[01:33:01] *** Joins: jmvanel (~jmvanel@117.6.163.56) [02:59:03] *** Quits: jmvanel (~jmvanel@117.6.163.56) (Read error: Connection reset by peer) [05:55:39] *** Joins: vin-ivar (uid74719@gateway/web/irccloud.com/x-ecgnaftqkyevfzta) [08:07:46] *** Quits: vin-ivar (uid74719@gateway/web/irccloud.com/x-ecgnaftqkyevfzta) (Quit: Connection closed for inactivity) [08:31:40] *** Joins: vin-ivar (~vinit@182.70.2.168) [11:52:11] *** Joins: jmvanel (~jmvanel@117.6.163.56) [12:20:57] inariksit, ping me when you have a minute [12:48:49] *** Joins: spectre (~fran@115.114.202.84.customer.cdi.no) [13:11:24] vin-ivar: now! [13:11:30] just going to get a cup of tea [14:46:20] whoops, sorry, I was driving [14:52:09] no problem :) I'm available now [14:53:00] cool [14:53:01] so [14:53:16] I've managed to write stuff that builds danish nouns and adjectives [14:53:28] (assuming worst case, noun and gender for mkN and five forms for mkA) [14:54:38] well, I have the script, anyway. running the nouns bit on the expanded monodix kills my laptop [14:56:15] http://paste2.org/gGkdkfKV <- nouns [14:56:34] http://paste2.org/PZEpaILc <- adjectives [14:56:38] does this look okay? [14:57:16] yep! ^^ [14:57:27] looks nice [15:07:35] yay [15:07:40] I was looking at aarne's lrec paper [15:07:50] on resource sharing between rbmt systems [15:09:49] i think using closely related languages could be pretty handy for building monodixes/concrete things in either framework [15:10:40] like here, I didn't have a source for english -> danish, and all the functions were english words, so I could convert the swedish thing to danish [15:11:01] there's likely to also be fewer inconsistencies with separable verbs and stuff [15:11:51] what do you reckon? [15:21:37] yeah, for monodixes definitely makes more sense than via english [15:39:49] btw, I just had a look at your latest (well, 10 days :-P) code in apertium-gf [15:40:24] sorry I've been pretty inactive with that! anyway it is a relevant thing and would be cool to continue [15:40:44] did you plan to have conversion between GF and apertium formats? [15:40:52] well tagsets [15:42:30] MorphDisambiguator doesn't print anything for me, btw [15:43:16] shit, I should update the repository with the scripts :s [15:43:26] pipe the MorphAnalyser output to it [15:44:23] ah ok [15:44:29] I'm thinking of building a sort of system that can take advantage of stuff that's in Apertium (shooting it into GF) or vice versa [15:45:21] yeah ok [15:45:44] if it's just lexicon building, many of the tags in apertium don't appear in the dictionary format [15:46:17] but if we'd like to have a conversion between an actual sentence, then it'd need all these apertium tags [15:46:50] actually one application for that could be what I'll start working on, inducing CG rules from GF grammars [15:49:26] yeah, that's what i thought as well [15:49:34] some sort of mapping [15:53:22] yeah [15:59:46] trying to visualise what it ought to look like [16:03:21] i'm also thinking of implementing/sharing some sort of lexical selection [21:11:22] inariksit: i've put the adjectives script on the repository [21:16:23] wondering what to work on next, probably tag translation [21:18:11] yeah, that would be nice [21:18:27] do you have any preference what language to start with? [21:18:40] if not, may I suggest spanish :-D because that would benefit my CG stuff [21:29:22] i'd prefer romance/germanic [21:29:30] so spanish works! [21:30:05] what exactly are you working on, with CG?