I've always felt that arbitrarily breaking a tie is just asking for trouble when ties actually occur since the choice that's made is rather likely to be wrong. At this point it seems fairly clear that definition-time warnings about method ambiguities are usually too annoying, especially between unrelated modules. It would probably be best to retain warnings for ambiguous methods within a module and make invocation of other ambiguous cases a runtime error.
On Mon, Apr 18, 2016 at 3:38 PM, Steven G. Johnson <[email protected]> wrote: > > > On Monday, April 18, 2016 at 10:38:28 AM UTC-4, Didier Verna wrote: >> >> >> Julia warns you when there's an ambiguity in method specificity, and >> picks one "arbitrarily" (according to the manual). I guess arbitrarily >> doesn't mean random. Is there a particular reason for not >> standardizing a tie-breaker (possibly the one currently in use) ? >> > > Because any practical tie-breaker rules are probably too complicated to be > useful? >
