On Thu, Sep 22, 2011 at 3:06 PM, C. Michael Pilato <cmpilato_at_collab.net> wrote:
> On 09/21/2011 06:07 PM, Hyrum K Wright wrote:
>> Roll rc4 this weekend (or as soon as the backport is merged) and run
>> that on svn.apache.org to satisfy the concerns of those wanting more
>> testing.
>>
>> When we are satisfied, I'll cut the final release from the same magic
>> revision as rc4, to simplify the signing procedure. This may be a
>> week or two after rc4 is released.
>
> +1. I get the sense that overall this release is a solid one, but I
> understand Ivan's concerns about the last-minute churn around the mergeinfo
> validation stuff. So, your proposed plan makes sense to me, Hyrum.
There seems to be something of a consensus around this, then, so I'll
run with it as soon as the validate-mergeinfo fixes are approved and
backported. I'll be on a plane most of tomorrow, so if it doesn't
happen before then, I'll roll rc4 on Saturday.
> One more thing, though: I would strongly prefer that you announce the
> tentative final tarball posting date at the same time that rc4 is announced.
> "Here's the rc4 release, and unless significant problems are found, we'll
> begin the process of releasing this code as 1.7.0-final on Sept 29..." That
> sort of thing. Let's avoid the "Hey, let's spit out a release later today"
> approach to things, please. Thanks!
I am more than happy to do so, especially since we'll already know
what the magic rev is, what the contents are, and that they have
already been tested (and will probably have a sense of at least some
of the known issues). Contrary to the start of the soak, at this
point, I feel very comfortable committing to a final release date,
given that there are fewer variables now.
-Hyrum
--
uberSVN: Apache Subversion Made Easy
http://www.uberSVN.com/
Received on 2011-09-22 16:48:30 CEST