Difference between revisions of "BPFK To-Do"

From Lojban
Jump to: navigation, search
(Grammar points to be decided upon)
m (Minor formatting)
Line 35: Line 35:
 
==Predicate interactions issues==
 
==Predicate interactions issues==
 
Includes sumtcita (since they are compressed predicates)
 
Includes sumtcita (since they are compressed predicates)
* Formalize fi'o and BAI. Currently they are all broken and terrible. Some BAI might need to be killed or heavily redefined to make sense / be useful. for fi'o there are two options... there are infinitely many predicates, so it's simply not possible to define interaction of fi'o with every predicate seperately. therefore i propose to use a single formula that works for all fi'o clauses. BAI can be considered special in that they are more precisely defined and don't need this general solution. (bau = fi'o se baupli). this means that the fi'o transformation of tags is but an approximation, not a full equivalence. co'e only word that is general enough==Mekso issues==
+
* Formalize fi'o and BAI. Currently they are all broken and terrible. Some BAI might need to be killed or heavily redefined to make sense / be useful. for fi'o there are two options... there are infinitely many predicates, so it's simply not possible to define interaction of fi'o with every predicate seperately. therefore i propose to use a single formula that works for all fi'o clauses. BAI can be considered special in that they are more precisely defined and don't need this general solution. (bau = fi'o se baupli). this means that the fi'o transformation of tags is but an approximation, not a full equivalence. co'e only word that is general enough
 +
==Mekso issues==
 
* Improve the BPFK definition for {ji'i} (see [https://groups.google.com/d/msg/lojban/XXO1L7zKGQE/VpgQ9JUfHBMJ this related discussion])
 
* Improve the BPFK definition for {ji'i} (see [https://groups.google.com/d/msg/lojban/XXO1L7zKGQE/VpgQ9JUfHBMJ this related discussion])
 
* Deciding upon the meaning of {ni}, and how to express amounts (zenba2, barda2, cnano2…)
 
* Deciding upon the meaning of {ni}, and how to express amounts (zenba2, barda2, cnano2…)

Revision as of 09:38, 3 September 2015

This is basically a free-form list of things the BPFK will eventually have to deal with, sooner or later. Feel free to add to it.

Meta-issues

  • la gleki:
    • This page is a mess.
    • Include discussions from other BPFK pages.

Grammar points to be decided upon

  • {ti citka be mi}
  • {lo nu brode ba brodo}
  • {fi do vecnu ti}
  • {jai frili fai ma}
  • {mi broda fa do}

Dictionary issues

  • Go through the gismu list and improve wordings as well as increase consistency between gismu where possible. Go through gismu errata list.
  • Lojban definitions of every cmavo
  • Consider experimental gismu to be made official (e.g. kanpe, kibro, etc)
  • Consider proposed cmavo and see if they warrant an official form

Type system issues

  • Type discussion. mono- vs poly-morphism
  • Define underdefined gismu places, e.g. skari2

Place structure issues

  • Formalization of {kau} in complete propositions (du'u, nu, main bridi…) and properties and amounts abstractions (ka, ni)
  • "at location" places in many gismu seem totally irrelevant to the relation. e.g. ckule, cange, malsi ... Should they be removed?
  • Decide how to fill "by standard" places

Predicate interactions issues

Includes sumtcita (since they are compressed predicates)

  • Formalize fi'o and BAI. Currently they are all broken and terrible. Some BAI might need to be killed or heavily redefined to make sense / be useful. for fi'o there are two options... there are infinitely many predicates, so it's simply not possible to define interaction of fi'o with every predicate seperately. therefore i propose to use a single formula that works for all fi'o clauses. BAI can be considered special in that they are more precisely defined and don't need this general solution. (bau = fi'o se baupli). this means that the fi'o transformation of tags is but an approximation, not a full equivalence. co'e only word that is general enough

Mekso issues

  • Improve the BPFK definition for {ji'i} (see this related discussion)
  • Deciding upon the meaning of {ni}, and how to express amounts (zenba2, barda2, cnano2…)

Unsorted issues

  • Formalize the tense system (?)
  • CAhA as sumtcita
  • scope weirdness in ZAhO definitions ("broda ba'o lo nu brode ~= ba'o broda ca lo nu brode"??)
  • subjunctivity formalization
  • position ne inner rel is outer with LE, but inner (part of name) with LA. is this desirable? (what if LE-LA merge goes through, still desirable?)
  • scope of NA
  • na'e ko'a
  • make the PEG the official grammar.
  • add uy and iy as BY
  • add a new FAhA that means "along a path". I suggest {pu'a} from {pluta} (ditch the {pluka} BAI). {mi cadzu pu'a lo rirxe}. {pu'a} has almost no usage (8 times and only on IRC)
  • discuss the jvojva
  • Have a discussion about the status of dialects.
  • Discuss logical connection of sentences with different illocutionay force
xorxes This reminds me that, if I recall correctly, we had different views when working on Xorban on whether it makes sense for two sentences with different illocutionary force to be logically connected. For me logical connection only applies to bare propositions, and not to propositions-in-use.
And My palpably deteriorating memory isn't pinging, so even if you recall correctly, I still don't. I do think that illocutionary operators can themselves be arguments of predicates (e.g. "Is it dinner time yet, for I'm hungry") tho, and since I take logical connectives to be predicates it follows that in principle I must allow illocutionary operators to be logically connected. I'm not sure how this bears on the current discussion, tho.
xorxes Yes, that was a good example. My view was that the underlying logic for that is:
  1. I (hereby) ask whether it is dinner time yet.
  2. The reason for my asking whether it is dinner time yet is that I am hungry.

1 and 2 are different propositions, but 2 happens to contain 1. Since both illocutionary acts make use of the same proposition, it is convenient to utter the sentence expressing that proposition only once, but that doesn't make the first illocutionary act (as opposed to just its propositional content) an argument of the predicate used in the second act.


Review change proposals