[CentOS] EL 6 rollout strategies? (Scientific Linux)

Mon May 16 09:32:15 UTC 2011
Dag Wieers <dag at wieers.com>

On Mon, 16 May 2011, Ron Blizzard wrote:

> On Mon, May 16, 2011 at 2:44 AM, Dag Wieers <dag at wieers.com> wrote:
>> On Thu, 12 May 2011, Johnny Hughes wrote:
>>> The ZERO release is always going to take longer than the others.
>> Past numbers debunks this myth:
>>     CentOS 4.0 took 23 days
>>     CentOS 5.0 took 28 days
>>     CentOS 6.0 is not released after 6 months.
> Why do you snip the explanations and ignore the arguments contained in
> the text you snipped? Why no mention of the time it took to get 3.1
> (not 3.0) out the door?

CentOS 3.0 was not released because the project was still in its infancy 
(cAos project). I don't think it makes sense to even use it as a point of 
reference (unless maybe to argue for a direct CentOS 6.1 release).

But that still makes Johnny's statement false by a large margin.

     "The ZERO release is always going to take longer than the others."

Also the whole explanation does not provide any reasoning why CentOS 5.6 
took 3 months. The QA team is not allowed to speak up or provide feedback, 
or they could loose their 'privilege'.

Sure CentOS 6.0 is a different beast, but CentOS 6.0 was delayed in favor 
of CentOS 5.6. So again, why would CentOS 6.1 be released quicker if 
CentOS 5.6 has a well-known process and non of the issues Johnny was 
pointing at ?

My question was very specific though.

> Why constantly cast CentOS in the darkest possible light?

I don't think that's what I am doing. I commended Johnny for his very 
quick CentOS 4.9 release, but I honestly can not praise a release that 
is 3 months or 6 months late (with no transparency to what is going on 
or how we could help).

But if anything brought up wouldn't be ignored or obfuscated, CentOS 
communication would be a lot more honest, and threads would be a lot 
shorter. It's because the discussion is being side-tracked that they are 
becoming larger and the essence is being repeated.

There was a recent thread on centos-devel which clearly demonstrated this. 
It took a long thread and real worls examples for the CentOS developers to 
finally acknowledge there was a problem, and acknowledge it could be fixed 
for CentOS 6. This thread could be 4 posts long if the response wouldn't 
be defensive by default.

(And just like this thread, I did not start it either and am hardly the 
largest contributor to the thread)

-- dag wieers, dag at wieers.com, http://dag.wieers.com/
-- dagit linux solutions, info at dagit.net, http://dagit.net/

[Any errors in spelling, tact or fact are transmission errors]