openjpa-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "robert burrell donkin" <robertburrelldon...@gmail.com>
Subject Re: Access to Open JPA
Date Mon, 15 May 2006 20:29:49 GMT
On 5/15/06, Patrick Linskey <plinskey@bea.com> wrote:
>
> Guys,
>
> I'd kinda like to keep things as locked-down as possible until we're
> closer to having functional code. So I'd rather just get people actually
> committing now who are looking at build environment set-up etc., so that
> we can minimize the number of moving parts. Does that sound reasonable?


(some tangental ramblings on open development...)

the primary method of oversight in ASF projects is by reading commit
records. this is something that podlings need to get used to. commits are
approved by lazy consensus: any commit can be -1'd (if there is sufficient
reason) and reverted. this may force a vote. note that this should happen
rarely in a healthy community. it's more polite to point out mistakes
without a veto.

social conventions are powerful. most people don't like looking stupid in
public. access control at the ASF is necessary for some extra security in
depth in the event of a compromise and to give privacy for confidential
information (for example, security).

podlings begin in a slightly atypical state. the typical apache contributor
starts by submitting patches. these patches are reviewed and committed by
existing committers. it is rare for the initial contributions of a developer
new to a project to be committed without changes. gradually, if the
developer learns from the feedback the patches should need fewer and fewer
changes. this builds trust and confidence until the developer is elected a
committer. at which stage, the review happens after the commit.

- robert

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message