On 06/18/2013 03:58 PM, Kevin A. McGrail wrote:
On 6/18/2013 9:34 AM, Axb wrote:
To be clear, if the polish involves changing the code, then you are
asking for an rc2.
The changes should be reflected in RC1 to avoid ppl having to trash
their Bayes data due to a data format change. This would not be very
kind and could scsare potential early adopters away and give the
feeling of this being very inmature, which it isn't.
I'm not going to bore you with details but in short, that just not how
ASF release procedures works. The rc1 is already created. Now we vote
on it. If you don't think it should be 3.4.0 because it needs new code
for example, vote -1 and we have to create an rc2 and repeat when it's
ready.
imo, we will need a RC2 (Mark?) so if that's the method -1
Though I seem to remember also that token generation in 3.4.0 for bayes
is different than 3.3.X so we likely need to warn people about that in
the release announcement as well.
I never used the 3.3 flavour so I can't say.
(went from 3.25 (with downports) to 3.4)
