On Mon, Oct 22, 2007 at 02:56:34PM -0700, Don Stewart wrote:
allbery:
On Oct 21, 2007, at 10:08 , Dougal Stanton wrote:
If it's not deemed useful for mainline I've packaged it as a contribution as well.
FWIW, if something *can* be expressed solely as a contrib, it generally *should* be. If it turns out to be a pattern used in many contrib extensions, then it could be a candidate for the base.
Yes, that's the process. New experiments go into the contrib code, if they're successful, they end up in the core.
I'd caveat that: if the experiments are successful, and would be more elegant in the core, they go in. I wouldn't call Tabbed unsuccessful, but se no reason for it ever to go into the core. -- David Roundy Department of Physics Oregon State University