Path: utzoo!attcan!uunet!tut.cis.ohio-state.edu!ucbvax!PSUVM.PSU.EDU!ART100
From: ART100@PSUVM.PSU.EDU ("Andy Tefft  725-1344", 814)
Newsgroups: comp.sys.apple2
Subject: Re: Kermit Installation Help!
Message-ID: <9010220459.AA13417@apple.com>
Date: 22 Oct 90 04:59:00 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 41

Since athena has ftp access, I'd recommend the following procedure:

1) ftp to shark.nosc.mil. Login as anonymous, any password
2) cd ker*mit.    (the asterisk and period are necessary)
3) get appprobns. This file is a text file
4) Split this up between binscii segments, into as many pieces as
   you need. Generally, individual binscii segments will be better
   if you are doing text capture, since this will mean a smaller chunk
   to be re-transmitted in case of an error.
5) capture the segments and run binscii on them
6) Use shrinkit to extract the files you need from the archive,
   and read all the readme files.

If you have and can use shrinkit and binscii this is the best method
because it gives you some error checking, can be undone  under
prodos (the way you want to do it must be done under dos 3.3, and
the binaries copied to prodos if you want to use it under prodos),
and the chunks are smaller << 26k.

There is no "easy" way to combine split files (i.e. universal).
Under prodos you could probably play with the "B" parameter,
i.e.

bload part2,a$900,ttxt
bsave part1,bxxxx,ttxt,a$900,lyyyy

where xxxx is the length of 'part1' as
shown in the catalog, and yyyy is the length of part2 as shown
in the catalog. If the files were small enough you could also

bload part1, a$900,ttxt
bload part2,axxxx,ttxt
create all,ttxt
bsave all,a$900,lyyyy

where xxxx is 1+the end of part1 in memory, and yyyy is the end of
part2 in memory.

I just tried the first method (using the B parameter) and it
worked just fine. You should try it out yourself with small
text files first.