# JewishNewYear date limitation

The documentation for the function "JewishNewYear" in the "calendar" package declares that the year argument must be between 1900 and 2100. No explanation is given for this limitation. The function continues to produce seemingly correct dates for years outside of that range. The Jewish calendar is mathematically determined, so there should be no reason for the algorithm to fail outside that range. Moreover, a recreation of the function using CalendarChange produces the same answers as JewishNewYear, for any Gregorian year:

JewishNewYear2[year_] :=  CalendarChange[{First[CalendarChange[{year, 8, 1},
Gregorian, Jewish]] + 1, 7, 1}, Jewish, Gregorian]


Anyone have any idea why the documentation would say that JewishNewYear doesn't work outside of 1900 to 2100?

-
Are leap year decisions necessarily mathematically defined? Perhaps this is a reason to specify what seem like arbitrary bounds on changes in century? – RM1618 Sep 24 '13 at 3:56
Yes, the entire calendar is mathematically defined. It used to be observation based, but it was defined mathematically over a thousand years ago and has not been touched since. – jmizrahi Sep 24 '13 at 4:45
Perhaps this has to do with AbsoluteTime[] which is defined as the number of seconds since January 1st, 1900. Perhaps accuracy can only be guaranteed until CE 2100? – RM1618 Sep 24 '13 at 4:48
I'm sure the support will be happy to tell you if that date range means anything. It sounds like they made a mistake, but only they can confirm it. If you ask, it would be great if you could report back/answer this question :) – C. E. Sep 24 '13 at 10:44
@jmizrahi Where is it documented that the calender is only defined from CE 1900 to CE 2100? I can't seem to find it? – RM1618 Sep 24 '13 at 18:07