Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 25 Oct 1997 21:53:03 GMT
From:      mouth@ibm.net (John Kelly)
To:        freebsd-stable@FreeBSD.ORG
Subject:   Please don't hurry
Message-ID:  <345260f6.1453370@smtp-gw01.ny.us.ibm.net>
In-Reply-To: <199710250440.VAA12327@hub.freebsd.org>
References:  <199710250440.VAA12327@hub.freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help

On Fri, 24 Oct 1997 21:40:01 -0700 (PDT), "Jordan K. Hubbard"
<jkh@time.cdrom.com> wrote:

> That's what I'll probably do.  When I checked into this, I realized
> that the ports team neglected to provide a version of the lynx package
> which matched the ports collection version this time around and hence
> it's missing from the INDEX, leading to its not being listed there at
> all in sysinstall's package installer menu. :-(
> 
> Oh well!
> 
> 					Jordan

May I make some suggestions for 2.2.6?

Use a longer beta cycle with new SNAPs cut weekly instead of daily,
and run that for seven to ten weeks.  I think many people like myself
who are not closely involved with development don't have time to do
any testing when the SNAPs are cut daily for only a couple of weeks.

Once you have a really good SNAP, cut a small production run CD set
and sell these to people who want to test out the new release.  Do at
least two of these, maybe even three or four, extending the beta cycle
to a much larger testing audience.

I would readily pay $30 or so to pre-subscribe to the beta CDs in the
hope of helping refine the 2.2.X series for a classic final release.
If several hundred users would pre-subscrible for them, you should be
able to cover the cost of producing and mailing them.

When you finally have a CD set that needs no more changes, go with
that one, unchanged, as the real production release.  Under no
circumstances should you make any changes, even trivial ones, between
the final beta CD and the real production run.  The risk of new bugs
introduced by last minute changes is too great.

I realize this suggestion would extend the beta cycle to three or four
months, perhaps unheard of for FreeBSD, but I would like to see the
final 2.2.X release as near to perfect as possible because it might be
a long time before I'm ready for anything in the 3.X series, even
after it goes stable.

While it's true that I can always CVSup fixes if the CD has bugs, if
you have a pack of production machines to install and maintain, it
sure is nice to just stick the CD in the drive, install, and be done
with it.

John





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?345260f6.1453370>