1 |
John Lawles wrote: |
2 |
> |
3 |
> The concept is that there would be a separate team maintaining |
4 |
> the stable tree. A precedent for this would be the Linux kernel's |
5 |
> two-track development. |
6 |
> |
7 |
|
8 |
Actually, in theory that is what already happens with stable keywording |
9 |
- or is supposed to happen. It is a different group of developers who |
10 |
maintain the stable tree (well, there is overlap, but some people wear |
11 |
multiple hats). |
12 |
|
13 |
Expat was a real mess - mainly because they break ABI and it doesn't use |
14 |
slotting. I'm not sure why slotting couldn't have been used in this |
15 |
case (others might be able to comment on this). |
16 |
|
17 |
What we need is a better mechanism of warning users that they're about |
18 |
to do something that could cause them major headaches. ELOG is useless |
19 |
when you find out after the fact. |
20 |
|
21 |
revdep-rebuild working better would also be nice. If it won't work the |
22 |
ebuild should abort prior to install with a link to a document |
23 |
indicating what will need to be done, and then users can read and |
24 |
understand it before they break things. |
25 |
|
26 |
> As a mere end-user, I am not recommending any particular |
27 |
> solution. I am suggesting that more effort be devoted to |
28 |
> (a) recognizing user frustrations, (b) proposing good solutions, |
29 |
> and (c) letting us know that you see the problems and that |
30 |
> solutions are being worked on. |
31 |
> |
32 |
|
33 |
I agree that this is something that does need better attention in |
34 |
general. I find it a bit frustrating to have to carefully check my |
35 |
emerge -auD lists to ensure that the whole system won't blow up after |
36 |
the install - and prior to last Aug I'd never even heard of expat... |
37 |
-- |
38 |
gentoo-project@l.g.o mailing list |