[LEAPSECS] BBC radio Crowd Science

[Eric R. Smith]

On 01/02/17 08:39 AM, Steve Summit wrote:
> On further reflection, I think we're all right.  For every
> let's-look-at-the-arithmetic argument that suggests we should
> use the "new" offset during the leap second, I can come up with
> one which suggests the opposite.  (Basically it depends on
> whether you come at the leap second "from below" or "from above".
> I'll send the longwinded details in a separate message, if anyone
> actually cares.)  So I'm right, and you're right, and Warner's
> right, and Steve Allen is especially right in his assertion that
> it's just inherently, fundamentally ambiguous.

The mapping between UTC and TAI is unambiguous. As Zefram pointed out 
earlier, one second after the end of the leap we have 
UTC=2017-01-01T00:00:01.0, TAI=2017-01-01T00:00:38.0. We can count 
backwards by half seconds like so:

    UTC=2017-01-01T00:00:00.5  TAI=2017-01-01T00:00:37.5
    UTC=2017-01-01T00:00:00.0  TAI=2017-01-01T00:00:37.0
    UTC=2016-12-31T23:59:60.5  TAI=2017-01-01T00:00:36.5
    UTC=2016-12-31T23:59:60.0  TAI=2017-01-01T00:00:36.0
    UTC=2016-12-31T23:59:59.5  TAI=2017-01-01T00:00:35.5
    UTC=2016-12-31T23:59:59.0  TAI=2017-01-01T00:00:35.0

To determine the offset during the leap second we need to find:

2017-01-01T00:00:36.5 - 2016-12-31T23:59:60.5

I think the most reasonable interpretation of that offset is +36. But in 
some sense it doesn't "really" matter, as long as whatever method you're 
using comes up with the correct labels for the seconds.

Regards,
Eric