From lojbab Sun Jul 24 01:03:21 1994
Date: Sun, 24 Jul 1994 01:03:15 -0400
From: Logical Language Group <lojbab>
Message-Id: <199407240503.AA01229@access1.digex.net>
To: nsn@krang.vis.mu.oz.au
Subject: lujvo files
Cc: lojbab@access.digex.net
Status: RO

Well, I can't seem to find you logged in on a variety of machines there.
There is a zip of files (how is that for a new collective term?) waiting
for you at ftp.digex.net  /pub/access/lojbab
a little less than 400K zipped

contents I think include
a) up-to-the-moment gismu list
b) lujvo list in unexpanded format (no keywords)
c)lujvo list in expanded format (keywords for each place)
d) expansion oblique keyword file
   the first word/hrase up to a semicolon is what is inserted
e) an up-to-6/24/94 list of possible Lojban words in the archives
   that might include new lujvo - relatively unprocessed
f) a set of changes to the lujvo list to choose the 'optimal form' as
   the main entry, with these lines as x-refs from whatever you had in the file
   (thus, any new entries should be added as a separate file so I don't have
    to run the old words thru the glosser again)

The unexapnded lujvo format is the ideal one for new words - we can auto-process
this into the exapnded format.  Any systematic abbrevaition is probably OK, but
your 1-2 letter gidsmu abbreviations were often incorrect (you used the rafsi
rather than the gismu at times, giving a dfifferent initial letter, and 
sometimes a totally unexpalinable letter).

I don't have a separate set of lujvo in the gismu file that is current - I've added a couple more int he last few days, I think, though I did send you a
list several months ago.  You may be able to pull them out fairly easily
yourself, since they all fit the pattern of being
1) after column 160
2) embedded in parens with an "=" before them

But I can glom together a list within a day if this is difficult.

I do have a couple of lujvo in the working English-order gismu file that are
not in here, but the numbers are under a couple of dozen, and I add new ones
all the time - sometimes even doing place strutures for them (not according
to the standard format alas, but REAL place structures in long-winded form)
I will collect these for you when you deem it appropropriate.  Many are nu+gismu
BTW.

The new list should have all se/te/ve/xe+gismu in it.  I still would
like to someday auto-generate nu/ka/ni+ gismu and gismu+gau/zu'e/zma/fri
and maybe a few others to greatly expand the list.  But undoubtedly not for 
first edition - the e-entries alone would fill a 1000 page book, I suspect.

I'll be on for a little while after mailig this reasding news. Else, send
me email when you have the ftp file so I can clear the directory - we are
running close to our 5 Meg limit.

lojbab 0100 EDT