Morgan Delagrange wrote:
> I'm not convinced that we should break from the mold
> of Apache development. I've seen good-intentioned
> contributions to my subproject that definitely did not
> meet production standards. However, while I favor 1a
> in theory, in practice I prefer the Taglibs method
> where there is a single pool of committers, but people
> are careful about contributing to an unfamiliar
> codebase. So my proposal is:
>
> (1c)
>
> There should be a single pool of Jakarta-util specific
> committers with the standard process for committer
> approval.
How about if, in practice, we give all the library committers karma to
the entire package library, but maintain a status file for each package
with a list of active committers. We could then encourage an etiquette
where a committer adds their name to the status file before making their
first commit to a package. This may sound like red tape, but I do
believe it would reinforce that each package is a product in its own
right, and needs to be handled on its own terms.
Does Taglibs do anything like this?
-Ted.
