[03:45:17] *** Quits: drbean (~drbean@124.219.83.127) (Quit: ZNC - http://znc.in) [03:47:38] *** Joins: drbean (~drbean@124.219.83.128) [04:00:00] *** Quits: drbean (~drbean@124.219.83.128) (Ping timeout: 264 seconds) [04:26:47] *** Joins: drbean (~drbean@124.219.83.197) [04:36:38] *** Quits: drbean (~drbean@124.219.83.197) (Ping timeout: 256 seconds) [05:13:17] *** Joins: drbean (~drbean@124.219.82.80) [05:14:08] *** Quits: drbean (~drbean@124.219.82.80) (Excess Flood) [05:15:31] *** Joins: drbean (~drbean@124.219.82.46) [12:42:32] *** Joins: BobL (82f1310b@gateway/web/freenode/ip.130.241.49.11) [13:05:41] *** Joins: Toni_ (82f1d185@gateway/web/freenode/ip.130.241.209.133) [13:06:01] *** Joins: liina (82f1310d@gateway/web/freenode/ip.130.241.49.13) [13:07:59] ^__^ [13:09:27] git clone https://github.com/GrammaticalFramework/gf-contrib.git [13:10:37] *** Joins: iwk (~user@flv49010.hum.gu.se) [13:13:15] git clone https://github.com/GrammaticalFramework/gf-contrib.git [13:14:23] *** Joins: LukeG (82f1310e@gateway/web/freenode/ip.130.241.49.14) [14:08:33] *** Quits: spectre (~fran@115.114.202.84.customer.cdi.no) (Ping timeout: 276 seconds) [14:17:01] *** Quits: Toni_ (82f1d185@gateway/web/freenode/ip.130.241.209.133) (Ping timeout: 246 seconds) [14:21:32] do you have your projects already? [14:21:40] I mean the final project of the course [14:22:58] Aarne has given 3 suggestions [14:23:23] Extended miniresource grammar, translation dictionary improvement or query language stuff [14:23:38] nice [14:24:00] so the translation dict improvement is something that is immediately useful [14:24:05] less funny/surreal translations [14:24:18] but the rest are more interesting to make [14:24:20] hehe, yeah, I noticed the need for that [14:24:25] :-P [14:24:25] but yes, probably [14:24:50] and anyway, just pick what you find nicest to do :) [14:24:54] *** Joins: crazydiamond (~crazydiam@178.141.72.205) [14:25:30] :) oh, and I just remembered I haven't yet handed in the dict improvements I've made over the last week [14:25:33] haha yeah [14:25:35] nobody gas [14:25:36] *has [14:25:46] ok, then at least I'm not alone :P [14:25:55] but it's ok, it's not really required, just for your own benefit [14:26:05] ok. would be nice to contribute tho [14:26:07] yeah [14:26:24] where do I find the explanations for the mysterious comments in the dicts? [14:26:29] hahah [14:26:30] like --comment=4 [14:26:41] that's from swedish? [14:26:44] I don't actually know [14:26:45] yep [14:26:48] oh, ok [14:26:50] some languages have the source [14:26:55] like wiktionary [14:27:05] and some just have --POSSIBLYERRORBAD [14:27:09] or something along the lines [14:27:11] :D [14:27:13] maybe it's something from saldo? [14:27:18] the --comment=4 [14:27:24] I think that's where most of the swe lexicon comes [14:27:29] perhaps. But the sources are clearly annotated elsewhere [14:27:32] like source=saldo [14:27:34] ah ok [14:27:41] you can ask aarne then [14:27:44] ok [14:28:57] *** Quits: iwk (~user@flv49010.hum.gu.se) (Remote host closed the connection) [14:46:46] *** Quits: liina (82f1310d@gateway/web/freenode/ip.130.241.49.13) (Ping timeout: 246 seconds) [14:50:19] *** Joins: spectre (~fran@dhcp658-ans.wifi.uit.no) [14:53:59] *** Quits: BobL (82f1310b@gateway/web/freenode/ip.130.241.49.11) (Quit: Page closed) [15:21:13] *** Quits: jmvanel (~jmvanel@jua06-2-88-163-170-224.fbx.proxad.net) (Quit: Quitte) [16:09:24] *** Quits: spectre (~fran@dhcp658-ans.wifi.uit.no) (Ping timeout: 245 seconds) [17:16:22] *** Joins: spectre (~fran@115.114.202.84.customer.cdi.no) [19:48:23] *** Quits: LukeG (82f1310e@gateway/web/freenode/ip.130.241.49.14) (Quit: Page closed) [20:59:34] *** Joins: BobL (02449a83@gateway/web/freenode/ip.2.68.154.131) [22:18:14] *** Quits: vin-ivar (~vinit@122.179.143.17) (Read error: No route to host) [22:27:11] *** Joins: jmvanel (~jmvanel@157.0.88.79.rev.sfr.net) [22:44:07] Hello nice experts in a cool Spring evening! A novice question: Would it be right to say that in the "lincat" segment of the Concrete Syntax file, we should only