Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 4 Aug 1997 00:43:04 +0200 (MET DST)
From:      Wolfgang Helbig <helbig@MX.BA-Stuttgart.De>
To:        jkh@time.cdrom.com (Jordan K. Hubbard)
Cc:        andreas@klemm.gtn.com, phk@FreeBSD.ORG, ports@FreeBSD.ORG, current@FreeBSD.ORG, asami@cs.berkeley.edu
Subject:   Re: Current is currently really a mess (was: Re: Tk/Tcl broken(?))
Message-ID:  <199708032243.AAA01485@helbig.informatik.ba-stuttgart.de>
In-Reply-To: <3326.870638153@time.cdrom.com> from "Jordan K. Hubbard" at "Aug 3, 97 12:55:53 pm"

next in thread | previous in thread | raw e-mail | index | archive | help
> > I think itīs not intended to make -current to a system, that
> > isnīt useable anymore for active developement work in the ports
> > area, or am I mistaken ???
> 
> You are mistaken.  The first and foremost "mission" of -current is to
> work on new ideas and implement new mechanisms that specifically DO
> break backwards compatibility (otherwise we could just do the work in
> -release and not worry about it, right?).  That's not to say that it's
> a goal unto itself to break such compatibility, but if you've got to
> do it (and this whole TCL thing is just a drop in the bucket compared
> to what I see for -current's future) then that's what -current is
> there for.  It's a testing ground, not a place for hosting the ports
> collection.

The one and only reason for any OS is to support applications,
which in FreeBSD means to a great extent the applications of the
ports collection.

The success of any OS is determined by the applications ready to run
on it (and by the hardware supported by it).

When I switched from a Linux distribution to FreeBSD a year ago it
was only because of the ease of use TeX and apsfilter.
(apsfilter saved my day when I was fighting to get my stuff printed
with Linux. Seeing it's author--Andreas Klemm--again in the FreeBSD
team convinced me that FreeBSD cannot be too bad :-) )

With this personal story I just want to stress the paramount
importance of the ports selection to the success of FreeBSD.  And
yes, -current is a testing ground--for the ports. If developing
the ports collection on -current is discontinued now, I don't see
*any* ports running on the future RELENG_3_0_RELEASE. And I don't
see any reason to switch to FreeBSD 3.0 from, say RedHat Linux, in
a nonserver environment.

> Perhaps we've lost sight of what the ports collection is there for
> also.  I always envisioned it as that "extra 10%", like comfortable
> seats in your car or airbags in your dashboard.  It makes the system
> much nicer to use and adds polish for the user base.  Putting effort
> into support ports for -current has, however, always struck me as a
> case of trying to add a wet bar and hot tub to an experimental racing
> craft.  What makes sense for a luxury yacht does NOT make sense for
> some prototype boat you're building just to test a new engine design.

But you have to water your boat. A prototype boat watered the first
time and drowning in the very moment makes a tragic picture!

> > In the past one of our goals was, to use -current as our playground
> > for new features, but as long as the system is still useable for the
> > other people using -current.
> 
> You are mistaken again.  The goal was to use -current as a playground
> for new features, period.  "Usability" is a purely subjective term and
> not really applicable here, certainly not where ports is concerned.
> If -current even compiles on a given day then I'm pretty happy and I
> doubt that anyone here would be rewarded by expecting more from it.

Well, I am using -current for half a year now, and ``make world''
completed at least 90% of the times I tried. Beside its fun using
(the taste of freedom and adventure, you know :-)) I have to use
it because I still cannot use the 2.2.2-GENERIC Kernel (remember
the CMD640b stuff?)

> I hate to say it, but I have now seriously come to question whether or
> not you're the sort of person who should be running -current at all.
> I realize that you want to play with SMP, but I think that maybe you
> should have held off on that until it was ready for the non-developer
> types.

I hate to say it, but discouraging ports development on -current
means the death of FreeBSD in the long run.

Wolfgang



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199708032243.AAA01485>