15:39:20 [timbl]
- it is a simple svn checkout
- 15:39:39 [timbl]
- http://dig.csail.mit.edu/2005/ajar/ajaw/Developer.html
- 15:39:57 [timbl]
- svn co --username dig https://svn.csail.mit.edu/dig/2005/ajar/ajaw/
- 15:51:27 [yosi_lap]
- 30-7-176:~/CVSROOT/WWW/2000/10/swap/pyjs yosi$ CLASSPATH="/devel/rhino/rhino1_6R5/js.jar" java org.mozilla.javascript.tools.shell.Main ../pyjs/testn3.js http://www.dajobe.org/2004/01/turtle/tests/test-03.ttl
- 15:51:27 [yosi_lap]
- # Parsing http://www.dajobe.org/2004/01/turtle/tests/test-03.ttl
- 15:51:27 [yosi_lap]
- org.mozilla.javascript.WrappedException: Wrapped java.io.FileNotFoundException: http://localhost/www.dajobe.org/2004/01/turtle/tests/test-03.ttl (../pyjs/testn3.js#32)
- 15:51:45 [yosi_lap]
- timbl, is there a right way to do this?
- 15:52:15 [yosi_lap]
- why did it stick in a localhost?
- 15:54:59 [lkagal]
- lkagal (n=lkagal1@30-5-196.wireless.csail.mit.edu) has joined #dig
- 15:58:26 [timbl]
- Stck in localhost? ah .. for working offlin, I have a local
- 15:58:35 [timbl]
- server which servs the stuff I have in CVS
- 15:58:40 [timbl]
- and also the main ontologies
- 15:59:00 [timbl]
- so I can work on a plane or on the cape by sticking n a localhost
- 15:59:10 [timbl]
- You can remove it or make it an env var
- 15:59:15 [timbl]
- teh latter sounds best
- 16:00:02 [timbl]
- Incidentally it isa useful \hack. the Mac OSX apacjhe you can tuen on with the system prefs and then put soft links i /Library/Web/Documents OWTTE
- 16:01:56 [yosi_lap]
- where is the system pref to turn apache off?
- 16:02:01 [yosi_lap]
- (it seems to be on)
- 16:49:54 [danbri]
- danbri (n=chatzill@unaffiliated/danbri) has joined #dig
- 17:06:36 [presbrey]
- presbrey (n=Joe@31-34-215.wireless.csail.mit.edu) has joined #dig
- 17:12:55 [soniaswim]
- soniaswim (i=soniaswi@30-7-172.wireless.csail.mit.edu) has joined #dig
- 17:22:35 [soniaswim]
- soniaswim has quit ()
- 17:24:10 [danbri]
- danbri has quit (Remote closed the connection)
- 17:39:34 [sandro]
- top
- 17:40:30 [sbp]
- bottom
- 17:40:32 [sbp]
- up
- 17:40:33 [sbp]
- down
- 17:40:37 [sbp]
- strange
- 17:40:39 [sbp]
- charmed
- 17:40:44 [sandro]
- sorry, wrong window.
- 17:40:47 [sandro]
- :-P
- 17:40:47 [sbp]
- :-)
- 17:41:12 [sbp]
- they should have an International Quark Day
- 18:07:43 [yosi_lap]
- sandro, I've never confused an irc window with a terminal before
- 18:08:43 [sbp]
- http://inamidst.com/code/shellbot
- 18:09:25 [sandro]
- I'm not sure if there's been some subtle change in how focus is handled, or what, but sometimes I find the text going to a window other than what I expect. I haven't isolated the issue.
