Date:         Mon, 14 Sep 1998 05:08:00 -0400
Reply-To:     Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Sender:       Lojban list <LOJBAN@CUVMB.BITNET>
From:         Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Subject:      Error (contradiction) discovered in the Book
To:           Multiple recipients of list LOJBAN <LOJBAN@CUVMB.BITNET>
Message-ID: <905764107.204733.0@listserv.cuny.edu>
Content-Length: 3587
Lines: 67


Nora found an honest to goodness error in The Complete Lojban
Language yesterday (and we have verified this with John Cowan).

In the discussion of "cei", Chapter 7, example 5.5, pg 151, and the
corresponding example for CEI in the selma'o catalog, the baselined grammar
will not parse the sentence as explained in the text.

Specifically, "cei
has "short-scope" and binds to the closest/smallest tanru unit possible.
Thus in example 5.5, to get the long-scope reading given in the translation,
one would need to surround the entire tanru by "ke... ke'e" brackets.

John, Nora, and I had instantaneous agreement that the baselined formal
grammar takes precedence over an example in the book (if at all possible,
which it is in this case), so the formal grammar determines the standard.

It is unfortunate, in making cei less useful in afterthought (you have to
have marked the beginning of the tanru string being assigned), but it
has the advantage of allowing cei to represent only a part of a larger tanru,
should someone ever want to do so.

But the book is "wrong" at least in having the example which contradicts
the formal grammar.  No way to talk around this one, like we have done for the
wording of the discussion of "ni" %^)

Since I am posting, I will let people iknow what "Lojban Central"is up to,
these days.  Of course I am primarily concerned with keeping book orders
going out as they come in, having finally gotten caught up with all
back orders a couple weeks ago, while getting our financial and business affairs
in order and getting our address list up to date so I can finally notify all
the non-Lojban List people of the book's publication.

Nora is again working on her multi-year Lojban Glosser project.  The prior
version of a few years ago was only minimally useful, putting out a word-for-
word translation with no recognition of the grammar.  She is now incorporating
an analysis of the parser output in order to identify in the gloss the
place structure position of each sumti, assigning an English preposition
or some other flag of that position to each sumti where possible in order
to give the gloss some semantic sense.

In order to test this, she needs a parser for the final baselined grammar,
which John Cowan had never produced.  So he is in turn working on such a
parser, though I am not sure how quickly he will have it tested and solid
enough for general release - we'll put it on the Web/ftp sites only when
he says it is good enough.

Book sales, BTW are at approximately 210 paid copies, which is about
half way to the break-even point.  I have yet to work out a settlement with
the printer based on the bad copies we found, but that will be getting my
 attention in the coming week (I hope).

lojbab
----
lojbab                                                lojbab@access.digex.net
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA                        703-385-0273
Artificial language Loglan/Lojban: ftp.access.digex.net /pub/access/lojbab
    or see Lojban WWW Server: href="http://xiron.pc.helsinki.fi/lojban/"
    Order _The Complete Lojban Language_ - see our Web pages or ask me.