<< Previous Message Main Index Next Message >>
<< Previous Message in Thread This Month Next Message in Thread >>
Date   : Mon, 27 Feb 1984 22:25:29 EST
From   : David Towson (CSD) <towson@amsaa>
Subject: Query to Multics people.

Multics people - I have been having an interesting exchange of messages with
David Cargo at Hi-Multics concerning the transfer of Simtel20 files stored in
ITS binary format (this includes all binary files, and ALL files in the SIGM
and CPMUG archives).  Apparently, Hi-Multics' FTP doesn't accommodate TENEX
mode
(which works so nicely for us here on the BRLNET UNIX machines).  That has left
David in the same awkward position we were in until we were told about TENEX
mode, namely, that of having to transfer the files using binary mode, and then
post-process them to get rid of the four padding zeros that follow the four
8-bit bytes in each 36-bit ITS binary word.  With the aid of a system wizard,
David now has a program that is a modified version of a proprietary system copy
routine, and which does the necessary post-processing.  Using this program, he
has successfully transferred ITS binary files from Simtel20.  He has then used
Keith Petersen's ITSCVT program (available as ITSCVT.HEX on Simtel20 in 
directory MICRO:<CPM.HEX>) to remove the four ITS "header-bytes".  (This is 
done after the program has been downloaded to a CP/M machine.)
     So my question to you Multics users is this:  What methods are you using
to transfer ITS binary files from Simtel20 ?  Does anyone have an FTP that will
transfer these files and automatically discard the padding bits ?  Or is there
an FTP option that has not been mentioned here, and which will do the job ?
     As the maintainer for this list, I get many queries.  By far, the most
frequent questions deal with the archives and how to FTP.  Not being a Multics
user, I am very much at a disadvantage in trying to help Multics people, and I
would greatly appreciate your assistance in providing me with information.  
The program David is now using is based on proprietary roots, and cannot,
therefore, be freely distributed.  He has suggested the possibility of creating
a difference file that could be used by legitimate users of the base-program
to produce the modified version.  Back before we learned about TENEX mode
(which solved our problem on the UNIX machines), we too used a post-processor.
It was a happy day when we were able to abandon it.  Is there a similar happy
solution for Multics users ?



Dave Towson
info-cpm-request@amsaa
<< Previous Message Main Index Next Message >>
<< Previous Message in Thread This Month Next Message in Thread >>