Right, this is exactly what I was thinking for the sandbox. I am the author of quite a few Atom extension drafts that I would like to play around with implementing. Having a sandbox to play around in is easier/better than messing around with a branch in that, unless the extension is broadly adopted, likely won't ever make it into the core project.
I'm fine either way, however. - James Garrett Rooney wrote: > On 6/8/06, Paul Querna <[EMAIL PROTECTED]> wrote: > >> What is the difference between a 'sandbox' and a >> branches/paul-is-trying-something-crazy? >> >> I don't see much reason to not just use branches/ for that purpose. > > For branches, I agree, an experimental branch to make radical changes > just to try them out is just that, a branch, and goes under branches/. > A sandbox, IIUC, is something different, it's for new work, not > modifications to existing code, and is for work that may or may not > ever make it into the trunk. > > For example, see > http://svn.apache.org/repos/asf/jakarta/commons/sandbox/ for the > Jakarta Commons sandbox. It's filled with random little projects that > may never really see production use. > > Depending on the type of work people do, we may never need a sandbox. > If the type of experimental work that goes on is such that it lends > itself to branch based development, then that makes perfect sense. > > On the other hand, I could see sandbox projects for implementing > specific Atom extensions that may or may not ever become widely used. > > -garrett >
