incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Kevan Miller <>
Subject Re: [PROPOSAL] Validation incubator for JSR-303 Bean Validation
Date Mon, 18 Jan 2010 23:30:35 GMT

On Jan 18, 2010, at 11:30 AM, Noel J. Bergman wrote:

> Kevan,
>> Time to restart/finish this discussion.
> I agree.  Seems that things have cooled off a bit.
>> Personally, I'd have been happy to see this move forward either way
>> 1) IP clearance with implementation work in Commons
> This works only if we're dealing with a CODEBASE and an existing ASF
> community that takes over it, albeit with the addition of a small number of
> new members who are part of, but not the entire, developer base for the
> code.
>> 2) Incubator project with graduation to Commons (or other TLP).
> This is the correct path if we need to incubate a community.
> Again, to summarize: we CLEAR code, we INCUBATE communities.
>> We seem to have talked our way into doing nothing.
> Option 3 is probably not the correct choice.  :-)


There's a sliding scale between those 2 options. And people may have their own evaluation
of where this instance falls. Summarizing the issues:

There's an existing Commons Validator component. However, it's been mostly dormant. Commons
provides the necessary oversight of the project, but there is small community of Commons committers
working on the Validator component. AFAICT, there is not enough interest in the Commons community
to implement the latest Bean Validation specification, on their own.

There's a partial (85%) implementation of the new JSR 303 Bean Validation Specification that
Agimatec would like to donate to the ASF. 

Using the initial committers list as a guide to the "community" that would be formed around
this codebase: there is one Commons PMC member/committer, 6 ASF committers (but not Commons
committers), and 2 non-ASF committers. 

I think we'd agree that a fair amount of community building will be required for this new
codebase and group of committers. A motivated TLP might be able to provide the necessary oversight
to build this new community. However, given the small makeup of the Commons Validator community,
I don't think it's reasonable to expect the Commons community to do this community building.

IMO, Incubator should support the creation of a new Validation project at incubator and let's
start the community building...

>> The Geronimo project is going to need a Bean Validation implementation for
> EE6
>> compliance. So, I'm confident that there is enough interest to create an
>> implementation at Apache. I'd prefer that it end up in Commons (since this
>> is really an SE technology). However, I can start discussions in the
> Geronimo
>> community, if that's what it takes.
> That would be great, but given the criteria above, which direction do you
> feel that this promotes?

It only provides an alternative TLP which might have the necessary interest in building a
community around this new codebase. If we are stuck in a stand-off between Commons and Incubator,
this might be an alternative. However, I certainly hope it doesn't come to this...

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