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

Year 0

Expand Messages
  • laartphoto
    From The ISO 8601 Date Format by Peter Meyer at http://serendipity.magnet.ch/hermetic/cal_stud/formats.htm ... Whether there are limits to the dates which
    Message 1 of 15 , Apr 27, 2002
    • 0 Attachment
      From "The ISO 8601 Date Format" by Peter Meyer at
      http://serendipity.magnet.ch/hermetic/cal_stud/formats.htm

      ------
      Whether there are limits to the dates which can be expressed.

      The year is expressed in the manner of the astronomers and
      chronologists. Year 0 is the year preceding year 1 (a.k.a. 1 B.C.).
      Year -1 is the year preceding year 0 (a.k.a. 2 B.C.). Generally,
      year -n is the year which began n+1 years before the beginning of
      year 1 (so year -n = year n+1 B.C.). With the use of this
      astronomical system of year-numbering there are no limits to the
      dates which can be expressed in this date format.
      ------

      I haven't found anywhere in the PDF of the standard where it says
      this (I'm still reading), but my opinion is that if you add a year 0
      you need to add positive and negative 0. But I hear people say, "but
      a positive and negative 0 is confusing". But these same persons are
      not confused by positive and negative 1, so what's the difference? I
      would very much prefer that we, as scientists and other professional
      thinking persons, use positive and negative 0 with years just as we
      would with positive and negative 0 degrees of temperature or whatnot,
      as in -0.5 degrees centigrade (oh, umm, 0,5 for the Europeans and
      such). It just doesn't make sense (to me) to only have one 0; "two or
      none" is what I say.


      As to the "comma or period (full-stop)" debate: I say period because
      comma is used for so many other things, like lists. On the other
      hand, comma should be used for fewer of the other things it's used
      for as well, for example "Comma Separated, Variable width" (CSV)
      files should actually use some other delimiter, I just don't know
      what. If a CSV file contains a time value with a comma it's going to
      be misinterpreted, one or the other format (or both) must be changed.

      If the ISO is trying to devise an international standard, they should
      pick one or the other.

      As to the concern over 00:65:00, I feel that that can certainly be a
      duration, as in "it lasted 00:65:00", but it makes far less sense
      as "I'll meet you at 00:65:00". Likewise, you might specify "it is
      now -00:65:00 from lift off", but not "I'll meet you at -00:65:00",
      just say "I'll meet you at 22:55".


      Following is a gripe with my boss, you may not want to read it or you
      may want to commiserate:

      A file that my employer receives from overseas contains time values
      as mmm.ss, where mmm is a number of minutes which may be greater than
      59 (in fact it may be in the thousands, it's a duration) and ss is
      the partial minutes. The question arose of whether ss is:

      1) an actual number of seconds, in which case the period is simply
      the non-U.S. delimiter as in hh.mm.ss in which case .50 means fifty
      seconds

      or

      2) the decimal point, in which case .50 means half a minute or thirty
      seconds

      I studied several files and concluded that the former is the case
      even though my boss said the latter is the case and would not be
      swayed.

      If the file used the standard then the delimiter should be a colon
      (mmm:ss) or not present (mmmss) in the former case and either the
      period (mmm.ss) or a comma (mmm,ss) in the latter case (of course,
      this is a CSV file and the comma could cause trouble).

      Ah, but the standard keeps saying "By mutual agreement of..." and I
      should assume that my boss and the sender of the file agreed on the
      latter because my boss says so.

      But the data just doesn't make sense that way, the biggest clue is
      that all the values for ss in all the files range only from 00 to 59,
      which they would not in the latter case.
    • bam
      ... No. Not at all. There is no need for two year-zeros. Only one year receives the number zero: it is the year commonly referred to as 1 BC. The year
      Message 2 of 15 , Apr 27, 2002
      • 0 Attachment
        laartphoto wrote:
        >...
        > I haven't found anywhere in the PDF of the standard where it says
        > this (I'm still reading), but my opinion is that if you add a year 0
        > you need to add positive and negative 0.

        No. Not at all. There is no need for two year-zeros.

        Only one year receives the number zero: it is the year
        commonly referred to as 1 BC. The year immediately prior
        to it is the year -1 (or 2 BC) and the one before that
        is -2 (or 3 BC).

        Regardless of whether the year number is positive, negative, or zero,
        the months and days ascend with time: YYYY-01-01 then YYYY-01-02 ...
        thru YYYY-12-31. This ascending sequence of dates is the same
        (modulo leap years) regardless of year sign.

        It is not necessary to begin the calendar in December for a year
        with a negative number (unless you are Merlin ;-).

        Unlike temperatures, there is no zero instant on a calendar,
        before which the months or days must roll backwards.
        Zero is merely a year number, as are -1, -2, -999,
        -3760 (when the Hebrew calendar begins),
        the dinosaur extinction of the year -65000000,
        or the pre-cambrian year -2E9.

