Re: Internet Time, GMT, UTC et.c.

1999-02-25 Thread John Shepherd

Slawomir K. Grzechnik wrote:


2. Paris meridian, or rather opposition to Greenwhich Meridian, was still
seriously raised on international conferences before the First World War.
French lost the Prime Meridan finally but won meters, liters and kilograms,
very useful BTW, not yet in the US. In order not to run out of causes for
holy wars we should prepare new in advance.


I was in France this last Summer and found that they are still selling maps
based on the Paris Meridian with the Latitude and Longitude measured in
grads (the Metric degree with 400 per full circle). In fact most of the
maps in the stores were this way with a few showing this and the UK system
as well.

Cheers,

John


Professor John P.G.Shepherd
Physics Department
University of Wisconsin-River Falls
410 S. 3rd. St.
River Falls,WI 54022

Phone (715)-425-3196, eve. (715)-425-6203
Fax (715)-425-0652

44.88 degrees N, 92.71 degrees W.



Re: Internet Time, GMT, UTC et.c.

1999-02-25 Thread Slawomir K. Grzechnik

Hello Malcolm, Richard, Gordon, Bill, John, John, Fernando et. al.

I must say that I track with pleasure this off topic subject. Yes, time was
and is very emotional issue.

1. I would not worry about Internet Time, the name is a marketing trick. In
reality time service on the Internet is taken very seriously these days.
Most servers use internally UTC (GMT) and are synchronized through the net
which is not that simple if you take into account variable route and time
of arriving packets. Yet more or less it is done fine thanks to smart IETF
engineers and few protocols they devised. Really important servers are
synchronized by chips receiving time signals. 

The only operating system that took time seriously from the very beginning
(early 70-ties) is Unix, which by default uses internally UTC. This by no
means impedes the displaying of dates and times in proper local time for
user convenience. Unix generally is not affected by the famous Year 2K
problem however applications written improperly are. All Unix systems count
seconds since Jan 1, 1970 00:00:00 UTC. Because many use 32 bits to store
seconds count their final date falls somewhere in 2038. This is no big deal
because 40 years is much too much time, at least for Linux hackers, to move
the system to 64 (or more) bits processors. This is our luck that Internet
is still based on Unix.

Linux (free Unix) calendar program cal has range 1- years. It does
include the transition to Gregorian Calendar which in the British Empire
took place in 1752. In Catholic countries the reform was introduced in
1582, as prescribed by Pope Gregory, and in Oct of that year 11 days were
skipped. There were reports of rioting in many places as people believed
that the Pope robbed them of 11 days of their lifes. 1600 was the first
centurial year that was a leap year in countries that accepted the calendar
reform in 1582. 

This is the output from Linux cal for Sep 1752 (Unix was originated on
American soil so 1752 for skipping days is correct)

   September 1752
Su Mo Tu We Th Fr Sa
   1  2 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30

The boldest reform of calendar was introduced during French Revolution and
stayed quite long. If I remeber well the Gregorian Calendar was
reintroduced by Napoleon in 1806, that is few years after his coronation
and next year after Austerlitz.

2. Paris meridian, or rather opposition to Greenwhich Meridian, was still
seriously raised on international conferences before the First World War.
French lost the Prime Meridan finally but won meters, liters and kilograms,
very useful BTW, not yet in the US. In order not to run out of causes for
holy wars we should prepare new in advance.

3. Richard gave http://www.rnw.nl/realradio/information/html/time.html to
search for explanation of times. I would add two more, at USNO and former
Greenwhich Observatory (yes, it was closed in Oct last year)

http://tycho.usno.navy.mil/systime.html
http://www.rog.nmm.ac.uk/leaflets/time/time.html

The second link contains lots of other short leaflets which may be of
general interest. To keep the long story short. In the past pendulum clock
prooved to be more accurate time keeper than the Sun. This prepared the
introduction of mean solar time. Today similar thing happened again because
of atomic clocks. TAI is atomic time with its well defined second as number
of specific oscillations of cesium atom. UTC uses same time scale or rate
and two conditions must be met

(*) UTC readings may differ from TAI by integer number of seconds
(**)UTC may not differ by more than 0.9s from time UT1 determined by mean
solar motion that includes some long range phenomena like motion of poles,
et.c.

Because of those two conditions UTC may be adjusted twice a year (Jun 30,
Dec 31) and a leap second is used if necessary. Adjustments justify
Coordinated in time name.

Old GMT is more or less UTC. The name is seldom used outside UK now.

Slawek



At 04:17 AM 2/24/99 +, you wrote:
Dear John, John ,Troy, Fernando et. al.

In my case Internet time is that time between 6pm(local) and 8am(local)
when the teleco charges are less per minute.
With prime surf-time being between midnight(local-civil)Friday and
midnight (local-civil)Sunday when the charges are least ( about a
loaf-of-bread per minute.) Regardless of when lunch-time is in the
big apple !!

The bill that the teleco sends me each month is much more important
than any other Bill ! So watch-out anyone sending big attachments.

Atleast the time on my dial is free. (see, I am trying to be on-topic!)

Now to the point of this missive :
The last gasp of a dying ( u mean dead ) British empire is/was
GMT.(also dead). The UT thing is an international got-up consortium
thingy. Dont blame that on us English, please :-) we have enough crosses
to bear as it is.
I dont want to start another 100-year war G but when was the last time
you heard of the Paris meridian,
whoosh --

Have fun,
Malcolm.
in SW England, happily my 

Re: Internet Time, GMT, UTC et.c.

1999-02-25 Thread Tom Semadeni

Hello Slawomir,

Tremendous post.

Slawomir K. Grzechnik wrote:

 In reality time service on the Internet is taken very seriously these days.
 Most servers use internally UTC (GMT) and are synchronized through the net
 which is not that simple if you take into account variable route and time
 of arriving packets. Yet more or less it is done fine thanks to smart IETF
 engineers and few protocols they devised. Really important servers are
 synchronized by chips receiving time signals.

How is the synchronization done to preserve precision and accuracy between
stations?  There are so many technologies nowadays which are dependent upon
consistently accurate and increasingly precise time (GPS, Broadcasting, Space
work, etc.) that synchronization among stations is, as you say, not trivial.
(UTC time updates at different locales would be dependent upon path lengths and
characteristics of the transmission media.)
Is signal dispersion an issue?  (i.e. difference between phase and group
velocity)
Are we into relativistic effects when transmitting/receiving between stations
with a relative velocity?

 Linux (free Unix) calendar program cal has range 1- years. It does
 include the transition to Gregorian Calendar which in the British Empire
 took place in 1752. In Catholic countries the reform was introduced in
 1582, as prescribed by Pope Gregory, and in Oct of that year 11 days were
 skipped. There were reports of rioting in many places as people believed
 that the Pope robbed them of 11 days of their lifes. 1600 was the first
 centurial year that was a leap year in countries that accepted the calendar
 reform in 1582.

 This is the output from Linux cal for Sep 1752 (Unix was originated on
 American soil so 1752 for skipping days is correct)

September 1752
 Su Mo Tu We Th Fr Sa
1  2 14 15 16
 17 18 19 20 21 22 23
 24 25 26 27 28 29 30

I recall reading (somewhere) that the riots were caused by the fact that
landlords charged one month of rent for October, 1582 while the tenants got only
20 days of accommodation.  I wonder what the money-lenders did!  How were dates
communicated between Catholic countries and British Empire countries during the
period 1582-1752?

Thanks again for a most informative and clarifying post.

Tom  Semadeni
Britthome Bounty   *
45.768* North   80.600* West