Subject: Re: Whats the point of this porting effort?
To: Lee Reynolds <email@example.com>
From: John Klos <firstname.lastname@example.org>
Date: 03/09/2001 04:20:10
Thoughts and opinions are important, and there are a lot of things people
can learn from examining the things around them.
While your post has good flame potential, I'd rather assume you really do
want to know what we think, so here is my reply:
> What is the point of the 68k port of NetBSD? Is it
> entirely academic? The x86 port of NetBSD will run on
> everything from the slowest 386sx-16 to the fastest
> Athlon. But because Apple dropped the 68k for the
> PowerPC, this port of NetBSD is limited to running on
> vintage computers.
Yes, but there are very, very many of these vintage computers. They are
literally found in the garbage (at least in New York City) and can be had
for, literally, $20.
> This wouldn't be so bad if the code base for NetBSD was becoming more
> efficient as time went by, thereby compensating for the increasing
> obsolescense of these systems. Unfortunately it is not. Every relese
> of NetBSD has been less efficient than the last. You don't see this
> on newer systems because the difference is not all that great, but on
> a platform that isn't being improved these small losses in efficiency
> add up. Just like Linux, OpenBSD, and FreeBSD, the primary
> development platform for NetBSD is the standard PC. When this code is
> ported to a 68k Mac, which in its day was a fast system, it runs like
> molasses through a soda straw.
I might be wrong, but I do not believe that PCs are the primary
development platform of NetBSD. The philosophy of NetBSD is to have a real
Unix operating system working as consistently as possible across the most
In addition, the code is not written on one machine and "ported" to
another. If it can't compile on all platforms, it isn't a part of NetBSD.
This, of course, refers to the non-hardware specific code.
> Five years or more ago this project was a good idea,
> there were lots of 68k macs still being used. In fact
> if Apple had been forthcoming with the technical
> details of how these systems worked then, NetBSD might
> be a really slick OS on the Mac now, with full color X
> and complete support for all devices on all the
> models. But Apple behaved like a spoiled brat, as
> usual, and so that didn't happen. So now I'm
> wondering what the point is anymore. If this is an
> academic exercise, or something people are doing
> because it is fun for them then great, but if its some
> kind of serious pursuit, then people might want to
> look at what exactly it is that they're accomplishing.
It would not be accurate to say that it is one thing or another; it is
many things. I have learned a tremendous amount working on the Mac 68k and
Amiga NetBSD ports, so education is one thing. When I teach other people
about Unix, I often encourage them to go out and buy a 68040 Mac, since
they are cheap and redily available.
> Creating a better version of unix for an obsolete
> platform that very few people use anymore isn't much
> of an achievement. Its about on par with creating a
> better 8 track tape. Unless the porting work can be
> applied elsewhere as well, such as to the PowerPC
> port, then just what good is it?
Tell that to the VAX port!
Also, development of m68k code also helps the other ports that run m68k -
NeXT, Sun3, Amiga, Atari, and so on. Active development still goes on for
the Amiga, so it would hardly be fair to say that this work is wasted on
an obsolete platform.
There are tens of thousands of m68k Macs out there, meaning that anyone
with a few dollars can learn all about Unix via NetBSD. What if you were
one of thousands of people who spent a couple of thousand dollars on an
m68k Mac and wanted to keep it and use it for something? I think that's
how lots of projects start - making use of something that has outlived
its more common use.
> I'm not trying to discourage anyone from doing the
> porting work if that is what they enjoy doing, or if
> they are learning valuable knowledge from doing so
> that can be applied to more useful project. I'm just
> wondering what the motivation behind it is at this
> point because I just can't see any practical use for
Well, that all depends on what you consider practical.
I doubt that I'm a typical example, but I run a modest business selling
hosting on a 68060 Amiga, and I have two 68040 Macs that I use for
compiling and testing in support of that Amiga, and a third that does
backup DNS and email. Also, I have set up and sold numerous 68040 Macs for
use as DNS servers, mail servers, and low volume web servers.
My home network of six machines all talk to the internet through a Quadra
800 with two ethernet interfaces doing IP NAT to a cable modem. I found
that Quadra in the trash; most of the Macs I have set up have been found
in the trash or given to me.
I just bought a Quadra 605 for $22, shipping included, from eBay because
the motherboard and power supply are physically small enough to fit,
literally, inside of a breadbox; one of my clients wants a mail server
that is completely unobtrusive and out of the way (New York City real
Where else can I get a cheap computer with ethernet and SCSI built in?
Macs have always had good quality hardware, and they are very reliable.
There are many, many good reasons for using them. It really depends on
what you're doing.
Get one, install NetBSD, and learn!
Help find a cure for MS!
Reeducate a Windows sysadmin!