> for the sake of those who don't read news.  I don't know if *BSD has
> the same problem with OOB messages that the author experienced.

Strictly speaking, 4.xBSD *invented* the OOB problems :-) That code is
only in rlogin, though, rsh is brain damaged in completely different
ways (the stderr back-connect for example.)  The problem, as I've come
to understand it (and posted about in comp.protocols.kerberos where
that message first appeared) is that the concept of "urgent data
pointer" is documented in the RFC's, and BSD corrupted that into an
"out of band data" mark.  [I'd appreciate it if someone could disabuse
me of this notion *with references to standards*...]