1 |
On Sun, 14 Aug 2005 17:13:14 -0500 |
2 |
Joe Menola wrote: |
3 |
|
4 |
> On Sunday August 14 2005 4:38 pm, Nick Rout wrote: |
5 |
> > Unstable does not really cut it IMHO. I am a gentoo enthusiast through |
6 |
> > and through, but plonking something in portage with a ~ beside it does |
7 |
> > not constitute a release of a recent version IMHO. |
8 |
> |
9 |
> Can you name any version of Linux where version upgrades go directly into |
10 |
> stable? |
11 |
> It's all about choice...the "latest n greatest" or "tried n true". |
12 |
> And that's how it is in any flavor of Linux I've tried, LFS included. |
13 |
|
14 |
I am not challenging anything of what the "biters" have said in relation |
15 |
to my post. I know the need for testing and i know that many people |
16 |
successfully run ~x86 systems for the latest and greatest. |
17 |
|
18 |
However, when I first used gentoo I was always the first in my LUG to |
19 |
have the latest kde, evolution, mplayer etc, and that was running x86 |
20 |
not ~x86. My perception is that gentoo is no longer first off the block |
21 |
with stable releases. |
22 |
|
23 |
I also realise that this is in no small part due to the huge increase in |
24 |
the number of packages in portage, the growing complexity of package and |
25 |
dependency management as both the tree and the "bloat" of complex |
26 |
software grows. |
27 |
|
28 |
However I also see constant complaints of people who contribute ebuilds |
29 |
for new versions or new packages to bugzilla and then have them sit |
30 |
there for months without any activity at all. |
31 |
|
32 |
Myself, I love gentoo and this is not a complaint, merely an observation |
33 |
that things are no longer as they were, and that there appear to be some |
34 |
factors slowing down the progression of packages in their path from |
35 |
bugzilla to stable. Maybe this is because the concepts of choice in |
36 |
gentoo lead to an environment where it is harder to track all possible |
37 |
combinations of those choices. After all if there were 50 use variables |
38 |
[1] there are 2^50 combinations of those variables being off or on. |
39 |
Combine that with variations to architecture, CFLAGS and whatever else |
40 |
you can configure and you have a much tougher development environment |
41 |
than one where everything is compiled into "we decide what you get" |
42 |
binaries. |
43 |
|
44 |
[1] i have no idea how many there are. |
45 |
|
46 |
Cheers. |
47 |
|
48 |
|
49 |
> |
50 |
> -jm |
51 |
> -- |
52 |
> gentoo-user@g.o mailing list |
53 |
|
54 |
-- |
55 |
Nick Rout <nick@×××××××.nz> |
56 |
|
57 |
-- |
58 |
gentoo-user@g.o mailing list |