| 1 |
On Fri, 09 May 2014 17:43:43 +0000 |
| 2 |
hasufell <hasufell@g.o> wrote: |
| 3 |
|
| 4 |
> I have problems believing in QA competence when I read comments like |
| 5 |
> these: |
| 6 |
> |
| 7 |
> https://bugs.gentoo.org/show_bug.cgi?id=473#c14 |
| 8 |
> https://bugs.gentoo.org/show_bug.cgi?id=473#c17 |
| 9 |
|
| 10 |
What's the point about quality assurance if we only add bugs but never |
| 11 |
fix them; we've already have a huge backlog of bugs, so, adding more of |
| 12 |
them isn't going to get them fixed any time soon. |
| 13 |
|
| 14 |
And if anyone has the server hardware and server administration skills |
| 15 |
available to set this up, nobody stops him; but you can't expect the QA |
| 16 |
team to set this up when we have a huge backlog of other bugs to do. |
| 17 |
|
| 18 |
> You even blocked me out of the bug. I mean... wtf is happening here? |
| 19 |
|
| 20 |
The bug is expecting a QA team response; not a random "lolwat?" |
| 21 |
comment, neither a random comment about locking down enter_bug.cgi that |
| 22 |
doesn't have to do with the bug at all. These aren't according to CoC. |
| 23 |
|
| 24 |
Thus the bug is temporarily marked as QA-only, awaiting a decision. |
| 25 |
|
| 26 |
> I'm not even sure the list can read those bug comments to understand |
| 27 |
> what I mean. I used to believe it's lack of manpower/time or even |
| 28 |
> laziness, but I am not sure anymore. |
| 29 |
|
| 30 |
You seem to be confused. consider the full image of what Tinderbox does |
| 31 |
as well as the current state of the QA related bugs in Bugzilla; while |
| 32 |
it is nice to have, it is not a reason to drop all of our other work. |
| 33 |
|
| 34 |
> Funnily... the meeting summary of march is lost where you were |
| 35 |
> supposed to vote on the tinderbox matter. |
| 36 |
|
| 37 |
Nothing to vote/discuss on, lack of server hardware and administration. |
| 38 |
|
| 39 |
> I am not sure either if that is "by accident" anymore. I pinged you |
| 40 |
> guys often enough. |
| 41 |
|
| 42 |
The people assigned to it are busy, they have been reminded about that; |
| 43 |
we try to adapt to a model to do it during the meeting, which happened |
| 44 |
to the very first and very last meeting and works out well. |
| 45 |
|
| 46 |
> No one seems to have the logs anymore. |
| 47 |
|
| 48 |
We do have the logs, just not the summary; created 23 days ago: |
| 49 |
|
| 50 |
https://gist.github.com/TomWij/9d033e0fbb5ce4e43568 |
| 51 |
|
| 52 |
> Anyway, I had less trouble getting responses from the old QA team. |
| 53 |
> When I ask you guys, I usually get one of these responses (or |
| 54 |
> similar): |
| 55 |
> * nothing, unless I ping you after 2 weeks again |
| 56 |
|
| 57 |
Similar to bugs, sometimes you don't get a response on the first try; |
| 58 |
this especially happens when everyone thinks that someone else of the |
| 59 |
team will reply, in the end everyone forgets it in the pile of e-mails. |
| 60 |
|
| 61 |
> * "probably not in QAs scope" |
| 62 |
|
| 63 |
QA gets CC-ed a lot on things that are outside our scope; these are |
| 64 |
either cases where the maintainer has to do something and QA has no |
| 65 |
policy to act on, ... |
| 66 |
|
| 67 |
> * "post on dev ML first, we don't know what to think" |
| 68 |
|
| 69 |
... cases that might be controversial and need a dev ML discussion ... |
| 70 |
|
| 71 |
> * "not sure if that's our business, appeal to council" |
| 72 |
|
| 73 |
... or cases other than Portage tree policy, consistency or breakage. |
| 74 |
|
| 75 |
> Not sure what's worth QA if they don't have strong opinions and try to |
| 76 |
> avoid to piss off people. |
| 77 |
|
| 78 |
We're not to be used as a weapon to piss off a maintainer; the QA team |
| 79 |
looks into the best interest of all developers, not the developer that |
| 80 |
happened to CC us first and wants to force the assignee to do a thing |
| 81 |
for which a policy does not even exist or there's no breakage or an |
| 82 |
uncontroversial inconsistency. |
| 83 |
|
| 84 |
> I used to ask QA for guidance beforehand on delicate matters before I |
| 85 |
> escalated elsewhere or chose a particular approach. That doesn't |
| 86 |
> really work anymore, instead people expect me to fire up the |
| 87 |
> bikeshedding dev-ML which is the worst place to get ideas and do |
| 88 |
> constructive brainstorming. |
| 89 |
|
| 90 |
If you view it as the worst place, it'll continue the worst place; get |
| 91 |
your hopes up, because it's actually only proper escalation that works. |
| 92 |
|
| 93 |
> I expect infra to unblock bug 473 now. This is unacceptable. |
| 94 |
|
| 95 |
Well, if you prefer improper escalation over proper discussion; fine. |
| 96 |
|
| 97 |
-- |
| 98 |
With kind regards, |
| 99 |
|
| 100 |
Tom Wijsman (TomWij) |
| 101 |
Gentoo Developer |
| 102 |
|
| 103 |
E-mail address : TomWij@g.o |
| 104 |
GPG Public Key : 6D34E57D |
| 105 |
GPG Fingerprint : C165 AF18 AB4C 400B C3D2 ABF0 95B2 1FCD 6D34 E57D |
Archives