Loading ...
Sorry, an error occurred while loading the content.

Re: [ISO8601] Problem with ISO 8601? Discussion on leap year

Expand Messages
  • hjwoudenberg@aol.com
    _http://en.wikipedia.org/wiki/Talk:Leap_year_ (http://en.wikipedia.org/wiki/Talk:Leap_year) Is there a 4000-year rule? It has been suggested (by the
    Message 1 of 25 , Mar 20, 2006
    • 0 Attachment
       

      Is there a 4000-year rule?

        It has been suggested (by the astronomer John Herschel (1792-1871) among others) that a better approximation to the length of the tropical year would be 365 969/4000 days = 365.24225 days. This would dictate 969 leap years every 4000 years, rather than the 970 leap years mandated by the Gregorian calendar. This could be achieved by dropping one leap year from the Gregorian calendar every 4000 years, which would make years divisible by 4000 non-leap years.

        This rule has, however, not been officially adopted.

    • NGUYEN Ivy
      I ve always thought that having a 4000-year rule would indeed make sense, but didn t know whether it would truly be more accurate. I guess it is... :-)
      Message 2 of 25 , Mar 20, 2006
      • 0 Attachment
        I've always thought that having a 4000-year rule would indeed make
        sense, but didn't know whether it would truly be more accurate. I
        guess it is... :-)


        On 20/03/06, hjwoudenberg@... <hjwoudenberg@...> wrote:
        >
        > http://en.wikipedia.org/wiki/Talk:Leap_year
        >
        >
        > Is there a 4000-year rule?
        >
        >
        > It has been suggested (by the astronomer John Herschel (1792-1871) among others) that a better approximation to the length of the tropical year would be 365 969/4000 days = 365.24225 days. This would dictate 969 leap years every 4000 years, rather than the 970 leap years mandated by the Gregorian calendar. This could be achieved by dropping one leap year from the Gregorian calendar every 4000 years, which would make years divisible by 4000 non-leap years.
        >
        > This rule has, however, not been officially adopted.
      • John Hynes
        ... This is incorrect. The Gregorian reform was not intended to match the mean tropical year, as is often erroneously stated. The Gregorian calendar was
        Message 3 of 25 , Mar 20, 2006
        • 0 Attachment
          > I've always thought that having a 4000-year rule would indeed make
          > sense, but didn't know whether it would truly be more accurate. I
          > guess it is... :-)

          This is incorrect. The Gregorian reform was not intended to match the
          mean tropical year, as is often erroneously stated. The Gregorian
          calendar was declared in the Inter Gravissimas, which specifically
          states that the purpose of the reform is to prevent the date of the
          vernal equinox from receding. The mean tropical year is often stated
          to be the same value as the VE year, but in fact is not. The mean
          tropical year is an average of all points on the zodiac, and differs
          in value from the mean period between successive vernal equinoxes, as
          can be easily determined by anyone from the times and dates of
          historical equinoxes. This value is approximately 365.24238 days, and
          thus more closely aligned to the Gregorian calendar year of 365.2425
          days than to the mean tropical year of 365.24219 days. This results
          in an error of less than one day in 8000 years, although the exact
          length of the day and year varies and is not known that far into the
          future.

          For more information, read the articles at
          http://www.hermetic.ch/cal_stud/cassidy/ or do the calculation
          yourself by simply averaging vernal equinoxes between any two years
          over a sufficiently long period. You can find a list of dates and
          times for 1452 to 2547 at
          http://www.newscotland1398.net/equinox/eqindex.html

          Average for past 200 years:
          2006 Mar 20 18:26 UT = JD 2453815.2680555
          1806 Mar 21 06:52 UT = JD 2380766.7861111
          2453815.2680555 - 2380766.7861111 = 73048.4819444 days
          73048.4819444 / (2006-1806) = 365.2424097 days

          Average for past 554 years:
          2006 Mar 20 18:26 UT = JD 2453815.2680555
          1452 Mar 20 11:52 UT = JD 2251470.9944444
          2453815.2680555 - 2251470.9944444 = 202344.2736111 days
          202344.2736111 / (2006-1452) = 365.2423711 days

          The average value will vary slightly depending on exactly which years
          are used, since the absolute value changes year to year, but it
          converges between 365.24237 and 365.24238 days, not 365.2422 days.

          John Hynes

          --- In ISO8601@yahoogroups.com, "NGUYEN Ivy" <nguyenivy@...> wrote:
          >
          > I've always thought that having a 4000-year rule would indeed make
          > sense, but didn't know whether it would truly be more accurate. I
          > guess it is... :-)
          >
          >
          > On 20/03/06, hjwoudenberg@... <hjwoudenberg@...> wrote:
          > >
          > > http://en.wikipedia.org/wiki/Talk:Leap_year
          > >
          > >
          > > Is there a 4000-year rule?
          > >
          > >
          > > It has been suggested (by the astronomer John Herschel (1792-1871)
          among others) that a better approximation to the length of the
          tropical year would be 365 969/4000 days = 365.24225 days. This
          would dictate 969 leap years every 4000 years, rather than the 970
          leap years mandated by the Gregorian calendar. This could be
          achieved by dropping one leap year from the Gregorian calendar
          every 4000 years, which would make years divisible by 4000 non-leap years.
          > >
          > > This rule has, however, not been officially adopted.
          >
        • John Steele
          It is unlikely in the 1580 s, Pope Gregory had the astronomical knowledge to think this way. However, the difference between the tropical year and the VE year
          Message 4 of 25 , Mar 21, 2006
          • 0 Attachment
            It is unlikely in the 1580's, Pope Gregory had the astronomical knowledge to think this way. However, the difference between the tropical year and the VE year (and years measured by the fall equinox and solstices) is a function of where the earth is in its precession cycle (period around 26500 years), which also determines which star is the North Star.
             
            The Gregorian reform probably set out to stabilize the VE year with the knowledge of the times. I would argue stabilizing the tropical year is the best way to fix it, averaged over a precession cycle, and has the advantage of stabilizing, on average, all the seasons.  It could also be argued as being "fairer" to people in the Southern Hemisphere, whose agricultural cycle is linked to the fall equinox, and to people of other faiths who are not concerned with Easter.  The fixing the calendar to a tropical year is completely consistent with a world calendar that serves everybody. Christians need to assess whether an extra 24 hours or so in the precision of Easter is worth having many people prefer other calendars.
             
