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

[Gordon Messmer]

On 05/15/2011 03:52 AM, Ron Blizzard wrote:
> On Sun, May 15, 2011 at 3:11 AM, Gordon Messmer<yinyang at eburg.com>  wrote:
>> As far as users know, all work on 6.0 was postponed to get 5.6 done.  At
>> the time of 5.6's release, it was the only release the team was working
>> on.  Work on 5 should have been something the team was quite familiar
>> with by that time.  If 5.6 took 3 months to finish, then Dag's question
>> is quite fair: why would we expect 6.1 to take so much less time?
>
> You're leaving out release 4.9.

No, I'm not.  4.9 was released just over a month after 5.6.  If 5.6 
couldn't be finished in that month, why would 6.1 be finished in only a 
month?

> You're also leaving out the fact that
> two major holidays occurred during the time *frame* that these three
> releases needed to be built.

I'm not sure which two you're referring to.  Are there going to be no 
holidays following the release of 6.1?

> You're also leaving out the fact (as
> mentioned by one of the developers) that they had to start from
> scratch on 6.0 -- that they'll be "set up" for 6.1 when it comes out.

I don't see how that's relevant.  They didn't have to start from scratch 
on 5.6, and that took three months.  Why would 6.1 take so much less 
time than 5.6?

> You're also leaving out the fact that SL had to rebuild the same three
> releases -- and they're still working on the last of those -- so the
> amount of time it's taking CentOS developers squares with the amount
> of time required by the SL developers.

No, I'm not.  Neither I nor Dag, as far as I saw, brought SL into the 
conversation at all.  The question is not whether CentOS can build 
releases in less time than SL, or even a reasonable amount of time.  The 
question that Dag posed was why users (or the release team) should 
expect 6.1 to be done in one month, when 5.6 took three and was a fairly 
well rehearsed process by that point.