[00:28:41] *** Joins: koo5 (~koom@236.152.broadband3.iol.cz) [00:39:09] *** Quits: koo5 (~koom@236.152.broadband3.iol.cz) (Ping timeout: 240 seconds) [05:13:28] *** Joins: JuanDaugherty (~jdaughert@98.4.124.117) [09:56:02] ayberkt: if it's necessary, then yes [09:56:44] CommonX is there just because it seems to be a common pattern, that SCs or Advs or whatnot tend to be just strings, but if any language needs more, then just override [10:55:28] *** Quits: JuanDaugherty (~jdaughert@98.4.124.117) (Quit: Ex Chat) [10:55:54] *** Joins: JuanDaugherty (~jdaughert@98.4.124.117) [11:30:37] Oh, there's documentation to the new C runtime now. [11:30:41] http://www.grammaticalframework.org/doc/runtime-api.html [11:35:55] I wonder why SC - embedded sentence or question, would be just strings. [11:36:26] mkSC takes S, QS and VP. [12:02:23] *** Joins: proteus-guy (~proteusgu@180.183.140.175) [12:20:11] *** Quits: JuanDaugherty (~jdaughert@98.4.124.117) (Remote host closed the connection) [14:19:50] *** Joins: JuanDaugherty (~jdaughert@98.4.124.117) [15:22:19] *** Joins: proteusguy (~proteus-g@180.183.140.175) [15:24:43] drbean: it's not really that it "should", I think it was just that someone noticed after writing a couple or ten languages in the RGL that oh this seems to be {s:Str} in all of them [15:25:03] so maybe it's just done all agreeing in standard average european [15:25:17] like "[that Mary sleeps] is good" [15:25:53] but if you need in some other language different form of the verb depending on if it's Mary or John who sleeps, or singular or plural, or whatever, then you need to override the SC lincat [15:25:58] and add e.g. an Agr field [15:26:31] or if it would inflect differently in "[that Mary sleeps] isn't good" [15:26:38] then it needs some table type in the s [15:36:05] *** Quits: proteusguy (~proteus-g@180.183.140.175) (Read error: Connection reset by peer) [15:37:31] *** Joins: proteusguy (~proteus-g@180.183.140.175) [15:54:23] *** Quits: proteusguy (~proteus-g@180.183.140.175) (Remote host closed the connection) [15:57:38] *** Quits: proteus-guy (~proteusgu@180.183.140.175) (Ping timeout: 240 seconds) [16:38:44] OK, I thought there would be problems assimilating [16:38:46] t [16:41:32] "that Mary sleeps" to NP, so that mkCl could overload like mkCl : NP -> VP -> Cl [16:43:08] But I see that there is a mkCl : SC -> VP -> Cl = PredSCVP [16:56:18] yeah :) [16:57:28] unrelated: if I choose Estonian-English in our great wide-coverage demo http://cloud.grammaticalframework.org/wc.html and type "ja", which is the conjunction "and" in estonian, I only get the translation "crap" [16:58:17] translation number 3: "Crap?" translation number 5 "? crap" translation number 7 "Crap? ?" [16:58:26] and translation number 9, "? crap?" [16:59:11] at least the GF translator seems to be self-aware [16:59:28] Someone is undoing your good work? [17:04:47] *** Joins: proteusguy (~proteus-g@2405:9800:bc10:1ca:90bf:67a9:8bc6:7d3d) [23:03:18] *** Quits: JuanDaugherty (~jdaughert@98.4.124.117) (Quit: Ex Chat) [23:05:59] drbean: heheh it wasn't very good to start with