Talk:Gregorian calendar

RFC for help with Julian calendar conversion algorithm

There's a somewhat technical mathematical discussion at the Julian day article related to how algorithms convert Julian date to calendar date. I think some of the people watching this article might be able to contribute. Here's the specific RFC section: Talk:Julian_day#Request_For_Comment_on_presentation_of_algorithms

Semi-protected edit request on 11 September 2021

"In addition, the reform also altered the lunar cycle used by the Church to calculate the date for Easter, because astronomical new moons were occurring four days before the calculated dates."

Please look this paragraph over carefully. I believe it is being stated backwards. Days have to be added to the lunar calendar (8 in 2500 years). That means that the calendar new moons were coming sooner than the astronomical new moons and not the reverse. It can be confusing. Gal2man (talk) 23:46, 11 September 2021 (UTC)

This will need some thought and people familiar with the topic will probably contribute in a day or two. Meanwhile I have disabled the edit request because that is for simple and specified changes which do not require consideration. Johnuniq (talk) 02:31, 12 September 2021 (UTC)

We can check with the 2021 Julian Easter. The Golden Number of 2021 is 8, and from the table in Date of Easter § Julian calendar, the Julian paschal full moon is Julian April 18. Which is Gregorian May 1 by adding the 13-day difference. (Searching the web verifies that Orthodox churches did celebrate Easter on the following Sunday, May 2 Gregorian.) Astronomically there was a full moon on April 26 (in my time zone), five days before the Julian paschal full moon. So the article is correct - "astronomical new moons were occurring four days before the calculated dates" in the 16th century, and often five days before in the 21st century. Indefatigable (talk) 19:15, 12 September 2021 (UTC)

Rather than using information from Wikipedia, I did a calculation for the paschal new moon for each year from 1570 to and including 1582, and 2021. I calculated the new moon using the Explanatory Supplement to the Ephemeris, the chapter "The Calendar", section E, "Ecclesiastical Calendars" (London: Her Majesty's Stationery Office, 1961). I obtained new moon information from the website of a retired NASA astronomer, Fred Espenak. (http://astropixels.com/ephemeris/phasescat/phases2001.html and related pages) I changed dates for 2021 to the Julian calendar. The times are UT, although one could argue Rome or Jerusalem would be better times to use.

Ecclesiastical Julian new moon and astronomical new moon

Year	Ecclesiastical New Moon	Astronomical New Moon	Difference
1570	11-Mar	6-Mar	5
1571	30-Mar	25-Mar	5
1572	19-Mar	14-Mar	5
1573	8-Mar	3-Mar	5
1574	27-Mar	9-Apr	5
1575	16-Mar	12-Mar	4
1576	4-Apr	30-Mar	5
1577	23-Mar	19-Mar	4
1578	12-Mar	8-Mar	4
1579	2-Apr	27-Mar	6
1580	20-Mar	15-Mar	5
1581	9-Mar	5 Mar	4
1582	28-Mar	24-Mar	4
2021	5-Apr	18-Apr	6

Jc3s5h (talk) 16:27, 13 September 2021 (UTC)

One further point is that in times gone by, the new moon was considered to occur the first day the waxing crescent moon was visible. By modern standards, the moon is completely dark at the time of the new moon. So saying the astronomical new moon was about 4 days earlier than the ecclesiastical new moon is consistent with the older meaning of a new moon. Jc3s5h (talk) 18:21, 14 September 2021 (UTC)

Thank you Jc3s5h for your thoughtful reply. Calendar questions are sometimes confusing to me. I did an excercise to verify your proof and help clarify it for me.

Take the length of a metonic cycle = 6940 days and divide by 235. That gives an average lunation of 29.5319 days which is longer than the actual average synodical lunation of 29.53059 days. That would indeed make the astronomical phenomena occur before the ecclesiastical dates.

Thank you again! Gal2man (talk) 20:33, 16 September 2021 (UTC)

Thank you Jc3s5h for your thoughtful reply. Calendar questions are sometimes confusing to me. I did an excercise to verify your proof and help clarify it for me.

Take the length of a metonic cycle = 6940 days and divide by 235. That gives an average lunation of 29.5319 days which is longer than the actual average synodical lunation of 29.53059 days. That would indeed make the astronomical phenomena occur before the ecclesiastical dates.

Thank you again! Gal2man (talk) 20:34, 16 September 2021 (UTC)

A more accurate excercise would be to use (365.25x19) = 6939.75 6939.75÷235=29.53085 29.53059<29.53085 Same result: astronomical precedes ecclesiastical.

Gal2man (talk) 23:27, 16 September 2021 (UTC)

I think the difference between the Gregorian and the Julian calendar should be 15 days rather than 13 days after year 2000, as the Gregorian calendar skips leap years during century years which are not multiples of 400. So, after the year 1500 it would have been leap year only in 400, 800 and 1200 which would give a difference of 12 days (15 minus 3) but now the calendar only moved forward ten days.

Besides that, a year is 365.2422 days to be more exact. To make it into 365.2422 days, a leap day should be added every year which is a multiple of 4 and not a century year + every year which is a quarter-millennium year (i.e. a multiple of 250) + every year which is a quarter-decamillennium year (i.e. a multiple of 2500); it would make the years 250, 750, 1250, 1750, 2250, 2500, 2750, 3250, 3750, 4250 and 4750 leap years = eleven "exceptional leap years" per half-decamillennium, which is exactly what should be needed to get an average year of 365.2422 days = 365 + 1211/5000. 213.65.211.63 (talk) 16:06, 9 March 2022 (UTC)

Gregorian Calendar.

