[YG Conlang Archives] > [engelang group] > messages [Date Index] [Thread Index] >
Jorge Llamb�as, On 30/08/2012 03:16:
On Wed, Aug 29, 2012 at 11:05 PM, And Rosta<and.rosta@hidden.email> wrote:Jorge Llamb�as, On 30/08/2012 02:34:Because so far only V('V)* are variables.I'd assumed V(V)* are variables, and we're writing<'> only because writing stuff in an obviously Lojbanish way stops us getting sidetracked by phonology discussion. I was further assuming that Lojban /'/ would not exist in Xorban, but that in V'Vthe<'> is intended as Lojban /'/ rather than<'>=<q>=[?].Well, I'm just trying to keep it simple for now. Formalizing the rules for the proposed restrictions on vowel sequences would take some work, and the result will end up looking quite messy, whereas V('V)* is straightforfard.
OK. I think my proposed restrictions on V sequences were pretty simple, but we don't need to deal with phonology for the time being. But anyway, in proposing "oi" I was assuming that it belonged to the syntactic category of variables.
See: http://xorban.wordpress.com/grammar/A complement is that which, when added to a formula, yields a formula? I can see why the term was chosen, but it's also confusing, since you'd expect it to mean something closer to its ordinary syntactic meaning -- e.g. binary operator has two complements, each of which is a formula. Given my background in syntax, I tend to think of syntactic structure in terms of combinatorial properties of words, so the "formula:= complement formula" rule seems weird -- still, it does the job of specifying the combinatorics.Any suggestions for better names for the rules? We've been constantly changing the names, so nothing is settled.
I would approach the rules differently, by defining the combinatorial properties of words, so the thing currently called "complement" wouldn't arise. I would normally have said a unary operator has one complement and a binary operator has two complements, but one could call them "dependents" instead, to avoid confusion. Basically, I'm influenced by the way natlangs work, on the grounds that ideally the loglang would be usable as if it were a natlang. But as with phonology, the syntactic description can come later. --And.