Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!wuarchive!gem.mps.ohio-state.edu!tut.cis.ohio-state.edu!uccba!uceng!dmocsny
From: dmocsny@uceng.UC.EDU (daniel mocsny)
Newsgroups: comp.arch
Subject: Re: Memory utilization & inter-process contention
Summary: Yearning for the Ultimate OS
Message-ID: <2089@uceng.UC.EDU>
Date: 5 Sep 89 09:44:14 GMT
References: <1114@aber-cs.UUCP> <278@baird.cs.strath.ac.uk>
Organization: Univ. of Cincinnati, College of Engg.
Lines: 33

In article <278@baird.cs.strath.ac.uk>, jim@cs.strath.ac.uk (Jim Reid) writes:
> A VM system general enough to perform well for most potential users and
> their applications on most potential hardware platforms is asking a lot.

And the cost of failing to give this to the users is a lot more.
Namely, OS fragmentation today costs users $$ billions/yr in lost
time, reduced software availability, artificially increased
complexity, and delays in quickly exploiting higher-performance
architecture.

This isn't pocket change, nor is it funny money. This is real wealth,
the kind you can take to the bank. Fragmentation takes this money out
of the users' pockets. Meaning that those users have less remaining in
their pockets to spend on more high-tech toys. The computer industry
is large enough now that it begins to tax the ability of society to
subsidize its past failures to develop and adhere to comprehensive
standards. Maximizing future computer industry growth depends on
removing as much nonessential complexity from the user as possible.

The best way to do this is to listen to what the users are screaming
for. From the user's standpoint, a computer is essentially a tool to
access the work of programmers. I.e., users don't buy computers
because they like the layout of the motherboard, the elegance of the
address decoders, etc. They buy computers to run software. The only
questions are: how much software will the box run, and how fast? The
ability of a programmer to generate wealth for users is proportional
to the number of boxes that can run that programmer's work. Therefore,
for the computer industry to deliver the most value to the customer
(and to consequently share in that value) it must take whatever steps
are necessary to stop destroying the work of programmers.

Dan Mocsny
dmocsny@uceng.uc.edu