cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dirk-Willem van Gulik <di...@webweaving.org>
Subject Re: [proposal] Private Branches
Date Fri, 30 Jul 2004 17:18:05 GMT

On Thu, 29 Jul 2004, Stefano Mazzocchi wrote:

..snipped inspiring text about using SNV features and perhaps
  changing things so it fits our patterns better now that we 'can'.
...

> My proposal is to create a private branch for every committer that wants
> it and place it in
>
>   http://svn.apache.org/repos/asf/cocoon/private/[id]
>
>   1) allow people to experiment in their own place
>   2) allow stuff to happen without giving the impression that it is a
> "fork" or that the status quo is in danger
>   3) give the feeling of some privacy

This would worry me somewhat; within the ASF we try to help the community
develop software and value that a project survive an individuals slant on
it. It is the fact that the whole community follows all commits (by and
large), there is comprehensive peer review and that the whole community
stands behind a release which gives the ASF code that unique slant on
quality and liability.

Encouraging[*] this would in my opinion:

->	give rise to personal slants

->	allow certain takes of the code to be less of a group
	product; but instead to be driven by 1 or 2 individuals

->	increase the risk that a wholesale chunk of code
	which is developed without sufficient peer review is
	subsequently introduced into the main brach without
	enough appropriate oversight by the whole community.

which seems contrary to a certain spirit I personally hold dear when it
comes to the ASF.

A more neutral name; say 'note pad', 'scratrchpad' or 'skunk works' would
perhaps be more appropriate; as one stresses that it is people working
_together_ that counts.

Finally I personally would expect that -if- the ASF see a situation where
significant chunks of code are developed outside the normal main peer
reviewed track, and see such chunks suddenly imported in a main branch
-then- I do expect a discussion about exactly what peer review and
oversight is, and what the limits between 2.0 license, CLA and a Software
Grant really is.

Dw

*: Of course we would never disallow this. People can always
	make their own private forks.

Mime
View raw message