Re: [vote] 0.4.0 beta 1

2009-08-13 Thread ant elder
On Wed, Aug 12, 2009 at 3:51 PM, Jonathan Ellisjbel...@gmail.com wrote:
 On Wed, Aug 12, 2009 at 8:41 AM, ant elderantel...@apache.org wrote:
 Its common to have nightly builds. Its common do beta's that have been
 voted on by a PMC. Those are quite different to a someone creating
 something called 0.4.0 beta 1 and advertising that to users as an
 unofficial release. Whats the problem following the ASF release
 procedures and letting the PMC vote on the artifacts before they're
 released? Even if nothing else it would help with graduation to show
 that happening.

 In my mind this beta will be relevant for about a week, at which point
 we'd have another beta or an RC with a lot of bug fixes.  If we go
 from waiting hours for the committers to waiting 3 days for the PMC,
 that's half its expected lifetime.  In other words I see this as
 closer to a glorified nightly than An Official Release.

 -Jonathan


Ok I wonder if this is just misunderstandings based on naming and
conventions. Could what you're calling here as an unofficial beta
release be the same as what other ASF projects might name a release
candidate? Is the 0.4.0 beta 1 just for other devs to review? If so
you probably don't need a vote. Also, whats the difference between
this and the latest Hudson nightly build?

   ...ant


Re: [vote] 0.4.0 beta 1

2009-08-13 Thread Jonathan Ellis
On Thu, Aug 13, 2009 at 1:20 AM, ant elderantel...@apache.org wrote:
 Ok I wonder if this is just misunderstandings based on naming and
 conventions. Could what you're calling here as an unofficial beta
 release be the same as what other ASF projects might name a release
 candidate? Is the 0.4.0 beta 1 just for other devs to review?

This is to get the feedback of people who are not routinely building
from svn, call them what you will.  (In my mind anyone who is not
building from svn is not really a dev, but if you want to include
people who file bug reports in that category, fine.)

 If so
 you probably don't need a vote. Also, whats the difference between
 this and the latest Hudson nightly build?

This would be a nightly that is considered unlikely to blow up in your
face _too_ obviously, which is not the case for all nightlies.

If the nightlies were archived I would be fine saying if you want to
test, use the nightly from 8-11 (guess we're up to at least 8-13 now);
it's relatively stable.  But they are not.

-Jonathan


Re: [vote] 0.4.0 beta 1

2009-08-12 Thread Ian Holsman

+1
great work jonathan!
On 11/08/2009, at 4:44 PM, Jonathan Ellis wrote:


Hi all,

Cassandra 0.4 is feature complete based on the roadmap I suggested a
few weeks ago.  We also have a mostly-empty issue tracker for 0.4.
(https://issues.apache.org/jira/secure/IssueNavigator.jspa?reset=truemode=hidesorter/order=DESCsorter/field=priorityresolution=-1pid=12310865fixfor=12313862 
)


I propose releasing an unofficial 0.4.0 beta based on current trunk to
signal that our api is stable, our disk format is stable, and we're
ready to get a wider group testing this.

+1 (binding) from me.

-Jonathan


--
Ian Holsman
i...@holsman.net





Re: [vote] 0.4.0 beta 1

2009-08-12 Thread ant elder
On Tue, Aug 11, 2009 at 9:44 PM, Jonathan Ellisjbel...@gmail.com wrote:
 Hi all,

 Cassandra 0.4 is feature complete based on the roadmap I suggested a
 few weeks ago.  We also have a mostly-empty issue tracker for 0.4.
 (https://issues.apache.org/jira/secure/IssueNavigator.jspa?reset=truemode=hidesorter/order=DESCsorter/field=priorityresolution=-1pid=12310865fixfor=12313862)

 I propose releasing an unofficial 0.4.0 beta based on current trunk to
 signal that our api is stable, our disk format is stable, and we're
 ready to get a wider group testing this.

 +1 (binding) from me.

