1 |
On Mon, Oct 30, 2017 at 7:50 PM, Andreas K. Huettel |
2 |
<dilfridge@g.o> wrote: |
3 |
> Am Dienstag, 31. Oktober 2017, 00:17:38 CET schrieb Michał Górny: |
4 |
>> |
5 |
>> We've put a significant effort to make it convenient to use GitHub to |
6 |
>> contribute to Gentoo. Many users have appreciated that, and so did many |
7 |
>> developers. I think we can call that a success. However, I don't think |
8 |
>> that's really successful enough. |
9 |
>> |
10 |
>> The main goal for using GitHub was to make it easy both for users to |
11 |
>> submit their contributions, and for Gentoo developers to review |
12 |
>> and merge them. However, for that to work out we'd actually have to have |
13 |
>> more Gentoo developers *care* and we don't have that. |
14 |
> |
15 |
> At the moment we have 22508 open bugs in bugzilla... should we maybe close |
16 |
> that too? |
17 |
|
18 |
++ |
19 |
|
20 |
Open bugs/PRs that are no longer relevant are noise, but open bugs/PRs |
21 |
that are still valid are actually signal. The more of these that you |
22 |
leave open, the greater the chance that one will get fixed. |
23 |
|
24 |
Sure, lots of ignored issues does say something about the project, but |
25 |
I think we also need to manage the expectation that volunteers will |
26 |
always fix issues the moment they're reported. If one checks in once |
27 |
every six months to clean up some bugs, that is still better than if |
28 |
they didn't check in at all. Simply closing real issues is treating |
29 |
the symptom and not the underlying issue (which probably can only be |
30 |
treated if people step up, and having PRs probably makes that easier |
31 |
if anything). |
32 |
|
33 |
The biggest concern (IMO) would be if an issue languishes as a PR when |
34 |
it might be fixed if it were a bug. I think some automation might |
35 |
help there, but I am skeptical of the idea that somebody who |
36 |
completely ignores a PR would suddenly start paying attention if it |
37 |
were a bug. |
38 |
|
39 |
-- |
40 |
Rich |