1 |
On Monday 02 February 2004 3:17 pm, Kurt Lieber wrote: |
2 |
> All -- |
3 |
> |
4 |
> I've posted GLEP 19 which talks about the inclusion of a new 'stable' tree |
5 |
> in portage that is updated on a periodic basis and only contains security |
6 |
> and major bugfixes out of cycle. |
7 |
> |
8 |
> http://www.gentoo.org/proj/en/glep/glep-0019.html |
9 |
> |
10 |
> Please take a moment to review the GLEP and offer any feedback or ask any |
11 |
> questions. |
12 |
> |
13 |
> --kurt |
14 |
|
15 |
Hi Kurt, |
16 |
|
17 |
First off, it's great that someone's finally done something about this. |
18 |
People have been talking about stable / unstable trees since before I first |
19 |
joined the project, with nothing actually done to implement it. |
20 |
|
21 |
The main suggestion I have is to drop the word 'stable'. It's an imprecise |
22 |
term that means different things to different people. For example, should an |
23 |
ebuild be marked stable because it works, or because the package itself |
24 |
works? It doesn't matter what the official policy is - you'll always have a |
25 |
gray area with the word 'stable'. |
26 |
|
27 |
If you want an 'enterprise'-grade tree, why not call it precisely that? ;-) |
28 |
Heh - then we can get a 'carrier'-grade tree too (yes, they really do call it |
29 |
that ;-) ;-) |
30 |
|
31 |
How are you going to manage whether something gets into your enterprise tree |
32 |
or not? I spent part of Saturday night fixing up mod_php ebuilds which |
33 |
someone had marked stable on an arch that was missing dependencies. It's |
34 |
just one example of the trouble we have getting people to use repoman. With |
35 |
the current tree - and expectations - these things don't matter that much. |
36 |
But for 'enterprise' level quality and higher ... |
37 |
|
38 |
Best regards, |
39 |
Stu |
40 |
-- |
41 |
Stuart Herbert stuart@g.o |
42 |
Gentoo Developer http://www.gentoo.org/ |
43 |
Beta packages for download http://dev.gentoo.org/~stuart/packages/ |
44 |
Come and meet me in March 2004 http://www.phparch.com/cruise/ |
45 |
|
46 |
GnuGP key id# F9AFC57C available from http://pgp.mit.edu |
47 |
Key fingerprint = 31FB 50D4 1F88 E227 F319 C549 0C2F 80BA F9AF C57C |
48 |
-- |