>> It sounds like just a little detail, but just about every single PC
>> I've run accross that was bought before 1995 had an Adaptek 1542
>> controller.  Partial 1542 support is a real turn-off.
>
>I've always thought it was a bit odd that someone who could go and
>put a thousand dollars worth of RAM in their machine couldn't afford
>another two hundred for a new PCI motherboard and NCR controller.
>But that's just me. :-)

*sigh* As Peter has already pointed out, one of the reasons people run
free OS's is because of the cost.  Maybe the memory was scavenged from
an older MB?  Maybe it was lent by a friend who upgraded his MB?

Telling people, "your hardware is too sucky to run NetBSD", when Linux
and FreeBSD work just fine on it is one of the reasons that NetBSD has
such a small user base.  And even worse, it's the worst sort of OS snobbery
("The OS I use only works on Gateway 2000s built between 9/94 and 11/94!
It's the best free Unix of all!").

Ok, that's not what Curt was saying.  My point is that we should want to
make NetBSD work on as many platforms as possible, to give people one less
reason to switch to other free Unixes.  This would help to increase our
user base.

Now before people start saying things like, "maybe we don't want a large
user base", let me point out the following things:

- I think most people will agree that the more people we have contributing
  to NetBSD, the better.

- Contributors come from the user base.  From this, one can figure that the
  more users we have, the more contributors we have.

- Thus, we can conclude that the more users we have, the more people we will
  have contributing the NetBSD, which is probably good.

Jordan has no doubt already figured this out for himself a long time ago.
I would hazard to guess that's why he's invested so much of his time and
efford into making FreeBSD so newbie-friendly.  A certain percentage of
these newbies will start writing device drivers, install documentation,
kernel hacker guides, etc etc, which is generally a good thing.  Who knows
what great kernel hacker we're excluding because he happens to have 24MB
on a machine with a 1542?  And this just helps to fuel the general attitude
that NetBSD is lame and Linux is much cooler (which tends to steer newbies
away - some people may argue that this is good, but let's not forget we
were _all_ newbies once!)

>> Well, I myself have been seriously, seriously tempted to switch to
>> FreeBSD.  The prospect of 32Megs of RAM and a working mmap...
>
>What's not working in 1.1's mmap? I've got INN running under 1.1 using
>mmap for both the active file and dbz, and I'd like to know what's about
>to go wrong. (Currently the active file on disk doesn't update, but I
>think I just need to put an msync() in ICDwriteactive().)

Well, to be honest, you shouldn't have to put an msync() in there at all,
but that's been beaten to death before.

Regarding "winning/losing" the free Unix war...

I'm not sure there's going to be a "winning/losing" in the free Unix war
(If there is, then Linux has already won - when I was explaining to a
coworker that I don't run Linux at home, but another free Unix, his
response was, "You mean there is another?").  I think they all have their
place.  However, one thing that _would_ be a loss would be if NetBSD
dried up because no one worked on it anymore; that's the biggest threat
to NetBSD right now, far more than Linux could ever be.

However, I'm not sure that's what will happen.  I see new people appear
occasionally on current-users, new people join core (and some people leave,
alas), new stuff gets integrated; it's obvious that NetBSD development isn't
dead by a long shot.  The whole OpenBSD business saddens me, moreso because
it's mostly ego-driven on both sides (but I wish them the best of luck).

--Ken