Message-Id: <9004201515.AA11672@mwunix.mitre.org>
Date: Friday, 20 Apr 1990 11:16:16 EST
From: mwvm.mitre.org!m16569 (Carl Burke)
To: snark.uu.net!lojban-list
Subject: Cmene for lojban lerfu; another approach
Status: RO

The use of the lerfu cmavo addresses the problem in part, but does not
address the entire problem, which arose largely from the need for clear
communications in a noisy environment.  Simply using the lerfu cmavo does not
provide sufficient redundancy to distinguish between, for example, 'by' and
'py'.
  One possible solution is to use the 'standard' phonetic alphabet as a base,
but lojbanize the words as lojban cmene.  For example, 'ALPHA' would become
'.elfas.' or '.alfas.'; the terminating consonant is arbitrary.  If this
alphabet is indeed an international standard, the approach seems reasonable.
  This approach has the advantage of utilizing cmene-space, which is much
larger than the available cmavo space, and does not require any grammar
changes.  The primary restriction is the need to place these cmene into either
'LA' descriptors or vocatives ('coi papas tcarlis' to greet pc); I'm not sure
about the permissable separation between components of cmene so you may need
to express each component letter as a separate sumti (la papas. la tcarlis.).

  Another problem is the presence of letters in lojban which do not have
English correlates.  These would require additional new cmene.

  Another possible solution, suggested by Bob, is to select extant lojban
bridi which are heavy in the sound desired and use those; the disadvantage is
in the deviation from existing standards, which should not be a problem IF you
are talking to another lojban speaker AND the phonetic cmene are self-
explanatory.

  The point is, there currently is no standard set of lerfu cmene
correspondingto the 'ALPHA, BRAVO, etc' for English; at least two good
mechanisms exist for implementing this, and anyone who is interested should
try thier hand at it and submit it for the community's approval/discussion.

  ******************** Now for another topic entirely ************************

  In response to Bob's E-Mail note prior to his machine's unfortunate surgery,
I have begun translating the machine grammar into a more accessible form of
BNF.  This process also includes a scrub of the grammar, to try to ensure that
what the grammar ACTUALLY expresses is what LLG WANTS the grammar to express.
I expect to have this project completed prior to LogFest; hopefully with a
decent draft by the end of April.  I hope someone else is also considering
thistask, as the more [intelligent and well-reasoned] commentary generated,
the cleaner and clearer the result will be.

  If anyone is interested in viewing the draft or samples thereof, please send
me a message (assuming you can get my E-mail address from the message header;
I've never memorized le malselcusku [le skami notci judri?]).

                                                Carl Burke
*
*        Carl