            I would argue fixing the calendar to the tropical year should be the "modern" goal regardless of exact intent in 1582; they're all dead.  Therefore we need to watch the error between the tropical year and the Gregorian year of 365.2425 days, and determine when a change to the calendar is warranted.
            John Hynes <john@...> wrote:
            > I've always thought that having a 4000-year rule would indeed make
            > sense, but didn't know whether it would truly be more accurate. I
            > guess it is... :-)

            This is incorrect.  The Gregorian reform was not intended to match the
            mean tropical year, as is often erroneously stated.  The Gregorian
            calendar was declared in the Inter Gravissimas, which specifically
            states that the purpose of the reform is to prevent the date of the
            vernal equinox from receding.  The mean tropical year is often stated
            to be the same value as the VE year, but in fact is not.  The mean
            tropical year is an average of all points on the zodiac, and differs
            in value from the mean period between successive vernal equinoxes, as
            can be easily determined by anyone from the times and dates of
            historical equinoxes.  This value is approximately 365.24238 days, and
            thus more closely aligned to the Gregorian calendar year of 365.2425
            days than to the mean tropical year of 365.24219 days.  This results
            in an error of less than one day in 8000 years, although the exact
            length of the day and year varies and is not known that far into the
            future.

            For more information, read the articles at
            http://www.hermetic.ch/cal_stud/cassidy/ or do the calculation
            yourself by simply averaging vernal equinoxes between any two years
            over a sufficiently long period.  You can find a list of dates and
            times for 1452 to 2547 at
            http://www.newscotland1398.net/equinox/eqindex.html

            Average for past 200 years:
            2006 Mar 20 18:26 UT = JD 2453815.2680555
            1806 Mar 21 06:52 UT = JD 2380766.7861111
            2453815.2680555 - 2380766.7861111 = 73048.4819444 days
            73048.4819444 / (2006-1806) = 365.2424097 days

            Average for past 554 years:
            2006 Mar 20 18:26 UT = JD 2453815.2680555
            1452 Mar 20 11:52 UT = JD 2251470.9944444
            2453815.2680555 - 2251470.9944444 = 202344.2736111 days
            202344.2736111 / (2006-1452) = 365.2423711 days

            The average value will vary slightly depending on exactly which years
            are used, since the absolute value changes year to year, but it
            converges between 365.24237 and 365.24238 days, not 365.2422 days.

            John Hynes

            --- In ISO8601@yahoogroups.com, "NGUYEN Ivy" <nguyenivy@...> wrote:
            >
            > I've always thought that having a 4000-year rule would indeed make
            > sense, but didn't know whether it would truly be more accurate. I
            > guess it is... :-)
            >
            >
            > On 20/03/06, hjwoudenberg@... <hjwoudenberg@...> wrote:
            > >
            > > http://en.wikipedia.org/wiki/Talk:Leap_year
            > >
            > >
            > > Is  there a 4000-year rule?
            > >
            > >
            > > It has been suggested (by the astronomer John Herschel (1792-1871)
            among    others) that a better approximation to the length of the
            tropical year would    be 365 969/4000 days = 365.24225 days. This
            would dictate 969 leap years every    4000 years, rather than the 970
            leap years mandated by the Gregorian calendar.    This could be
            achieved by dropping one leap year from the Gregorian calendar  
            every 4000 years, which would make years divisible by 4000 non-leap years.
            > >
            > > This rule has, however, not been officially  adopted.
            >




          • hjwoudenberg@aol.com
            In a message dated 3/21/2006 1:21:28 A.M. Central Standard Time, john@hynes.net writes: This is incorrect. The Gregorian reform was not intended to match the
            Message 5 of 25 , Mar 21, 2006
            • 0 Attachment
              In a message dated 3/21/2006 1:21:28 A.M. Central Standard Time, john@... writes:
              This is incorrect.  The Gregorian reform was not intended to match the
              mean tropical year, as is often erroneously stated. 
              Interesting
               
              There is a grievous error in Leroy Doggett's "Calendars" chapter from the Explanatory Supplement to the Astronomical Almanac. This same error also occurs in http://www.ast.cam.ac.uk/pubinfo/leaflets/leapyear/leapyear.html, Royal Greenwich Observatory pamphlet #48 (LEAP YEARS).
               
              So says, Simon Cassidy who studied mathematics in his native England and works as a software engineer on medical devices in California.
            • hjwoudenberg@aol.com
              In a message dated 3/21/2006 7:04:17 A.M. Central Standard Time, johnmsteele@yahoo.com writes: Christians need to assess whether an extra 24 hours or so in the
              Message 6 of 25 , Mar 21, 2006
              • 0 Attachment
                In a message dated 3/21/2006 7:04:17 A.M. Central Standard Time, johnmsteele@... writes:
                Christians need to assess whether an extra 24 hours or so in the precision of Easter is worth having many people prefer other calendars.
                There are two branches of Christianity;
                Those who conceive of Easter and the symbol as one.
                Those who think of Easter as a symbol a sign, therefore the day is not critical.
                It seems to me the former should all observe Easter based on the time in Israel, not local time, which could be a different day. 
              • Duncan Bath
                In a message dated 3/21/2006 7:04:17 A.M. Why this discredited, potentially ambiguous, date (time) form? ANYWHERE? D. ... From: hjwoudenberg@aol.com To:
                Message 7 of 25 , Mar 21, 2006
                • 0 Attachment
                  "In a message dated 3/21/2006 7:04:17 A.M."
                  Why this discredited, potentially ambiguous, date (time) form?
                  ANYWHERE?
                  D.
                   
