On 2011 Sep 18, at 14:10, Poul-Henning Kamp wrote: > The fundamental problem is that a vast majority of the worlds > software is written as if leap seconds simply do not exist. > > The fact that they do, are horribly expensive to test, and tend not > to get tested because *recently* they have not happened a lot, is > merely icing on the cake.
Handling leap seconds, and testing them, become trivial if their representation is changed. Anybody whose system uses the tzdata and tzcode that comprise the zoneinfo scheme can test them, repeatedly. But in their "leap smear" blog post what Google has done is to affirm what I had written in http://www.ucolick.org/~sla/leapsecs/right+gps.html that is to day, as currently implemented it takes a huge amount of resources in order to test and maintain a workaround. The engineers know that GPS time is a more suitable representation for use by POSIX, but the standards and compliance folks object to the use of GPS time. Fixing the standards can be done without breaking the connection between clock and calendar because the mechanism has already been coded by Arthur David Olson and the tz list folks. It takes only a week or so of being subscribed to the tz mail list to see that they routinely handle far more difficult scenarios than leap seconds. -- Steve Allen <s...@ucolick.org> WGS-84 (GPS) UCO/Lick Observatory Natural Sciences II, Room 165 Lat +36.99855 University of California Voice: +1 831 459 3046 Lng -122.06015 Santa Cruz, CA 95064 http://www.ucolick.org/~sla/ Hgt +250 m _______________________________________________ LEAPSECS mailing list LEAPSECS@leapsecond.com http://six.pairlist.net/mailman/listinfo/leapsecs