Channel #semsol: Logs

This is a public chat log generated from the #semsol IRC channel.

11:09:45 enzo: hi
11:12:19 enzo: i was wondering, does arc uses redland ?
11:13:00 bengee: no, arc is a stand-alone, pure-php implementation
11:13:20 bengee: (well, +mysql)
11:13:58 enzo: by the way bengee, which tool do you use to create a rdf/owl file ?
11:14:45 bengee: I usually auto-generate them from instance data
11:15:12 enzo: what does it mean by instance data ?
11:15:46 enzo: the fact is I'd like to create a thesaurus, but i don't know where to begin: create it with Protégé, then uses arc to create database in order to query it
11:15:54 bengee: I wouldn't start building an app by creating the model/ontology
11:15:58 enzo: or another method...
11:16:17 enzo: but for a website, it's best to create the thesaurus first, no ?
11:16:39 bengee: worth a read: http://www.betaversion.org/~stefano/linotype/news/93/
11:16:59 enzo: ok i read it now
11:17:29 enzo: ah it's a general paper
11:17:49 enzo: well i need to do my structure first
11:18:10 enzo: you don't do the same thing bengee?
11:20:04 bengee: if you need the thesaurus first, then using some thesaurus editor makes sense
11:20:15 enzo: you know some bengee?
11:20:38 bengee: only he ones I mentioned
11:20:40 enzo: I've seen ThManager, but it's deprecated, and Protégé, good, but complex
11:21:26 bengee: maybe I should build one..
11:21:30 enzo: ok, so if i create a good owl with protégé, i'll be able to load it in arc in order to construct a database automatically in order to use sparql or query the base manually ?
11:22:10 bengee: yes, you can load owl/rdf into arc and then sparql the ata
11:22:18 bengee: s/ata/data/
11:22:44 enzo: sparql use mysql as a backend ?
11:22:52 bengee: arc does
11:23:28 enzo: ok, and you know how to use a powerful websearch engine on your website that benefits from this thesaurus ?
11:23:40 enzo: for example, if i search dad, it will search daddy, etc...
11:23:45 bengee: no
11:24:20 bengee: people on #swig might know one
11:24:35 enzo: ok i ask, maybe lucene, solR do this
11:38:44 enzo: bengee: do you if this is possible, with arc, to have a thesaurus in a database (given by arc), in order for arc to create the rdf file on the fly ?
11:39:54 bengee: you mean, if ARC can export your thesaurus as RDF again? yes
11:40:20 enzo: and arc gives the structure to use in mysql ?
11:40:38 enzo: the fact i find protégé very complex to create a simple thing as a thesorus
13:27:56 enzo: after googling a lot, i've seen that Jena and sesame are good candidate to manage RDF, someone has an idea of benchmark between arc and those tools ?
13:29:49 bengee: they are faster and more scalable than arc
13:40:04 enzo: ok
13:40:42 enzo: i've read in arc, that it's more data first than structure first, ok, but how do you add rdf data easily bengee ?
13:41:00 bengee: via sparql
13:41:07 enzo: i've seen Sesame 2 Windows Client, I don't know if it works with last version of sesame, but it's quite convenient
13:41:30 enzo: so you do a php loop to do the equivalent of a sql INSERT query, that's it ?
15:40:29 enzo: bengee: you know if the developers are very active on arc ?