        All years begin in January and end in December.


        > But I hear people say, "but
        > a positive and negative 0 is confusing". But these same persons are
        > not confused by positive and negative 1, so what's the difference? I
        > would very much prefer that we, as scientists and other professional
        > thinking persons, use positive and negative 0 with years just as we
        > would with positive and negative 0 degrees of temperature or whatnot,
        > as in -0.5 degrees centigrade (oh, umm, 0,5 for the Europeans and
        > such). It just doesn't make sense (to me) to only have one 0; "two or
        > none" is what I say.

        >
        > As to the "comma or period (full-stop)" debate: I say period because
        > comma is used for so many other things, like lists. On the other
        > hand, comma should be used for fewer of the other things it's used
        > for as well, for example "Comma Separated, Variable width" (CSV)
        > files should actually use some other delimiter, I just don't know
        > what. If a CSV file contains a time value with a comma it's going to
        > be misinterpreted, one or the other format (or both) must be changed.
        >
        > If the ISO is trying to devise an international standard, they should
        > pick one or the other.

        Why not pick something else, to avoid perpetuating the problem?
        There is no reason to entangle the time standard in the dispute
        regarding decimal points (errr, uh, decimal indicator or whatever),
        since years do not need commas (and times above one second do not
        need decimals).

        The ASCII (ISCII) character set is rich enuf to supply another
        symbol without overloading (and ambiguating) the comma or period.

        I'd also like to see the ISO 8601 dash (as in YYYY-MM-DD)
        changed to something else, since dash is already doubly
        overloaded as both a minus sign (as in year -999) and an
        interval indicator as in (15-20 minutes).

        My use and promotion of ISO 8601 have both been hampered by
        the unfortunate choice of the dash (minus sign and interval
        and interval indicator) as a date separator.

        I often substitute a slash.
        Writing today as 2002/04/27 (or '02/04/27) makes very clear
        to everyone that a date is indicated (while promoting the
        YMD sequence
      • Morris, Mike
        laartphoto is right that There is no need for two year-zeros , but his comments actually miss the point: The concept of +/- 0 is inherently non-sensical. You
        Message 3 of 15 , Apr 27, 2002
        • 0 Attachment
          laartphoto is right that "There is no need for two year-zeros", but his
          comments actually miss the point: The concept of +/- 0 is inherently
          non-sensical.

          You wouldn't use any other value twice on a number line. The fact that there
          is a +5 and a -5 is not a repetition of the number 5, it's an "overloading"
          of the symbol "5". Positive five and negative five represent two separate
          and distinct values, defined as being 5 units to the right or left,
          respectively, of zero on a number line. Likewise, in the example of "+0.5"
          and "-0.5", the appearance of a zero in both expressions is a matter of
          symbology; the two expressions actually represent different values.

          On the other hand, "Positive" or "Negative" zero are inherently
          contradictory, not because of symbology, but because zero is the point on
          the number line that separates positive values from negative values.

          To use the concepts of plus/minus zero, you have only two options in
          defining them: either they're identical values, which renders the usage of
          "+/-" moot... or they're different quantities, which breaks the definition
          of zero.

          Actually, I suppose you could also say that +0 is an arbitrarily small
          positive fractional value, and -0 is an arbitrarily small negative
          fractional value, but then neither value is actually zero at all.
        • laartphoto
          A) I agree that the minus sign (-) is overused. Perhaps *all* symbols for mathematical operations are poor choices for delimiters. B) Being American, I still
          Message 4 of 15 , Apr 28, 2002
          • 0 Attachment
            A) I agree that the minus sign (-) is overused. Perhaps *all* symbols
            for mathematical operations are poor choices for delimiters.

            B) Being American, I still like the decimal point (.).

            C) I'm the one who *is* suggesting the use of both positive and
            negative years 0000. How is overloading the symbol 0000 any different
            from overloading the symbol 0005? The 0 in +0.5 and -0.5
            indicates "no whole units" from the zero point, this would be the
            same for "0 whole years" from the zero point.

            D) I'm almost ready to agree that one year 0000 makes the most sense,
            but not quite. I do see that this means they should both be leap
            years, and that's bad, but that's a flaw in specifying time in
            reference to astronomical events, which is also bad.

            E) Oh, Merlin and I kinda like the idea of -0000-01 expressing the
            month before the zero point, i.e. December of 2 B.C. It's very
            mathematical. Well, OK, I would also prefer to use 00 for the first
            month of the year and 00 for the first day of the month, it just
            makes more sense; they mean "0 whole months from the beginning of the
            year" and "0 whole days from the beginning of the month" respectively.
            But *that* change isn't likely to occur in my lifetime.

