Path: utzoo!attcan!uunet!wuarchive!zaphod.mps.ohio-state.edu!ub!bowen
From: bowen@cs.Buffalo.EDU (Devon E Bowen)
Newsgroups: comp.os.minix
Subject: Re: Re^2: Where is Minix headed?
Keywords: Minix for grad courses?
Message-ID: <1990Oct17.145952@cs.Buffalo.EDU>
Date: 17 Oct 90 18:59:52 GMT
References: <33514@nigel.ee.udel.edu> <7956@star.cs.vu.nl> <612@eds1.UUCP>
Sender: news@acsu.Buffalo.EDU
Reply-To: bowen@cs.Buffalo.EDU (Devon E Bowen)
Organization: State University of New York at Buffalo/Comp Sci
Lines: 19
Nntp-Posting-Host: castor.cs.buffalo.edu
Originator: bowen@castor.cs.buffalo.edu


In article <612@eds1.UUCP>, cdh1@eds1.UUCP (C. Daniel Hassell) writes:
> I propose that we solve this by not pestering Andy to amend his elegant
> code, but rather develop an enhanced Minix, or perhaps Internet Minix,
> that has the features WE want.

I've been thinking a lot about this lately, too. I'm considering using
Minix in an OS course next semester so I can certainly appreciate the need
to keep it clean. But as a user, I'd really like it to be as usable as
possible.

It seems we would need to keep multiple versions of the OS, though. One
for each type of system. Because each system is going to be able to support
different things. For example, I've just got an 8088 laptop. I certainly
don't want kernel space wasted for features that aren't going to help me
on this machine type. I would be willing to coordinate something along
these lines for my machine type, though.

Devon