[LEAPSECS] CCTF reported to CIPM last month and it's interesting!

[Poul-Henning Kamp]

	https://www.bipm.org/documents/20126/280440045/11.+NMI+directors%2BMS+meeting-oct2025-CCTF+presentation-V5/b620db0a-bd29-2692-ebe9-83cf4ab7a0a2

Optical second to page 33, leapsecond from there to p58, then lunar time.

The proposed CGPM resolution (pg. 58) is 1h DUT1 tolerance from May 20 2027.

They hold the door open to 2028, and must decide before jan 13th
where the CGPM convocation goes out.  No reason given for why 2028
might be preferable.

Everybody (cited) wants "as soon as possible".

I note the uncertaintely in ER modelling:

The summary plot on pg, 50&57 show that the probability of a negative
leap-second in 2029 is already more than 10%, even assuming the
Director uses his discretion to delay it as long as possible.

And the "spaghetti" plot on pg. 49 shows TAI-UT1 from 36s…39s
already in 2030, and more than 32s…43 in 2035.

That can be fairly and precisly be summarized as "We have NO idea what
goes on with Earth Rotation" :-)

I guess we're done here ?

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.