Re: [lojban] Fu'ivla of class MLONGENA

["Jorge Llambias" <jjllambias@hotmail.com>]

la djan cusku di'e

> Brivla ending in CVCV are almost certainly legal.  "y" is not permitted
> in fu'ivla in any case, and the fifth-letter restriction prevents collision
> with the type-3 algorithm.  So this looks good.

Actually, the restriction on "r" is either insufficient or
unnecessary. If "r" is banned, "n" should be banned too, because
it plays the same role as "r" for gismu whose fourth letter is "r".
At lest "rn" in fourth-fifth place should be banned.
On the other hand, there is no need to ban either, because the
sixth letter is a vowel so no conflict is ever possible with
type-3 fu'ivla that must have a consonant there.

> An early version of the
> current type-3 fu'ivla algorithm proposed a CVCV pseudo-rafsi for each
> gismu, so that instead of kulnrblgaria (Bulgarian culture) we would have
> blgariakulu or the like.

Sounds nice, but it needs additional constraints, no?
If -kani was the pseudo-rafsi for {klani} , I couldn't
use {gauskani} for Gauss, becuase it breaks into {gau skani}.

> This scheme had the advantage of putting the more general term at the end,
> analogous to tanru order. It was rejected because making people learn
> a whole new set of pseudo-rafsi seemed like a Bad Thing.

It seems to me that the restrictions necessary at the start of
such words would also be quite complicated.

co'o mi'e xorxes


________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com