Return to BSD News archive
Path: sserve!newshost.anu.edu.au!munnari.oz.au!constellation!osuunx.ucc.okstate.edu!moe.ksu.ksu.edu!vixen.cso.uiuc.edu!howland.reston.ans.net!agate!agate.berkeley.edu!cgd
From: cgd@eden.CS.Berkeley.EDU (Chris G. Demetriou)
Newsgroups: comp.os.386bsd.misc
Subject: Re: Status on discussed merge between NetBSD and FreeBSD
Date: 14 Nov 93 10:00:33
Organization: Kernel Hackers 'r' Us
Lines: 75
Message-ID: <CGD.93Nov14100033@eden.CS.Berkeley.EDU>
References: <JKH.93Nov13222001.2@whisker.lotus.ie> <crt.753292942@tiamat.umd.umich.edu>
<CGD.93Nov14085627@eden.cs.berkeley.edu> <CGHs3y.Au2@kithrup.com>
NNTP-Posting-Host: eden.cs.berkeley.edu
In-reply-to: sef@kithrup.com's message of Sun, 14 Nov 1993 17:18:55 GMT
In article <CGHs3y.Au2@kithrup.com> sef@kithrup.com (Sean Eric Fagan) writes:
>And what's with the "us" vs. "them"? Between that and a couple of the
>core netbsd people shouting insults at anyone who disagrees with them,
>I'm surprised you have anyone helping you at all anymore.
i don't see anyone shouting usually, except you, sean. and invariably
you can't take it if people shout, or even peep, back. Hmm, or was
that intentionally bouncing mail...?
perhaps the "us vs. them" comes from being pretty-much blindsided by
the announcement? i dunno, i worked many long and hard hours on trying
to get the two systems merged. I would have liked maybe 10 minutes
notice of what was going on. Or perhaps some details from Jordan
as to why FreeBSD decided to make this announcement...
>>i get get their commit messages via e-mail, in the same way that many of
>>them get ours.)
>
>Real useful after you've removed the ability for all or most of them to
>check the CVS files, isn't it?
i'm sorry; NetBSD simply doesn't have resources to support the development
of NetBSD *and* the development of another operating system, as well.
Remember: we're not making any money off of this, and thus far i've sunk
about 15k into hardware for this project. There are simply not the
resources for people who aren't doing development to eat disk space and
CPU cycles on lamp for the purposes of sucking changes into their
own system, so they can claim it's "better," and not give us properly-
attributed credit for the work we've done. The day that i started
seeing postings on FreeBSD mailing lists which had no body, and only
a subject which said "Has anybody picked up XXX from NetBSD yet?"
where XXX was the "neat feature of the day," i lost all faith in
FreeBSD's development process.
The fact that, for the most part, the FreeBSD "developers" used their
accounts on sun-lamp *only* for the purpose of snarfing revisions of
our code and migrating it to FreeBSD revolted me, and it still does.
You can't call it "development" if you don't have a coherent understanding
of what's going on. That's exactly what i meant about the shared libraries.
It's not a matter of whether or not they have to debug the changes they've
made. The point is that the changes that went in WEREN'T TESTED, they
broke in HORRIBLE WAYS (the last of which are only being discovered),
and it's taken them two weeks to figure them out. If you're going to make
a change that major, TEST IT FIRST; we do!
It's a similar (sad) story with the "4M vs. the VM system" changes:
I've been using 386bsd and then NetBSD on a 4M machine, doing system
builds and whatnot with no problems (other than heavy swapping 8-).
Then along comes David, hacking on the VM system, and lo an behold,
suddenly, 4M machines break. did they back out the changes? no.
How long have they known about the bug? weeks, now, i believe it is...
>(A little note: I am biased. I will not work on netbsd anymore due to the
>attitudes of deraadt and mycroft, and a bit of cgd; stuff he did last night
>has further increased that ... *anger*. So take it all with a grain of salt;
>unlike what cgd and others would like you to believe, both systems do work,
>quite well, and are perfectly usable for development. Of course, I'm still
>running 386bsd, and my system has now been up for 8 days; that's better than
>my SCO system managed at times, so what do I know? :))
I never, ever said that both systems "didn't work", or that both
systems weren't "perfectly usable". On the contrary, as noted, that post
was made to dispell the long-standing and completely incorrect notion
that NetBSD *WASN'T* "perfectly usable." Thank you for helping me
make my point.
chris
--
chris g. demetriou cgd@cs.berkeley.edu
smarter than your average clam.