Re-aligning the Ubuntu Developer Process

Michael Bienia michael at bienia.de
Fri Jul 8 17:44:03 UTC 2011


On 2011-07-05 11:30:09 -0700, Jono Bacon wrote:
Hello,

first it would be good if this mail also went to the DMB even if it is
just to collect initial input from the TB. Otherwise it looks like the
DMB will only get notified about the outcome/decision without any real
chance to comment.

> I want to propose a few changes to our developer process. Recently there
> has been concerns expressed that the process is difficult to get
> through, even for people with extensive knowledge.

Can you be more specific which "requirements" are seen to difficult?
But I agree that some parts could be improved but I don't any ideas that
would work out.

> Today I feel our process is too focused on showcasing work as opposed to
> harnessing reputation - I have always felt that a +1 from a respected
> community member holds more weight than a list of bugs fixed on a wiki
> page. I believe that adjusting our process to focus on reputation will
> streamline it and reduce the risk of good people getting rejected
> because their body of work was not in the specific form that a DMB
> member expects.

I'd be willing to give good testimonials a heigher weight if there were
more testimonials in the applications. And a lack of comments or
testimonials can be seen as a bad integration into the development
community.

[...]
> With this in mind I would like to propose two modifications to the
> current assess process for each of these areas:
> 
>      1. Core-Dev Testimonials - for each candidate they should gather
>         testimonials from core-devs. These testimonials should account
>         for the vast majority of the assessment. I believe that if
>         someone gets two core-dev +1s for approval, there should be a
>         good reason if the DMB wants to reject the candidate.
>      2. Brief Core-Devs - we should make it clear that getting a +1 from
>         a core-dev is key part of the assessment, and core-devs should
>         expect to provide honest and fair testimonials as part of they
>         work as a core-dev. We could even provide some kind of
>         application queue for those requesting testimonials.

Doesn't this move the problem towards core-dev?
Some applicants have a nice long list of testimonals which makes
processing an application easier but there are also applicants with only
a few testimonials (2 or 3) which makes it harder to judge how well an
applicant is integrated into the development community.

I remember a recent case where an applicant for PPU rights considered
about aborting his applications because he couldn't get any
testimonials.

How should that application queue work? I read in some testimonials that
the commenter doesn't have an opinion about an applicant because he
didn't sponsored enough work from the applicant. How should a random
core-dev who didn't work together with the applicant judge his
reputation?


> As such, I would recommend that the process would work like this:
> 
>       * A candidate wishes to apply to be a dev. They outline their body
>         of work in a wiki page and request testimoniald from core-devs.
>       * In the meeting the DMB look at the testimonials first and if
>         there at least are two core-devs who +1 the candidate the
>         application should be considered approved unless there is a good
>         reason to suggest this confidence is unwarranted.
>       * If a candidate can't get testimonials from core-devs, it is the
>         DMBs discretion if the candidate should be approved.

I think that will make the DMB useless as we don't need a whole team to
check if an applicant got two +1 from core-devs.

If an applicant can't get two +1 from core-devs on which basis should he
get approved nonetheless? Apparently the applicant doesn't have enough
reputation yet to get a +1 from a core-dev. How can he then be suitable
for upload rights?

> I believe that optimizing this process around reputation (in the form of
> +1s from core-devs) will help streamline the approval process, and I
> have confidence that core-devs won't arbitrarily +1 people as they would
> not want to cast their own reputations in doubt. This should result in a
> high-confidence level, and ensure that good people with decent bodies of
> experience seen by other community members get approved.

I think if a +1 from core-dev counts already as "half-accepted" than
some core-devs will be more hesitant to give his testimonials at all or
give testimonials that would count as "0" (neither approval nor
rejection) which don't help the applicant either.

>From the time when the MC processed applications by mail I still
remember that we had to repeatedly ping sponsors and ask for
comments/testimonials. I don't believe that this has improved now with
the wiki page and it won't be easier in future either.

I believe applicants will get bottlenecked on waiting on two core-devs
confident enough about an applicant to give their +1.

> Thoughts?

A more general comment:
I've always had a hard time when dealing with applications from
Canonical employees as I had a feeling that an approval is "expected" or
at least some requirements waived through.
Perhaps this is just a feeling I've without any real background but it
makes dealing with those applications not easier knowing that those
applicants "need" the upload rights for their day work and on the other
hand don't judge Canonical employees and community members different.

Michael



More information about the technical-board mailing list