CFV: New OpenJDK Committer: Martin Balao

[Andrew Hughes]

On 28 June 2018 at 07:42, Volker Simonis <volker.simonis at gmail.com> wrote:
> Hi Andrew,
>
> I totally agree that we need more OpenJDK developers in the security
> area and I'm sure Martin is a perfect fit for this role.
>
> But the process document [1] clearly states that a "Contributor should
> make at least eight significant contributions to that Project before
> being nominated". From the references you've provided I can only see
> five changes contributed by Martin. I'd therefore like to kindly ask
> you to withdraw this CFV and postpone it until Martin has reached at
> least the required minimum number of contributions.
>
> Sorry for nit-picking, but I think we should all play by the same rules.
>
> Best regards,
> Volker
>
> [1] http://openjdk.java.net/projects/#project-committer
>

Hi Volker,

Thanks for bringing this up. I don't think it's nit-picking; we should all
try and be on the same page when it comes to such things.

I wasn't aware of the document you refer to. My only reference
when writing the original e-mail, and others in the past, has been
the bylaws [0], which have no such prescription. I tend to concur
with what Mario and Andrew Dinn have already said so well, in that
this is intended as guidance, rather than a strict rule.

It is hard to interpret it as such without also defining "significant" in
some absolute way. Generally, what is regard a significant patch
by one person may not be by another. I can also easily see how
someone could easily produce more than eight patches of low
complexity in the time it may take them to produce one of a higher
complexity.

Moreover, I do not see the need for such strong barriers on making
someone a committer. If we were considering the post of reviewer,
I may be more stringent, but all we are offering is the ability to push
approved patches to the repositories. Unless we believe Martin
to be a rogue agent, I don't see a value in this. We are not suggesting
that he should be able to freely push patches without approval.
All delaying his approval as committer achieves is creating more
tedious work for others, as they then have to repeat his work of applying
and building the patch before pushing it on his behalf. I do think we
need to keep practicality and accessibility in mind, as well as strict
conformance to the rules.

What prompted me to post this yesterday is I also opened a similar
vote for Severin on 8u [1]. There, we have the even more bizarre situation
that someone who can push to OpenJDK 9 and later can not push to
OpenJDK 8, because, while such permissions are automatically carried
to new versions of the JDK project, they are not applied retrospectively
to older versions. This brought Martin to mind, and, to be fair to him, for
his part, he rather modestly thought it was too soon to be proposed.
Being aware of the work he has done, and is in the process of doing,
I thought otherwise.

It may well be that, by the time the two week voting period has expired,
eight patches have been committed. My experience is that this has more
to do with the availability of reviewers in the security space than anything
else. If however, you still disagree and wish to veto, I bear you no hard
feelings on the matter.

[0] http://openjdk.java.net/bylaws#committer
[1] http://mail.openjdk.java.net/pipermail/jdk8u-dev/2018-June/007588.html

Thanks,
-- 
Andrew :)

Senior Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

Web Site: http://fuseyism.com
Twitter: https://twitter.com/gnu_andrew_java
PGP Key: ed25519/0xCFDA0F9B35964222 (hkp://keys.gnupg.net)
Fingerprint = 5132 579D D154 0ED2 3E04  C5A0 CFDA 0F9B 3596 4222