On 21/08/12 00:27, Karanbir Singh wrote:
> On 08/21/2012 12:22 AM, Ian wrote:
>> I seriously think this is the wrong place to discuss CentOS
>> specifically, don't you?
> I dont think this is about CentOS, its more about what the overall
> impression being carried forward is.
Me neither. Wires crossed maybe.
>> But, I'll humour you... if somehow, Delay= CloneReleaseTime -
>> SrpmReleaseTime = 0 (i.e. no delay) then I fail to see where your
>> testing time is.
>>
>> If you consider:-
>>
>> Delay = CloneReleaseTime - SrpmReleaseTime -
>> AcceptedRebuildAndIndependentTestTime = 0 then maybe.
> is it zero ? how did you verify that ? and how did you audit the process
> that was used to get from here to there during that
> what-you-assume-to-be negative time.
>
> your argument is based on the flawed assumption that we release stuff
> before Red Hat does. that would be - absurd, if anything.
No, as you say that would be absurd. I think my maths illustration has
served to confuse more than clarify. As I said elsewhere, I consider
delay to be the time between Redhat dropping the goods and the clone
dropping the good. Some "delay" or "lag" is inevitable, by the nature of
the beast. My original point was that it was a bit unfair to call the
other person out as FUD for the statement....
"delays in issuing bug fixes are good for you - no delay means they
did not test the stuff before pushing it out."
The whole argument is a bit pointless anyway, because what "testing"
means is not defined .e.g. automated vs manual, etc. I am not saying the
OP is right or wrong, just that calling FUD without explanation is not
helpful IMHO.
>> To be fair to the OP, if you are going to call someone out on FUD, you
>> really IMHO ought to give a lot more detail as to why, otherwise you are
>> merely perpetuating your own FUD.
> How about you just look through the last few posts from me on this list ?
>
I don't regularly read the list, but this thread caught my eye in
particular. IMHO if you have something that is relevant to this thread
it should be linked to, as thread context will be lost when it hits the
archive, and for the benefit of people who aren't able to read each and
every post. :)
Last time I touched on the subject of release process on a certain other
list, I was told to mind my own business and find a new distribution...
and here I am. ;)
|