off-by-one-hour error in OST Docket No. 2005-22114

[Paul Eggert]

Paul Schauble <Paul.Schauble at ticketmaster.com> writes:

> Doesn't the first proposal have the problem that there is no 2:59:59 EST,
> the Eastern zone having changed to DST?

No, because the affected locations were observing EST without DST.

It is a bit tricky, in that the "simpler" solution here relies on the
fact that there's no DST before the change.  The "more complicated"
solution is a recipe that will work when the DOT makes more-typical
changes, e.g., in North Dakota, which observes DST both before and
after the change.