On Mar 7, 2011 10:56 AM, "Ray Denenberg, Library of Congress" <[log in to unmask]
> From: SaaĹˇha MetsĂ¤rantala
>> Sent: Friday, March 04, 2011 12:40 PM
>> > January 3 of 2003 can be represented as 2003-01-03.
>> Yes, as long as the day number is below 60 and / or it is known whether
>> the year is a leap year or not, which is not always known. Let's
>> consider the following example:
>> The representation of y30000-070 as year-month-day depends on whether
>> year 30 thousand will be a leap year or not, which is not determined
>> yet, ....
> First let's digress to the discussion we had about how to represent a year greater than 9999. Recall that in earlier drafts we had arbitrarily chosen "six-digits" for any extended date. That was challenged and the consensus was that any year to be represented with ISO 8601 would be four digits and there would be no representation of years greater than four digits using 8601 representation. That simplifid matters considerably, but left open the question of how indeed to represent years later than 9999. We agreed that we would represent them using scientific notation (and implicit in the agreement was that for this feature we would deviate from ISO 8601). That led to the question of how to represent an actual date when the year is greater than 9999, for example, January 1, 10001. And the consensus was that there is no need to represent more than a year - that a specific date that far away would be meaningless.
> So to your point, the spec does not address "day 70 of year 30000" or "March 10, 30000", etc. and so, the leap year issue is only a concern for years 9999 or less.