                  ----- Original Message -----
                  Sent: Tuesday, March 21, 2006 12:39 PM
                  Subject: Re: [ISO8601] Re: Problem with ISO 8601? Discussion on leap year

                  In a message dated 3/21/2006 7:04:17 A.M. Central Standard Time, johnmsteele@... writes:
                  Christians need to assess whether an extra 24 hours or so in the precision of Easter is worth having many people prefer other calendars.
                  There are two branches of Christianity;
                  Those who conceive of Easter and the symbol as one.
                  Those who think of Easter as a symbol a sign, therefore the day is not critical.
                  It seems to me the former should all observe Easter based on the time in Israel, not local time, which could be a different day. 
                • Tex Texin
                  It would help if you clarified which grievous error you are referring to. Tex Texin Internationalization Architect, Yahoo! Inc. ... From:
                  Message 8 of 25 , Mar 21, 2006
                  • 0 Attachment
                    Message
                    It would help if you clarified which grievous error you are referring to.
                     
                     
                    Tex Texin
                    Internationalization Architect,   Yahoo! Inc.
                     
                     
                    -----Original Message-----
                    From: ISO8601@yahoogroups.com [mailto:ISO8601@yahoogroups.com] On Behalf Of hjwoudenberg@...
                    Sent: Tuesday, March 21, 2006 9:32 AM
                    To: ISO8601@yahoogroups.com
                    Subject: Re: [ISO8601] Re: Problem with ISO 8601? Discussion on leap year

                    In a message dated 3/21/2006 1:21:28 A.M. Central Standard Time, john@... writes:
                    This is incorrect.  The Gregorian reform was not intended to match the
                    mean tropical year, as is often erroneously stated. 
                    Interesting
                     
                    There is a grievous error in Leroy Doggett's "Calendars" chapter from the Explanatory Supplement to the Astronomical Almanac. This same error also occurs in http://www.ast.cam.ac.uk/pubinfo/leaflets/leapyear/leapyear.html, Royal Greenwich Observatory pamphlet #48 (LEAP YEARS).
                     
                    So says, Simon Cassidy who studied mathematics in his native England and works as a software engineer on medical devices in California.
                  • Tex Texin
                    there are more than the 2 branches methinks.... Anyway, the date for easter is based in part on a lunar event so it is not clear to me that the time in Israel
                    Message 9 of 25 , Mar 21, 2006
                    • 0 Attachment
                      Message
                      there are more than the 2 branches methinks....
                       
                      Anyway, the date for easter is based in part on a lunar event so it is not clear to me that the time in Israel or the 24 hour span makes much of a difference, but then I don't subscribe to any religion.
                       
                       
                       
                      Tex
                      -----Original Message-----
                      From: ISO8601@yahoogroups.com [mailto:ISO8601@yahoogroups.com] On Behalf Of hjwoudenberg@...
                      Sent: Tuesday, March 21, 2006 9:39 AM
                      To: ISO8601@yahoogroups.com
                      Subject: Re: [ISO8601] Re: Problem with ISO 8601? Discussion on leap year

                      In a message dated 3/21/2006 7:04:17 A.M. Central Standard Time, johnmsteele@... writes:
                      Christians need to assess whether an extra 24 hours or so in the precision of Easter is worth having many people prefer other calendars.
                      There are two branches of Christianity;
                      Those who conceive of Easter and the symbol as one.
                      Those who think of Easter as a symbol a sign, therefore the day is not critical.
                      It seems to me the former should all observe Easter based on the time in Israel, not local time, which could be a different day. 
                    • John Steele
                      It s not my posting, but I believe it is a direct quote from Simon Cassidy. He has been beating the drum of VE vs. tropical year for years. In a very narrow
                      Message 10 of 25 , Mar 21, 2006
                      • 0 Attachment
                        It's not my posting, but I believe it is a direct quote from Simon Cassidy. He has been beating the drum of VE vs. tropical year for years.
                         
                        In a very narrow sense, he is technically correct, although the error is not "grievous." IMO, as argued in an earlier post, when considering "big picture," Cassidy is flat wrong.  However, the point is certainly debatable and multiple positions can be argued. Jean Meeus' book "Astronomical Algorithms" has a good chapter on season drift.
                         
                        The topic might be a little outside the scope of ISO8601, however.

                        Tex Texin <tex@...> wrote:
                        It would help if you clarified which grievous error you are referring to.
                         
                         
                        Tex Texin
                        Internationalization Architect,   Yahoo! Inc.
                      • John Steele
                        True on the lunar event, which is about a 4 week uncertainty. But the two methods have perhaps a 24 hour effect on the vernal equinox. However, the Gregorian
                        Message 11 of 25 , Mar 21, 2006
                        • 0 Attachment
                          True on the lunar event, which is about a 4 week uncertainty. But the two methods have perhaps a 24 hour effect on the vernal equinox.  However, the Gregorian method just "assumes" the equinox is on a fixed date, the 21st. The rule could as easily be the the Sunday after the first full moon after the vernal equinox (which it almost is anyway).

                          Tex Texin <tex@...> wrote:
                          there are more than the 2 branches methinks....
                           
                          Anyway, the date for easter is based in part on a lunar event so it is not clear to me that the time in Israel or the 24 hour span makes much of a difference, but then I don't subscribe to any religion.
                           
                           
                           
                          Tex
                          -----Original Message-----
                          From: ISO8601@yahoogroups.com [mailto:ISO8601@yahoogroups.com] On Behalf Of hjwoudenberg@...
                          Sent: Tuesday, March 21, 2006 9:39 AM
                          To: ISO8601@yahoogroups.com
                          Subject: Re: [ISO8601] Re: Problem with ISO 8601? Discussion on leap year

                          In a message dated 3/21/2006 7:04:17 A.M. Central Standard Time, johnmsteele@... writes:
                          Christians need to assess whether an extra 24 hours or so in the precision of Easter is worth having many people prefer other calendars.
                          There are two branches of Christianity;
                          Those who conceive of Easter and the symbol as one.
                          Those who think of Easter as a symbol a sign, therefore the day is not critical.
                          It seems to me the former should all observe Easter based on the time in Israel, not local time, which could be a different day. 

                        • piebaldconsult
                          ... The part with which I take issue is where it says: If the year is the last year of a century, eg. 1800, 1900, 2000... Those, to me, (along with year
                          Message 12 of 25 , Mar 21, 2006
                          • 0 Attachment
                            > In a very narrow sense, he is technically correct, although the
                            >error is not "grievous." IMO, as argued in an earlier post, when
                            >considering "big picture," Cassidy is flat wrong.

