| 1 |
On Thu, Dec 14, 2017 at 2:32 PM, Kristian Fiskerstrand <k_f@g.o> wrote: |
| 2 |
> On 12/14/2017 09:21 PM, R0b0t1 wrote: |
| 3 |
>> It seems like lagging stability is due to a lack of resources. I do |
| 4 |
>> not know a single person who would be able to run only stable |
| 5 |
>> packages. |
| 6 |
> |
| 7 |
> I run stable only on most of my systems. |
| 8 |
> |
| 9 |
|
| 10 |
That is fine, but this thread exists because at least the OP thinks |
| 11 |
stabilization is not happening quickly enough, likely because there |
| 12 |
are not enough people working on it. Allowing stabilization work from |
| 13 |
mixed systems might allow more people to help. |
| 14 |
|
| 15 |
>> They seem to move too slowly, and people switch to unstable |
| 16 |
>> packages because they contain bugfixes and sometimes new features. |
| 17 |
> |
| 18 |
> slow isn't necessarily a problem, as long as security fixes are handled. |
| 19 |
> There is some balancing for large performance gains, but most existing |
| 20 |
> systems are scaled based on the current estimates so it would only be |
| 21 |
> relevant for the up sizing of the server park for growth needs etc. |
| 22 |
> |
| 23 |
>> |
| 24 |
>> Could the criteria for stability be reconsidered? Mixed systems might |
| 25 |
> |
| 26 |
> why would it? |
| 27 |
> |
| 28 |
|
| 29 |
Per the question posed by OP the current state of affairs does not |
| 30 |
seem to be working, and I have tried to point out one likely cause. If |
| 31 |
it's hard to justify the criteria for stability then maybe the |
| 32 |
criteria don't make sense. |
| 33 |
|
| 34 |
>> not be supported, but save for cases of ABI/API breakage (which can |
| 35 |
>> happen when transitioning from stable->stable) I do not know why the |
| 36 |
>> packages would not play well with each other. I am sure there are |
| 37 |
>> examples where things have blown up, but it seems like expecting that |
| 38 |
>> to be the case isn't helping. |
| 39 |
> |
| 40 |
> There are plenty of cases where this fails in miserable ways, so thats |
| 41 |
> not a good idea (not to mention the dependency hell from it). That said, |
| 42 |
> you can have a stable chroot, or just use a VM for testing etc. |
| 43 |
> |
| 44 |
|
| 45 |
Can you be specific? Human memory is biased towards negative |
| 46 |
experiences. If it's hard to actually describe the multitude of issues |
| 47 |
that mixed systems cause then it is very likely mixed systems do not |
| 48 |
cause many issues. |
| 49 |
|
| 50 |
Personally, I have very few problems due to my mixed system, and less |
| 51 |
than I would have on a stable system. |
| 52 |
|
| 53 |
Cheers, |
| 54 |
R0b0t1 |
Archives