Gentoo Archives: gentoo-project

From: "Michał Górny" <mgorny@g.o>
To: gentoo-project@l.g.o
Cc: qa <qa@g.o>, comrel <comrel@g.o>
Subject: Re: [gentoo-project] [PATCH] glep-0048: Provide clear rules for disciplinary actions
Date: Thu, 18 Apr 2019 12:10:28
Message-Id: fe4729d9bb00f58b5eaa3890897cc20eabbe6707.camel@gentoo.org
In Reply to: Re: [gentoo-project] [PATCH] glep-0048: Provide clear rules for disciplinary actions by Alec Warner
1 On Fri, 2019-04-12 at 11:30 -0400, Alec Warner wrote:
2 > On Fri, Apr 12, 2019 at 10:40 AM Michał Górny <mgorny@g.o> wrote:
3 >
4 > > Update the wording of GLEP 48 to provide clear information on what kind
5 > > of disciplinary actions QA can issue, and in what circumstances they can
6 > > be exercised. Remove the unclear reference to ComRel that is either
7 > > meaningless or violation of scope.
8 > >
9 >
10 > Is there a particular driver for this change? E.g. have you been
11 > dissatisfied with the current procedure, or perhaps Comrel is not acting on
12 > your referrals?
13 >
14
15 Ok, you wanted a driver, so here's some data for the most recent QA ban.
16
17 QA finished voting within one day (~8 hours, to be precise), and
18 requested ComRel to implement the ban in the European evening
19 of the same day. ComRel needed whole 4 days to vote for implementing
20 this. In the end, according to the data presented to us two ComRel
21 members have voted against implementing this, and three wanted the ban
22 length changed.
23
24 This proves the existence of two problems.
25
26 Firstly, the presence of ComRel in the pipeline has delayed the ban four
27 days. You might consider this insignificant if you presume that
28 the purpose of the ban is to satisfy somebody's petty vengeance, as it
29 is frequent for ComRel cases. However, I believe that the ban makes
30 only sense if it aims to discipline the developer in question,
31 and disciplining makes sense only if you act reasonably promptly.
32
33 The later the ban is established, the more likely it will have adverse
34 effect of 'I understood my mistakes, I apologized, I improved my
35 behavior and you're banning me *now*'. If there were no ComRel
36 in the pipeline, the ban would be set ~4 days earlier, and finished ~4
37 days earlier instead of putting the developer in silly position
38 of knowing that he's been banned but waiting for another body to confirm
39 it.
40
41 Secondly and more importantly, ComRel has failed to vote unanimously.
42 This clearly indicates that at least some of the ComRel developers
43 do not respect QA's jurisdiction over technical issues, and believe that
44 they have better authority to judge the technical actions of developers.
45 If that is the case, I'm wondering why those developers haven't
46 volunteered to join QA and express their opinion in the initial vote,
47 and instead chose to try to override QA's decision.
48
49 In my opinion, the developers who voted 'no' should simply give up their
50 ComRel hats because they have clearly abused their ComRel position,
51 in order to make judgment outside their jurisdiction, and therefore
52 ridiculed the role given to them by GLEP 48.
53
54 Of course, unless they believe that QA has abused its powers in making
55 the vote. But then, I would have to ask why those ComRel members
56 decided to address this by silently voting 'no' rather than bringing
57 this problem to light.
58
59 To summarize, I believe the arbitrary and artificial presence of ComRel
60 in GLEP 48 is simply harmful. Furthermore, if ComRel was indeed added
61 because of potential for unprofessional behavior of QA team,
62 the evidence proves that such additional verification should be added
63 for ComRel instead.
64
65 --
66 Best regards,
67 Michał Górny

Attachments

File name MIME type
signature.asc application/pgp-signature

Replies