                            The part with which I take issue is where it says:
                            "If the year is the last year of a century, eg. 1800, 1900, 2000..."

                            Those, to me, (along with year 0000) are all _first_ years of centuries.
                          • Tex Texin
                            ... The first year is year one (Jan 1 to Dec 31, 0001). Hence the year ending in 00 is the last year. 1-100 101-200, etc. year 0 would be the last year of 99BC
                            Message 13 of 25 , Mar 21, 2006
                            • 0 Attachment
                              > The part with which I take issue is where it says:
                              > "If the year is the last year of a century, eg. 1800, 1900, 2000..."
                              >
                              > Those, to me, (along with year 0000) are all _first_ years of
                              > centuries.

                              The first year is year one (Jan 1 to Dec 31, 0001). Hence the year ending in
                              00 is the last year.
                              1-100
                              101-200, etc.

                              year 0 would be the last year of 99BC to 0.
                            • John Hynes
                              16th century astronomers, including Clavius, Lilius and Dee, certainly knew how long the year was. The ROG pamphlet someone referred to (now at
                              Message 14 of 25 , Mar 22, 2006
                              • 0 Attachment
                                16th century astronomers, including Clavius, Lilius and Dee, certainly
                                knew how long the year was.

                                The ROG pamphlet someone referred to (now at
                                http://www.nmm.ac.uk/server/show/conWebDoc.349) may not be in
                                "grievous" error, but it clearly contradicts itself. It states, "The
                                year is defined as being the interval between two successive passages
                                of the Sun through the vernal equinox...The year as defined above is
                                called the tropical year and it is the year length that defines the
                                repetition of the seasons. The length of the tropical year is
                                365.24219 days." It should be noted that ROG is today a museum, and
                                not a working observatory.

                                The motivation for the Gregorian reform was religious. (The best way
                                to determine Easter is a separate issue that has been much discussed
                                elsewhere.) If religion is not important, then why "fix" the calendar
                                at all? So what if the seasons slip a bit thousands of years in the
                                future? The vernal equinox now varies from year to year, and can fall
                                on the 19th, 20th or 21st of March. Will people in the year 10000
                                care that it sometimes falls on the 18th? In the year 4000, the
                                average difference will be only about a quarter of a day from what it
                                is now, so dropping a day would create a larger discrepency than
                                leaving it alone!

                                For 1600 years (2000 years in some places) few seemed to care that the
                                seasons started about one day earlier every 130 Julian calendar years
                                or so. The ancient Egyptians did not bother with leap years at all
                                for thousands of years, and let the seasons slip through all the
                                months. And the modern Islamic calendar slips on average about one
                                day for every month!

                                So if people living 1994 years or 26500 years from now think that they
                                need to change the calendar, they can do so, but I doubt that they
                                would appreciate us doing it for them now, since, after all, we'll all
                                be dead, along with our "modern" goals!

                                It has been claimed that the original Julian calendar was supposed to
                                have the equinox be March 25, and the other seasons similarly shifted,
                                but Gregory reset the calendar to what it was hundreds of years later.
                                If we really wanted to fix the calendar, we should take a few more
                                days out.

                                John Hynes

                                --- In ISO8601@yahoogroups.com, John Steele <johnmsteele@...> wrote:
                                >
                                > It is unlikely in the 1580's, Pope Gregory had the astronomical
                                knowledge to think this way. However, the difference between the
                                tropical year and the VE year (and years measured by the fall equinox
                                and solstices) is a function of where the earth is in its precession
                                cycle (period around 26500 years), which also determines which star is
                                the North Star.
                                >
                                > The Gregorian reform probably set out to stabilize the VE year
                                with the knowledge of the times. I would argue stabilizing the
                                tropical year is the best way to fix it, averaged over a precession
                                cycle, and has the advantage of stabilizing, on average, all the
                                seasons. It could also be argued as being "fairer" to people in the
                                Southern Hemisphere, whose agricultural cycle is linked to the fall
                                equinox, and to people of other faiths who are not concerned with
                                Easter. The fixing the calendar to a tropical year is completely
                                consistent with a world calendar that serves everybody. Christians
                                need to assess whether an extra 24 hours or so in the precision of
                                Easter is worth having many people prefer other calendars.
                                >
                                > I would argue fixing the calendar to the tropical year should be
                                the "modern" goal regardless of exact intent in 1582; they're all
                                dead. Therefore we need to watch the error between the tropical year
                                and the Gregorian year of 365.2425 days, and determine when a change
                                to the calendar is warranted.
                                > John Hynes <john@...> wrote:
                                > > I've always thought that having a 4000-year rule would indeed make
                                > > sense, but didn't know whether it would truly be more accurate. I
                                > > guess it is... :-)
                                >
                                > This is incorrect. The Gregorian reform was not intended to match the
                                > mean tropical year, as is often erroneously stated. The Gregorian
                                > calendar was declared in the Inter Gravissimas, which specifically
                                > states that the purpose of the reform is to prevent the date of the
                                > vernal equinox from receding. The mean tropical year is often stated
                                > to be the same value as the VE year, but in fact is not. The mean
                                > tropical year is an average of all points on the zodiac, and differs
                                > in value from the mean period between successive vernal equinoxes, as
                                > can be easily determined by anyone from the times and dates of
                                > historical equinoxes. This value is approximately 365.24238 days, and
                                > thus more closely aligned to the Gregorian calendar year of 365.2425
                                > days than to the mean tropical year of 365.24219 days. This results
                                > in an error of less than one day in 8000 years, although the exact
                                > length of the day and year varies and is not known that far into the
                                > future.
                                >
                                > For more information, read the articles at
                                > http://www.hermetic.ch/cal_stud/cassidy/ or do the calculation
                                > yourself by simply averaging vernal equinoxes between any two years
                                > over a sufficiently long period. You can find a list of dates and
                                > times for 1452 to 2547 at
                                > http://www.newscotland1398.net/equinox/eqindex.html
                                >
                                > Average for past 200 years:
                                > 2006 Mar 20 18:26 UT = JD 2453815.2680555
                                > 1806 Mar 21 06:52 UT = JD 2380766.7861111
                                > 2453815.2680555 - 2380766.7861111 = 73048.4819444 days
                                > 73048.4819444 / (2006-1806) = 365.2424097 days
                                >
                                > Average for past 554 years:
                                > 2006 Mar 20 18:26 UT = JD 2453815.2680555
                                > 1452 Mar 20 11:52 UT = JD 2251470.9944444
                                > 2453815.2680555 - 2251470.9944444 = 202344.2736111 days
                                > 202344.2736111 / (2006-1452) = 365.2423711 days
                                >
                                > The average value will vary slightly depending on exactly which years
                                > are used, since the absolute value changes year to year, but it
                                > converges between 365.24237 and 365.24238 days, not 365.2422 days.
                                >
                                > John Hynes
                                >
                                > --- In ISO8601@yahoogroups.com, "NGUYEN Ivy" <nguyenivy@> wrote:
                                > >
                                > > I've always thought that having a 4000-year rule would indeed make
                                > > sense, but didn't know whether it would truly be more accurate. I
                                > > guess it is... :-)
                                > >
                                > >
                                > > On 20/03/06, hjwoudenberg@ <hjwoudenberg@> wrote:
                                > > >
                                > > > http://en.wikipedia.org/wiki/Talk:Leap_year
                                > > >
                                > > >
                                > > > Is there a 4000-year rule?
                                > > >
                                > > >
                                > > > It has been suggested (by the astronomer John Herschel (1792-1871)
                                > among others) that a better approximation to the length of the
                                > tropical year would be 365 969/4000 days = 365.24225 days. This
                                > would dictate 969 leap years every 4000 years, rather than the 970
                                > leap years mandated by the Gregorian calendar. This could be
                                > achieved by dropping one leap year from the Gregorian calendar
                                > every 4000 years, which would make years divisible by 4000 non-leap
                                years.
                                > > >
                                > > > This rule has, however, not been officially adopted.
                                > >
                                >
                                >
                                >
                                >
                                >
                                > SPONSORED LINKS
                                > Computer internet security Computer internet business
                                Computer internet access Computer internet help How to format a
                                computer hard drive Computer internet connection
                                >
                                > ---------------------------------
                                > YAHOO! GROUPS LINKS
                                >
                                >
                                > Visit your group "ISO8601" on the web.
                                >
                                > To unsubscribe from this group, send an email to:
                                > ISO8601-unsubscribe@yahoogroups.com
                                >
                                > Your use of Yahoo! Groups is subject to the Yahoo! Terms of
                                Service.
                                >
                                >
                                > ---------------------------------
                                >
                              • Tex Texin
                                John, After I finish building and patenting my time travel machine, it will be important to have precise calendar dates going back and forth +/- 10000 years.
                                Message 15 of 25 , Mar 22, 2006
                                • 0 Attachment
                                  John,