            F) Upon closer reading of the PDF, section 4.3.2.1, the standard
            invokes "... by agreement of the partners..." in reference to years
            before 1582. This means you all and I only need to agree if we intend
            to exchange data containing such dates.
            And of course none of this affects day-to-day life and the ability of
            people to catch trains.
            However, there is still room for the exchange of ideas on an academic
            level so I'll continue...


            G) Perhaps the idea of positive and negative should be scrapped and
            we should stick with BCE and CE, with both year 0 CE and 0 BCE?
            Having the positive and negative signs (to me) implies mathematical
            purity as in E) above.

            H)
            I believe that in order for there to be a true "zero point" on the
            time scale we *must* use positive and negative 0000. +0000 means the
            first positive year and -0000 means the first negative year, just as
            1 A.D. and 1 B.C. do. [0000-01] is equivalent to January of 1 B.C.
            and the previous month, December of 2 B.C., is [-0000-12]. This makes
            [0000-00-00 00:00:00] or conversely [-0000-00-00 24:00:00] be the
            very instant, i.e. "point", of midnight between the two.

            In the B.C./A.D. manner of years, there is a zero point and years
            with the same ordinal are the same interval from that point.
            If there is only one year 0000 and we consider [0000-00-00 00:00:00]
            to be the zero point but the previous year is -0001, then +x and -x
            are *not* the same interval from the zero point and things *then*
            become confusing.
            What happens if we have one year 0000 but consider 0001 and -0001 to
            both be the same interval from the zero point? Is [0000-07-01
            00:00:00] or conversely [0000-06-30 24:00:00] the zero point? Or is
            there no "zero point" at all?

            If you insist on a scale with a zero point and values stretching out
            on both sides thereof then for every positive value there *must*
            logically be a corresponding negative value. Yes, [+0] and [-0] are
            simply different ways of specifying the same point on the scale, and
            saying [-0] by itself has an aura of absurdity. The points half of
            some unit on either side of the zero point are [+0.5] and [-0.5],
            meaning "no whole units and half of a unit distant from the zero
            point" (in whichever direction). In such values the [-0] portion has
            meaning, even though we could simplify the reference to [-.5].
            Likewise we should be able to say "no whole millenia, no whole
            centuries, no whole decades, no whole years, and one half of a year
            before the common era" and specify it as [-0000.5]. To me the
            reference [-0001.5] as you suggest for the same time-point looks
            like "one whole year and a half of a year before the common era".

            I've been writing this for a couple of hours now and just realized
            that the benefit of one year 0000 (that none of you mentioned) is in
            determining time intervals by simply subtracting. The interval from
            0001 to 0000 should be 1 year, likewise the interval from -0001 to
            0000 should be 1 year, and the interval from 0001 to -0001 is 2
            years. But again this a flaw introduced by using units based on
            astronomical events. The interval from 0000-01 to -0001-12 should
            only be 1 month, not 13 months. Unless a purely mathematical system
            as mentioned in E) is used we can't use math to find the interval. As
            mentioned somewhere in my readings (regarding the interval between
            [xxxx-W01] and [yyyy-02] for instance) on this you need to convert
            everything to the same units (generally seconds) to do the math.

            <SOAPBOX>
            This whole mess came about because our ancient forebears
            (understandably) were not sophisticated enough to *measure* time, but
            only counted time intervals based on the obvious cycles of years,
            months, and days. We (or at least *I*) now know better, but we're
            stuck with it and need to make do. At some point in the future we
            will need to devise a better method of specifying times, especially
            when space travel becomes commonplace.
            Such a method of measuring time should avoid the usage of arbitrary
            and incompatible units such as year, month, and day; just as the
            metric system (SI) avoids whacky units like tablespoons and teaspoons.
            </SOAPBOX>

            Now I must go to bed.

            --- In ISO8601@y..., "Morris, Mike" <Mike.Morris@k...> wrote:
            > laartphoto is right that "There is no need for two year-zeros", but
            his
            > comments actually miss the point: The concept of +/- 0 is inherently
            > non-sensical.
            >
            > You wouldn't use any other value twice on a number line. The fact
            that there
            > is a +5 and a -5 is not a repetition of the number 5, it's
            an "overloading"
            > of the symbol "5". Positive five and negative five represent two
            separate
            > and distinct values, defined as being 5 units to the right or left,
            > respectively, of zero on a number line. Likewise, in the example
            of "+0.5"
            > and "-0.5", the appearance of a zero in both expressions is a
            matter of
            > symbology; the two expressions actually represent different values.
            >
            > On the other hand, "Positive" or "Negative" zero are inherently
            > contradictory, not because of symbology, but because zero is the
            point on
            > the number line that separates positive values from negative values.
            >
            > To use the concepts of plus/minus zero, you have only two options in
            > defining them: either they're identical values, which renders the
            usage of
            > "+/-" moot... or they're different quantities, which breaks the
            definition
            > of zero.
            >
            > Actually, I suppose you could also say that +0 is an arbitrarily
            small
            > positive fractional value, and -0 is an arbitrarily small negative
            > fractional value, but then neither value is actually zero at all.
          • Harry Shipley
            ... [snip] ... The point is mentioned in the footnotes pp5,6. I think the word proleptic is missspelt. But the method of counting using negative numbers is
            Message 5 of 15 , Apr 28, 2002
            • 0 Attachment
              > -----Original Message-----
              > From: laartphoto [mailto:LAartPhoto@...]
              > Sent: 2002-Apr-27 08:25
              > To: ISO8601@yahoogroups.com
              > Subject: [ISO8601] Year 0
              >
              [snip]

