[CF-metadata] mixed Julian-Gregorian calendar

From: Jonathan Gregory <j.m.gregory>
Date: Mon, 17 Dec 2012 18:22:34 +0000

Dear Steve

This isn't the central issue, but I wonder if I have missed your point. I
agree that there is a gap in the legal dates in the real-world mixed Julian-
Gregorian calendar, but this is just an inconvenient problem of translation,
isn't it? There is no discontinuity in real-world time!

1 days since 1582-10-14 = 0 days since 1582-10-25

because in the real-world calendar 14 Oct 1582 is followed by 25 Oct 1582.
If you code a time coordinate in "days since 1582-10-1" as 0, 1, 2, 3, ... 22
it should be translated into dates of 1582-10-1, 2, 3, ..., 14, 25, ..., 31.
This is what really happened. How could we stamp it out?

I note that cal(1) has the date change in Sep 1752, when it happened in England

$ cal 9 1752
   September 1752
Su Mo Tu We Th Fr Sa
       1 2 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30

but anyone who works with the dates of real historical events should be aware
of the need to put them into the same calendar.

Cheers

Jonathan


----- Forwarded message from Steve Hankin <steven.c.hankin at noaa.gov> -----

> Date: Fri, 14 Dec 2012 15:45:27 -0800
> From: Steve Hankin <steven.c.hankin at noaa.gov>
> User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:16.0) Gecko/20121026
> Thunderbird/16.0.2
> To: Jonathan Gregory <j.m.gregory at reading.ac.uk>
> CC: cf-metadata at cgd.ucar.edu
> Subject: Re: [CF-metadata] CF calendars (was: problem with times in PSD
> dataset)
>
>
> On 12/14/2012 9:35 AM, Jonathan Gregory wrote:
> >Dear Cecilia, Steve et al.
> >
> >Steve is right that mostly we use the Gregorian calendar. That is what I meant
> >mostly when I said that the default is the calendar we use. The real world
> >is mixed Julian-Gregorian, and I don't think dealing with this calendar is an
> >issue only for Renaissance historians. I can't give you examples, but I
> >think it is conceivable or likely that at some point people would want to
> >record real-world data in CF earlier than the Renaissance, or have already
> >done so. For instance, what about astronomical data, such as the dates of
> >eclipses. These are real-world events, on precise dates which are translated
> >into the mixed Julian-Gregorian calendar.
>
> Hi Jonathan,
>
> If scientists somewhere have encoded the dates of these historical
> events as data(*) using a mixed Gregorian-Julian calendar Lord help
> 'em. Those poor folks have to face an 11 day discontinuity in their
> own data, as well as in ours. I'm not meaning to be snarky. I just
> want to stamp out this pesky calendar issue. It has been tripping
> us up for too many years.
>
> Your points below are definitely the real guts of the discussion,
> but in this email I am addressing just the one single point. I'm
> afraid we will never heal ourselves from this virus if we do not
> eradicate it from our thinking.
>
> - Steve
>
> (*) attaching a date to an historical narrative is different from
> using a date as a time coordinate. It's metadata versus data.
>

----- End forwarded message -----
Received on Mon Dec 17 2012 - 11:22:34 GMT