Experimental IRC log swig-2007-10-08

Available formats: content-negotiated html turtle (see SIOC for the vocabulary)

Back to channel and daily index: content-negotiated html turtle

These logs are provided as an experiment in indexing discussions using IRCHub.py, Irc2RDF.hs, and SIOC.

13:53:49<TipTop>hello all
13:54:57<TipTop>i am trying to find ElementBlock in jena
13:55:18<TipTop>did it disappear?
14:16:59<AndyS>(answered in #jena - ElementBlock went away a long time ago)
14:37:52<timbl>Andys, good note on N3 dialects. Small corection about @base etc
14:38:04<timbl>It is not the onlt hing which can affect file concatenation.
14:38:17<timbl>Cwm assumes by default @prefix : <#>.
14:38:44<timbl>This is just so useful for teeny files that it is IMHO worth uttig in the spec.
14:39:40<timbl>But it means that if you want concatenate files, you have to put @bae <some abs uri>. @prefix : <#>. between them to remove previous context.
14:42:52<AndyS>timbl - will note - was more thinking of turtle where @base is the only thing IIRC
14:44:27<AndyS>Where's the assumed "@prefix : <#>" documented for N3? (I used n3.n3 mainly)
14:44:38<timbl>Does turtle not have any binding to : by default?
14:44:53<AndyS>Not as far as I ca see from the doc.
14:45:10<timbl>n3.n3 is a grammar -- it can't tell you tokenizing or whitespec or semantics
14:46:11<timbl>The spec is http://www.w3.org/DesignIssues/Notation3.html
14:47:03<AndyS>-> The empty prefix "" is by default , bound to the empty URI "".
14:47:10<timbl>Anothr =er issue: the < around URIs conventionally are so you can split them across lines
14:48:59<AndyS>Which is : bound to? <> or <#> ?
14:56:15<AndyS>Re: line splitting and <> - what's the rule - ignore whietspace or ignore newlines only?
14:56:37<timbl>Goof point. <#>.
14:56:46<AndyS>Got it - whitespace.
14:57:04<timbl>I'm not even sure of cwm implemenst the line splitting in <>
14:57:36<timbl>In principel I think it is a godo idea, as it gets around huge lines in files, esp. wit some of the sarql queries etc we get nowadays
14:57:43<timbl>but I don't think we have a test case
15:13:09<timbl>ACTION clarifies http://www.w3.org/DesignIssues/Notation3.html
15:18:18<dajobe>no, turtle does not bind : by default
15:22:52<timbl>I would suggest it as an improvement.
15:23:23<timbl>It makes little test files simpler
15:23:44<timbl>like :Joe :mother :Jane.
15:23:47<dajobe>hmm
15:25:08<dajobe>this is what I have right now for base uris/prefixes http://www.dajobe.org/2004/01/turtle/#sec-uris
16:37:08<EliasT>Anybody that can ping AndyS? http://sparql.org/sparql is down.
17:17:18<Wikier>hi
17:21:56<danja>phenny, tell bblfish s/Tallis/Talis & thanks!
17:21:59<phenny>danja: I'll pass that on when bblfish is around.
17:22:09<bblfish>ah thanks
17:22:12<phenny>bblfish: 17:21Z <danja> tell bblfish s/Tallis/Talis & thanks!
17:23:47<bblfish>done
17:24:14<danja>ta
17:24:28<danja>gnu-ish, yep - I think it was timbl reckoned it didn't pass his test of independent invention
17:24:50<danja>I dunno, gpl has been pretty useful
17:24:51<bblfish>I am going to get onto the talis platform. But first priority is Semantic Address Book again
17:25:47<danja>cool
17:26:47<danja>ACTION would like to hook that up to the Platform :-)
18:11:51<kidehen>bblfish: when will you resume the addressbook work?
18:12:50<kidehen>bblfish: we are already producing RDF Views based on Baetle re. Bugzilla :-) We also already have an Address Book Data Space that is waiting to mesh with the likes of your project and bengees :-)
18:13:36<kidehen>danja: I still don't grok how Virtuoso and/or ODS will mesh with Talis platform though? I do remember you mentioning some kind of RDF store abstraction.
18:14:03<kidehen>danja: cannot discern said abstraction from Talis platform material on the Wiki etc..
18:14:18<kidehen>danja: I see Talis as a variant of EC2 with an RDF bent. Please correct me if I am wrong etc..
18:14:56<kidehen>danja: It is also an RDF store (REST based) which kinda makes it alternative storage to most RDF stores, at least based on current material
18:16:00<kidehen>danja: for instance, how would Talis mesh with Virtuoso? Based on the fact that you can shove RDF into Virtuoso via Curl for instance (as you do currently with Talis Platform)?
18:16:10<kidehen>danja: I see the way in, but don't see the way out :-)