              > I haven't found anywhere in the PDF of the standard where it says
              > this (I'm still reading), but my opinion is that if you add a year 0
              > you need to add positive and negative 0. But I hear people say, "but
              > a positive and negative 0 is confusing". But these same persons are
              > not confused by positive and negative 1, so what's the difference? I
              > would very much prefer that we, as scientists and other professional
              > thinking persons, use positive and negative 0 with years just as we
              > would with positive and negative 0 degrees of temperature or whatnot,
              > as in -0.5 degrees centigrade (oh, umm, 0,5 for the Europeans and
              > such). It just doesn't make sense (to me) to only have one 0; "two or
              > none" is what I say.
              >
              The point is mentioned in the footnotes pp5,6. I think the word "proleptic"
              is missspelt. But the method of counting using negative numbers is surely
              standard mathematics, and at least the mathematicians have adopted a
              consistent system. The method allows arithmetic to move from positive only
              to positive and negative without a glitch.

              Harry Shipley

              >
              [snip]
              >
            • bam
              ... This is silly and irrelevant! If Merlin wants December to be represented by month -1 then I suppose that s OK, but this extension would apply to all
              Message 6 of 15 , Apr 28, 2002
              • 0 Attachment
                laartphoto wrote:
                > ...
                > E) Oh, Merlin and I kinda like the idea of -0000-01 expressing the
                > month before the zero point, i.e. December of 2 B.C. It's very
                > mathematical.

                This is silly and irrelevant!

                If Merlin wants December to be represented by "month -1" then I suppose
                that's OK, but this extension would apply to all years, not just to
                years before the year zero! (This month, then, would be -9 of the year 2003.)

                I don't see much point in it, but there is no real objection to
                measuring BACKWARD time with negative intervals.

                However, the year number has nothing to do with any of this!

                The year preceeding the year whose number is 1 is the year whose
                number is zero; the year before that has a number of -1;
                the year -2 preceeds it; and so forth.

                Each year, regardless of whether its number is positive or negative,
                contains twelve months which can be numbered from 1 to 12 -- with
                month 1 coming chronologically BEFORE month 2, etc. This is not
                affected by the sign of the year. The days of a month proceed
                FORWARD in time, from 1 to 31 (or 30 or 29 or 28), regardless
                of whether the number of the year is positive, negative, or zero.

                Year numbering does NOT reverse the sequence of months or days.

                Month 1 PRECEEDS month 2, regardless of year number.
                Day 2 PRECEEDS day 2, regardless of year number or month number.

                Time does not run backwards during a year simply because it
                happens to be prior to some arbitrary point at which the year
                number becomes positive!

                The number of a year is an integer.
                The number of a year is an ORDINAL number. (Look it up!)
                It is not unreasonable to extende the ordinal numbering of years
                to years prior to the year 1, but using zero for the year before it,
                and negative numbers before that. However, that cause calendars to
                suddenly run backwards just because a year preceeds the arbitrarily-
                chosen zero year.

                The number of the month within any year normally runs (FORWARD) from 1 to 12.
                Even if negative numbers are used to indicate a month last year,
                the days of THAT MONTH still run FORWARD from 1 to 28,29,30, or 31,
                regardless of whether the month is numbered. You may use negative
                numbers to indicate days of LAST MONTH, if you wish, but that is
                not affected by whether the year number is positive or negative
                (and it is not affected by whether the year numbers cross zero).

                You must distinguish between numbering and intervals.
                Numbering merely indicates a sequence, with negative numbers
                occurring (and diminishing) before zero and positive numbers
                occurring (and increasing) after zero. When a year is given
                a number, that year number applies to the entire year but does
                not affect the way months or days are identified within that year.
                (Similarly, the minute number of any hour is not affected by
                AM or PM, nor by whether the hour is before or after midnight.)

