Experimental IRC log swig-2008-03-25

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.

04:57:34<puff>Hm, so #swig is for the semantic web? What channel is for the swig.org tool?
05:00:16<adu>puff: I don't think swig.org has a channel here
05:02:57<puff>Ah well, thanks anyway. Good luck with the semantic web.
05:03:33<adu>thanks, If you See What I Mean
05:05:05<puff>Whatever the in joke is, I don't get it.
05:05:16<puff>ACTION waves.
08:26:12<EtnaRosso>morning all
08:26:15<EtnaRosso>morning all
08:39:58<mahound>morning EtnaRosso
13:01:22<kwijibo>Hi, if you are exporting rdf from one store to another, using sparql to get the rdf out, but spreading it out over multiple requests (paging through the store with offset and limit), so as to cope with larger datasets, and the rdf uses some bnodes, is it possible to export the data without the risk of creating duplicates of the bnode resources?
16:13:47<bblfish>bengee?
16:36:41<bblfish>http://www.thearchitect.co.uk/weblog/archives/2008/03/000493.html
16:37:12<bblfish>A:| blog post: Delivering Data Portability (Part 2) - Sharing Contacts Between Social Networks
16:37:58<bblfish>A: "Microsoft announced the release of the Windows Live Contacts API, which web developers can use to enable their users to transfer and share their Windows Live Contacts in a safe and secure way. Starting today, Microsoft is working with Facebook, Bebo, Hi5, Tagged and LinkedIn to exchange functionally-similar Contacts APIs, allowing all partners to create a safe, secure two-way street for users to move their relationships between our
16:38:55<bblfish>A: talk of API makes me think this is non RESTful
16:39:41<bblfish>A: all that is needed to get foaf and linked data to compete is a way to have resources return different representations depening on the client identity
16:55:08<mahound>bblfish: "A RESTful API that works on Address Book objects."
16:55:23<mahound>http://dev.live.com/contacts/
16:55:35<mahound>B:| Windows Live Contacts API
16:55:45<mahound>the name "Windows live" scares me
16:56:27<bblfish>mahound: yes
16:56:35<bblfish>so it's a centralised service right?
16:57:30<mahound>seems so
16:57:45<mahound>"The Contacts API offers a more granular control over the user experience and allows users to grant and revoke permission to their data on a site-by-site basis."
16:57:55<mahound>this sounds so... OpenID
17:03:02<mahound>as for A, i disagree with the author
17:03:16<mahound>it's not a huge step for social networking
17:04:45<mahound>it's a social network cartel :)
17:13:30<bblfish>A: Ok so mahound lets me know that this is restful http://dev.live.com/contacts/ , but the problem remains that it seems very centralised
17:25:22<bengee>ACTION is back
17:26:48<danbri>sounds like the google contacts api
17:26:48<phenny>danbri: 22 Mar 00:32Z <iand> tell danbri yes re oauth
17:26:52<danbri>:)
17:41:02<bblfish>bengee: hi
17:41:12<bengee>hi
17:41:27<bengee>the username part has to be something simple, for http's sake
17:41:27<bblfish>So I like the idea of RDFAuth
17:41:43<bengee>but that doesn't rule out the idea of using pgp somehow
17:42:00<bblfish>well the idea is to link it to foaf somehow
17:42:16<bblfish>so that one can also get information about the person
17:42:33<bblfish>this would be openid simplified to the max
17:43:05<bblfish>but noting stops me having my foaf at http://bblfish.net/
17:43:08<bblfish>encoded as rdfa
17:43:33<bblfish>that may be a better id then: http://bblfish.net/#henry
17:44:27<bblfish>Anyway, this is getting really urgent
17:44:51<bblfish>I think it is the only argument people have against foaf for the moment: no protection
17:45:23<bengee>ACTION googles for pgp via php on shared hosts
17:46:24<bengee>hmm, doesn't seem to be too easy
17:47:20<bblfish>As I mentioned, RDF is flexible, this could also work for other cryptographic keys
17:48:27<bblfish>mhh. being flexible there, would require the header to specify the entryption mechanism
17:57:50<bengee>a pgp verifier could be used as an rdfauth endpoint, I guess
17:58:38<bblfish>a cool.
18:00:00<bblfish>Well I'd be really interested to see if we can get something sensible and get something going with Beatnik. IT would be good to get the Tabulator people involved too
18:00:10<bengee>a sort-of stateless endpoint that doesn't issue tokens for known users, but does a little bit of http to verify a token
18:01:58<bblfish>http://lists.w3.org/Archives/Public/semantic-web/2008Mar/0178.html
18:02:13<bblfish>We are just discussing C for those interested
18:06:42<bblfish>bengee: do you know oAuth
18:06:43<bblfish>?
18:07:02<bengee>looked at it a couple of times
