commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Niall Pemberton <>
Subject Re: [pool] time to move groupId?
Date Sat, 16 Oct 2010 17:39:09 GMT
On Sat, Oct 16, 2010 at 6:20 PM, Phil Steitz <> wrote:
> On 10/16/10 12:52 PM, James Carman wrote:
>> On Sat, Oct 16, 2010 at 12:23 PM, Niall Pemberton
>> <>  wrote:
>>> Consistency is good, but deciding something based purely on
>>> *consistency* rather than the merits of the situation is mindless.
>> Trying to keep things consistent is being mindful of the user's needs.
>>  The consistency is part of the situation.  When you go to pool
>> version 3, you'll have:
>> 1.x - commons-pool:common-pool with package org.apache.commons.pool
>> 2.x - org.apache.commons:commons-pool with package
>> org.apache.commons.pool2
>> 3.x - org.apache.commons:commons-pool3 with package
>> org.apache.commons-pool3
> From what Dennis says above, it is not obvious to me that the artifactId
> change for v 3 will be necessary - i.e., just bumping the version number
> will allow multiples to be on the classpath at the same time.  Can someone
> answer this question definitively?  If it will be necessary to move to pool3
> later, then I agree with James that we may as well do it now.

What James says wrt to pool3 is correct. Maven *resolves* dependencies
to pick a single version of a groupId/artifactId based on the
dependency tree. If not people would have continued problems with
multiple versions of the same artefact in their classpath.

>> Makes lots of sense to the users.
>> Again, it comes down to "you don't work on this project so stay out of
>> its damn business" and "the release manager is doing the work so they
>> can do whatever the hell they want."  So, why don't we just do what
>> Incubator does and have a mini-PMC for each subproject since you don't
>> want other members of the Commons PMC butting their noses into a
>> project they don't actively participate in?
> I don't think that is what Niall meant and its certainly not how I see
> things.  A great strength of our community is that we *welcome* input for
> one another as we make decisions at the component level. We all benefit from
> that. The tricky bit is to agree on what we standardize across components.
>  We have a long tradition of letting component communities and those who
> step up to RM releases make a lot of decisions independently.  That does not
> mean that we shouldn't listen to feedback from the community or provide it
> when we have something to say.  Nor does it mean that we are not *all*
> responsible for *all* of our components.  Sometimes we disagree and
> sometimes it takes a while to get to consensus, but IMO we are *much* better
> off managing Commons as one community.


>> At this point, I really don't care what you guys do.  In the grand
>> scheme of things, it has absolutely zero impact on me what you do with
>> this code.  I'm tired of arguing about this stuff all the time.
> I am sorry you feel like that atm.

me too.


> Phil

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message