1 |
On Tue, 2004-02-03 at 05:04, Dan Armak wrote: |
2 |
> A real separate cvs branch seems like a lot of extra work; most updates going |
3 |
> into the stable branch will probably also go into the main tree. What am I |
4 |
> missing? |
5 |
|
6 |
I would think the better way would be a separate CVS branch in which |
7 |
only specific ebuilds are added. I tend to look at the "stable" tree as |
8 |
a separate entity from the "regular" portage tree. |
9 |
|
10 |
> About keyword naming, I agree with Stuart's note elsewhere in the thread that |
11 |
> 'stable' is misleading. I also want to ask how the transition of |
12 |
> arch-->~stable:arch-->stable:arch is different from the existing transition |
13 |
> of ~arch-->arch. |
14 |
|
15 |
I see no point in implementing *any* new keywords. In fact, I could see |
16 |
instead *removing* the ~arch from the "stable" branch and keeping arch |
17 |
only. After all, we should not be adding any "testing" ebuilds to the |
18 |
stable tree. |
19 |
|
20 |
I like the idea of having the stable tree be separate from the updates. |
21 |
In fact, I pretty much see this as a requirement. The updates tree |
22 |
*could* use ~arch, especially in the case of new exploits which require |
23 |
new package versions from the upstream authors to resolve. |
24 |
|
25 |
Also, can we drop the idea of "stable"? It does not fit the audience |
26 |
that it seems we're shooting for at all. I would think "enterprise" is |
27 |
much more fitting, as suggested by others before myself. |
28 |
|
29 |
> If it isn't different and it's just a matter of the package being more and |
30 |
> more tested and used and proven without known (unfixed) bugs/vulnerabilites, |
31 |
> I don't think it's appropriate to create keywords by adding several modifiers |
32 |
> to an arch's name (~ and stable). We're not really combining the properties |
33 |
> of ~ and 'stable', and might as well assign stability levels with keywords |
34 |
> like 0:x86 for ~x86, ..., 3:x86 for stable:x86. |
35 |
> |
36 |
> Or, what is the difference? The GLEP doesn't actually explain the meaning of |
37 |
> 'stable' marking - the uncertainty Stuart refers to. |
38 |
|
39 |
This is the initial proposal for the GLEP mainly to get comments and to |
40 |
get the ball rolling from our developers and the community. As I see |
41 |
it, pretty much anything in the GLEP is subject to change. |
42 |
|
43 |
> One possible distinction is: stable status is given to a package that is |
44 |
> widely enough used and respected in the big bad world and has no known bugs, |
45 |
> as opposed to a package that's in portage for a month and has no bugs but |
46 |
> hasn't actually seen much use or been a target for attempted attacks. The |
47 |
> latter would never move beyond a regular arch keyword. |
48 |
> Some ebuilds might perhaps never be considered for the stable tree at all |
49 |
> because the target audience demonstrably isn't interested in them (based also |
50 |
> on actual usage data after the tree is up). |
51 |
|
52 |
I agree with this completely. I see no reason at all for things such as |
53 |
games to be added to the enterprise Gentoo. If a user really wants |
54 |
them, they can grab the ebuild from the "regular" portage tree and add |
55 |
it to their overlay. I would see enterprise Gentoo as a stable |
56 |
platform for use in commercial environments, and by users which value |
57 |
stability over the most current packages. This would allow Gentoo to |
58 |
fit a much larger audience, especially since our "Enterprise" version |
59 |
would still be free for all to use. |
60 |
|
61 |
> Both these are an RFC more than a suggestion; I want to understand the GLEP's |
62 |
> idea, not propose an alternative of my own. |
63 |
|
64 |
-- |
65 |
Chris Gianelloni |
66 |
Developer, Gentoo Linux |
67 |
Games Team |
68 |
|
69 |
Is your power animal a pengiun? |