                An interval may be positive or negative, to indicate direction,
                just like a vector. Two months ago, the month was called March.
                The days of last March were numbered from 1 to 31 in the FORWARD
                direction. The days of next March, 2003, are numbered from 1 to 31
                in the forward direction, too. That numbering will remain true,
                even when, in a few years, March 2003 (2003-03) will be in the past.


                >....
                > G) Perhaps the idea of positive and negative should be scrapped and
                > we should stick with BCE and CE, with both year 0 CE and 0 BCE?
                > Having the positive and negative signs (to me) implies mathematical
                > purity as in E) above.

                Nooooooooooo!

                There is no 0 CE, and there is no 0 BCE.
                (That's part of the problem with that arbitrary, archaic system.)

                There is absolutely no need for the silly abbreviations
                (CE, AD, BC, BCE, etc.) either. Integers are well defined.
                Use integers to number years in sequence -- regardless of
                which year is arbitrarily used as an origin, by assigning
                it the number zero. Only ONE year can be assigned the
                number zero. The year before is assigned the number -1.

                The interval from any date in the the year -1 to the same date
                in the year +1 is exactly two years. (The interval from the
                last day in the year -1 to the first day in the year +1 is
                a year and a day.)
              • i18nGuy Tex Texin
                Laartphoto: I think you are onto something with 2 names for the same point. But why stop with two? Why not have 10 names for the same point or more? We can
                Message 7 of 15 , Apr 29, 2002
                • 0 Attachment
                  Laartphoto:

                  I think you are onto something with 2 names for the same point. But why
                  stop with two?
                  Why not have 10 names for the same point or more?

                  We can have +0, -0, A0, B0, Laartphoto0, Tex0...
                  It's a great idea.
                  Maybe we can start a registry and charge people who want to have their
                  name given to the point?
                  Like those services that sell star names...

                  --
                  -------------------------------------------------------------
                  Tex Texin
                  mailto:Tex@...
                  http://www.i18nGuy.com
                  -------------------------------------------------------------
                  For a compelling demonstration for Unicode:
                  http://www.i18nguy.com/unicode/unicode-example-intro.html
                • Harry Shipley
                  ... It s like logarithms. Where you have tables for dealing with the fractional parts (calendar of log tables) which are basically positive, it is an
                  Message 8 of 15 , Apr 29, 2002
                  • 0 Attachment
                    > -----Original Message-----
                    > From: bam [mailto:websitedevelopment@...]
                    > Sent: 2002-Apr-29 06:37
                    > To: ISO8601@yahoogroups.com
                    > Subject: Re: [ISO8601] Re: Year 0
                    >
                    >
                    > laartphoto wrote:
                    > > ...
                    > > E) Oh, Merlin and I kinda like the idea of -0000-01 expressing the
                    > > month before the zero point, i.e. December of 2 B.C. It's very
                    > > mathematical.
                    >
                    > This is silly and irrelevant!
                    >
                    > If Merlin wants December to be represented by "month -1" then I suppose
                    > that's OK, but this extension would apply to all years, not just to
                    > years before the year zero! (This month, then, would be -9 of
                    > the year 2003.)
                    >
                    > I don't see much point in it, but there is no real objection to
                    > measuring BACKWARD time with negative intervals.
                    >
                    > However, the year number has nothing to do with any of this!
                    >
                    > The year preceeding the year whose number is 1 is the year whose
                    > number is zero; the year before that has a number of -1;
                    > the year -2 preceeds it; and so forth.
                    >
                    > Each year, regardless of whether its number is positive or negative,
                    > contains twelve months which can be numbered from 1 to 12 -- with
                    > month 1 coming chronologically BEFORE month 2, etc. This is not
                    > affected by the sign of the year. The days of a month proceed
                    > FORWARD in time, from 1 to 31 (or 30 or 29 or 28), regardless
                    > of whether the number of the year is positive, negative, or zero.
                    >
                    > Year numbering does NOT reverse the sequence of months or days.
                    >
                    > Month 1 PRECEEDS month 2, regardless of year number.
                    > Day 2 PRECEEDS day 2, regardless of year number or month number.
                    >
                    > Time does not run backwards during a year simply because it
                    > happens to be prior to some arbitrary point at which the year
                    > number becomes positive!
                    >
                    > The number of a year is an integer.
                    > The number of a year is an ORDINAL number. (Look it up!)
                    > It is not unreasonable to extende the ordinal numbering of years
                    > to years prior to the year 1, but using zero for the year before it,
                    > and negative numbers before that. However, that cause calendars to
                    > suddenly run backwards just because a year preceeds the arbitrarily-
                    > chosen zero year.
                    >
                    > The number of the month within any year normally runs (FORWARD)
                    > from 1 to 12.
                    > Even if negative numbers are used to indicate a month last year,
                    > the days of THAT MONTH still run FORWARD from 1 to 28,29,30, or 31,
                    > regardless of whether the month is numbered. You may use negative
                    > numbers to indicate days of LAST MONTH, if you wish, but that is
                    > not affected by whether the year number is positive or negative
                    > (and it is not affected by whether the year numbers cross zero).

