25 Oct
2010
25 Oct
'10
2:51 p.m.
On Oct 25, 2010, at 2:44 PM, Sergiu Dumitriu wrote:
On 10/25/2010 02:26 PM, Vincent Massol wrote:
Indeed. I think I messed up the end date. what about a RC2 for the 8th of Nov.
2.6RC1: 1st of Nov
2.6RC2: 8th of Nov
2.6 final: start 15th, done for the 18th at latest
Thanks
-Vincent
On Oct 25, 2010, at 2:21 PM, Sergiu Dumitriu wrote:
I'm +0 on either date, my statement was just a "if possible", since 18
days between a RC and the final release is a more than what we usually do. On 10/25/2010 12:32 PM, Vincent Massol wrote:
End to end, it's actually 3.5 weeks (I don't know why I put the date of 18th of
Nov. I think I wanted to put 15th initially but that slipped for some reason... anyway
let's keep 18th but with no slippage ;) ie we should start releasing no later than the
15th, if not earlier ;)).
Hi Sergiu,
On Oct 21, 2010, at 1:08 AM, Sergiu Dumitriu wrote:
> On 10/20/2010 09:31 AM, Vincent Massol wrote:
>> The idea would be to have the following dates:
>> - 2.6RC1: 1st of November 2010 (there's no M1)
>> - 2.6 final: 18h of November
>
> I'd move the RC one week later.
I actually think that 3 weeks is already almost too much actually, since the longest the
likeliest people will introduce new features and not do stabilization.
WDYM 3 weeks? When did you start counting? 1st is next week, so
just one
week between 2.5 and 2.6RC1. Also, trunk became 2.6 on the 14th, so that
would be only two weeks since the 2.6 version was introduced, I really
don't understand where the 3 weeks before the RC came. Note that I only
proposed moving the RC date, not the final date.
I thought you were proposing to move the final date.
I'm +0 to move only the RC by one week (although this gives less time for
stabilization and may result in more features being added in). A few days instead of a
week would be better. We need the stabilization time as much as possible IMO. One week
less means less stabilization on the new features add for RC1. Note that
Raluca thinks she'll have finished the Recent Activity feature for the 1st of Nov.
Thanks
-Vincent
>> In addition this gives less chance for people to test the stabilization
release(s) (people only test when there are final releases).
>>
>> I'd really like to keep the proposed dates. Please let me know if it's ok
with you since you're the only one who voiced a concern on the date.
--
Sergiu Dumitriu
http://purl.org/net/sergiu/