[LEAPSECS] W1K GPS rollover for some time servers

[Warner Losh]

> On May 5, 2015, at 9:19 PM, Rob Seaman <seaman at noao.edu> wrote:
> 
> On May 5, 2015, at 6:16 PM, Warner Losh <imp at bsdimp.com> wrote:
> 
>> This is an excellent example of the unintended consequences of leap seconds,
>> and the ways they insinuate themselves into non-obvious parts of the code.
> 
> ...and is it somehow impossible that there might be unintended consequences from permitting UTC to drift away from actual Universal Time?  There are very obvious *intended* consequences that will negatively impact astronomical, aerospace and navigational systems.

Engineering is the choice of which consequences you want to have. Leap seconds are nearly impossible to implement correctly. It is time to look at other solutions to the synchronization problem that can be implemented correctly. The only people that care are people that point things at the sky. Sadly, many of these are proprietary and may break if there’s a change. Things are already broken in other areas, but since those areas don’t affect pointing at stars, it is hard to get traction to find some common ground.

Warner

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 842 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <https://pairlist6.pair.net/pipermail/leapsecs/attachments/20150505/e8662137/attachment.pgp>