hadoop-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Doug Cutting <cutt...@apache.org>
Subject Re: Release plans
Date Fri, 19 Feb 2010 22:49:32 GMT
Eli Collins wrote:
>> My concern is more about when the next branch from trunk will be made.
> 
> Does there need to be a dependency?

No.  I just wanted to note that, from my point of view, releasing from 
the existing 0.21 branch is not sufficient, that, regardless, we still 
need to release from trunk soon, and we need a schedule going forward 
for when future branches from trunk will be made.

Aside from resolving compatibility issues (more on that below) what we 
perhaps need are one or two release master volunteers, for a release 
based on trunk and perhaps for a release based on the 0.21 branch.

> Could the vote on whether 22 is backwards compatible with 20 be
> independent of what we call 1.0?

We minimally need to declare whether releases are major or minor.  The 
convention has been that all 0.X releases are major, 1.0 is major, 1.1, 
1.2, etc. are minor, 2.0 is major, 2.1, 2.2, etc are minor and so on. 
We could amend this convention, but things might get confusing.  For 
example, we could declare that we just have a sequence of numerically 
increasing releases, some major, some minor, but that you can't tell 
which is which from the number.  Or we could, like Sun did with Java, 
move the decimal point, and say that 0.20 is just 2.0, and that, instead 
of 0.21 our next release is 2.1.  Or we could, as I suggested in my 
previous message, retroactively consider 0.20 to be 1.0, and then make 
our next release 1.1.  None are particularly attractive options.

Doug

Mime
View raw message