1 |
On Fri, Apr 12, 2019 at 11:10:39AM -0500, William Hubbs wrote: |
2 |
> On Fri, Apr 12, 2019 at 04:40:43PM +0200, Michał Górny wrote: |
3 |
> > Update the wording of GLEP 48 to provide clear information on what kind |
4 |
> > of disciplinary actions QA can issue, and in what circumstances they can |
5 |
> > be exercised. Remove the unclear reference to ComRel that is either |
6 |
> > meaningless or violation of scope. |
7 |
> > |
8 |
> > According to the old wording, QA could request 're-evaluating commit |
9 |
> > rights' from ComRel. This is very unclear, and has been a source of |
10 |
> > confusion more than once. Firstly, it is unclear whether ComRel merely |
11 |
> > serves as a body executing the QA team's decision, or whether it is |
12 |
> > supposed to make independent judgment (which would be outside its |
13 |
> > scope). Secondly, it suggests that the only disciplinary action |
14 |
> > possible would be 're-evaluating commits rights' which sounds like |
15 |
> > an euphemism for removing commit access permanently. |
16 |
> > |
17 |
> > The new wording aims to make things clear, and make QA disciplinary |
18 |
> > actions independent of ComRel. Explanation for the individual points |
19 |
> > follow. |
20 |
> |
21 |
> I would support this because in the very earliest days of the qa team, |
22 |
> qa actions were independent of ComRel. |
23 |
> |
24 |
> I have no specific comrel action that I have issues with, but QA actions |
25 |
> should be independent of ComRel. If there are any reasons to belive that |
26 |
> the qa team abuses this, that would be a point where ComRel or the |
27 |
> council could get involved. |
28 |
|
29 |
There are things I would definitely change about the past though. |
30 |
|
31 |
In the past, the qa lead was the one who could go directly to |
32 |
infrastructure on his own and ask for people to be blocked. I think the |
33 |
only way this should happen is after a majority vote on the qa team. |
34 |
|
35 |
Also, this definitely should not be the first step the qa team takes. |
36 |
|
37 |
William |