directory-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alex Karasulu <akaras...@apache.org>
Subject Re: shared M1 release, or not ?
Date Wed, 09 Feb 2011 15:25:14 GMT
On Wed, Feb 9, 2011 at 3:14 PM, Emmanuel Lecharny <elecharny@gmail.com> wrote:
> Hi guys,
>
> we have had a convo yesterday night with Alex about the current shared M1
> state. We mainly talked about the need for a release now, or later. Let me
> give you some feedback about what we discussed.
>
> Fist of all, Alex thinks that we still have one or 2 weeks of work to
> cleanup shared. I agree. He also told me that if we decide to go for a
> release now, he will 0+ it.

Just to clarify my reasons for a +0 on a 1.0-m1 release of shared.
It's a +0 and not a 0 or -1 because I don't want to hold you guys back
if you want the release. If you want it you got it. Just make sure the
legal aspects are right and you have my support. Plus going through
the exercise is a good thing.

Why not a +1? It does not meet my personal standards for viability and
I'm unwilling to put my name behind it. This does not mean we suck.
It's just not ready, but that's another story in just a couple weeks.

There are other sensible reasons for the +0 rather than a +1:

(1) We're close to having a stable useful product out: in about 2
weeks or so. We waited this long, what's another 2 weeks?

(2) With all the sub-projects, voting time, and release preparations
it will consume a good week to get M1 out the door. We started a
couple days ago so if we do the math this is not so unrealistic an
estimate. Then what? We start this process again for m2 right after
the m1 is done?

(3) I'd like to offer our users a single consolidation jar without
dependencies so they can use that as the complete LDAP client API
product. Right now in the present configuration we have a disorderly
set of separate jars. It will be a PITA for our users, and how many
are we doing this for?

(4) I don't want peeps to start using it for 2 weeks then have them
deal with moderate to large changes. For example between m1 and m2
their maven configurations will change (not just version info) as well
as some APIs. If two weeks will prevent this then I think it's best to
wait.

So why bother is my thought if in 2 weeks they get the close to final
configuration. However if this is something you feel is something you
must do right now then go for it. I don't think it makes any
difference at all, but I want those wanting it to have what they want.

> Note that I'm *not* pushing for such a release,

There is a push but that's OK. Nothing wrong with that. Having
consistent releases is a really really good thing. But right now I see
it as pointless due to where we are and where we will be in 2 weeks.

We're wasting man power that should be put towards the effort instead
of releasing just to release.

but it may also be a good
> thing to test the release process.

This is the main reason why I will vote +0 (if the legals are in
order), though the product does not meet my personal standards.

Regards,
Alex

Mime
View raw message