Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!rice!rice!sun-spots-request
From: auspex!guy@uunet.uu.net (Guy Harris)
Newsgroups: comp.sys.sun
Subject: Re: "talkd" problems with 386i and SPARC
Keywords: Software
Message-ID: <1990Jul31.030829.14150@rice.edu>
Date: 30 Jul 90 18:06:48 GMT
Sender: sun-spots-request@rice.edu
Organization: Sun-Spots
Lines: 19
Approved: Sun-Spots@rice.edu
Originator: spots@titan.rice.edu
X-Sun-Spots-Digest: Volume 9, Issue 289, message 11
X-Refs:  Original: v9n278

>Can anybody explain this ?

Yes, the SunOS "talk" program is based on the 4.2BSD "talk" program, whose
messages are sent over the wire in a machine-dependent format.  The
original Sun version had some heuristics to try to make it work between
VAXes and 68K-based Suns; those heuristics would probably have made it
work between 68K-based and 80386-based Suns as well.

Then the heuristics were changed to make it work between 68K-based and
SPARC-based Suns; this may have broken its ability to work between
68K-based Suns and VAXes, and/or may not have increased its ability to
work between SPARC-based Suns and VAXes or 386 machines.

>Can I correct this problem with a different software package or command ?

Yes, throw out the SunOS "talk" software on all your Suns and grab the
4.3-tahoe "talk" software instead.  It's in the "AT&T-free BSD software"
archives on "uunet"; in 4.3BSD, Berkeley fixed it to send stuff over the
wire in a much more machine-independent fashion.