Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!gatech!hao!husc6!harvard!ksr!alcatraz!benson
From: benson@alcatraz.ksr.com (Benson Margulies)
Newsgroups: news.misc
Subject: Re: Re: Answers to Frequently Asked Questions (Last changed: 31 May 1987)
Message-ID: <169@ksr.UUCP>
Date: Fri, 19-Jun-87 09:49:26 EDT
Article-I.D.: ksr.169
Posted: Fri Jun 19 09:49:26 1987
Date-Received: Mon, 22-Jun-87 01:43:48 EDT
References: <560@unmvax.UNM.EDU> <1180001@hpcea.HP.COM>
Sender: nobody@ksr.UUCP
Reply-To: benson@ksr.UUCP (Benson Margulies)
Organization: Kendall Square Research, Cambridge MA
Lines: 16


First of all, the spelling is Multics, not MULTICS. Us alumni of the
system get nasty in a hurry when people make that mistake.

As as for large tables in core, the real Multics system that a few
lucky people still have in spite of Honeywell's having killed it
(after 15 years of trying) know that its a damned sight better at some
things, including making large amounts of memory available to user
programs, than Unix. How would you like a 2**36 bit segmented process
address space, with dynamic linking, a robust file system (no fchk on
every boot), and direct mapping of files into the address space?

I could go on, but I'm in the wrong newsgroup.
Benson I. Margulies                         Kendall Square Research Corp.
harvard!ksr!benson			    All comments the responsibility
ksr!benson@harvard.harvard.edu		    of the author, if anyone.