1 |
On Wednesday 03 May 2006 23:02, lordsauronthegreat@×××××.com wrote |
2 |
about 'Re: [gentoo-user] why firefox is so slow?': |
3 |
> I've asked on multiple occassions for a comparison to a Debian paradigm. |
4 |
> My first distro was Debian. So a comparison to Debian would do me a |
5 |
> world of good. |
6 |
|
7 |
ARCH ~= STABLE |
8 |
|
9 |
but more volatile -- there's no real release version, packages are |
10 |
maintained separately unless they are packaged together upstream or depend |
11 |
on one another so you'll still get regular updates, but the packages have |
12 |
been though some amount of testing by the devs and those running: |
13 |
|
14 |
~ARCH ~= TESTING |
15 |
|
16 |
though there's less testing, packages don't have to survive being masked |
17 |
for any period of time so you will occasionally get an package that would |
18 |
only be suited for "Sid". It's still exceedingly rare for these packages |
19 |
to screw your system, but it is possible especially if you don't read the |
20 |
e(info|notice|warn)s that are output during the install or become lax in |
21 |
your config file updating. Anything truly experimental will be masked |
22 |
while the devs (and brave users) test internally. |
23 |
|
24 |
If you do run ~arch and can spend some time reporting you successes and |
25 |
failures, consider becoming an arch tester, assuming your arch needs more. |
26 |
|
27 |
package.mask'd, profile masked, or missing keyword ~= SID (or worse) |
28 |
|
29 |
These packages are either experimental with changes that are likely to |
30 |
break the system, or have been known to break someones system, possibly |
31 |
including data loss. Some development teams *cough*KDE*cough* are quite |
32 |
flexible with the "experimental" label -- I had kde 3.5 installed with few |
33 |
to no problems for (what seemed like) months before they were unmasked. |
34 |
|
35 |
Of course, ebuilds in the forums and bugzilla (and anywhere else other than |
36 |
the standard portage tree) are like foreign repositories. Gentoo devs |
37 |
(and most of the people on the mailing lists) can't and won't provide |
38 |
support for those packages or problems that may/might be caused by them. |
39 |
|
40 |
-- |
41 |
"If there's one thing we've established over the years, |
42 |
it's that the vast majority of our users don't have the slightest |
43 |
clue what's best for them in terms of package stability." |
44 |
-- Gentoo Developer Ciaran McCreesh |