svn commit: r591140 - /incubator/pig/branches/types/

[ga...@apache.org]

Author: gates
Date: Thu Nov  1 13:22:39 2007
New Revision: 591140

URL: http://svn.apache.org/viewvc?rev=591140&view=rev
Log:
branch for adding types to pig

Added:
    incubator/pig/branches/types/
      - copied from r591139, incubator/pig/trunk/

Re: svn commit: r591140 - /incubator/pig/branches/types/

[Doug Cutting <cu...@apache.org>]

Olga Natkovich wrote:
> Since we are new to Apache, I was not aware of the rules on when the
> branches can be created.

That's to be expected.  So please do not take my remarks personally.

I know of no rules specific to branches, but common sense says that a 
subversion branch is for shared development.  Sharing implies multiple 
committers, and committers must not make project decisions outside of 
the mailing lists.  The project needs to operate transparently to 
provide a level playing field, so that no contributors have privileged 
information.  Its permissible and inevitable that contributors will have 
off-list discussions, but any such discussions must be presented to 
other contributors before they are acted on, so that all have an equal 
opportunity to participate in decisions.

> Actually, people on my team asked that maybe
> you could give us a brief overview of the process to avoid this kind of
> situations in the future.

Suggested readings include:

http://www.apache.org/foundation/how-it-works.html#management
http://www.apache.org/dev/committers.html
http://www.cubiclemuses.com/cm/files/incubating_communities.pdf

The last is a nice presentation about project incubation.

But the incubator is all about learning while doing.  Don't worry if you 
make a few mistakes.  Each community develops its own standards within 
guidelines passed on by mentors.  We're just beginning that process.

Doug

RE: svn commit: r591140 - /incubator/pig/branches/types/

[Olga Natkovich <ol...@yahoo-inc.com>]

Doug,

I asked Alan to put this into a separate branch. We have frozen our tree
in preparation for the release of code into apache. Meanwhile we were
doing development work which is pretty major changes to support types
and we needed some place to incrementally save the changes until they
are ready for integration into the trunk.

Since we are new to Apache, I was not aware of the rules on when the
branches can be created. Actually, people on my team asked that maybe
you could give us a brief overview of the process to avoid this kind of
situations in the future.

Alan is in the process of putting documentation for his work into Wiki.
Once it is there, we can have further discussion.

Olga

-----Original Message-----
From: Doug Cutting [mailto:cutting@apache.org] 
Sent: Thursday, November 01, 2007 2:46 PM
To: pig-dev@incubator.apache.org
Subject: Re: svn commit: r591140 - /incubator/pig/branches/types/

Can you please explain why you're branching?  I'd normally expect
creation of a branch to be discussed on the dev list first.  To me, a
branch implies the need for collaborative development of changes that
are incompatible with trunk development.  Is this the case?  If so, then
there first needs to be a public discussion with potential
collaborators.  To plan a branch in private would be collusion, and a
branch with a single developer doesn't usually make sense.

Thanks,

Doug

Re: svn commit: r591140 - /incubator/pig/branches/types/

[Doug Cutting <cu...@apache.org>]

Can you please explain why you're branching?  I'd normally expect 
creation of a branch to be discussed on the dev list first.  To me, a 
branch implies the need for collaborative development of changes that 
are incompatible with trunk development.  Is this the case?  If so, then 
there first needs to be a public discussion with potential 
collaborators.  To plan a branch in private would be collusion, and a 
branch with a single developer doesn't usually make sense.

Thanks,

Doug