Role of the Sponsorship Queue

[Emmet Hikory]

Daniel Holbach wrote:
> On 04.03.2010 04:57, Emmet Hikory wrote:
>>     Not quite, and in two ways.  Firstly, I don't think there is any
>> way that we can determine if someone is an Ubuntu Developer or not
>> without asking them (as there's no clear way to distinguide
>> Prospective Developers from arbitrary users, and no clear way to
>> distinguish Contributing Developres who didn't happen to chat with the
>> DMB from other Ubuntu Members).  Secondly, I'd really prefer to see
>> sponsors discovering bugs in the sponsors queue that ought be in the
>> reiviewers queue leave a bug comment like the following:
<...>
> The unfortunate thing is: it's the reality.
>
> 1) Attach a patch to a bug of a random package: effectively nobody will
> notice or care.
> 2) "Make it a debdiff", get it into the Sponsoring Queue: it will
> probably be applied.
>
> In your mail above you acknowledge how hard it is to make the
> distinction between the two "categories of contributors". The proposal
> to not allow patches from "non-developers" into Sponsoring does not even
> try to solve the problem of those 2000 waiting patches, but, hey, "we
> ought solve" it.
>
> What is this proposal and theoretical distinction good for? The proposal
> to not allow patches from "non-developers" into the Sponsoring Queue
> does not solve any problems we have.

    Let me put this a different way.  I think the entire concept of a
sponsorship queue is completely without merit unless we intend to
differentiate between classes of patches using this queue.  Asking
random folk to learn how to subscribe some team is just adding extra
burden on patch submitters.  Let's drop it completely if we're not
differentiating.

    Now, I happen to believe that it's worth having a sponsors queue.
I remember when we didn't have one.  We created it specifically so
that those developers who didn't yet have upload rights to a given
package could get priority for reviews of their candidate packages
(which at the time primarily consisted of collections of the
previously unreviewed patches).  If we don't wish to fast-track
uploads for developers, that's fine, but I believe that we'll end up
back in the days when people just repeatedly posted on IRC when they
had a debdiff, and I think that's a waste of the submitters time and
the time of anyone reading iRC (either in real-time or in logs).

-- 
Emmet HIKORY