Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 7 Mar 96 10:29:50 MET
From:      Greg Lehey <lehey.pad@sni.de>
To:        hackers@freebsd.org (Hackers; FreeBSD)
Subject:   Comparing FreeBSD and other OSs
Message-ID:  <199603070933.KAA18914@nixpbe.pdb.sni.de>

next in thread | raw e-mail | index | archive | help
The current FreeBSD/Linux comparison is only part of a larger
question: which operating system should I install on my PC?
Stretching the term "Operating System" to include program loaders like
DOS, we have at least the following choices on standard PC hardware:


  DOS
  Windows 95%
  Windows NT
  OS/2
  Linux
  Xenix
  SCO "UNIX"
  UnixWare
  Solaris 2.x
  BSD/OS
  FreeBSD
  NetBSD
  OpenBSD (?)

I've tried to list these roughly in order of publicity; no flames
please, note the word "roughly".  One way or another, FreeBSD is quite
a way down the list.  What questions do people ask?  I assert that the
most common one is "which OS do I choose".  In most cases the implicit
answer is "none, DOS", but in some cases people really do care.  We
should be able to help them come to a good decision--*not* necessarily
FreeBSD.  

Another question is the one most people have been discussing in the
past day or two: "Which system has the best performance?"  As people
have indicated, this is so tied in to hardware that it's impossible to
give a really good answer.  Certainly it pays to be fair to each
system and show up its strengths (and weaknesses).

A third question might be "ease of use".  This is a hot potato, since
it relates to previous experience.  For many people, the choice might
be DOS.  For the rest of us, though, it's more difficult to choose.  I
don't know if it's even worth following up on this one.

Looking at the two first questions, I think it would be presumptous of
us to want to make a choice by ourselves.  I think we should at least
contact the other groups (yes, even Microsoft) and for a group to hack
up some criteria for comparing the systems.  Possibly a magazine would
be interested, if we can come to a consensus on which magazine would
be neutral enough.  I expect that many vendors (particularly Microsoft
and IBM) would either not want to participate or impose such
ridiculous conditions that we wouldn't be able to work with them, but
I think we should at least show the good will.

That brings us to the next point: what criteria?  I don't even know
how to start, but here are some things that spring to mind:

 - Stability (how do we measure it?  Maybe including ease of getting
   bugs fixed)
 - Performance (what do we measure?)
 - Ease of installation
 - Ease of use
 - Price/performance ratio :-)
 - Ease of administration and maintenance (including installing new
   versions of the system)

These are, of course, only labels on which to hang a whole lot more.
In particular, such a group would have to establish criteria for the
measurements.

This is a pretty vague suggestion, but I think it will draw more
attention to our professional attitude than if we start a one-sided
attack on Linux, however professionally we go about it.

Comments? Flames?
Greg




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199603070933.KAA18914>