[time-nuts] FW: Bulletin C number 30
M. Warner Losh
imp at bsdimp.com
Tue Jul 5 11:44:46 EDT 2005
In message: <126.96.36.199.0.20050705100343.0412ac70 at mjs.alientech.net>
: No, you didn't. What you and Warner HAVE demonstrated is that you
: _chose_ the wrong time coordinate system for your
How do you know this? I've demonstrated no such thing. I told you no
such thing in any of my posts. You have made an assumption that turns
out to be wrong.
: Instead of using TAI, which doesn't have leap
: seconds, you chose to use UTC, which does.
You would be wrong there. We *DO* use TAI for our internal time
keeping. The trouble with that is two fold. One: GPS receivers tend[*]
to give you time in UTC and you need to convert the one to the other.
Second: Users want to see the UTC time on their atomic clocks, time
code counters, etc. So you're stuck displaying UTC. Both of these
are reasons for needing to know the leap seconds involved. And No,
the users aren't interested in TAI time, so displaying it instead is
not an alternative.
Internally, all the software I've written uses time scales without
leap seconds. However, that doesn't get away from any problem except
the 't1 - t2' problem you have in utc.
: You're now arguing that
: the single differentiating factor between UTC and TAI should be
: removed to resolve issues cause by your poor choice. You've offered
: nothing in the way of rational argument as to why that should
: happen, except that it would be convenient for YOU. That is not
One can easily draw false conclusions from a false premise. Since I
do not use a poor timescale in my applications, but have indeed gone
the TAI route, I can tell you first hand how much leap seconds
complicate things in many different ways.
[*] Some GPS receivers will give you GPS week/second of week time,
which you can recover TAI from (well, at least partially given the
1024 week roll over problem). However, if you just have TAI time, you
can't display UTC to the users. You can get the leap second offset
from the GPS stream, but that is delivered relatively infrequently in
the almanac. Some GPS receivers will remember these values from their
last power on, others won't. Those that don't take 20 minutes to know
what time it really is (in the UTC time scale)... Another
complication of leap seconds. One can, of course, work around these
issues by knowing when you last knew the offset and making educated
guesses about the likelihood of the number changing, but given the
rate of the earth's roational change, these guesses will be obsolete
in 50 to 100 years.
More information about the time-nuts