                    It's like logarithms. Where you have tables for dealing with the fractional
                    parts (calendar of log tables) which are basically positive, it is an
                    advantage if you can use them when the overall number is negative. So
                    instead of -4.7, we call it bar5 point 3 ie )0.3 more than (positive)
                    than -5. Bar 5 is written with a line or bar over the top. This keeps the
                    fractional bits positive.

                    >
                    > You must distinguish between numbering and intervals.
                    > Numbering merely indicates a sequence, with negative numbers
                    > occurring (and diminishing) before zero and positive numbers
                    > occurring (and increasing) after zero. When a year is given
                    > a number, that year number applies to the entire year but does
                    > not affect the way months or days are identified within that year.
                    > (Similarly, the minute number of any hour is not affected by
                    > AM or PM, nor by whether the hour is before or after midnight.)
                    >
                    > An interval may be positive or negative, to indicate direction,
                    > just like a vector. Two months ago, the month was called March.
                    > The days of last March were numbered from 1 to 31 in the FORWARD
                    > direction. The days of next March, 2003, are numbered from 1 to 31
                    > in the forward direction, too. That numbering will remain true,
                    > even when, in a few years, March 2003 (2003-03) will be in the past.
                    >
                    >
                    > >....
                    > > G) Perhaps the idea of positive and negative should be scrapped and
                    > > we should stick with BCE and CE, with both year 0 CE and 0 BCE?
                    > > Having the positive and negative signs (to me) implies mathematical
                    > > purity as in E) above.
                    >
                    > Nooooooooooo!
                    >
                    > There is no 0 CE, and there is no 0 BCE.
                    > (That's part of the problem with that arbitrary, archaic system.)
                    >
                    > There is absolutely no need for the silly abbreviations
                    > (CE, AD, BC, BCE, etc.) either. Integers are well defined.
                    > Use integers to number years in sequence -- regardless of
                    > which year is arbitrarily used as an origin, by assigning
                    > it the number zero. Only ONE year can be assigned the
                    > number zero. The year before is assigned the number -1.
                    >
                    > The interval from any date in the the year -1 to the same date
                    > in the year +1 is exactly two years. (The interval from the
                    > last day in the year -1 to the first day in the year +1 is
                    > a year and a day.)
                    >

                    I have seen on the web (and I can't recall where) an attempt to use CE to
                    imply the astronomical method of counting. It also seemed to use BCE to
                    count in the same way as BC, which I found confusing.

                    Harry Shipley
                  • jsears
                    ...and a slash may also represent a division sign, or per ;-) Jon Sears
                    Message 9 of 15 , Apr 29, 2002
                    • 0 Attachment
                      ...and a slash may also represent a division sign, or "per" ;-)
                      Jon Sears

                      bam wrote:

                      >
                      > My use and promotion of ISO 8601 have both been hampered by
                      > the unfortunate choice of the dash (minus sign and interval
                      > and interval indicator) as a date separator.
                      >
                      > I often substitute a slash.
                      > Writing today as 2002/04/27 (or '02/04/27) makes very clear
                      > to everyone that a date is indicated (while promoting the
                      > YMD sequence
                    • laartphoto
                      Umm, of course you can give many names to each point, I call one of the points my birthday . You don t need to call that point that, you can and probably do
                      Message 10 of 15 , Apr 29, 2002
                      • 0 Attachment
                        Umm, of course you can give many names to each point, I call one of
                        the points "my birthday". You don't need to call that point that, you
                        can and probably do call a diffeent point "my birthday". On the other
                        hand we probably agree pretty well on a point we call "when JFK was
                        shot".

                        --- In ISO8601@y..., i18nGuy Tex Texin <tex@i...> wrote:
                        > Laartphoto:
                        >
                        > I think you are onto something with 2 names for the same point. But
                        why
                        > stop with two?
                        > Why not have 10 names for the same point or more?
                        >
                        > We can have +0, -0, A0, B0, Laartphoto0, Tex0...
                        > It's a great idea.
                        > Maybe we can start a registry and charge people who want to have
                        their
                        > name given to the point?
                        > Like those services that sell star names...
                        >
                        > --
                        > -------------------------------------------------------------
                        > Tex Texin
                        > mailto:Tex@i...
                        > http://www.i18nGuy.com
                        > -------------------------------------------------------------
                        > For a compelling demonstration for Unicode:
                        > http://www.i18nguy.com/unicode/unicode-example-intro.html
                      • i18nGuy Tex Texin
                        Apparently my sarcasm didn t help refine the discussion. I am of course trying to understand why you want more than one name for a single point. It satisfies
                        Message 11 of 15 , Apr 29, 2002
                        • 0 Attachment
                          Apparently my sarcasm didn't help refine the discussion.
                          I am of course trying to understand why you want more than one name for
                          a single point.
                          It satisfies no requirement that I can think of.

