Hi,
At 04:06 3/3/01 -0800, [EMAIL PROTECTED] wrote:
>Geir, I'm sure your DBCP will be ok in both models, and the result will be
>as "productised" as you want it to be. After all any component can become
>a real jakarta project ( like Ant did )
>and we can't usurp the right to
>create projects ( even "mini-projects" ) from the PMC.
I hadn't though about it in this way before - and I got to think - why not
? ;)
Presumably the library components that bubble to the top of the heap do it
because of merit. So if component foo has widespread use - it is likely
because it is good. There could even be an internal voting system setup so
components that are stable, feature complete etc have lots of votes. (And
to lend weight the names of voters/using projects are displayed on vote
tally).
>From what I gather the PMC deciding whether to accept projects are acting
as gatekeepers/protectors. They are meant to make sure a project is
technically sound/will be successful and that the participants exhibit the
Apache spirit. Projects are "paid" by being able to use Apache name (and
all benefits that brings) while Apache is paid by a successful project
because it increases Apache name. Apaches "pay" however is delayed till the
project is successful. So in effect the PMC is meant to mitigate risk.
In the case of a successful library component there is very very low risk.
The component is successful (hence technically sound) and it is built by
existing apache committers (and thus presumably done in Apache spirit). And
with the pushing down of power philosophy it fits right in.
Anyways thats another viewpoint ;)
Cheers,
Pete
*-----------------------------------------------------*
| "Faced with the choice between changing one's mind, |
| and proving that there is no need to do so - almost |
| everyone gets busy on the proof." |
| - John Kenneth Galbraith |
*-----------------------------------------------------*
