[04:16:02] *** Joins: drbean (~drbean@124.219.83.42) [05:51:44] *** Quits: drbean (~drbean@124.219.83.42) (Ping timeout: 260 seconds) [05:52:27] *** Joins: drbean (~drbean@124.219.83.42) [06:11:03] ResEng's infVP : VVType -> VP -> Anteriority -> CPolarity -> Agr -> Str is written to parse "pretend not to notice" but mkVP overloads ExtraEng's ComplVV, which assumes CPos. However, ExtensionsEng has ComplVV : VV -> Ant -> Pol -> VP -> VP; which also uses infVP [06:11:05] It does the right thing. [06:59:36] *** Quits: Flammie (~flammie@sandbox.multilingua.uni-hamburg.de) (Ping timeout: 244 seconds) [10:24:15] drbean: aha, so the CPolarity and Anteriority are parameters, and Ant and Pol are abstract syntax categories? [10:25:34] if it's in ResEng and its type includes params or opers, rather than abstract syntax types, it doesn't sound like it should be visible to users [10:28:10] if there is the same function (ComplVV) in ExtraEng and ExtensionsEng, that sounds just like different developments; Extra was originally a RGL thing, just a place where a resource grammarian could put constructions that are not in the core RGL shared by all [10:28:24] Extensions is specifically meant for the wide-coverage translator [10:29:07] if Extra was unified for all languages (at which point it should stop being called Extra, and just distribute the functions to other modules), Extensions wouldn't have to define its own ComplVV [10:29:26] but for some languages that don't have a ComplVV in Extra, it's only defined in Extensions [10:29:53] so if they're different for english, maybe someone just didn't realise there's one already in extra [10:47:31] *** Joins: daherb (~daherb@cse-principia.cse.chalmers.se) [13:10:46] *** Quits: drbean (~drbean@124.219.83.42) (Ping timeout: 250 seconds) [13:16:28] *** Joins: drbean (~drbean@124.219.83.41)