Date:         Wed, 23 Sep 1998 09:17:37 -0600
Reply-To:     kingcats@earthlink.net
Sender:       Lojban list <LOJBAN@CUVMB.BITNET>
From:         rapplean <kingcats@EARTHLINK.NET>
Subject:      Re: Word frequency
To:           Multiple recipients of list LOJBAN <LOJBAN@CUVMB.BITNET>
Message-ID: <906564015.1022941.0@listserv.cuny.edu>
Content-Length: 1473
Lines: 35


Sorry about the last one.  I just mysteriously discovered a key
combination that automatically sends a document.

Robin Turner wrote:

> > This process would of course be much more efficient if one could start
> > memorizing the most used words first.
>
> It's actually very easy - you just take every text you have and feed it into a
> concordancing program.  I could do this if people want, but I'm not sure how
> ...
> if you are interested in general communicative competence, then it would
 probably
> be more appropriate to use a frequncy count for your native language,
> then translate this into Lojban.

On this subject re the language tapes, I have two thoughts.  The first
one is that the count of the stuff discussed on the newsgroup is
probably the best count that we are likely to get until a married lojban
speaking couple winds up with a naturally-speaking lojbanist.

The second thought is that "hey, didn't they make a frequency count and
weighted average to come up with the original list of gismu?"  If we
want something that's close to normal conversational language, that
should do the trick, shouldn't it?

Robert Rapplean