18:16:41<kidehen>danja: unless I can SPARQL CONSTRUCT ?
18:19:37<bblfish>kidehen: cool
18:56:34<danja_>kidehen, quick response (food on table!)
18:56:53<danja_>"Talis as a variant of EC2 with an RDF bent" - yep, not a bad summary
18:57:10<danja_>SPARQL CONSTRUCT - yep, can do
18:57:44<danja_>" how Virtuoso and/or ODS will mesh with Talis platform though?" - tricky one that :-)
18:58:12<danja_>I was thinking that at a base level both systems deal in linked data
18:59:04<danja_>but above that, we know each systems local conventions for URI construction etc, so e.g. cross-system should be easier to set up
18:59:09<danja_>biab
19:03:37<sr>is foaf:based_near [ a geo:Point; geo:long "bar"; geo:lat "foo"; dc:description "Brussels, Belgium'; ] good ?
19:05:56<sr>or should I use vcard:Country instead? (plain-english location because I am generating html from foaf with danja's sparql2hcard.xsl)
19:10:49<kidehen>danja: so the common ground is linked data. There is not concept of abstracted RDF data storage in Talis anymore, just confirming what I think to be the case etc..
19:58:51<DanC>bingo! I've got my 10 years of quicken data in sqlite. 15 years, maybe even.
19:59:03<DanC>sqlite> select Accounts.id, Accounts.name, total(Splits.subtot) from Accounts, Transactions, Splits where Splits.trx = Transactions.id and Transactions.acct = Accounts.id group by Transactions.acct;
19:59:05<DanC>^ a net worth report
20:01:56<sbp>ACTION briefly entertains ideas of comparing annual expenditures against the consumer price index using RDFery...
20:06:43<DanC>ACTION commits http://homer.w3.org/~connolly/projects/quacken/file/f1a45c56333c
20:09:09<DanC>ok, so I used my owlsql stuff for the schema... http://homer.w3.org/~connolly/projects/quacken/raw-file/c4b726456185/qdb.html
20:09:11<DanC>I wonder if I can convince django to work with that, or if I have to re-code the schema in python
20:11:45<DanC>that was wild... I tried to go to django.org ... I got a reply from an "OpenDNS guide". I chose into opendns... I dind't realize I was getting something like the verisign sitefinder while I was at it. hmm.
20:35:44<DanC>ACTION adapts owl2sql.xsl to owl2django.xsl easily enough... tests it out...
20:36:00<DanC>EnvironmentError: Environment variable DJANGO_SETTINGS_MODULE is undefined.
20:37:09<DanC>ACTION backs up to http://www.djangoproject.com/documentation/0.96/tutorial01/ ...
20:39:01<DanC>today's an MIT holiday, so I'm hacking on wishlist items
20:39:34<DanC>hmm... time for a name for this thing. I think quacken is taken
20:39:57<DanC>personal finance manager... pfm... quick personal finance manager.... pqfm ... too hard to type
20:40:03<DanC>piql
20:40:30<DanC>qfm
20:40:36<chimezie>.ety money
20:40:37<DanC>quick finance manager
20:40:47<phenny>"c.1290, 'coinage, metal currency,' from O.Fr. moneie, from L. moneta 'mint, coinage,' from Moneta, a title of the Roman goddess Juno, in or near whose temple money was coined; perhaps from monere 'advise, warn' (see monitor), with the sense of 'admonishing goddess,' [...]" - http://etymonline.com/?term=money
20:45:51<DanC>ACTION publishes http://homer.w3.org/~connolly/projects/owlsql/
20:50:00<chimezie>I came across a decent article on using mercurial to manage changesets off an SVN trunk, but seem to have misplaced it.
20:52:05<DanC>I'm a happy hgsvn user
20:52:24<DanC>hmm... django is asking if I want to make a superuser account. hmm.
20:52:57<chimezie>ahh yes, hgsvn - that rings a bell
21:03:09<DanC>oh... hmm... django has projects and apps... I think I'm making an app.
22:01:50<DanC>ooh... I think I just got my quicken data into a django app...
22:03:43<DanC>hmm... invalid literal for int() with base 10: '10/23/07'
22:03:45<DanC>not handling dates correctly.
22:07:16<DanC>Exception Location: /var/lib/python-support/python2.5/django/db/backends/util.py in typecast_date, line 46
22:12:31<pjenvey>hrmm django
22:14:51<pjenvey>ACTION recommends pylons + sqlalchemy but is biased
22:15:19<pjenvey>sounds like you might be using their admin interface anyway, which can be nice if it does what you need
22:23:45<DanC>django seems to be working; I don't think I want to know about the competition. (except that I like genshi...)

Back to channel and daily index: content-negotiated html turtle