1 |
Begin forgot to cc gentoo-dev message: |
2 |
|
3 |
On Sunday, Dec 15, 2002, at 13:56 Europe/Brussels, Rainer Groesslinger |
4 |
wrote: |
5 |
|
6 |
> For example Maik "blizzy" Schreiber told me about |
7 |
> http://gentoo-stable.iq-computing.de which is something like a "voting |
8 |
> system" but almost nobody is using it (for example mozilla 1.2.1 only |
9 |
> has |
10 |
> one vote although many thousand people are using it - with success) |
11 |
> and if |
12 |
> you take a look at the ebuild you see that every mozilla ebuild with |
13 |
> version |
14 |
> 1.2.1 has the keyword ~x86 - so stable users don't get it although |
15 |
> there's |
16 |
> no reason for calling Mozilla 1.2.1 "unstable"... |
17 |
|
18 |
AFAIK gentoo-stable currently is in the process of being integrated |
19 |
into gentoo.org (and into bugzilla?). gentoo-stable.iq-computing.de is |
20 |
not official. It is only open to public for testing and for comments |
21 |
(from mostly gentoo devs, maybe some users). It would probably be a bit |
22 |
too soon to have many thousands of people use a system that is still |
23 |
under heavy development. You should ask maik about this in person, |
24 |
cause I don't know everything about its current state. But I do know it |
25 |
will be officially announced on www.gentoo.org if it is ready for |
26 |
everyone. |
27 |
|
28 |
> In my opinion http://gentoo-stable.iq-computing.de should be a |
29 |
> more-or-less |
30 |
> official voting system for the packages or gentoo stable will end like |
31 |
> debian stable and I don't think Gentoo wants to go *that* stable :) |
32 |
|
33 |
gentoo-stable was needed for the following reason: we needed a |
34 |
mechanism to be able to track which (unstable) ebuilds users have |
35 |
installed/tested and run great (or crashed their system). (Right now |
36 |
lot of devs mark an ebuild with ~x86 ~ppc ~alpha ~sparc ~mips... and |
37 |
never receive input from users to inform them if an ebuild works great |
38 |
or doesn't. That's why gentoo-stable is being created.) but like I |
39 |
said, it still is in the development/testing/open for comments stage |
40 |
and is probably not production ready (again...you should ask maik) |
41 |
|
42 |
> There are just not enough users and feedback pushing unstable packages |
43 |
> to |
44 |
> stable from what I see... |
45 |
|
46 |
it's not in production stage yet so there's no need to announce this |
47 |
officially to everyone yet. We want user input, but I can imagine Maik |
48 |
doesn't want a few thousand people reporting everything. We don't want |
49 |
to slashdot maik's server. (That doesn't mean that you can't have a |
50 |
look at it and send your comments to this list or even to maik if you |
51 |
find a bug). Again: ask maik about the development status, but I think |
52 |
it's not production ready yet and in testing stage. |
53 |
|
54 |
> There was/is talk about package.mask being removed in the future - |
55 |
> good idea |
56 |
> but I think it should look like this |
57 |
> |
58 |
> stable: KDE 3.0.5 |
59 |
> unstable: KDE3.1RC5 |
60 |
> |
61 |
> stable: Mozilla 1.2.1 |
62 |
> unstable: Mozilla 1.3a |
63 |
> |
64 |
> and so on...In short: Gentoo stable should be as close as possible to |
65 |
> what |
66 |
> the developers of the various applications call "stable" - why not |
67 |
> believe |
68 |
> them ? ;p |
69 |
|
70 |
It's not really the app only that should be called 'stable' before it |
71 |
can appear as stable in gentoo. If the ebuild is broken, or does some |
72 |
weird stuff (or just needs to be tested) a stable app can be called |
73 |
unstable. Gentoo is a metadistribution, that means that instead of |
74 |
sending users applications in binary form we give users executable |
75 |
(readable) instructions to build the Gentoo binaries themselves. These |
76 |
instructions can be called unstable if they are only recently |
77 |
introduced into portage and need to be tested first. but otherwise the |
78 |
stable/unstable thing does follow what the developers call their |
79 |
application (unless we think what they say is incorrect (happens when |
80 |
users report that the application breaks) |
81 |
|
82 |
> Currently the package.mask carries packages which have a right to be |
83 |
> called |
84 |
> unstable, e.g. XFree 4.2.99 and so on... |
85 |
> But the stable/unstable situation in some ebuilds is a bit confusing |
86 |
> and |
87 |
> leading in the wrong direction if continued like this ? |
88 |
|
89 |
Can you give an example of these ebuilds? |
90 |
|
91 |
I know there are some apps which are called unstable by their |
92 |
developer, and stable by us. But this happens only after a long period |
93 |
of testing/running the app. |
94 |
|
95 |
> Of course every distribution needs to test individual things, make some |
96 |
> changes here and there...And to avoid a bad stable tree I highly |
97 |
> suggest |
98 |
> using blizzy's system... |
99 |
|
100 |
Maik will be glad to hear that. But keep in mind that the system is |
101 |
probably only up for testing and not yet in production. |
102 |
|
103 |
> Just my opinion about current stable/unstable things... |
104 |
> Rainer |
105 |
|
106 |
Pieter |
107 |
|
108 |
-- |
109 |
Pieter Van den Abeele |
110 |
pvdabeel@g.o - pvdabeel@××××××.be |
111 |
|
112 |
|
113 |
-- |
114 |
gentoo-dev@g.o mailing list |