Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 14 Apr 1997 21:21:22 +0200
From:      j@uriah.heep.sax.de (J Wunsch)
To:        freebsd-hackers@freebsd.org (FreeBSD hackers)
Cc:        abelits@phobos.illtel.denver.co.us
Subject:   Re: Commercial vendors registry
Message-ID:  <19970414212122.VV36149@uriah.heep.sax.de>
In-Reply-To: <199704140653.BAA00534@dyson.iquest.net>; from John S. Dyson on Apr 14, 1997 01:53:02 -0500
References:  <Pine.LNX.3.95.970413203337.6395A-100000@phobos.illtel.denver.co.us> <199704140653.BAA00534@dyson.iquest.net>

next in thread | previous in thread | raw e-mail | index | archive | help
As John S. Dyson wrote:

(I have somehow missed the article John was replying to, thus i'm
commenting both here.)

> > there is one "distributor" for FreeBSD -- Walnut Creek, but it doesn't
> > really organize or supports anything of that kind).
> >
> I am not involved with WC, but I heard that they do have support.

Yep, they now officially announce that they support their customers,
on the CD cover.  They did `inofficial' support for quite some time
before.

> > and not fall apart instantly, distribution that supports upgrades
> > that could be done by user with minimal knowledge about OS internals
> >
> You mean the kernel of the week syndrome?

Nope, John, we seriously lack a good upgrade system.  Jordan has been
throwing some ideas around, i've also got something in mind regarding
the /etc merges, but nobody ever really put any energy into organizing
a usable upgrade system.  The existing `upgrade' option in sysinstall
is nothing else but a stop-gap measure.  It basically works, and gives
people some form of an upgrade path, but it's far from being a Good
Thing.

> >   Both Linux and FreeBSD change fast, although FreeBSD comes in one
> > monolithic distribution, and any attempt to get something fixed throws
> > user into -CURRENT (no pun intended but it seems appropriate) with all its
> > instability and experiments around.
> >
> Not true, 2.2.X is a new released codebase.  It isn't -current.  Things get
> fixed in the 2.2.X base.

Right, and i'd also like to point out that there's always the
possibility to read the commit logs and extract any patch you'd like
at least via the WWW interface.  Not to speak of the CVS tree, as John
already mentioned.

Mind you, i'm also operating a small ISP here, that's where my private
internet access goes.  The servers there run FreeBSD (big surprise :),
and of course, we aren't maintaining them on any sort of a patch-of-
the-week basis.  However, we occasionally apply security and
functional patches to the system as required, without doing a `make
world', or other heavy-weight actions.  One example of a functional
patch that was quickly applied was Bruce's fix for the broken SLIP
behaviour.

I still remember the days when i've been developing on commercial
systems, and the many times where i've been seriously wanting this
ability, to fix just one thing that was urgent for me, without
upgrading an entire machine.  Now, i can even get the source patches
by the operating system ``vendor''.

> >   FreeBSD technically is a nice OS. Organization of its development and
> > distribution looks umm... unhealthy.

Funny.  You didn't notice that most of our userbase who know both,
Linux and BSD, and have chosen BSD in the end, did it since they found
its development better organized?  Only few people care about
benchmarks or that.

> You mean a central group of people who are trying to maintain quality and
> branding (FreeBSD)?   Or a bunch of distributions with a bunch of different
> combinations of shared libs and apps (and kernel versions, Linux)?  I prefer
> a coherent development path/group.  It is pretty good that we have
> 70+- committers that can modify the tree directly, and don't have chaos.
> In fact, we are pretty well organized.

I've also done commercial software development at my previous
employer, and i must now say that the degree of organization, the CVS
repository maintenance (hi Peter :), the quality of the resulting
code, the general mutual agreement of how things are done, etc. are
_way_ advanced compared to the commercial development.  For the
latter, there was a constant time pressure leading to releasing code
where you as the developer knew that it wasn't ready for prime-time
(and trust me, this hurts, and finally gets you into an apathic state
about the quality of your code).  There often have been heated
discussions about relatively minor things like a style guide.  The
internationally distributed FreeBSD project has currently spent maybe
5 hours into discussing stylistic issues, around the time when we
emitted style(9).  That's with 70-odd developers.  At said software
development company, we were about 15 developers, distributed at two
locations 650 km away, and we've been spending hours and hours into a
style guide, with loads of paper where style(9) is just nothing
compared to.  Needless to say, we could never agree on anything
generally acceptable, simply since the chief developer attempted to
press his ideas of good style to all the other people, totally
ignoring that stylistic problems are personality problems you cannot
regulate up to the last TAB.  And, it was never possible to even use
CVS there, since some of the upper-level guys were too conservative
about it.  RCS was all that was to be, and i finally (by a time when
i've already been about to leave them) proved that they've got a
problem by locking some of the sources for the duration of my own
development when i knew that another group of people also needed to
access them. :-] (By that time, i've already been a FreeBSD committer,
and thus knew how much better things were going with CVS.)

-- 
cheers, J"org

joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE
Never trust an operating system you don't have sources for. ;-)



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19970414212122.VV36149>