Exactly correct. Until the need for optimization shows up, this is all about curiosity, not implementation need.
On 3/14/08 11:31 AM, "Jeff Eastman" <[EMAIL PROTECTED]> wrote: > Not to diminish their > importance, the impact is well localized and subject to optimization in the > future if we can get empirical data that indicates they need it. So far, I > think the results have been somewhat surprising and counter-intuitive.
