Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!ccplumb
From: ccplumb@watmath.UUCP
Newsgroups: comp.edu
Subject: Getting Computer Resources (Re: software engineering)
Message-ID: <12912@watmath.UUCP>
Date: Sun, 19-Apr-87 17:28:45 EST
Article-I.D.: watmath.12912
Posted: Sun Apr 19 17:28:45 1987
Date-Received: Sun, 19-Apr-87 18:37:40 EST
References: <340@ndsuvax.UUCP> <1986@cwruecmp.UUCP> <12881@watnot.UUCP> <176@brandx.klinzhai.RUTGERS.EDU>
Reply-To: watmath!ccplumb (Colin Plumb)
Organization: U of Waterloo, Ontario
Lines: 65
Confusion: U. of Waterloo, Ontario

webber@klinzhai.RUTGERS.EDU (Webber) says
(in <176@brandx.klinzhai.RUTGERS.EDU>):
>In article <12881@watnot.UUCP>, I (ccplumb@watmath.UUCP) wrote:
>>Speaking as a first-year CS student, I could conceive of writing a
>>Unix kernel.  It might take me years, but I don't think the people
>>who write them are a higher order of being or anything.

>I don't think you want to speak too much about writing Unix kernals
>single-handedly without first establishing some operating system
>specific background.  A good case can be made that no person or
>group of persons has successfully written an operating system of
>the size of unix yet (e.g., unix has existed for at least 10 years
>and people are still reporting bugs (other operating systems benefit
>from being badly documented enough that people can't recognize the
>bugs :-) ).

I must admit that I am a bit weak in OS background.  I read about
the Unix kernel, but I haven't had a chance to study the code yet.
Still, I could certainly get a time-sharing OS of some description
running.

If you define `success' as a complete absence of bugs, then I don't
think I could write such a thing without a major advance in correctness
proving.  I might add, however, that fewer authors leads to a more
coherent philosophy (one of the reasons for the early success of Unix),
which can reduce the number of subtle interaction bugs.

I might point out RMS to people who doubt that a Unix kernel *can* be
written single-handed.  No, he hasn't produced a kernel yet, but I
haven't any doubt that he'll do it.  I'm not nywhere near his level,
but I'm not without aspirations.

>There are two classic approaches to getting extra computing resources.
>One is to get `hired' by the local system staff.  This gets done by
>demonstrating a knowledge and interest in what they do and a
>willingness to work with them.  The other way is to offer to aid one
>of your faculty in their research (while they may or may not have
>money, they don't suffer from the problem of not being able to use
>people that they can't pay for).  Again, this gets done by
>demonstrating a knowledge and interest in what they do and a
>willingness to work with them.  Of course, in figuring out which way
>to go, you should determine whether you want to establish vocational
>credentials or academic credentials.

Well, first of all, my nefarious schemes to get more resources more directly
have had various effects on the systems staff, from amusement through concern
to serious annoyance.  Also, around here, they've made an agreement to only
hire co-op students for summer jobs.  I'm not one of them.

Still, I'm getting as far into the system as I can.  I'm currently
unilaterally insatlling (with bug fixes and improvements) software from
the net, learning about the innards of X, and learning VLSI design so I
can try to actually design a CPU I've been dreaming about for a year or
so.

>Enjoy.

Oh, I'm doing so.

>------------------ BOB (webber@aramis.rutgers.edu ; BACKBONE!topaz!webber)
--
	-Colin Plumb (watmath!ccplumb)

Silly quote:
Anybody who marries her would stand out like a sore thumb.