(In reply to comment #0)

at least within some sensible time limit.

If this is time-based, a very good eye should be kept on the Apaches' server times remaining synchronized. API login throttling once broke pretty badly because one server's clock was a few hours off, but if this ever gets implemented, even a few minutes of clock drift would be bad (newbies typically don't take the "ah, this is probably just clock drift, let's resubmit in the hopes of hitting another Apache" approach).

happy.melon.wiki wrote:

If "sensible time limit" happened to coincide exactly with the maximum age of the recentchanges table, it would make things much easier :D "if everything that needs to be deleted is still in the table, then go" would be reliable (and it would all need to be deleted anyway).

herd wrote:

This seems like a dupe of bug 14325 (though that is more about making pages deletable if the current user is the only author) or bug 17309 (which is more about users deleting pages in their own namespace).

FT2.wiki wrote:

Maybe some discussion of which of these circumstances a user should be able to delete their own addition, would be sensible.

ie, should a non-admin user be allowed to delete a page where they are the sole editor:

1/ in their userspace, on all pages, or never
2/ if it has had one, up to <N>, or any number of edits
3/ for a limited amount of time, or any time

Useful abilities, but their deployment would be a policy matter for individual projects. Those would probably be the most obvious parameters.