                                  After I finish building and patenting my time travel machine, it will be
                                  important to have precise calendar dates going back and forth +/- 10000
                                  years.


                                  Tex Texin

                                  > -----Original Message-----
                                  > From: ISO8601@yahoogroups.com
                                  > [mailto:ISO8601@yahoogroups.com] On Behalf Of John Hynes
                                  > Sent: Wednesday, March 22, 2006 3:01 AM
                                  > To: ISO8601@yahoogroups.com
                                  > Subject: [ISO8601] Re: Problem with ISO 8601? Discussion on leap year
                                  >
                                  >
                                  > 16th century astronomers, including Clavius, Lilius and Dee,
                                  > certainly knew how long the year was.
                                  >
                                  > The ROG pamphlet someone referred to (now at
                                  > http://www.nmm.ac.uk/server/show/conWebDoc.349) may not be in
                                  > "grievous" error, but it clearly contradicts itself. It
                                  > states, "The year is defined as being the interval between
                                  > two successive passages of the Sun through the vernal
                                  > equinox...The year as defined above is called the tropical
                                  > year and it is the year length that defines the repetition of
                                  > the seasons. The length of the tropical year is 365.24219
                                  > days." It should be noted that ROG is today a museum, and
                                  > not a working observatory.
                                  >
                                  > The motivation for the Gregorian reform was religious. (The
                                  > best way to determine Easter is a separate issue that has
                                  > been much discussed
                                  > elsewhere.) If religion is not important, then why "fix" the
                                  > calendar at all? So what if the seasons slip a bit thousands
                                  > of years in the future? The vernal equinox now varies from
                                  > year to year, and can fall on the 19th, 20th or 21st of
                                  > March. Will people in the year 10000 care that it sometimes
                                  > falls on the 18th? In the year 4000, the average difference
                                  > will be only about a quarter of a day from what it is now, so
                                  > dropping a day would create a larger discrepency than leaving
                                  > it alone!
                                  >
                                  > For 1600 years (2000 years in some places) few seemed to care
                                  > that the seasons started about one day earlier every 130
                                  > Julian calendar years or so. The ancient Egyptians did not
                                  > bother with leap years at all for thousands of years, and let
                                  > the seasons slip through all the months. And the modern
                                  > Islamic calendar slips on average about one day for every month!
                                  >
                                  > So if people living 1994 years or 26500 years from now think
                                  > that they need to change the calendar, they can do so, but I
                                  > doubt that they would appreciate us doing it for them now,
                                  > since, after all, we'll all be dead, along with our "modern" goals!
                                  >
                                  > It has been claimed that the original Julian calendar was
                                  > supposed to have the equinox be March 25, and the other
                                  > seasons similarly shifted, but Gregory reset the calendar to
                                  > what it was hundreds of years later. If we really wanted to
                                  > fix the calendar, we should take a few more days out.
                                  >
                                  > John Hynes
                                • piebaldconsult
                                  ... ending in ... Ah, no, that makes no sense, is much less convenient, and ISO 8601 does it the other way... ISO 8601:2000, section 5.2.1.3 (a), (b), and (c)
                                  Message 16 of 25 , Mar 22, 2006
                                  • 0 Attachment
                                    --- In ISO8601@yahoogroups.com, "Tex Texin" <tex@...> wrote:
                                    >
                                    > > The part with which I take issue is where it says:
                                    > > "If the year is the last year of a century, eg. 1800, 1900, 2000..."
                                    > >
                                    > > Those, to me, (along with year 0000) are all _first_ years of
                                    > > centuries.
                                    >
                                    > The first year is year one (Jan 1 to Dec 31, 0001). Hence the year
                                    ending in
                                    > 00 is the last year.
                                    > 1-100
                                    > 101-200, etc.
                                    >
                                    > year 0 would be the last year of 99BC to 0.
                                    >

                                    Ah, no, that makes no sense, is much less convenient, and ISO 8601 does
                                    it the other way...

