incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "William A. Rowe Jr." <>
Subject Re: Insanity. Apache Incubator should be about education (was: [PROPOSAL][VOTE] Subversion)
Date Tue, 10 Nov 2009 18:08:40 GMT
Greg Stein wrote:
> On Tue, Nov 10, 2009 at 03:48, William A. Rowe, Jr. <> wrote:
>> Quite frankly, all svncorp releases could, with reasonable documentation
>> [read: mailing list archives, CLA's and code grant] be licensed as ASF
>> releases under the AL 2.0, irrespective of their internal artifact
>> copyright statements.
> I doubt it. Those old releases are signed tarballs. We can't "reach
> in" and alter the LICENSE file without re-signing the whole tarball,
> and I think that would be a very bad idea.

We don't.  I didn't say re-licensed; I said additionally licensed.  It's as
simple as putting the tarballs into a directory which says "XYZ are further
licensed under the Apache License 2.0".  Nothing needs to be altered to give
users a license.

>> A proviso that 1.7.0 won't be approved without running it through RAT,
>> either pre or post graduation seems sufficient.  The process is better
>> documented than 95% of ASF project release processes, so there's no issue.
> RAT can be run right now, and the podling can work against its
> results. No issue there. The *release* of "something" is my pain
> point.

+1; although we both know that extra artifacts 'appear' magically during
most assembly processes, and that has bit us before.

> And yes, the PMC that will manage the svn project can/should have a
> responsibility to use RAT. But if you "make that rule", then you
> better impose it upon every PMC here at the ASF. That's effectively
> what you're saying :-)

No, I'm saying give SVN a pass on demonstrating the [already demonstrated]
ability to have an effective release process; *contingent* upon running RAT
on the first release artifact created after graduation.  That's what I am

>> But ranting against your perception of Incubator's failure to EDUCATE and
>> TEACH podlings how the ASF environment works is really quite disappointing,
>> coming from you.
> Look at the context. Being asked to throw together some bits for a
> "release". Oh, just any bits will do. But wait, since they aren't
> quite proper, you don't really have to announce it to users. ... come
> on, that is not education. That isn't teaching anybody anything.

We don't disagree.  So stick around long enough to make a real release that
the Incubator PMC can validate, or come to a reasonable exception that the
Incubator can accept.  But don't go flying off into rants about process that
the board has *charged* the Incubator with defining and enforcing :)

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

View raw message