                          It also causes needless problems, since when I take the difference of
                          two numbers I now need to accomodate producing two results potentially
                          or decide which of two to use. +0 or -0.
                          It gets especially ugly when I take the difference of +0 and -0.
                          Providing for this would break most api and software.
                          tex


                          laartphoto wrote:
                          >
                          > Umm, of course you can give many names to each point, I call one of
                          > the points "my birthday". You don't need to call that point that, you
                          > can and probably do call a diffeent point "my birthday". On the other
                          > hand we probably agree pretty well on a point we call "when JFK was
                          > shot".
                          >
                          > --- In ISO8601@y..., i18nGuy Tex Texin <tex@i...> wrote:
                          > > Laartphoto:
                          > >
                          > > I think you are onto something with 2 names for the same point. But
                          > why
                          > > stop with two?
                          > > Why not have 10 names for the same point or more?
                          > >
                          > > We can have +0, -0, A0, B0, Laartphoto0, Tex0...
                          > > It's a great idea.
                          > > Maybe we can start a registry and charge people who want to have
                          > their
                          > > name given to the point?
                          > > Like those services that sell star names...
                          > >
                          > > --
                          > > -------------------------------------------------------------
                          > > Tex Texin
                          > > mailto:Tex@i...
                          > > http://www.i18nGuy.com
                          > > -------------------------------------------------------------
                          > > For a compelling demonstration for Unicode:
                          > > http://www.i18nguy.com/unicode/unicode-example-intro.html
                          >
                          > Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

                          --
                          -------------------------------------------------------------
                          Tex Texin
                          mailto:Tex@...
                          http://www.i18nGuy.com
                          -------------------------------------------------------------
                          For a compelling demonstration for Unicode:
                          http://www.i18nguy.com/unicode/unicode-example-intro.html
                        • Morris, Mike
                          I m a very casual reader of this list, and I must admit that my earlier posting didn t take the subject matter context into account. I still think it would be
                          Message 12 of 15 , Apr 29, 2002
                          • 0 Attachment
                            I'm a very casual reader of this list, and I must admit that my earlier
                            posting didn't take the subject matter context into account. I still think
                            it would be a VBTI (Very Bad Thing Indeed) to use + and - zero in any
                            context, but after some reflection I do now understand better what
                            laartphoto probably meant.

                            The problem came, for me, by missing the distinction between "points" and
                            "intervals". In a mathematical context, zero is, of course, a single point;
                            hence my comments about the silliness of having +/-0.

                            However, in a calendaring context, the norm is to use numbers to name
                            intervals, not points. Hence "2002" is the collection of all points for
                            which:
                            2001 < X < 2003

                            In that context it is at least possible to have an interval named "+0",
                            which could be the unit interval beginning at 0, and a distinct interval
                            "-0", which would be the unit interval ending at 0.

                            But, as I said, still a VBTI, IMHO... certainly your point about breaking
                            most software is well taken!

                            I'm also ignorant of what is probably the crux of the issue: Does 8601
                            define, imply or endorse a year zero? Or are years 1 and -1 contiguous?



                            -----Original Message-----
                            From: i18nGuy Tex Texin [mailto:tex@...]
                            Sent: Monday, April 29, 2002 3:38 PM
                            To: ISO8601@yahoogroups.com
                            Subject: Re: [ISO8601] Re: Year 0


                            Apparently my sarcasm didn't help refine the discussion.
                            I am of course trying to understand why you want more than one name for
                            a single point.
                            It satisfies no requirement that I can think of.

                            It also causes needless problems, since when I take the difference of
                            two numbers I now need to accomodate producing two results potentially
                            or decide which of two to use. +0 or -0.
                            It gets especially ugly when I take the difference of +0 and -0.
                            Providing for this would break most api and software.
                            tex
                          • i18nGuy Tex Texin
                            Mike, interesting comments. If plus or minus were a notation for indicating an interval denoted by a starting or ending point, then I can see some purpose,
                            Message 13 of 15 , Apr 30, 2002
                            • 0 Attachment
                              Mike, interesting comments.
                              If plus or minus were a notation for indicating an interval denoted by a
                              starting or ending point, then I can see some purpose, although I would
                              recommend using different symbols, and it wouldn't depend on the value
                              0.

                              Also, please note that there are intervals that span 0, so what is -1 to
                              +1? Is it a plus or minus interval?

                              8601 does have a year 0. The years go ...-3, -2, -1, 0, + 1, +2...
                              0 is a leap year btw.

