Experimental IRC log swig-2008-01-24

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.

02:25:42<nwalsh>This morning I reported some problems I was having with cwm
02:25:49<nwalsh>DanC suggested that @forall would help
02:26:01<nwalsh>Anyone got a pointer to documentation for @forall
02:26:08<nwalsh>Several random attempts haven't worked
02:27:47<nwalsh>Ahhh...
02:27:57<nwalsh>Because it's spelled "@forAll"
08:20:53<mahound>hello
08:26:04<drewp>hi mahound
09:22:12<danieljohnlewis>Good Morning swiggers
10:16:03<Wikier>hi
10:21:44<danieljohnlewis>hi Wikier
14:21:00<edsu>cygri: ping
14:21:22<cygri>edsu: pong
14:22:47<edsu>cygri: you are richard cyganiak right?
14:23:19<edsu>i imagine phenny can tell me or something too :)
14:23:46<cygri>yes that's me ... not sure if phenny knows ;-)
14:23:48<edsu>cygri: edsu =~ ed summers, lurking on dbpedia list, following discussion of skos:subject with interest
14:24:48<edsu>cygri: i was the one asking if dc:subject had been dismissed instead of using skos:subject
14:25:24<edsu>umm, i mean, if skos:subject had been prefered over dc:subject
14:27:43<cygri>i was originally in favor of skos:subject, but am now reconsidering after a closer reading of the SKOS Core guide
14:29:01<edsu>cygri: i feel a bit guilty, because i mentioned how skos:subject has been intentionally left out of the latest editors draft for the skos reference
14:29:22<edsu>cygri: but i'm glad you cc'ed the skos lists, so that people are aware of the usage in dbpedia
14:29:31<edsu>cygri: since it's still kind of up in the air
14:29:40<edsu>cygri: from what i can tell
14:32:49<masaka>the idea to relate any resource and concept is ok, but that relation doesn't seem 'subject'
14:35:13<cygri>masaka, i tend to agree now
14:35:15<edsu>masaka: yeah, i'm kind of biased coming from the library-world, where resources are typically bibliographic objects
14:35:45<edsu>masaka: can you express why it doesn't seem that way?
14:36:59<masaka>ah, i'm not native English user, but 'subject' or 'topic' is not applicable to people or city etc.
14:37:23<masaka>i guess, it's simply dc:relation
14:40:24<cygri>masaka: i think dc:relation is too weak, and dc:subject suffers from the same problem as skos:subject
14:41:24<cygri>edsu: it's good that we have the issue out in the open now ... it's not such a big deal for dbpedia actually, because we can happily create our own property dbpedia:category or so, we do that all the time
14:41:40<masaka>yes, so you might define a new property as subproperty of dc:relation
14:42:10<edsu>cygri: i wonder, would you make it a subproperty of dc:subject like skos:subject?
14:42:49<masaka>i think it should not be subproperty of dc:subject
14:43:25<edsu>which is how skos:subject was set up in the first place :)
14:44:59<cygri>edsu: i think it could be a super-property of skos:subject
14:45:31<cygri>masaka: regarding your point that skos:subject is not necessary in SKOS ...
14:45:33<edsu>cygri: that would be an odd switch :)
14:46:29<masaka>cygri: you mean twitter message ?
14:46:32<cygri>... i think that vocabularies should be well-rounded packages that cover all basic needs of a specific application, and i think that a bit of duplication with other vocabularies is not a problem ... therefore, i think skos:subject should stay in SKOS
14:46:47<cygri>masaka, yes
14:47:09<edsu>cygri: agreed
14:47:41<masaka>cygri: yes it could be a part of SKOS, but not essential, i thought
14:48:05<masaka>because, we need other terms to describe a resource anyway
15:01:42<john-l>What sorts of resource descriptions are we trying to make (that do not fit the semantics of `skos:subject`)?
15:04:12<masaka>an example discussed in dbpedia-ml is how to relate dbpedia:Tim_Berners-Lee and dbpedia:Category:Living_people
15:04:34<john-l>masaka: Thanks, I was trying to find the original example.
15:05:07<john-l>That wouldn't be rdf:type?
15:06:58<masaka>it could be, but it raises another problem to use wikipedia category as a Class
15:08:41<cygri_>john-l, the DBpedia categories are derived from the wikipedia category system, and they don't work as classes for various reasons (e.g. cycles in the hierarchy)
15:09:11<john-l>cygri_: Ah, thank you.
15:09:33<john-l>ACTION remembers reading about cycles in the WP hierarchy somewhere in the WP docs
15:10:29<cygri_>it's quite funny, if you start at Berlin and go up the hierarchy, you end up in categories such as “Jesus”, “Asia” and “Mexico”
15:12:05<jaa>Jesus is a city in Montreal
15:16:30<john-l>Yeah, WP categorization could definitely use a proprietary property. :)
15:18:16<john-l>*Some* of the WP categories seem like classes; it would be useful if WP could make some sort of distinction.
15:19:11<john-l>ACTION imagines URIs like .../Category:Guilds and .../Class:Cities_in_Germany
15:19:27<masaka>sorry john-l, what is WP ?
15:19:36<john-l>Apologies; I mean Wikipedia.
15:19:54<masaka>i see, thank you :)
18:15:34<DanC>ACTION waves
18:15:50<iand>ACTION waves
18:16:02<HarryH>ACTION waves
18:19:03<DanC>ACTION tends to a few other things for a sec...
18:19:09<john-l>HarryH: Good afternoon.
18:19:17<john-l>HarryH: Did you still need DocBook help?
19:01:56<duck1123>Has anyone noticed that Triplr will screw up conversion of large RDF/XML files to turtle?
23:22:52<yvesr>http://www.bbc.co.uk/ontologies/programmes/
23:23:12<yvesr>A: An ontology for brands/series/episodes/broadcasts/etc.
23:26:17<yvesr>A: Quite a lot of linked data to come, from the BBC Programmes team

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