Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 30 Aug 2007 18:05:08 -0400
From:      John Baldwin <jhb@freebsd.org>
To:        John-Mark Gurney <gurney_j@resnet.uoregon.edu>
Cc:        cvs-src@freebsd.org, src-committers@freebsd.org, peterjeremy@optushome.com.au, Alfred Perlstein <alfred@freebsd.org>, cvs-all@freebsd.org, yar@comp.chem.msu.su, eischen@vigrid.com, kensmith@cse.buffalo.edu, "M. Warner Losh" <imp@bsdimp.com>
Subject:   Re: cvs commit: src/lib/libc/gen fts-compat.c fts-compat.h
Message-ID:  <200708301805.10011.jhb@freebsd.org>
In-Reply-To: <20070829205537.GB977@funkthat.com>
References:  <Pine.GSO.4.64.0708251703550.19091@sea.ntplx.net> <200708271530.45290.jhb@freebsd.org> <20070829205537.GB977@funkthat.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wednesday 29 August 2007 04:55:37 pm John-Mark Gurney wrote:
> John Baldwin wrote this message on Mon, Aug 27, 2007 at 15:30 -0400:
> > On Monday 27 August 2007 02:57:41 pm Alfred Perlstein wrote:
> > > Sorry for top posting, but...
> > > 
> > > I agree very strongly with Warner, in short, if possible, reducing the
> > > number of major gotchas of running current will make our developer
> > > and early adopters a lot happier.
> > > 
> > > It will help FreeBSD.
> > > 
> > > One of the things that turns me off to FreeBSD is the feeling that
> > > I get that certain people take some kind of pride in forcing users
> > > to go through dangerous and complex hoops in order to run current.
> > > 
> > > It shouldn't be so if the overhead of making it easier is so small.
> > 
> > It has zero to do with pride, but it does have a lot to do with allowing 
HEAD 
> > to be a branch for development as opposed to the stable branches that are 
> > more intended with deployment.  Those are quite different feature sets.
> 
> Isn't that what perforce is for?
> 
> Also, we NEED users to be running -current...  How many times has a release
> happened and people complained that it didn't get the testing it needed
> to find bug xyz?  If we take the attitude that only developers should
> be running -current we'll continue w/ unhappy -stable releases like we
> have in the past...
> 
> Not that Kris doesn't do a great job finding bugs, but other people
> are good at finding bugs too, and we shouldn't lock them out...

How many custom kernels from p4 do you run from other people's branches?

Putting stuff into HEAD is how we get wider exposure for stuff beyond a 
developer's own boxes (or a few other people who will test it).  We can't 
restrict change so much in HEAD that we can't do development there.  Also, 
for better or for worse, a lot of the peer review kicks in when stuff is 
committed to CVS, so churn in HEAD is quite common and it's going to continue 
to be there in the future.

-- 
John Baldwin



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200708301805.10011.jhb>