On Fri, Nov 8, 2013 at 8:38 AM, Sean Busbey <[email protected]>wrote:
> I believe that's a level of compatibility we haven't had in the past. Are > we sure we want to restrict ourselves like that? > I think its a good goal that causes less confusion over time. It certainly does not have to be a hard and fast rule. I think API changes in bug fix release should have a strong justification. > > -Sean > > > On Thu, Nov 7, 2013 at 9:49 PM, Josh Elser <[email protected]> wrote: > > > That's fine. I wasn't sure if that was going to be overkill or not. > > > > I hadn't considered wanting to run 1.5.1 code against 1.5.0. I guess I > > can't argue with that though. > > > > > > On 11/7/13, 9:28 PM, Keith Turner wrote: > > > >> I don't think this change needs to impact that API. I suspect just > adding > >> a cache w/ a really short timeout (like 50ms to 100ms) will give the > >> performance benefit we are looking for. Also making API changes in a > bug > >> fix release means its possible to write something for 1.5.1 that will > not > >> work w/ 1.5.0 > >> > >> > >> > >> > > > -- > Sean >
