Noel J. Bergman wrote:
The current tally is extremely close (9 +1 vs. 8 -1 binding)
I don't want to close an issue with such a small margin.
I suggest that we should not change policy on anything like this lack of
concensus. I do, however, suggest that pressure be put on Maven to enforce
signing.
As no-maven still hasn't been justified, and when we authorized releases
we had not explicitly called out specific channels, so I'm wondering what
the "change" you refer to actually is :)
We seem to believe that Maven releases were "prohibited" but I don't ever
recall a vote on that. I do recall a policy change when we said "no more
people.a.o release artifacts!" and shifted it all to www.a.o/dist/incubator
But I suggest we don't close the vote until 10 days have passed, people do
have strong opinions, but not all of the opinions have been registered.
When it's this close, we need to be sensitive to the fact that some folks
are traveling or absorbed in work, on vacation, etc.
I think Daniel Kulp did a much better job that I did asking "what is the
basis or justification for this policy?" And the only two incubator
specific answers I've seen are "because we want incentive for projects
to graduate" which is non sequitur if releases are allowed at all, and
"because users will come to rely on incubating project artifacts" which
is an issue with the developer who creates the dependency to an -incubating'
artifact, and disclaimer to the consumer is that developer's issue.
Of course, the disclaimer does land in the META-INF/ tree along with any
licensing, correct?
Bill
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
