improving the incubator release process [WAS Re: [VOTE] publish openjpa 0.9.6-incubating release]

[robert burrell donkin <ro...@gmail.com>]

On 11/17/06, Marc Prud'hommeaux <mp...@apache.org> wrote:
> Robert-
>
> We'll correct these issues asap. Once corrected artifacts are
> uploaded, I assume it will necessary to re-start the vote on the open-
> jpa-dev list before re-starting the vote here. Please correct me if I
> am wrong.

there have been recent discussion around this topic of release process
in the incubator. the process is neither quick nor smooth. IMHO this
is bad for everyone.

IMHO better communicating and tooling is required from the IPMC but
also the long and slow feedback loop needs to be cut.

IMO the IPMC should be responsible for the technical content of the release only

IMO it is unreasonable for the IPMC to review the semantic qualities
of a release. in other words, it is not reasonable for the IPMC to
judge whether the application or library works well enough to be
released, whether all important bugs have been fixed or whether the
mix of features to be included is right. these should be matters for
the community  and the PPMC.

following this line of reasoning, it would not be unreasonable to
allow a re-cut using lazy consensus without another formal release
vote by the PPMC. this would probably significantly reduce the time
between initial PPMC approval and IPMC signoff. it would mean that the
release process would need to change after graduation.

i'd be interested to hear other opinions on this

- robert

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: improving the incubator release process

[Eddie O'Neil <ek...@gmail.com>]

  I agree that we should instill Best Practice, which (IMHO) means
that podlings should learn a "real" Apache release process including
voting on new bits before graduating into the wild.

  More today than ever, a podling controls the smoothness its release
cycle given improved Incubator documentation and better tooling.  RAT
helped OpenJPA resolve lots of issues before a release was brought
before the IPMC.  There were issues raised that weren't resolved
(headers on the *.rsrc files) and the Sun XSD (sorry I missed that --
thanks for catching it!).

  Tools that enable PPMCs to obtain feedback about their releases are
Good Things, and we should require a RAT report for votes brought to
the IPMC.  For those that haven't used it, RAT is *super* easy to
checkout, build, and run -- great work Robert -- and adds virtually
zero overhead to the release process.  I'm a big fan and it makes my
job as a Mentor much easier.

  So, I agree that we need smoothness in the release process but
should stick to our release guidelines while enabling PPMCs to address
their own structural issues.

My $0.02.
Eddie



On 11/18/06, Noel J. Bergman <no...@devtech.com> wrote:
> robert burrell donkin wrote:
>
> > Marc Prud'hommeaux wrote:
> > > We'll correct these issues asap. Once corrected artifacts are
> > > uploaded, I assume it will necessary to re-start the vote on the open-
> > > jpa-dev list before re-starting the vote here. Please correct me if I
> > > am wrong.
>
> > there have been recent discussion around this topic of release process
> > in the incubator.
>
> I consider Roy's response to be a correct one.  The vote is on the release
> artifacts.  Although we can argue whether or not that is up to each PMC, we
> should be instilling what is at least a Best Practice to the incubating
> projects.
>
> > the long and slow feedback loop needs to be cut.
>
> If projects would run RAT and provide that report along with the request,
> and do so only after satisying RAT, it would go a long way to streamlining
> the process.
>
> > IMO the IPMC should be responsible for the technical content of the
> release only
>
> That's primarily what we do.
>
> > IMO it is unreasonable for the IPMC to review the semantic qualities
> > of a release. in other words, it is not reasonable for the IPMC to
> > judge whether the application or library works well enough to be
> > released
>
> When have you seen that be a serious talking point in our votes here?
>
>         --- Noel
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

RE: improving the incubator release process

["Noel J. Bergman" <no...@devtech.com>]

robert burrell donkin wrote:

> Marc Prud'hommeaux wrote:
> > We'll correct these issues asap. Once corrected artifacts are
> > uploaded, I assume it will necessary to re-start the vote on the open-
> > jpa-dev list before re-starting the vote here. Please correct me if I
> > am wrong.

> there have been recent discussion around this topic of release process
> in the incubator.

I consider Roy's response to be a correct one.  The vote is on the release
artifacts.  Although we can argue whether or not that is up to each PMC, we
should be instilling what is at least a Best Practice to the incubating
projects.

> the long and slow feedback loop needs to be cut.

If projects would run RAT and provide that report along with the request,
and do so only after satisying RAT, it would go a long way to streamlining
the process.

> IMO the IPMC should be responsible for the technical content of the
release only

That's primarily what we do.

> IMO it is unreasonable for the IPMC to review the semantic qualities
> of a release. in other words, it is not reasonable for the IPMC to
> judge whether the application or library works well enough to be
> released

When have you seen that be a serious talking point in our votes here?

	--- Noel



---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org