- 18:09:26 [timbl]
- pwd
- 18:09:48 [sbp]
- /web/freenode/#dig
- 18:40:44 [timbl]
- Ralphs - do you know, is there a creative commons namespace for RDF ? not evident from their website. For the licence relationship
- 18:43:22 [DanC]
- hmm... I thought the RDF details were linked from the human-readable licenses, but I can't confirm at http://creativecommons.org/licenses/by/3.0/
- 18:44:09 [DanC]
- no "RDF" on http://wiki.creativecommons.org/Developers
- 18:44:26 [sbp]
- /licenses/by/3.0/: see the source
- 18:44:38 [sbp]
- uses "http://creativecommons.org/ns#" as a namespace in hot-comment RDF
- 18:44:51 [DanC]
- google finds http://validator.creativecommons.org/
- 18:45:40 [sbp]
- http://creativecommons.org/licenses/by/3.0/ is said to be a cc:License instance with cc: bound to that namespace... let's see if the namespace has further clues...
- 18:45:55 [sbp]
- yep, it lists cc:license
- 18:46:59 [sbp]
- and the namespace uses RDDL and links to http://web.resource.org/cc/schema.rdf as the schema
- 18:47:20 [timbl]
- AH, RDDL
- 18:47:24 [timbl]
- goes it grddl
- 18:47:31 [timbl]
- http://creativecommons.org/ns#licencse
- 18:47:40 [sbp]
- there's no @profile in there, nope
- 18:47:41 [timbl]
- is what we need, it seems.
- 18:47:54 [timbl]
- I wonder what it would take to get one there so tabr would grok it
- 18:49:05 [sbp]
- shame there's no normative RDDL -> GRDDL transform
- 18:49:29 [sbp]
- archaic in-the-wild RDDL, at least
- 18:53:24 [sbp]
- not sure to whom you'd send feedback requesting the addition of @profile... ironically the page is licensed under a Creative Commons Attribution 3.0 License, but there's no mention of to whom that attribution is to be made! the same problem beplagues me when using images from Flickr
- 18:55:37 [DanC]
- flickr is unclear where feedback belongs? odd; I've never seen an anonymous flickr photo
- 18:56:31 [sbp]
- not anonymous, but username only
- 18:57:00 [sbp]
- presumably people who list their usernames but not their real names are fine with being credited by their username only...
- 18:57:12 [soniaswim]
- soniaswim (i=soniaswi@128.30.7.172) has joined #dig
- 19:02:08 [DanC]
- I see
- 19:11:20 [soniaswim]
- soniaswim has quit ()
- 19:29:08 [presbrey]
- presbrey has left #dig
- 19:31:46 [timbl]
- Yosi? I checked stufff in
- 19:38:16 [timbl]
- Sonia? My FOAF file has a cc:license now
- 19:38:28 [timbl]
- oops she isn;t on iRC
- 20:23:40 [RalphS]
- RalphS has quit ("bye for today")
- 20:28:30 [DanC]
- er... where did all the tabulator tabs go? "Find all" doesn't seem to do anything
- 20:29:53 [DanC]
- hmm... presbey seems to be working on SPARQL update back-end stuff... did we make an issue for that? I don't see it in http://dig.csail.mit.edu/issues/tabulator/
- 21:01:10 [yosi_lap]
- timbl, I see the test file js_results.html does not include the turtle tests yet
- 21:03:54 [timbl]
- ummm
- 21:05:36 [timbl]
- $ grep -c description js_results.n30
- 21:05:36 [timbl]
- 137
- 21:05:37 [timbl]
- $ grep -c errorMessage js_results.n30
- 21:05:37 [timbl]
- 51
- 21:05:37 [timbl]
- $
- 21:07:05 [timbl]
- Looks as though the report formatter might not get new tests?
- 21:07:47 [timbl]
- $ grep -c "<tr" js_results.html
- 21:07:47 [timbl]
- 80
- 21:07:49 [timbl]
- hmmm
- 21:09:06 [timbl]
- Maybe it fails on any http ones
- 21:10:44 [yosi_lap]
- they should still show up
- 21:11:27 [yosi_lap]
- what happens if you run ``make js_results.html'' again?
