| 1 |
On Sun, Nov 27, 2016 at 5:56 AM, Raymond Jennings <shentino@×××××.com> wrote: |
| 2 |
> On Tue, Nov 22, 2016 at 12:06 AM, Alice Ferrazzi <alicef@g.o> wrote: |
| 3 |
>> |
| 4 |
>> What about maintainers that are away without writing it in their |
| 5 |
>> maintainer bug ? |
| 6 |
>> After how many days of no replay can be fair to touch their package ? |
| 7 |
> |
| 8 |
> If a developer lapses long enough and doesn't use devaway properly to mark |
| 9 |
> any waylaid packages as touchable...then that's probably an example of an |
| 10 |
> even bigger fish to fry that undertakers would handle anyway. |
| 11 |
|
| 12 |
A maintainer can be actively doing other work and still not respond to |
| 13 |
a stable request bug. The only thing the undertakers could do about |
| 14 |
it is get rid of them, which stops the work they were actively doing |
| 15 |
and doesn't make the situation with the bug they were ignoring any |
| 16 |
better. |
| 17 |
|
| 18 |
Are devs supposed to ignore stable request bugs? No. Has anybody |
| 19 |
come up with a way to make them not do it? Unfortunately not. Part |
| 20 |
of the issue is that some devs are just somewhat antisocial and prefer |
| 21 |
to do their own thing. For the most part as long as they're not |
| 22 |
actually actively making trouble for others we tend to accept this, |
| 23 |
since the only visible change to getting rid of them is less stuff |
| 24 |
getting done (the stuff they passively ignored still ends up being |
| 25 |
passively ignored). |
| 26 |
|
| 27 |
This is why we tend to favor procedures that don't block progress by |
| 28 |
default. Just set a timeout. If the maintainer doesn't respond |
| 29 |
within x days then stabilization can proceed. Maybe make an exception |
| 30 |
for @system. We do similar things when devs want to touch each |
| 31 |
other's packages; if you don't get a response the assumption is that |
| 32 |
you can just go ahead. |
| 33 |
|
| 34 |
-- |
| 35 |
Rich |
Archives