Stefan Bodewig wrote:
> At first I thought I wouldn't have a real opinion so wanted to see
> other people's responses first. After having slept over it, I realize
> I do have an opinion ...
:-)
> The short form: +1 for having ant.apache.org with exactly one
> codebase, Ant. -1 for a new container project named ant.apache.org.
> +0 for a new container project that is *not* named Ant but contained
> the Ant codebase. My concern is that we must not dilute the name Ant.
Ok, I can see your point.
> To me it is a rather hard decision to make as I'm wearing several hats
> and have to balance that in some way.
>
> The ASF member may see benefit for the ASF if we get rid of as many
> container projects as we can (at the cost of swamping the board, but
> that's another issue).
>
> As a PMC member, it may make live easier for the next Jakarta PMC if
> codebases would leave the area where the Jakarta PMC has to monitor
> each and every commit. Ant may not be a problem for the current PMC
> (Geir is the only PMC member who is technically not an Ant committer),
> but I'm sure the next Jakarta PMC won't have as many Ant committers on
> it as the current one. Then again, Ant's codebase is probably one of
> the best monitored codebases you can find at Apache, and this won't
> change in the future.
Yup.
> As a member of the Ant community, I fail to see what the benefit for
> the Ant community would be. I pretty much doubt that Ant could be any
> more successful than it is right now - Ant doesn't need increased
> visibility at all.
Yes, it's not about visibility.
> Finally, as an Ant committer, I just now found out that I do have an
> opinion. This one is based on a mail by Roy sent to reorg@ and the
> Jakarta PMC list and I want to share the relevant excerpt:
>
> On Thu, 24 Oct 2002, Roy T. Fielding <fielding@apache.org> wrote:
>
>
>>The concept of a PMC, and the reason that anyone having a vote on
>>the project code-base should be a member of the PMC, is to provide
>>legal protection to those people as individuals. Not being on a PMC
>>(as defined by the bylaws) means that each and every decision made
>>by those committers is outside the scope of Apache's legal
>>protection, which in turn means that if a mistake is made (or some
>>asshole lawyer just feels like it), any suit against the committer
>>actions (such as infringement of some unknown patent) would have to
>>be defended by the committers on their own. The ASF would be able
>>to defend the code itself, but not the people whose actions were
>>outside the PMC.
>
>
> with that, I pretty much feel that every Ant committer needs to be in
> a PMC to gain the protection she/he deserves. To do that, Ant would
> have to become a top-level project, thus my +1 above.
I agree, this is exactly the point.
> Starting another container project will result in the same problems
> the existing containers face - the PMCs are legally responsible for
> more code than they can monitor. If there was a new container, it
> should be extremely small. Collecting Gump, Maven, Centipede, Ant,
> Antidote whatever would certainly be too much for this new project's
> PMC.
Ok, in fact as Peter correctly points out a "federation" would be much
better.
So ATM it's simple about making an Ant PMC and pushing Ant to
ant.apache.org.
--
Nicola Ken Barozzi nicolaken@apache.org
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
---------------------------------------------------------------------
--
To unsubscribe, e-mail: <mailto:ant-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:ant-dev-help@jakarta.apache.org>
|