Re: SUN vs Lisp machine hardware differences?

[Christopher Garrigues <7thSon@spar-20.arpa>]

    Date: Wed, 29 Jul 87 17:32:56 PDT
    From: larus%paris.Berkeley.EDU@berkeley.edu (James Larus)

    [ . . . ]

	 Oh yea, one more thing.  If you've ever had to write network code
	 you'll appreciate 1> The quality of Symbolics' code (don't laugh --
	 their competitors are even worse!) and 2> the fact that you don't have
	 to write with foreign-function interfaces and worry about unboxing
	 packets.

    Having tried to use the "quality" Symbolics network code on the same
    network as other manufacture's machines, I can only assume that you
    are kidding.

As a former Symbolics support person, who has had to diagnose network
problems and maintain network code, I have to support Gumby's claim that
Symbolics' network code is really better than that of many other
vendors.  Aside from the relative ease of writing servers and clients
with the generic network system, I should also point out that in most
(admittedly not all) cases where Symbolics and another vendor weren't
communicating, it turned out to be the other guy who was violating the
spec.  For some reason, users tend to assume that just because there are
more Un*x boxes out there that they must be doing The Right Thing.

    [ . . . ]