                              tex

                              "Morris, Mike" wrote:
                              >
                              > I'm a very casual reader of this list, and I must admit that my
                              > earlier
                              > posting didn't take the subject matter context into account. I still
                              > think
                              > it would be a VBTI (Very Bad Thing Indeed) to use + and - zero in any
                              > context, but after some reflection I do now understand better what
                              > laartphoto probably meant.
                              >
                              > The problem came, for me, by missing the distinction between "points"
                              > and
                              > "intervals". In a mathematical context, zero is, of course, a single
                              > point;
                              > hence my comments about the silliness of having +/-0.
                              >
                              > However, in a calendaring context, the norm is to use numbers to name
                              > intervals, not points. Hence "2002" is the collection of all points
                              > for
                              > which:
                              > 2001 < X < 2003
                              >
                              > In that context it is at least possible to have an interval named
                              > "+0",
                              > which could be the unit interval beginning at 0, and a distinct
                              > interval
                              > "-0", which would be the unit interval ending at 0.
                              >
                              > But, as I said, still a VBTI, IMHO... certainly your point about
                              > breaking
                              > most software is well taken!
                              >
                              > I'm also ignorant of what is probably the crux of the issue: Does 8601
                              > define, imply or endorse a year zero? Or are years 1 and -1
                              > contiguous?
                              >
                              > -----Original Message-----
                              > From: i18nGuy Tex Texin [mailto:tex@...]
                              > Sent: Monday, April 29, 2002 3:38 PM
                              > To: ISO8601@yahoogroups.com
                              > Subject: Re: [ISO8601] Re: Year 0
                              >
                              > Apparently my sarcasm didn't help refine the discussion.
                              > I am of course trying to understand why you want more than one name
                              > for
                              > a single point.
                              > It satisfies no requirement that I can think of.
                              >
                              > It also causes needless problems, since when I take the difference of
                              > two numbers I now need to accomodate producing two results potentially
                              > or decide which of two to use. +0 or -0.
                              > It gets especially ugly when I take the difference of +0 and -0.
                              > Providing for this would break most api and software.
                              > tex
                              >
                              > Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

                              --
                              -------------------------------------------------------------
                              Tex Texin
                              mailto:Tex@...
                              http://www.i18nGuy.com
                              -------------------------------------------------------------
                              For a compelling demonstration for Unicode:
                              http://www.i18nguy.com/unicode/unicode-example-intro.html
                            • laartphoto
                              Message 14 of 15 , Apr 30, 2002
                              • 0 Attachment
                                <
                                I'm also ignorant of what is probably the crux of the issue: Does 8601
                                define, imply or endorse a year zero? Or are years 1 and -1
                                contiguous?
                                >

                                ISO 8601 only specifies formats for expressing time-related data
                                values that are defined elsewhere. It also specifically says that the
                                standard does not apply to expressing time periods before the year
                                1582. Unfortunately, section 4.3.2.1 has a note saying that year 0
                                should be a leap year.

                                I'll have to look up ISO 30-1 to see what it says.
                              • i18nGuy Tex Texin
                                Hi, Actually 8601 specifies the format and is very clear on handling years before 1582, including the specification of year 0 and leap years before 1582 to
                                Message 15 of 15 , Apr 30, 2002
                                • 0 Attachment
                                  Hi,
                                  Actually 8601 specifies the format and is very clear on handling years
                                  before 1582, including the specification of year 0 and leap years before
                                  1582 to year 0 and earlier. The expanded representation shows the sign
                                  of the year and the century.

                                  However, the standard says dates before 1582 should only be used by
                                  agreement of the partners in the information interchange.
                                  tex

                                  laartphoto wrote:
                                  >
                                  > <
                                  > I'm also ignorant of what is probably the crux of the issue: Does 8601
                                  > define, imply or endorse a year zero? Or are years 1 and -1
                                  > contiguous?
                                  > >
                                  >
                                  > ISO 8601 only specifies formats for expressing time-related data
                                  > values that are defined elsewhere. It also specifically says that the
                                  > standard does not apply to expressing time periods before the year
                                  > 1582. Unfortunately, section 4.3.2.1 has a note saying that year 0
                                  > should be a leap year.
                                  >
                                  > I'll have to look up ISO 30-1 to see what it says.
                                  >
                                  > Yahoo! Groups Sponsor
                                  > ADVERTISEMENT
                                  > [Image]
                                  >
                                  >
                                  >
                                  > Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

                                  --
                                  -------------------------------------------------------------
                                  Tex Texin
                                  mailto:Tex@...
                                  http://www.i18nGuy.com
                                  -------------------------------------------------------------
                                  For a compelling demonstration for Unicode:
                                  http://www.i18nguy.com/unicode/unicode-example-intro.html
                                Your message has been successfully submitted and would be delivered to recipients shortly.