|
|
|
|
|||||
|
||||||
From: Fredrik Noon (no email)
Date: Mon Aug 11 2003 - 15:41:42 EDT
Paul Hirose <> writes:
> This old message from 1988 indicates leap second bugs have occurred
> many times. I'm not sure I believe that story about the French,
> though.
>
> http://www-mice.cs.ucl.ac.uk/multimedia/misc/tcp_ip/8801.mm.www/0022.html
An excerpt from the above link:
The earth's rotational poles wander over a roughly circular path
every 14 months or so. This is caused by changes in the earth's mass
distribution (solar tides, seasonal atmosphere mass shifts, melting
snow caps, etc). The amplitude of the polar variation is roughly the
size of a baseball diamond, and it causes small changes in the
observer's true latitude and longitude -- therefore affecting the
observer's astronomical observations on the order of 30 ms.
This seems rather basic: does it mean that ones latitude and longitude
isn't fixed, but wanders over this baseball diamond-sized area? Are
the latitude and longitude lines on a (relatively small-scaled) map
the "average latitude" and "average latitude," or am I missing the
point here?
Btw., I'm not in the "time business" myself, but, as a programmer, I
don't see the point of getting rid of UTC. Sure, software can have
bugs, but this proposal won't result in bug-free software: indeed, it
will probably create confusion in the short-term. Time is a
complicated thing regardless, and (as another poster commented) the
leap second business is probably the _easiest_ thing one needs to deal
with (albeit it involves unpredictable, unforcastable elements that
can't be precalculated to arbitrary points in the future). The bottom
line is that programmers should rigorously test such code, but often
don't (for a variety of reasons). It's the same story in other fields
as well.
There's a funny parallel to the Julian Calendar. I'm an Orthodox
Christian in a parish that uses the Julian Calendar for all liturgical
purposes, but over the course of the 20th Century many jurisdictions
went over to the Gregorian Calendar for most purposes (except the
Paschal Cycle). Now, Pope Gregory XIII thought he'd be stopping the
procession of the equinoxes, and those Orthodox that originally made
the change did so to promote Ecumenism, but the "common man's"
argument these days among Orthodox "New Calendarists" seems to be that
the Gregorian Calendar is simply "more accurate," because it better
matches the Tropical Solar Year (as opposed to, say, the Sidereal Year
or some such measurement). The proposal to abandon UTC seems to go
the other way: decouple time from the movement of the Earth. I
presume that the "common man" who would object to seeing his "Spring
holiday" end up in the Fall (unless he changed hemispheres!) would
also object to breakfast time ending up at midnight. Granted, all
this will take a *very* long time to become noticeable, but I think
the principle should be considered: people in the modern West are used
to an Earth-based time standard, and the "common man" wouldn't see
much use in having anything else. UTC seems to be a good compromise.
*And* if it can upset the French holiday schedule, that's an added
bonus! ;>
(As I've said, I'm not in the "time business," so may have a few
errors in my understanding of the workings here. Actually, I'm not
even in the "navigation business": I enjoy mathematical tinkering and
measurment, and can go on vicarious sea voyages on this list without
having to deal with salt water!)
/Fredrik
+-----------------------------------------------------------------+
| Symeon | Fredrik Noon, Senior Software Engineer |
| | Hifn, Inc. |
| www.noon.org | +408/399-3630 (USA) |
|-------------------+---------------------------------------------|
| pgp key: <ftp://ftp.noon.org/fcn.asc> DH/DSS 4093/1024:7840AC55 |
| fingerprint: 5FB8 29B0 7F6D 24FA 6AD4 07E4 022E 0C58 7840 AC55 |
+-----------------------------------------------------------------+
|