                                    ISO 8601:2000, section 5.2.1.3 (a), (b), and (c) (which is under
                                    truncation, which of course is not allowed by :2004...) makes it clear
                                    that the two high digits of a four digit year represent the century.
                                    Otherwise an interval like "2000-01-01/01-01-01" would equate to "2000-
                                    01-01/1901-01-01" !!

                                    Forget what the original intent of the Gregorian calendar may have
                                    been, it was _sooo_ last millenium, we can do better, after all we now
                                    have year 0000 and leap seconds!

                                    It is far simpler to divide the year by 100 to get the century.
                                    Although, yes, the -00 century would seem to have only 99 years or the
                                    year 0000 would be in two centuries (+00 and -00), but so what?!

                                    One can even say that 1582 is the "first" year and that Gregorian
                                    centuries run from xx82 through xy81!

                                    And don't get me back to saying there should be a year -0000 separate
                                    from year +0000 !
                                  • piebaldconsult
                                    ... be ... 10000 ... I can t see using the Gregorian (or any yet known) calender for that, you ll need something much better. And remember to think four-
                                    Message 17 of 25 , Mar 22, 2006
                                    • 0 Attachment
                                      > After I finish building and patenting my time travel machine, it will
                                      be
                                      > important to have precise calendar dates going back and forth +/-
                                      10000
                                      > years.

                                      I can't see using the Gregorian (or any yet known) calender for that,
                                      you'll need something much better. And remember to think four-
                                      dimensionally! If you jump ahead a day you want to know _exactly_ where
                                      the Earth (or whatever) will be at that time.

                                      Ah, what a blessing that time travel is _not_ possible!
                                    • Tex Texin
                                      Piebald, we were doing so well in agreeing recently too.... oh well. I see you subscribe to Humpty Dumpty s philosophy that words mean exactly what you want
                                      Message 18 of 25 , Mar 22, 2006
                                      • 0 Attachment
                                        Piebald, we were doing so well in agreeing recently too....
                                        oh well.

                                        I see you subscribe to Humpty Dumpty's philosophy that words mean exactly
                                        what you want them to mean.

                                        That's fine, but changing meaning does present a problem for clear
                                        communication to others.

                                        I don't think the designers of 8601 had in mind to change the meaning of
                                        "century" or "millenia" etc. when they offered to call the 2 digits in the
                                        8601 framework the century digits.

                                        We have enough trouble getting buy-in to 8601 without also redefining
                                        English, even though it simplifies the calculation...

                                        ah well....
                                        tex


                                        > -----Original Message-----
                                        > From: ISO8601@yahoogroups.com
                                        > [mailto:ISO8601@yahoogroups.com] On Behalf Of piebaldconsult
                                        > Sent: Wednesday, March 22, 2006 8:46 PM
                                        > To: ISO8601@yahoogroups.com
                                        > Subject: [ISO8601] Re: Problem with ISO 8601? Discussion on leap year
                                        >
                                        >
                                        > --- In ISO8601@yahoogroups.com, "Tex Texin" <tex@...> wrote:
                                        > >
                                        > > > The part with which I take issue is where it says:
                                        > > > "If the year is the last year of a century, eg. 1800,
                                        > 1900, 2000..."
                                        > > >
                                        > > > Those, to me, (along with year 0000) are all _first_ years of
                                        > > > centuries.
                                        > >
                                        > > The first year is year one (Jan 1 to Dec 31, 0001). Hence the year
                                        > ending in
                                        > > 00 is the last year.
                                        > > 1-100
                                        > > 101-200, etc.
                                        > >
                                        > > year 0 would be the last year of 99BC to 0.
                                        > >
                                        >
                                        > Ah, no, that makes no sense, is much less convenient, and ISO
                                        > 8601 does
                                        > it the other way...
                                        >
                                        > ISO 8601:2000, section 5.2.1.3 (a), (b), and (c) (which is under
                                        > truncation, which of course is not allowed by :2004...) makes
                                        > it clear
                                        > that the two high digits of a four digit year represent the century.
                                        > Otherwise an interval like "2000-01-01/01-01-01" would equate
                                        > to "2000- 01-01/1901-01-01" !!
                                        >
                                        > Forget what the original intent of the Gregorian calendar may have
                                        > been, it was _sooo_ last millenium, we can do better, after
                                        > all we now
                                        > have year 0000 and leap seconds!
                                        >
                                        > It is far simpler to divide the year by 100 to get the century.
                                        > Although, yes, the -00 century would seem to have only 99
                                        > years or the
                                        > year 0000 would be in two centuries (+00 and -00), but so what?!
                                        >
                                        > One can even say that 1582 is the "first" year and that Gregorian
                                        > centuries run from xx82 through xy81!
                                        >
                                        > And don't get me back to saying there should be a year -0000 separate
                                        > from year +0000 !
                                        >
                                        >
                                        >
                                        >
                                        >
                                        >
                                        >
                                        > Yahoo! Groups Links
                                        >
                                        >
                                        >
                                        >
                                        >
                                        >
                                        >
                                      • piebaldconsult
                                        ... meaning of ... in the ... Where do you see decade, century, or millennium _defined_ as beginning at any particular point?
                                        Message 19 of 25 , Mar 23, 2006
                                        • 0 Attachment
                                          > I don't think the designers of 8601 had in mind to change the
                                          meaning of
                                          > "century" or "millenia" etc. when they offered to call the 2 digits
                                          in the
                                          > 8601 framework the century digits.

                                          Where do you see decade, century, or millennium _defined_ as
                                          beginning at any particular point?

                                          http://merriamwebster.com/dictionary/decade
                                          http://merriamwebster.com/dictionary/century
                                          http://merriamwebster.com/dictionary/millennium

                                          Year 0000 is the _first_ non-negative year.

