[LEAPSECS] draft revision of ITU-R TF.460-6

[Warner Losh]

On Dec 8, 2011, at 1:04 PM, Zefram wrote:


> Warner Losh wrote:

>> TAI is a paper clock.

> 

> In that sense, so is UTC.  The proposed revision leaves intact the

> sentence about TAI and UTC differing by an integral number of seconds.


True, but UTC(k) is realized in real-time, while nobody does a TAI(k) in real time.


>> UTC, on the other hand, is actively measured and computed in real-time

>> and physical devices are steered to be on-time and on-frequency for UTC.

> 

> These devices serve equally well as a realisation of TAI.

> 

> I'm failing to see any relevant difference between TAI and UTC here, other

> than that the author of the amendment wants his mitts on the name "UTC".


One could easily construct a TAI time scale, but the national labs just construct a UTC one.

I'm not saying that this is a good argument.  I'm just explaining the distinction that the time guys are trying to make.

Warner