| 1 |
On 28/01/2014 14:37, Steven J. Long wrote: |
| 2 |
> I concur that "QA should be focusing on making stable, actually stable, |
| 3 |
> not more bleeding edge." That's not a "performance" issue as you put it, |
| 4 |
> except in management nuspeek. It's the whole bloody point of the distro, |
| 5 |
> in overarching terms: to test and stabilise robust ebuilds. That process |
| 6 |
> is what leads to better software, not staying at the "bleeding-edge" |
| 7 |
> and forgetting about robustness since "a new version is out." |
| 8 |
|
| 9 |
|
| 10 |
+1 |
| 11 |
|
| 12 |
Nice to see a dev echo my sentiments almost word for word exactly. |
| 13 |
|
| 14 |
9 years later I'm still here, still running Gentoo on all my hosts (over |
| 15 |
10 at last count excluding VMs). Why? Because Gentoo |
| 16 |
just.works.right.every.single.time, even on ~arch - and that is an |
| 17 |
amazing accomplishment for an distro never mind a USE based one. |
| 18 |
|
| 19 |
If I want bleeding edge I'll use funtoo or exherbo or unmask everything |
| 20 |
-9999. If I want the latest new! improved! shiny! crap re-implemented |
| 21 |
yet again and badly, there's Ubuntu or nightlies from rawhide. |
| 22 |
|
| 23 |
The joy of Gentoo is that it works on just about anything. Stable |
| 24 |
well-tested code continues to just work for the most part even on |
| 25 |
slacker arches even if the ebuild is years old. When stable is just a |
| 26 |
bit too stable for a specific case, we have overlays and |
| 27 |
/usr/local/portage/cat/pkg. |
| 28 |
|
| 29 |
This is why Gentoo works so well, because the weird arches still get to |
| 30 |
play on the same playground with the other kids. I work at a carrier ISP |
| 31 |
and you'd be pleasantly surprised to see just how many gentoo-powered |
| 32 |
vendor POC blackboxes come through the office from vendors wanting to |
| 33 |
sell their network magic. Business seems to have cottoned onto the idea |
| 34 |
that gentoo let's you stop wasting time with make and rather fire off |
| 35 |
emerge, doesn't matter what the silicon is. |
| 36 |
|
| 37 |
Slow arches is the price for supporting everything out there. But so |
| 38 |
what? If slow_arch_X is stuck on some old version of an @system package, |
| 39 |
who cares? It's not like portage will pick it for an amd64 box. An old |
| 40 |
ebuild is a file, it sits next to 178,477 files and does no harm, it |
| 41 |
only gets used on hardware that needs it. |
| 42 |
|
| 43 |
|
| 44 |
|
| 45 |
-- |
| 46 |
Alan McKinnon |
| 47 |
alan.mckinnon@×××××.com |
Archives