                                          This from Wikipedia:

                                          "In 2000 the International Organization for Standardization clarified
                                          ISO 8601 to use the astronomical year numbering system, which could
                                          be interpreted as retrospectively endorsing all the people who had
                                          celebrated the new century a few months earlier. Also, decades are
                                          almost always considered as starting with the "0" year and named
                                          accordingly ("1960s", etc.)."

                                          There are simply two camps: we who do things the easy way, and you
                                          who, umm don't. And it doesn't really matter anyway.

                                          But if I get a time machine I'll go back and give Gregory a kick in
                                          the patoot.
                                        • John Hynes
                                          1901-2000 is a century. So is 1900-1999. 1962-2061 is also a century. We re just talking about different centuries. The so-called twentieth (20th) century
                                          Message 20 of 25 , Mar 24, 2006
                                          • 0 Attachment
                                            1901-2000 is a century. So is 1900-1999. 1962-2061 is also a
                                            century. We're just talking about different centuries.

                                            The so-called twentieth (20th) century was 1901-2000 A.D., and the
                                            21st is 2001-2100. The first year of these centuries end with 01, the
                                            second year with 02, etc. This is using ordinal numbers, counting
                                            from Year 1 A.D., which was named after the fact.

                                            The nineteen-hundreds (1900s) was the century 1900-1999 CE, and the
                                            2000s are 2000-2099. Using cardinal numbers, elapsed years are
                                            counted from Year 0 CE, which was named after the fact.

                                            It is not about right or wrong, but apples and oranges. It's not
                                            about "the" century, but "which" century. Where this is not clear, it
                                            should, of course, be specified. ISO 8601 uses cardinal numbers for
                                            centuries and years, but ordinal numbers for ordinal dates (1-365) and
                                            week dates (W1-W53,1-7).

                                            I read John Herschel's Outlines of Astronomy and noticed that he makes
                                            a distinction between "current" and "elapsed" enumeration:

                                            "(915.) The Gregorian rule is as follows:— The years are denominated
                                            as years current (not as years elapsed) from the midnight between the
                                            31st of December and the 1st of January immediately subsequent to the
                                            birth of Christ, according to the chronological determination of that
                                            event by Dionysius Exiguus."

                                            "(927.) The determination of the exact interval between any two given
                                            dates, is a matter of such importance, and, unless methodically
                                            performed, is so very liable to error, that the following rules will
                                            not be found out of place. In the first place it must be remarked,
                                            generally, that a date, whether of a day or year, always expresses the
                                            day or year current and not elapsed, and that the designation of a
                                            year by A.D. or B.C. is to be regarded as the name of that year, and
                                            not as a mere number uninterruptedly designating the place of the year
                                            in the scale of time. Thus, in the date, Jan. 5. B.C. 1, Jan. 5 does
                                            not mean that 5 days of January in the year in question have elapsed,
                                            but that 4 have elapsed, and the 5th is current. And the B.C. 1,
                                            indicates that the first day of the year so named, (the first year
                                            current before Christ,) preceded the first day of the vulgar era by
                                            one year. The scale of A.D. and B.C., as already observed, is not
                                            continuous, the year 0 in both being wanting; so that (supposing the
                                            vulgar reckoning correct) our Saviour was born in the year B.C. 1."

                                            John Hynes

                                            --- In ISO8601@yahoogroups.com, "piebaldconsult" <PIEBALDconsult@...>
                                            wrote:
                                            >
                                            > --- In ISO8601@yahoogroups.com, "Tex Texin" <tex@> wrote:
                                            > >
                                            > > > The part with which I take issue is where it says:
                                            > > > "If the year is the last year of a century, eg. 1800, 1900, 2000..."
                                            > > >
                                            > > > Those, to me, (along with year 0000) are all _first_ years of
                                            > > > centuries.
                                            > >
                                            > > The first year is year one (Jan 1 to Dec 31, 0001). Hence the year
                                            > ending in
                                            > > 00 is the last year.
                                            > > 1-100
                                            > > 101-200, etc.
                                            > >
                                            > > year 0 would be the last year of 99BC to 0.
                                            > >
                                            >
                                            > Ah, no, that makes no sense, is much less convenient, and ISO 8601 does
                                            > it the other way...
                                            >
                                            > ISO 8601:2000, section 5.2.1.3 (a), (b), and (c) (which is under
                                            > truncation, which of course is not allowed by :2004...) makes it clear
                                            > that the two high digits of a four digit year represent the century.
                                            > Otherwise an interval like "2000-01-01/01-01-01" would equate to "2000-
                                            > 01-01/1901-01-01" !!
                                            >
                                            > Forget what the original intent of the Gregorian calendar may have
                                            > been, it was _sooo_ last millenium, we can do better, after all we now
                                            > have year 0000 and leap seconds!
                                            >
                                            > It is far simpler to divide the year by 100 to get the century.
                                            > Although, yes, the -00 century would seem to have only 99 years or the
                                            > year 0000 would be in two centuries (+00 and -00), but so what?!
                                            >
                                            > One can even say that 1582 is the "first" year and that Gregorian
                                            > centuries run from xx82 through xy81!
                                            >
                                            > And don't get me back to saying there should be a year -0000 separate
                                            > from year +0000 !
                                            >
                                          • piebaldconsult
                                            ... When was that written? Given such short-comings we now consider that there _was_ a year 0000, thereby allowing us to correct the incongruence between
                                            Message 21 of 25 , Mar 24, 2006
                                            • 0 Attachment
                                              > The so-called twentieth (20th) century was 1901-2000 A.D., and the
                                              > 21st is 2001-2100. The first year of these centuries end with 01, the
                                              > second year with 02, etc. This is using ordinal numbers, counting

                                              > one year. The scale of A.D. and B.C., as already observed, is not
                                              > continuous, the year 0 in both being wanting; so that (supposing the
                                              > vulgar reckoning correct) our Saviour was born in the year B.C. 1."

                                              When was that written?

                                              Given such short-comings we now consider that there _was_ a year 0000,
                                              thereby allowing us to correct the incongruence between "1800s"
                                              and "nineteenth century".

                                              Why accept one without the other?

                                              What the ancients believed has no power; should we still believe that
                                              the Earth is flat and the sun revolves around it simply because that
                                              was once held to be true?

