*BSD News Article 92473


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!news.mel.connect.com.au!munnari.OZ.AU!news.ecn.uoknor.edu!feed1.news.erols.com!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!ix.netcom.com!enews.sgi.com!news.be.com!news1.crl.com!nntp.crl.com!usenet
From: "Jordan K. Hubbard" <jkh@FreeBSD.org>
Newsgroups: comp.unix.bsd.freebsd.misc,comp.unix.bsd.bsdi.misc,comp.sys.sgi.misc
Subject: Re: no such thing as a "general user community"
Date: Tue, 01 Apr 1997 00:14:40 -0800
Organization: Walnut Creek CDROM
Lines: 94
Message-ID: <3340C3F0.41C67EA6@FreeBSD.org>
References: <331BB7DD.28EC@net5.net> <333EE416.ABD322C@FreeBSD.org> <5hn00k$dio@fido.asd.sgi.com> <333F45A6.41C67EA6@FreeBSD.org> <5hplcv$6lf@fido.asd.sgi.com>
NNTP-Posting-Host: time.cdrom.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 3.01 (X11; I; FreeBSD 2.2-STABLE i386)
Xref: euryale.cc.adfa.oz.au comp.unix.bsd.freebsd.misc:38181 comp.unix.bsd.bsdi.misc:6530 comp.sys.sgi.misc:29569

Steve Alexander wrote:
> Let me point out that I don't own a computer and have no real opinion about any
> of the free UNIX versions, before making a few points...  I work at

I'll keep that in mind. :-)

>         - Your statement above would look a lot more compelling if it just said
>           "FreeBSD" instead of "NetBSD," "FreeBSD," and "BSDI."  If

Sure it could, but I already admitted we can't say that and so I'm
hardly going to be dishonest and present a united front where none
exists, right?  I can't change reality to fit my vision of a perfect
picture.

>         - An established code base is just as likely to be an albatross as it

In theory, yes.  In theory, a new code base could also be a festering
breeding ground for new bugs and half-baked ideas.  What are we proving?
That the potential for evil exists? :-)  Fortunately, FreeBSD's code
base is no more an Albatras than Linux's code base is half-baked (there
are a lot of good ideas there).

> it seems like what you're really competing on is:
>                 - performance (+ scaling)
>                 - reliability
>                 - hardware platform/driver support
>                 - availability of third-party applications

More or less, but these are also broad categories.  What is
"reliability" a function of, for example?  The care taken during release
engineering? The quality of the kernel?  The robustness of the
applications?  Or maybe a combination of all 3?  To get things right
across the board requires the unified effort of many individuals and the
application of sound development practices.  That is why I took special
care to note such "metadetails" as source code management and unified
bug tracking - I can't imagine trying to do a coherent operating system
release *for the long term* without them, it would simply be too much
"grunt work" and I'd prefer to stay away from operating systems
altogether than spend my time that way.

Performance is also another broad category, with both microbenchmark and
macrobenchmark numbers to compare and frown over for years to come. 
Suffice it to say that all groups are working hard on enhancing
performance and that measuring effective enhancement is also difficult,
so be wary of any quick claims about performance.

FreeBSD is adding support for new hardware every day, and NetBSD still
remains a very viable solution for those looking for a truly
cross-platform solution.  Sure, it would be nice if you could get
everything rolled into one nice unified package, but it's not like
having to choose between two highly capable BSD derived operating
systems is any significant hardship. ;)

And finally, as far as the applications are concerned, it's certainly
FreeBSD's point of view that we should just run everyone else's
applications and not make ISVs jump through any hoops they don't want
to.  I've run the Linux version of StarOffice, Word Perfect for SCO and
any number of other miscellaneous apps for non-BSD systems on my FreeBSD
box. Except for some early glitches which have now been fixed, they work
just fine and making sure that they continue to work fine is definitely
one of our stated goals.

>           So, why not show that FreeBSD can crush Linux on things like
>           performance, if that's the case.  If not, what other

Actually, I really don't truly care all that much if Linux or FreeBSD
are a few milliseconds behind the other on any given week.  It's like
the land-speed record; someone is always trying to break it and the
numbers change every year (my money's on Craig Breedlove again, BTW).
What's far more important to me is that the FreeBSD project members
continue to get along well together, find reasonable agreement on
FreeBSD's common goal set and continue to use sound judgement in how and
where they make changes to the system.  Assembling a team of individuals
with just the right combination of significant commercial experience and
youthful vigor over these last 3+ years has been no easy task, and now
that we've taken our lumps and learned many of our lessons the hard way
it's hardly the time to say "well, we certainly did a fine job with this
team and it seems to be working out nicely.  OK, time for us all to quit
and do something else. :-)"

>           were stable, the source control system used by the developers would
>           be the last thing that would affect my decision process.  As a
>           customer, I just don't care.

And as a customer, why should you?  But as a developer, especially a
volunteer one, I need to work with a group which fits my needs and
expectations just as you as a user need to find software which is a
similarly good match for you.  The source code is where I'll be spending
a LOT of my time, and it makes every difference in the world to me how
it's organized and modified.

-- 
- Jordan Hubbard
  FreeBSD core team / Walnut Creek CDROM.