On Fri, Mar 13, 2015 at 1:01 PM, Stefan Karpinski <[email protected]> wrote: > Also, given that Julia time is based on UT, Time is well-defined – a Time is > conceptually an equivalence class of DateTimes that differ by exactly an > integral number of days. If you think of a Date as the set of DateTimes that > occur during the same UTC day, Date x Time is naturally isomorphic to > DateTime.
I'm not an expert, but doesn't the existence of leap seconds in universal time mean that the last statement is false? For certain dates, it will sometimes be necessary to check if they harbor leap seconds, so I don't see how there is a clean factorization of DateTime into Date x Time. Thanks, Jiahao Chen Staff Research Scientist MIT Computer Science and Artificial Intelligence Laboratory
