[ntp:hackers] smearing the leap second

Warner Losh imp at bsdimp.com
Fri Jul 10 16:22:41 UTC 2015

> On Jul 10, 2015, at 10:11 AM, Mike S <mikes at flatsurface.com> wrote:
> On 7/10/2015 11:56 AM, Warner Losh wrote:
>> Coupled with the attitude of “it’s only a second, why should I care?”
>> It seems the whole enterprise is working around the fact that leap
>> seconds break the fundamental assumption that we all learned in
>> school that minutes always have 60 seconds. An assumption that goes
>> back a very long time.
> That goes hand-in-hand with the fundamental linkage of human time with Sol, which goes back even further. When civil time moved from GMT to UTC, maintaining that linkage was given priority to maintaining the 60 second convention. Now there are people who want to reverse that decision to fix their own flaws which they've had 45 years to fix, without regard to the fact that it will break the systems of people who rely on that linkage. Selfish in the extreme.

One could easily argue that anything that takes 45 years to fix made the wrong trade off in the first place because it breaks too many deep-seated assumptions. Resisting this data to preserve the flawed system seems equally selfish in the extreme. We will not solve this argument by characterizing the other side’s position as “selfish” because that brush tars both ways.

The shift to atomic time is the real reason there’s a discrepancy. Even the “UTC” system itself has changed since it started in the early 1960s from the rubber seconds and ~100ms leaps to what we have to day, presumably at the cost to the current incumbent players (I know that LORAN C operations shifted from UTC time to LORAN time in 1972 because of this shift, for example). If the current standard is not workable, despite 45 years of trying, perhaps it is time to reevaluate the standard. Of course, that “if” is a big one, and one where there’s much spirited debate.


-------------- 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: <http://lists.ntp.org/pipermail/hackers/attachments/20150710/078f3fb3/attachment-0001.sig>

More information about the hackers mailing list