                                              Get with the twenty-first century!
                                            • Tex Texin
                                              There is the realm of all possibilities, and then there is current convention. So the first year could be zero or one. The first century could be 0-99 or
                                              Message 22 of 25 , Mar 24, 2006
                                              • 0 Attachment
                                                There is the realm of all possibilities, and then there is current
                                                convention.

                                                So the first year could be zero or one.
                                                The first century could be 0-99 or 1-100.
                                                Or it could be 100-199, with 0-99 being the zeroth century.

                                                (The rule being chop off the digits in the above the tens place, so that
                                                00/00-00/99 is the zeroth century.)

                                                But 500 years of counting from one should count for something.
                                                (Was that a pun? I am not sure.)

                                                So what century is this?
                                                Everyone off of this list says it is the 21st, since 1-100 was the first.

                                                If we are to count from zero, then we have to explain to the world why we
                                                now call it the 20th.

                                                geez. I'll go along with this after Herman implements it in hardware.

                                                tex
                                              • piebaldconsult
                                                ... first. It s still the 21st, but it started with year 2000, not 2001. The first century was the zero-hundreds? When measuring distance the first foot begins
                                                Message 23 of 25 , Mar 24, 2006
                                                • 0 Attachment
                                                  > So what century is this?
                                                  > Everyone off of this list says it is the 21st, since 1-100 was the
                                                  first.

                                                  It's still the 21st, but it started with year 2000, not 2001. The
                                                  first century was the zero-hundreds?

                                                  When measuring distance the first foot begins with the first inch, at
                                                  offset zero, not the second, at offset one.

                                                  If you count the year 0000 as being the final year in the final BCE
                                                  century, then you can find a BCE year's century simply by dividing
                                                  the BCE year by 100 (and how often do you have to do that?), but that
                                                  is not true of CE years. On the other hand, counting year 0000 as the
                                                  first year of the first CE century, you can determine the century of
                                                  a CE year (which is a lot more useful) simply by dividing by 100, but
                                                  not so for BCE years (and who cares?). Of course without year 0000
                                                  neither is true. On the other (third) hand I would simply say that
                                                  you determine the decade, century, millennium, etc. of a year in
                                                  either era by division. And I don't give a flying crap that one or
                                                  more will end up a year short or that year 0000 will be in two such
                                                  divisions, the economy won't fail because of it.

                                                  Of course Marilyn Vos Savant (sp) said it well in 1999 when she said
                                                  she'd celebrate the new millennium twice.
                                                • John Hynes
                                                  ... 1849, but that does not affect his point about current v. elapsed enumeration, especially since what he wrote is still valid. ... Who is we ? Some,
                                                  Message 24 of 25 , Mar 25, 2006
                                                  • 0 Attachment
                                                    --- In ISO8601@yahoogroups.com, "piebaldconsult" <PIEBALDconsult@...>
                                                    wrote:
                                                    > When was that written?

                                                    1849, but that does not affect his point about "current" v. "elapsed"
                                                    enumeration, especially since what he wrote is still valid.

                                                    > Given such short-comings we now consider that there _was_ a year 0000,
                                                    > thereby allowing us to correct the incongruence between "1800s"
                                                    > and "nineteenth century".

                                                    Who is "we"? Some, including astronomers, now call the year before
                                                    Year 1 CE "Year 0" and give negative numbers to previous years, but
                                                    this is far from universal. The long-existing standard of AD/BC or
                                                    CE/BCE with no year 0 is still in wide use.

                                                    Neither usage is correct or incorrect. They are simply two different
                                                    standards. Astronomical year numbering has not replaced ordinal year
                                                    numbering, but exists alongside it. The two standards are equivalent
                                                    for the years 1 and following, but differ on previous years. The use
                                                    of BC and BCE is still common, moreso, in fact, than negative AD or CE.

                                                    > What the ancients believed has no power; should we still believe that
                                                    > the Earth is flat and the sun revolves around it simply because that
                                                    > was once held to be true?

                                                    This is not a debate about belief, but semantics. And BC year
                                                    numbering is still held as "true" by the vast majority TODAY. In
                                                    fact, the main rival to the use of BC is not negative years, but BCE,
                                                    which does not affect the actual numbering of the years.

                                                    Ordinal numbering is widely used in timekeeping. We use ordinal
                                                    numbers for month (1-12), day of the month (1-31), day of the year
                                                    (1-366), day of the week (1-7), clock hour (1-12) and year (1-).
                                                    Cardinal numbering is used for minutes and seconds, and sometimes in
                                                    other contexts by computer programmers and astronomers. However,
                                                    counting months 0-11 has not replaced counting them 1-12, even in ISO
                                                    8601.

                                                    You may believe that the entire world has stopped using BC/BCE year
                                                    numbering, and embraced the year 0 and negative CE years, but this is
                                                    a false belief. The truth is that BOTH standards are in concurrent
                                                    use, that neither one is correct or incorrect, but just different, and
                                                    that the traditional year numbering is far more common than
                                                    astronomical year numbering.

                                                    John Hynes
                                                  • piebaldconsult
                                                    Oy, this is so off-topic. My point is that if you use year 0000, then you should have it as the first year of the current era, not the last year of the
                                                    Message 25 of 25 , Mar 25, 2006
                                                    • 0 Attachment
                                                      Oy, this is so off-topic.

                                                      My point is that if you use year 0000, then you should have it as the
                                                      first year of the current era, not the last year of the preceding era.

                                                      The whole point of using year 0000 is to make calculating certain
                                                      things simple, like:

                                                      What is the duration between timepoints that don't lie in the same
                                                      era? Just subtract.

                                                      Which years are leap years? Years evenly divisble by 4, but not 100,
                                                      etc. (you know the rules). Without year 0000, this is not true of
                                                      years in the preceding era (which may not matter).

                                                      What is the decade, century, millennium, etc. of a year? Just divide
                                                      by the appropriate value. Without year 0000 this is either not true
                                                      or both eras (erae?) begin with a short period (which is OK with me).
                                                      With year 0000 this can be true of one era (in which case, why would
                                                      you not choose the current era?) and the other can either begin with
                                                      a short period or not have this property.
                                                    Your message has been successfully submitted and would be delivered to recipients shortly.