1 |
On 10/24/13 10:02 AM, Rich Freeman wrote: |
2 |
> The situation causing issues is a different one. |
3 |
> |
4 |
> 1. Tree contains libfoo-1. |
5 |
> 2. Maintainer introduces libfoo-2, dropping keywords due to |
6 |
> significant changes. |
7 |
> 3. Major archs keyword libfoo-2. Tree is generally updated to work |
8 |
> with libfoo-2. |
9 |
> 4. Maintainer wants to drop libfoo-1, but libfoo-2 is not keyworded |
10 |
> on all the minor archs libfoo-1 is. |
11 |
> 5. Maintainer logs KEYWORDREQ for libfoo-1, and it is ignored. |
12 |
> |
13 |
> This leaves us with several options: |
14 |
> a. Don't let the maintainer remove libfoo-1. |
15 |
> 1. Maintainer continues to care for libfoo-1 - extra work. |
16 |
|
17 |
This can be especially hard if upstreams start removing support for |
18 |
libfoo-1, or removing support for their package versions that depend on |
19 |
or even require libfoo-1. |
20 |
|
21 |
> 2. Maintainer is required to care for libfoo-1 but doesn't - complaints/etc. |
22 |
|
23 |
Exactly - being "required" to maintain something doesn't make that |
24 |
person actually maintain it. |
25 |
|
26 |
> 3. Maintainer gets tired of dealing with QA and stops maintaining |
27 |
> libfoo entirely. Now major arch users are at a loss, and the minor |
28 |
> arch users are no better off. |
29 |
|
30 |
The worst of both worlds. I hope that doesn't happen often but yeah, |
31 |
people should realize this is a very plausible reaction to such pressures. |
32 |
|
33 |
> b. Let the maintainer remove libfoo-1. Minor arch users have large |
34 |
> numbers of packages break, have cascading keyword removal (no better, |
35 |
> really). |
36 |
> |
37 |
> I'm open to suggestions for other options if anybody has them. I'm |
38 |
> suggesting to allow the option of b. Maintainers can still choose to |
39 |
> do 1a, and minor arch users are really no worse off than if they were |
40 |
> to do 1b/c. |
41 |
|
42 |
Totally agreed. Maybe last rite hard mask libfoo-1 before removal to |
43 |
"give people time" to start maintaining it. It'll probably make it even |
44 |
more clear that there is no one to do that. |
45 |
|
46 |
Paweł |