From LOJBAN%CUVMB.BITNET@UBVM.CC.BUFFALO.EDU Sun Nov  5 14:16:48 1995
Received: from VMS.DC.LSOFT.COM (vms.dc.lsoft.com [205.186.43.2]) by locke.ccil.org (8.6.9/8.6.10) with ESMTP id OAA14120 for <cowan@LOCKE.CCIL.ORG>; Sun, 5 Nov 1995 14:16:46 -0500
Message-Id: <199511051916.OAA14120@locke.ccil.org>
Received: from PEACH.EASE.LSOFT.COM (205.186.43.4) by VMS.DC.LSOFT.COM (LSMTP for OpenVMS v1.0a) with SMTP id 8C537F00 ; Sun, 5 Nov 1995 15:12:13 -0400
Date:         Sun, 5 Nov 1995 11:07:44 -0800
Reply-To: "John E. Clifford" <pcliffje@CRL.COM>
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
From: "John E. Clifford" <pcliffje@CRL.COM>
Subject:      rafsi
To:         lojban list <lojban@cuvmb.cc.columbia.edu>
Status: OR

Oops! (Night thoughts)  Mark,l 's rafsi proposal does fail the
internal resolution test in its present form: CAr+trAC for
example is not distinguishable from CArt+rAc.  I ball-park teh
number of problematic CCCs at above 500, so pretty likely to
have actual cases.  To resolve this, mark,l will need either
glue at the CVC-four-letter rafsi joins or seriously restrict the
legitimate CVCs or, perhaps, go to only four-letter rafsi, with
joins as needed.  The first and last further diminish the
advantage of his system over the present long-rafsi system, the
second makes the chances of complete coverage smaller.  The
last version is, in fact, just the long rafsi format with an
occasional dropped schwa.  Glueing all CVC-fourletter joins
is not strcitly necessary, of course, but specifying which ones
don't need the y would make for an even more complicated set
of rules added to the present  complexity.
pc>|83