- 21:11:39 [timbl]
- ah, the /localhost/
- 21:11:49 [timbl]
- .html up to date
- 21:12:06 [timbl]
- js testn3.js http://www.dajobe.org/2004/01/turtle/tests/test-12.ttlorg.mozilla.javascript.WrappedException: Wrapped java.io.FileNotFoundException: http://localhost/www.dajobe.org/2004/01/turtle/tests/test-12.ttl (testn3.js#33)
- 21:12:43 [timbl]
- it is the /local kludge
- 21:14:09 [yosi_lap]
- my box takes an extremely long time on test14.ttl --- not surprising, it is 10000 triples
- 21:15:20 [timbl]
- !!
- 21:15:38 [timbl]
- Not the sort fo thing to encourage oen to do the regression test!
- 21:16:19 [timbl]
- Not 13?
- 21:16:36 [timbl]
- Mine is hanging on 13
- 21:17:30 [timbl]
- maybe the log is behind. #13 has 82 bytes
- 21:18:26 [timbl]
- yes, it iddoinh 14
- 21:18:32 [timbl]
- is doing 14
- 21:18:50 [timbl]
- 216kB
- 21:19:49 [davidli]
- davidli has quit (kubrick.freenode.net irc.freenode.net)
- 21:20:04 [davidli]
- davidli (i=dli@MACGREGOR-FIVE-FIFTY-NINE.MIT.EDU) has joined #dig
- 21:20:39 [timbl]
- Seems a long time to me
- 21:21:09 [timbl]
- I think that test deserves to be in the performance test section.
- 21:21:28 [yosi_lap]
- it's 10000 triples
- 21:21:34 [yosi_lap]
- cwm takes a while on it
- 21:22:09 [timbl]
- same parser :)
- 21:22:34 [timbl]
- We haven't tested whether the predctive parser is quicker
- 21:22:41 [yosi_lap]
- not quite the same parser
- 21:22:44 [timbl]
- it could be .. less backtracking
- 21:23:54 [timbl]
- Not quite the same? I wondr how different they are.. how different the js and python runtimes and compilation are
- 21:24:04 [timbl]
- I have to go. I may comment out #14!
- 21:24:29 [timbl]
- Only 10k triples.
- 21:25:07 [yosi_lap]
- 30-7-176:~/SWAP/pyjs yosi$ time cwm http://www.dajobe.org/2004/01/turtle/tests/test-14.ttl > /dev/null
- 21:25:07 [yosi_lap]
- Try getting python-xml from http://downloads.sourceforge.net/pyxml/PyXML-0.8.4.tar.gz
- 21:25:08 [yosi_lap]
- real 2m16.289s
- 21:25:08 [yosi_lap]
- user 0m29.652s
- 21:25:08 [yosi_lap]
- sys 0m1.262s
- 21:27:35 [timbl]
- gtg
- 21:28:18 [timbl]
- ps: 7489 p2 R 11:50.71 java org.mozilla.javascript.tools.shell.Main ../pyjs/testn3.js http://www.dajobe.org/2004/01/turtle/tests/test-14.ttl
- 21:28:25 [timbl]
- 12 mins
- 21:28:35 [timbl]
- That is js in java
- 21:29:00 [timbl]
- Should pute timing measurements into tab'r parse
- 21:29:46 [yosi_lap]
- see you on Friday?
- 21:29:55 [yosi_lap]
- (if I end up going in on Friday)
- 21:30:39 [timbl]
- y
- 21:30:45 [timbl]
- at tami?
- 21:31:16 [timbl]
- timbl has quit ()
- 21:32:38 [yosi_lap]
- yosi_lap has quit ()
- 21:57:05 [lkagal]
- lkagal has quit ()
- 21:59:01 [davidli]
- davidli has quit ()
- 23:40:42 [timbl]
- timbl (n=timbl@146-115-66-146.c3-0.lex-ubr1.sbo-lex.ma.cable.rcn.com) has joined #dig