That's a great post, Tom, showing examples of the variety in calendar usage. It's truly fascinating in a way that appeals to my inner "geek". It also provides a great sample of reasons to consider DATETIME2 simply a data type and nothing more. It holds values of dates following certain construction rules (number of days in a month, which is a leap year....), but does not attempt to codify the varieties of calendar adjustments made in different locales over the centuries. I think that's what you were saying too, in the last sentence of the first paragraph.

I think that's a fair thing for MS to have done. If I'm storing dates of historical events in the eighteenth century, then I ought to make any duration calculations aware of the Old Style/New Style break in 1752, or any other anomalies like the Scottish/English calendar difference.