1 |
On Tue, Jul 06, 2021 at 09:02:14AM +0200, Ulrich Mueller wrote: |
2 |
> >>>>> On Mon, 05 Jul 2021, Aaron Bauman wrote: |
3 |
> |
4 |
> > On Mon, Jul 05, 2021 at 11:42:36PM +0200, Ulrich Mueller wrote: |
5 |
> >> The point is that banning an EAPI doesn't have any noticeable effect. |
6 |
> >> For example, if you look at EAPI 0 (banned on 2016-01-10) and EAPI 4 |
7 |
> >> (banned on 2018-04-08), there's neither a cusp nor any change of slope |
8 |
> >> visible for the curves plotted here: |
9 |
> >> https://www.akhuettel.de/~huettel/plots/eapi.php |
10 |
> |
11 |
> > We have had the issue in the past. Taking 1 minute to vote on a ban of |
12 |
> > an EAPI in a council meeting is a minute task and enforces the goals |
13 |
> > of the project. |
14 |
> |
15 |
> But we don't actually enforce the ban. Doing so would mean adding the |
16 |
> EAPI to eapis-banned in layout.conf, which in turn would imply that |
17 |
> commits of (new or existing) ebuilds with a banned EAPI would be |
18 |
> rejected. |
19 |
> |
20 |
> Ulrich |
21 |
|
22 |
I know we don't technically enforce the ban as there is no mechanism to |
23 |
do so. It is a terrible in-between. |
24 |
|
25 |
So, it is a formality that takes a minute to vote on. Then, if someone |
26 |
should be an asshole and keep committing things that are deprecated it |
27 |
can be enforced by the QA team or other developers. |
28 |
|
29 |
I ask to keep the formality until a technical solution is in place as we |
30 |
*have* had the issue before. While it is not statistically significant |
31 |
it did cause undue pain for others. |
32 |
|
33 |
-Aaron |