Should the 0.0075 not be 0.75 in the following “ The Gregorian reform shortened the average (calendar) year by 0.0075 days to stop the drift of the calendar with respect to the equinoxes.”? 192.145.145.236 (talk) 09:59, 14 April 2022 (UTC)

The current value is correct. The change from Julian to Gregorian meant dropping 3 leap days over a 400-year period. 3/400 = 0.0075 days per year average. Favonian (talk) 10:37, 14 April 2022 (UTC)

Semi-protected edit request on 18 June 2022

Change Hegira to Hijrah اخسجہ (talk) 12:43, 18 June 2022 (UTC)

Comment: makes sense, given that the relevant article is Hijrah not Hegira. --John Maynard Friedman (talk) 12:56, 18 June 2022 (UTC)

 Done as has already been done for the other requests by same editor today.--John Maynard Friedman (talk) 20:47, 18 June 2022 (UTC)

Semi-inaccuracy in top section

The number 365.2422 is the current length of the "mean tropical year", but the Gregorian calendar reformers weren't trying to approximate that (I'm not sure that the concept of mean tropical year was known in 1582). Since one of the main motivations of the Gregorian reform was to correct the date of Easter, and Easter is defined in terms of the vernal equinox, they were trying to approximate the vernal equinox year, or time interval separating vernal equinoxes, which is not exactly the same (the equinox interval is mentioned prominently near the beginning of our tropical year article). According to Tropical year#Different tropical year definitions, the current length of the vernal equinox year is 365.242374 days, and this number is more relevant for judging the accuracy of the Gregorian calendar than 365.2422. The devisers of the Revised Julian Calendar ignored this when they made the average year length be 365.242 days (a disimprovement with respect to the vernal equinox year -- they were more desperate to show their independence from Catholics by having their calendar not be exactly the same, than thinking about the historical function of the calendar with respect to Christianity). AnonMoos (talk) 09:49, 14 October 2022 (UTC)

:Time for a WP:rs then. How accurately could astronomers of the day measure the tropical year? And the reformers had to consider what would be acceptable and explicable to a substantially illiterate and inumerate "flock". The algorithm is good enough. IMO, the current statement in the lead is also good enough for our purposes: the "perfect" is described in detail in the body. We know that the orbit of planet earth is not a perfect metronome, so approximations will always be needed. For almost all of the people for almost all of the time, the niceties are entirely background radiation. --𝕁𝕄𝔽 (talk) 10:14, 14 October 2022 (UTC)

Reading on mobile, I see I didn't quite appreciate your question, my apologies. Would it meet your objection if the current more closely approximating the 365.2422-day 'tropical' or 'solar' year that is determined by the Earth's revolution around the Sun. were rephrased as more closely approximating the true duration of the time between vernal equinoxes, which is a little less than than the 365.25 days in the Julian calendar. I think we can say that without breaking WP:OR (and the present lead doesn't even say what the Julian figure is, so needs adding).

BTW, we shouldn't really say determined by the Earth's revolution around the Sun. since at the time the sun went around the earth. --𝕁𝕄𝔽 (talk) 17:10, 14 October 2022 (UTC)

Where angels fear to tread

I decided to be bold and give effect to this discussion so that the lead would read

The Gregorian calendar is the calendar used in most parts of the world.^ref It was introduced in October 1582 by Pope Gregory XIII as a modification of, and replacement for, the Julian calendar. The principal change was to space leap years differently so as to make the average calendar year 365.2425 days long, which is a little less than than the 365.25 days in the Julian calendar but which more closely approximates the true duration of the time between vernal equinoxes.

but that introduces duplication in close succession, because the next para (after stating the rule) reads

There were two reasons to establish the Gregorian calendar. First, the Julian calendar assumed incorrectly that the average solar year is exactly 365.25 days long, an overestimate of a little under one day per century, and thus has a leap year every four years without exception. The Gregorian reform shortened the average (calendar) year by 0.0075 days to stop the drift of the calendar with respect to the equinoxes.

Would it be too terse to delete There were two reasons to establish the Gregorian calendar. First, the Julian calendar assumed incorrectly that the average solar year is exactly 365.25 days long, an overestimate of a little under one day per century, and thus has a leap year every four years without exception. ? --𝕁𝕄𝔽 (talk) 13:26, 18 October 2022 (UTC)

No, not delete. How about There were two reasons to establish the Gregorian calendar. First, the Julian calendar overestimated the length of the year by a little under one day per century, and thus has a leap year every four years without exception. Better? --𝕁𝕄𝔽 (talk) 15:06, 18 October 2022 (UTC)

Swapping description and lead sections?

Would it be better to move a bunch of the stuff in "Description" into the lead section, and move what's now in the lead section to a "History" section? The lead section right now really assumes you already are very familiar with the Gregorian calendar, which reads a bit odd to me. AapeliV (talk) 00:32, 1 January 2023 (UTC)

Semi-protected edit request on 4 February 2023

Please remove this sentence:

Because the date of Easter is a function – the computus – of the date of the (northern hemisphere) spring equinox

and add this one:

Because the date of Easter is a function – the computus – of the date of the spring equinox in the northern hemisphere

While it's slightly longer, it flows better without the parenthetical phrase. 192.180.91.15 (talk) 11:48, 4 February 2023 (UTC)

 Done – small jars tc 12:24, 4 February 2023 (UTC)