1 |
On Mon, Jan 19, 2015 at 4:33 AM, Sergey Popov <pinkbyte@g.o> wrote: |
2 |
> |
3 |
> Do not get me wrong, Patrick. You, as QA team member, can touch other's |
4 |
> packages without prior noticing, if fixing serious issues involved. But |
5 |
> with great power comes great responsibility. Please, use your power more |
6 |
> wisely next time. |
7 |
> |
8 |
|
9 |
This wasn't a QA commit. |
10 |
|
11 |
QA modifications should always be noted as such in the |
12 |
commit/changelogs/etc. If you revert/etc one of these commits expect |
13 |
to be called on it, and you had better have a REALLY good reason for |
14 |
it. You're best off pinging somebody in QA if you have an issue with |
15 |
a QA commit, and working with them. If somebody feels QA commits are |
16 |
being abused they should reach out to the QA lead, or ultimately |
17 |
Comrel/Council if things can't be worked out - as you say with great |
18 |
power comes great responsibility. |
19 |
|
20 |
Commits made by people who happen to be members of QA that aren't |
21 |
labeled as QA commits are the same as commits made by any random |
22 |
developer, and should generally follow the same processes. That means |
23 |
working out things 1:1 if possible, and if not going through the |
24 |
normal Comrel process. |
25 |
|
26 |
The QA team really had nothing to do with this commit. I don't think |
27 |
bringing up QA is particularly helpful here. |
28 |
|
29 |
However, in general developers should always work with maintainers |
30 |
when modifying their packages, especially for things like bumps. It |
31 |
isn't always practical to consult with individual developers for |
32 |
tree-wide work, but this kind of work should generally be announced on |
33 |
the appropriate lists and so on, and will often use tracker bugs and |
34 |
the like. It sounds like neither was done here. However, in these |
35 |
sorts of situations it is probably better to let Comrel do their job |
36 |
(and appeal to Council if you're unsatisfied with it), rather than |
37 |
having everybody bicker on the list. |
38 |
|
39 |
-- |
40 |
Rich |