[gentoo-qa] QA power to suspend access [WAS: Re: Roll-call for the team members and requesting a new election]

[gentoo-qa] QA power to suspend access [WAS: Re: Roll-call for the team members and requesting a new election]

[Mark Loeser]

[-- Attachment #1: Type: text/plain, Size: 1528 bytes --]

Splitting this since Piotr brought up something I wanted to.

Piotr Jaroszyński <p.jaroszynski@gmail.com> said:
> On 21 August 2010 19:46, Diego Elio “Flameeyes” <flameeyes@gmail.com> wrote:
> > The role of the lead in this project cannot be simply a "supervising"
> > one, given that it's the only member of the team that devrel is
> > interested in hearing from (situation that, by itself, should probably
> > be changed); it requires actual leading and decision-taking, especially
> > given the current situation where many policies have never been written
> > down, and our documentation resources for developers are scattered and
> > outdated.
> 
> I agree that this should probably be changed. I somehow recall 2 QA
> members being able to make decisions on behalf of QA if quick action
> is necessary.

This was brought up in a recent council meeting as well.  I just haven't
had a chance to write up an email to the list since I've been on
vacation.

2 ideas were presented:

1) Have someone else that those powers can be delegated to (a co-lead or
something)

2) If 2 QA members rule for suspension of access, then it is done.

#2 seems to alleviate any bottlenecks that could happen.  What does
everyone else think?  This can wait until after whatever is decided with
the election, but the conversation should take place regardless.

-- 
Mark Loeser
email         -   halcy0n AT gentoo DOT org
email         -   mark AT halcy0n DOT com
web           -   http://www.halcy0n.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-qa] QA power to suspend access [WAS: Re: Roll-call for the team members and requesting a new election]

[Diego Elio “Flameeyes” Pettenò]

Il giorno sab, 21/08/2010 alle 14.11 -0400, Mark Loeser ha scritto:
> 
> #2 seems to alleviate any bottlenecks that could happen.  What does
> everyone else think?  This can wait until after whatever is decided
> with
> the election, but the conversation should take place regardless. 

I agree that #2 sounds the best.

In case we all agree on that, should we ask devrel nicely to write that
down on their own policy or do we have to pass through council?

-- 
Diego Elio Pettenò — “Flameeyes”
http://blog.flameeyes.eu/

If you found a .asc file in this mail and know not what it is,
it's a GnuPG digital signature: http://www.gnupg.org/

Re: [gentoo-qa] QA power to suspend access [WAS: Re: Roll-call for the team members and requesting a new election]

[Mark Loeser]

[-- Attachment #1: Type: text/plain, Size: 1531 bytes --]

Diego Elio “Flameeyes” Pettenò <flameeyes@gmail.com> said:
> Il giorno sab, 21/08/2010 alle 14.11 -0400, Mark Loeser ha scritto:
> > 
> > #2 seems to alleviate any bottlenecks that could happen.  What does
> > everyone else think?  This can wait until after whatever is decided
> > with
> > the election, but the conversation should take place regardless. 
> 
> I agree that #2 sounds the best.
> 
> In case we all agree on that, should we ask devrel nicely to write that
> down on their own policy or do we have to pass through council?

Well, we seem to have an agreement from the majority here, of 2.  I'll
push this through the council at the next meeting since it was my
assignment to bring the discussion up.  The exact text that will be
replace is:

  If a particular developer persistently causes breakage, the QA team may
  request that devrel re-evaluates that developer's commit rights.
  Evidence of past breakages will be presented with this request to
  devrel.

It will now become:

  If a particular developer persistently causes breakage, the QA team
  can request infra to revoke the developer's commit rights, and will
  contact devrel to help mediate the case.  The QA team can request that
  the commit rights be revoked either by the lead asking, or 2 members
  of the QA team requesting such an action be taken.

Thanks,

-- 
Mark Loeser
email         -   halcy0n AT gentoo DOT org
email         -   mark AT halcy0n DOT com
web           -   http://www.halcy0n.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-qa] QA power to suspend access [WAS: Re: Roll-call for the team members and requesting a new election]

[Petteri Räty]

[-- Attachment #1: Type: text/plain, Size: 672 bytes --]

On 09/03/2010 06:12 PM, Mark Loeser wrote:
> 
>   If a particular developer persistently causes breakage, the QA team
>   can request infra to revoke the developer's commit rights, and will
>   contact devrel to help mediate the case.  The QA team can request that
>   the commit rights be revoked either by the lead asking, or 2 members
>   of the QA team requesting such an action be taken.
> 

It doesn't require infra as DevRel has the needed powers to shut down
access. DevRel is needed any way to process the retirement. I think
infra should only be the fallback if DevRel people with needed access
are not available fast enough.

Regards,
Petteri


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 900 bytes --]

Re: [gentoo-qa] QA power to suspend access [WAS: Re: Roll-call for the team members and requesting a new election]

[Diego Elio Pettenò]

[-- Attachment #1: Type: text/plain, Size: 1143 bytes --]

Il giorno ven, 03/09/2010 alle 19.43 +0300, Petteri Räty ha scritto:
> 
> It doesn't require infra as DevRel has the needed powers to shut down
> access. DevRel is needed any way to process the retirement. I think
> infra should only be the fallback if DevRel people with needed access
> are not available fast enough.

If you couldn't notice, we're explicitly trying to get this done without
DevRel intervention at first.

Right now we have something along these lines:

QA team member finds the screwup → reports to QA lead → ask nicely the
devrel read → devrel can sit thinking about it.

In the mean time, the person who was originally caught can keep
committing, eventually bringing up a problem worse than the previous
one, and so on so forth.

What _I_ would like to see for QA? (and think others would agree)

QA team member finds the screwup → track down another one that agrees on
the problem being big enough → they ask infra to suspend access → devrel
can either negotiate for the reinstatement or proceed with retirement.

-- 
Diego Elio Pettenò — Flameeyes
http://blog.flameeyes.eu/


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 490 bytes --]

Re: [gentoo-qa] QA power to suspend access [WAS: Re: Roll-call for the team members and requesting a new election]

[Petteri Räty]

[-- Attachment #1: Type: text/plain, Size: 809 bytes --]

On 09/03/2010 10:06 PM, Diego Elio Pettenò wrote:
> Il giorno ven, 03/09/2010 alle 19.43 +0300, Petteri Räty ha scritto:
>>
>> It doesn't require infra as DevRel has the needed powers to shut down
>> access. DevRel is needed any way to process the retirement. I think
>> infra should only be the fallback if DevRel people with needed access
>> are not available fast enough.
> 
> If you couldn't notice, we're explicitly trying to get this done without
> DevRel intervention at first.
> 
> Right now we have something along these lines:
> 
> QA team member finds the screwup → reports to QA lead → ask nicely the
> devrel read → devrel can sit thinking about it.
> 

I didn't propose allowing DevRel to refuse. I find it simpler to have
less people involved.

Regards,
Petteri


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 900 bytes --]