use certain types to declare categories, the types being only "String", "Record", and "Table"? [22:50:21] hmm I'm not sure if there's a syntax restriction from using a param also, but yeah those make sense [22:54:04] Thanks Inari. Would it be right to say "if I use a 'param' for defining categories in 'lincat' then it should end as, for example, a 'Str'? [22:54:37] Kind = Number => Str [22:54:52] Number (coming from 'param') [22:55:24] wait a sec, I'll get to this later [22:55:26] 10mins or so [22:55:33] Thanks! [23:02:51] strings are not params [23:02:55] params are finite [23:03:26] I mean, you can define e.g. param Colour = Blue | Red | Green [23:03:36] and those are the only 3 possible things of type Colour [23:03:49] whereas for a Str you can have all possible strings in the world [23:03:55] your name or the wikipedia dump [23:04:10] or the wikipedia dump + one more space after that [23:05:54] also, for instance, in tables, the keys can only be params [23:06:03] so you can have a table Colour => Str [23:06:07] but not Str => Colour [23:06:10] or Str => Str [23:06:48] so, simple strings and tables have the thing in common that there's something that will end up being a string eventually [23:07:23] for english adjective you need just a Str, for french adjective you need Gender => Number => Str [23:07:39] and both of those, strings and tables, can be present in records [23:08:13] (ok also records can be inside records, but if we're just keeping things simple, you don't see them so often) [23:08:13] Wow very interesting. [23:08:49] I see [23:09:09] so, for the purposes of having a lincat [23:09:24] you'll want to have something that will eventually be a string [23:09:41] in practice, all categories in the GF resource grammar library are records [23:09:52] even if just {s : Str} [23:10:33] then there's things like {s : SomeParam => Str ; anotherUsefulField : SomeOtherParam } [23:11:51] I see [23:12:06] (I'm just writing a test grammar) [23:12:32] Thank you, very cool [23:16:06] hahah ok it compiles but I can't form sentences [23:16:14] hehe should've made it in the cloud for handy sharing [23:16:29] :) [23:17:47] Thanks [23:18:03] ok let's see [23:18:08] I can publish it if you're interested [23:18:15] anyway did I answer your question? :-D [23:18:22] if not, just ask again, I just got carried away [23:18:48] Yes please, that would be very nice to have it. Yes you fully answered my whole question. [23:19:10] cool :) [23:19:17] It's up to me to practice :) [23:23:37] ok, look for TotallyUselessGrammar http://cloud.grammaticalframework.org/gfse/ [23:23:52] the point is just that the lincats for Cat1 and Cat2 can be param types [23:24:16] but it's not very useful, and the strings are just hardcoded in the MkS1 and MkS2 functions [23:25:30] if you want to see some point there, it's that even though Cat1 and Cat2 don't have any textual form, the choice of the argument affects which MkS is used [23:25:50] but it could be done much more nicely if you just have a single MkS function, and in it you pattern match on the argument [23:29:06] Great! Now I take some time to understand your description and the grammar... [23:29:35] heheh, you don't need to take that grammar as an example of anything nice [23:29:59] just that "ok it can be done, a param type can also be in a lincat" [23:30:32] Yes, I understand... it is very interesting the way you describe it. [23:33:02] I added a new concrete syntax, go and look for a newer TotallyUselessGrammar [23:33:14] okay [23:33:25] now I only have MkS2 left, and it always takes Cat2 [23:33:39] also, there is Fun2 which is "natively" a Cat2 [23:33:54] and there's a Fun3 which makes Cat1 into Cat2 [23:33:57] sorry my imaginative names [23:34:16] and at this point, a value of type Cat2 can be either Hargle or Bargle [23:38:16] very cool... thanks Inari, I shall take the time to enjoy the basics now... it is very useful the way I have this grammar to look at now. I will ask more questions when I am a bit more prepared :) [23:38:30] ok! [23:38:36] thanks!! [23:38:42] no problem :)