Re: [Libreoffice-qa] QA Field in FDO - What's the Purpose (discussion)

Joel Madero píše v Po 18. 03. 2013 v 11:38 -0700:
> During our last QA call we had a continuing discussion about the QA
> field that Rainer began a few weeks back [1]
>
>
> Basically the conversation is "what's the point of the QA field"
>
>
> There are multiple opinions:
>
>
> 1. QA Field should represent exactly what Dev field represents, ie. QA
> contact is responsible for everything QA related, if further testing
> is asked for, bibisect etc...QA contact should do it

I think that this is the intuitive explanation.


> 2. QA contact is just a way to signal "QA has looked at this, priority
> has been set (most of the time)" but there is no obligation for the QA
> contact to monitor bug further - usually they will as they are
> receiving emails and are familiar with the bug but other QA people are
> able to update and get further details to devs.

IMHO, this is not intuitive. Also you would get too many mails from
bugzilla. It would be hard to distinguish what need your reaction
(bugtriaging) and what is related to the other stages (fixing).


> 3. QA contact is a combination of #1 and #2, they should do
> "everything in their power" to keep up with the bug and if they are
> unable to provide something, should find someone who can.

I am afraid that it would not work as it did not work with developers.
Some persons might be overhelmed. It does not solve the problem when a
person stops contributing at all.

I personally would be afraid to become fully responsible for more than
200[*] bugs because then I would not be able to react on time, get
stressed, and disappointed with that job. 

[*]The number depends on how much time you have to spend with the project,
how much you are perfectionist, etc.


I would suggest to keep the two things separated and use something else
to mark prioritized bugs, for example whiteboard status. We might get
rid of the whiteboard status in the long run, when people get used to
prioritization and most confirmed bugs have reasonable priority and
severity.

We need not think about perfect solution here. Bugzilla newer will be
100% prioritized. Some users will change it even for already
prioritized bugs. Even bug triagers need not be objective because it is
often not easy to know how many users are affected and what are all
side effects.

I think that prioritization is worth the effort but we should not be
too strict about it. Otherwise, we spend more time fighting about
priorities than with solving the bugs :-)


Best Regards,
Petr