 -Jonathan


What do you mean by an unofficial release? A release needs follow
the ASF release procedures and be voted on so needs svn tags, and
distributions to review etc right?

   ...ant


Re: [vote] 0.4.0 beta 1

2009-08-12 Thread Jonathan Ellis
On Wed, Aug 12, 2009 at 5:49 AM, ant elderant.el...@gmail.com wrote:
 On Tue, Aug 11, 2009 at 9:44 PM, Jonathan Ellisjbel...@gmail.com wrote:
 What do you mean by an unofficial release? A release needs follow
 the ASF release procedures and be voted on so needs svn tags, and
 distributions to review etc right?

I mean that I will put a tarball up in my personal space for people to
test but it will not go through the full mentor gauntlet nor will it
be recommended for more than temporary use.

There was a long discussion about this back in May in the Cassandra
0.3 RC is out thread:
http://mail-archives.apache.org/mod_mbox/incubator-cassandra-dev/200905.mbox/browser

My impression was that the consensus from the mentors, including
Matthieu and Bertrand, was that it's both okay and common to do
nightly builds or betas like this, as long as it's not implied that
these are Official ASF Releases.

-Jonathan


Re: [vote] 0.4.0 beta 1

2009-08-12 Thread ant elder
On Wed, Aug 12, 2009 at 2:21 PM, Jonathan Ellisjbel...@gmail.com wrote:
 On Wed, Aug 12, 2009 at 5:49 AM, ant elderant.el...@gmail.com wrote:
 On Tue, Aug 11, 2009 at 9:44 PM, Jonathan Ellisjbel...@gmail.com wrote:
 What do you mean by an unofficial release? A release needs follow
 the ASF release procedures and be voted on so needs svn tags, and
 distributions to review etc right?

 I mean that I will put a tarball up in my personal space for people to
 test but it will not go through the full mentor gauntlet nor will it
 be recommended for more than temporary use.

 There was a long discussion about this back in May in the Cassandra
 0.3 RC is out thread:
 http://mail-archives.apache.org/mod_mbox/incubator-cassandra-dev/200905.mbox/browser

 My impression was that the consensus from the mentors, including
 Matthieu and Bertrand, was that it's both okay and common to do
 nightly builds or betas like this, as long as it's not implied that
 these are Official ASF Releases.

 -Jonathan


Its common to have nightly builds. Its common do beta's that have been
voted on by a PMC. Those are quite different to a someone creating
something called 0.4.0 beta 1 and advertising that to users as an
unofficial release. Whats the problem following the ASF release
procedures and letting the PMC vote on the artifacts before they're
released? Even if nothing else it would help with graduation to show
that happening.

   ...ant


Re: [vote] 0.4.0 beta 1

2009-08-11 Thread Arin Sarkissian
+1 from me as well

On Tue, Aug 11, 2009 at 6:13 PM, Jun Raojun...@almaden.ibm.com wrote:
 +1 for getting the beta out.

 Jun
 IBM Almaden Research Center
 K55/B1, 650 Harry Road, San Jose, CA 95120-6099

 jun...@almaden.ibm.com


 Jonathan Ellis ---08/11/2009 01:45:33 PM---Hi all, Cassandra 0.4 is feature
 complete based on the roadmap I suggested a


 From:
 Jonathan Ellis jbel...@gmail.com
 To:
 cassandra-dev@incubator.apache.org
 Date:
 08/11/2009 01:45 PM
 Subject:
 [vote] 0.4.0 beta 1
 



 Hi all,

 Cassandra 0.4 is feature complete based on the roadmap I suggested a
 few weeks ago.  We also have a mostly-empty issue tracker for 0.4.
 (https://issues.apache.org/jira/secure/IssueNavigator.jspa?reset=truemode=hidesorter/order=DESCsorter/field=priorityresolution=-1pid=12310865fixfor=12313862)

 I propose releasing an unofficial 0.4.0 beta based on current trunk to
 signal that our api is stable, our disk format is stable, and we're
 ready to get a wider group testing this.

 +1 (binding) from me.

 -Jonathan