cheap GPS

Sean Donelan SEAN at SDG.DRA.COM
Mon Aug 23 02:34:42 UTC 1999


>It's too early to say with high confidence, but you had to stretch pretty
>far away from the question you asked (will it cause time problems with
>NTP, etc) to find a problem with the GPS rollover.  General reports are
>that it went better than expected and that few critical systems were
>impacted. If you had asked whether anyone would be hit by the rollover, I
>would have said of course there were going to be systems and people hit.
>It may be fun to watch the PR wars that result.

I carefully avoided asking about NTP or any specific uses of the time
signals.  My question was would the GPS rollover cause more or fewer
problems than when NIST set daylight savings time wrong in 1992.
Approximately 600 people have reported problems, so far, due to the GPS
rollover.  NIST claimed fewer than a dozen reported problems when daylight
savings time was set incorrectly in 1992.

    12 < 600

Both are pretty small numbers, but as you point out people aren't very
good at understanding extremely rare risks.  You need to be carefull when
you say XYZ will cause almost no problems, you didn't mean they specifically
wouldn't have a problem with XYZ.  See airplane crash, nuclear plant
meltdown, etc.  It is a PR problem, and technical people aren't always
the best ones at explaining how a risk affects the public.
-- 
Sean Donelan, Data Research Associates, Inc, St. Louis, MO
  Affiliation given for identification not representation




More information about the NANOG mailing list