Re: [Cbor] draft-ietf-cbor-date-tag-02 - handling of time zone offsets

[Carsten Bormann <[e-mail]>]

> On 2020-07-09, at 06:04, Jim Schaad <[e-mail]> wrote:
> 
> I agree that I don't want to deal with differences in calendars.  

Tag 100 is calendar agnostic, as long as the calendar has an equivalent to what is 1970-01-01 in the Gregorian calendar, and can count earth days back/forward from that.

Tag 1004 has the Gregorian calendar baked in, with leap years etc.  That also means that 1582-10-04 (100(-141428)) is immediately followed by 1582-10-15 (100(-141427)), and even more caution is needed before this date (proleptic Gregorian calendar).  But we don’t have to explain all that, just point out that there is some complexity with the proleptic Gregorian calendar.

Tag 100 dates inherit some of that complexity if combined with/converted to/from Gregorian dates, more so if the conversion actually is to/from a date-time anchored to a time zone (civil time).  That is also pretty much all we need to say about that.

> It might have been easier to go with the Jewish calendar but that would not be at all well known.  The conversions on dates should assume that the calendars are the same (and reasonable).

I’d go with the French revolutionary calendar — ten-day weeks (properly split into five workdays and five-day weekends, of course) would be so much more natural.  But I digress…

Grüße, Carsten