18:07:14<bblfish>does it look anything like what we are doing?
18:07:17<bengee>but it seems to follow openid's redirect dance
18:07:20<bblfish>ah ok.
18:07:39<bengee>at least to generate the tokens
18:07:46<bblfish>Could you reply to that email on foaf dev?
18:07:51<bengee>after that you can do server-side communication
18:08:02<bengee>yeah, will do
18:08:07<bblfish>I need to look at it... thanks :-)
18:10:27<yvesr>http://blog.last.fm/2008/03/25/fingerprinting-and-metadata-progress-report
18:10:43<yvesr>D:| Fingerprinting lookup open-source software by last.fm
18:11:45<yvesr>D: Example result http://moustaki.org/resources/lastfm.xml
18:13:20<yvesr>D: The results are a bit disapointing, compared to a plain metadata lookup against musicbrainz, in most of the cases - I guess it is mainly due to their "quest for the ultimate canonical title", whereas they should emphasise having unique URIs for unique things (artists, tracks, albums, etc.)
18:56:54<danja__>ndw, you around?
18:57:49<danja__>phenny, tell ndw http://norman.walsh.name/robots.txt for http://norman.walsh.name/knows/who/#norman-walsh as found in timbl's foaf
18:57:49<phenny>danja__: I'll pass that on when ndw is around.
19:11:59<ndw>danja__, yes
19:11:59<phenny>ndw: 18:57Z <danja__> tell ndw http://norman.walsh.name/robots.txt for http://norman.walsh.name/knows/who/#norman-walsh as found in timbl's foaf
19:12:39<ndw>Hmmm
19:17:12<timbl>?
19:17:31<timbl>now a good uri for u ndw ?
19:18:08<ndw>uhm. well, actually knows/who/norman-walsh is better but there's supposed to be an owl sameAs
19:18:17<ndw>and I do need to publihs the updated robots.txt
19:19:34<danja__>ACTION prefers irc to twitter :-)
19:19:42<ndw>ok, new robots.txt pushed
19:19:47<ndw>robots, have at it!
19:20:37<danja__>ew, eclipse fell over
19:21:09<shellac>danja__: permgen full?
19:21:24<shellac>ACTION has been suffering that recently
19:22:31<danja__>dunno
19:23:57<danja__>ACTION not sure where permgen lives
19:24:13<shellac>sun java 6 and eclipse don't play well together
19:24:26<danja__>ah - I'm on 6
19:24:58<danja__>(not that I've read the manual since 1.3)
19:25:43<danja__>ACTION waiting for bot to get to ndw's foaf
19:26:02<ndw>ndw's foaf is /foaf.rdf
19:26:49<danja__>ok...
19:28:22<danja__>that'll have to wait until I update my own foaf then
19:32:22<danja__>ACTION runs bot on ndw's foaf
19:32:59<danja__>boo, think I must have broken something, it's not discovering any more triple
19:33:48<danja__>oh wait, there's some
19:34:02<danja__>so why didn't it see any rdf at http://purl.org/net/eric/webwho.xrdf
19:35:04<danja__>.seen em
19:35:04<phenny>Sorry, I haven't seen em around.
19:35:25<danja__>his foaf's served as text/plain
19:35:39<danja__>. http://www.w3.org/People/EM/webwho.xrdf
19:36:15<danja__>my bot is a bit narrow-minded at the moment...
19:40:47<timbl>phpht .. $ curl -I http://www.w3.org/People/EM/webwho.xrdf
19:40:47<timbl> text/plain
20:01:29<bblfish_>I heard the Jena team were doing some relational db to linked data mapping
20:01:36<bblfish_>an open source project
20:01:53<bblfish_>so that you can query relational DBs with SPARQL
20:10:03<timbl>really?
20:34:05<mahound>ls
20:34:07<TedThibodeauJr>phenny ask bblfish please post a pointer to Jena RDB-2-LOD rumor stuff?
20:34:14<mahound>(delete that)
20:34:15<TedThibodeauJr>phenny, ask bblfish please post a pointer to Jena RDB-2-LOD rumor stuff?
20:34:15<phenny>TedThibodeauJr: I'll pass that on when bblfish is around.
20:34:17<TedThibodeauJr>that.
21:58:46<tommorris>http://upcoming.yahoo.com/event/462878/
21:58:59<tommorris>E:| Portable London 0.96b - DataPortability Lunch Meet-up (upcoming.org / London)
21:59:07<tommorris>E: "Informal lunch for contributors of the DataPortability Project and those interested to learn more about it."
21:59:28<tommorris>E: Sunday April 6 2008 - 12.30pm - 3.00pm
22:00:56<tommorris>ACTION has committed a Scratchpad Driveby
22:02:44<tommorris>hey, iand! are you coming to http://upcoming.yahoo.com/event/462878/ ?
22:08:45<iand>hmmm tommorris, not sure
22:08:58<iand>need to check with family
22:26:22<KjetilK>http://animaldiversity.ummz.umich.edu/
22:26:36<KjetilK>F:|Animal Diversity Web
22:27:07<KjetilK>F: I just discovered it has OWL, which seems to make a good candidate for Linking Open Data
22:27:32<KjetilK>F: E.g. http://spire.umbc.edu/ontologies/taxa/Ochotona_curzoniae.owl
22:28:06<KjetilK>(I was writing an article about that little guy, which I was fortunate to bump into in Tibet :-) )
22:31:15<KjetilK>ACTION looks at the host names and realises they might be two different efforts, but interesting nevertheless

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