[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Releasing 1.6

From: Stefan Sperling <stsp_at_elego.de>
Date: Thu, 7 Aug 2008 11:28:30 +0200

On Thu, Aug 07, 2008 at 10:20:42AM +0200, Stefan Sperling wrote:
> On Wed, Aug 06, 2008 at 10:15:30PM -0700, Hyrum K. Wright wrote:
> > [moving this to dev@ - apologies to those joining us mid-program]
>
> It's lacking one important bit of context (see below).
>
> > > On Thu, Aug 7, 2008 at 5:19 AM, Justin Erenkrantz <justin_at_erenkrantz.com> wrote:
> > >> Let's just create 1.6 *today* that has that feature and move on.
>
> > >> There is no reason that we need to wait six months to get this feature
> > >> in the hands of our users.
>
> > >> Now, that said, since CollabNet apparently really wants to see this
> > >> feature out,
>
> Which feature?

Right, now that I found the corresponding thread on the
svn-full-committers mailing list, I can answer this myself:
The desired features are the improvements made to the authentication
credentials caching behaviour, including the plaintext password prompt,
gnome-keyring and KDEwallet support, and the encrypted caching of SSL
client certificate passphrases.

But that does not really matter.

The real question is whether we want to cut 1.6 sooner rather
than later, with the current set of completed features in trunk.
That is, whether we want to shorten the planned period of time
between major releases for this release cycle because there is
demand for some features which are already in trunk, and which
are more or less completed.

Sort of the reverse we did for 1.5 -- rather than delaying the
release for a feature, we cut the release early because some
features are complete and desired by users :)

I would not object, given that Hyrum and Justin said they would
not mind putting in the bulk of the work of cutting the release.

But there's at least one remaining issue with one of the completed
features in trunk which I was planning to fix before 1.6 is branched.
See issue #3226: http://subversion.tigris.org/issues/show_bug.cgi?id=3236
But I won't have time to work on this during the next few weeks.

Also, Senthil has asked for review of some rather important-looking
patches for some of the new auth-related features we have in trunk.
See this mail which points to the relevant threads:
http://subversion.tigris.org/servlets/ReadMsg?listName=dev&msgNo=141525

I don't care whether we handle these issues in trunk or an a 1.6.x
branch, but I think we should get them done before releasing 1.6.0
should we decide to do so.

Another issue that comes to mind is whether we can really afford
to phase out 1.5.x that quickly. Not everyone will want to upgrade
again in a few weeks time because 1.6 is out. Do we want to maintain
two backport branches, 1.5.x and 1.6.x, at least for a while?
If so, can we afford the workload of doing this?
OTOH , I'd expect the 1.5 -> 1.6 transition to be relatively
painless for most users given the relatively small changeset,
so this may not really be much of an issue.

Stefan

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: dev-help_at_subversion.tigris.org
Received on 2008-08-07 11:28:02 CEST

This is an archived mail posted to the Subversion Dev mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.