Print

Print


Perhaps we need "date zones", much like time zones, to make seasons (in the MM field = SS) less ambiguous where necessary:

* ±CCYY-SS±MM-DD

  * 2018-41+02-20 = the first 3-month quarter of the year 2018, 
                    starting 2 months and 20 days after 1 January, 
                    i.e. on 21 March 2018 
                    -- or should that be 2018-31+31-21?
  * 2018-41-00-11 = the first 3-month quarter of the year 2018, 
                    starting no month and 11 days before 1 January 2018, 
                    i.e. on 21 December 2017, 
                    would be equivalent to 2018-31-011
  * Assume "21" instead of "41" above for the drafted ISO/EDTF value.

* ±CCYY-SS±MM

  * 2018-32+01 = the second 4-month third-year of the year 2018, 
                 starting 1 month later than 1 May, 
                 i.e. it runs from 1 June through 30 September
  * might not work well for negative offsets 
    because -SS-MM could be indistinguishable from -MM-DD
  * Assume "38" instead of "32" above for the drafted ISO/EDTF value.

* ±CCYY-SS±DDD

  * 2018-41-011 = the first 3-month quarter of the year 2018, 
                  starting 11 days before 1 January 2018, 
                  i.e. on 21 December 2017, 
                  could be equivalent to 2018-31-00-11
  * Assume "21" instead of "41" above for the drafted ISO/EDTF value.

* ±CCYY-SS±wWW-D

  * 2018-41-W00-1 = the first 13-week quarter of the year 2018, 
                   starting no week and 1 day before the Monday of the first week 
                   (2018-W01-1 = 1 January), 
                   i.e. on Sunday, 31 December 2017
  * Assume "21" instead of "41" above for the drafted ISO/EDTF value.

* ±CCYY-SS±wWW
  * 2018-42+W00 = the second 13-week quarter of the year 2018,
                  starting the Monday of the 14th week (2018-W14-1),
                  i.e. on 2 April 2018
  * Assume "22" instead of "42" above for the drafted ISO/EDTF value.

This, however, would definitely have to wait for the next revision of ISO 8601-2. I don't know whether there are any plans to update EDTF independently from its ISO adaptation.

Nick Bart:
> 
> Since season intervals using seasons 21 to 24, even if ambiguous, serve a useful purpose (a view that has never been challenged, I think) and, if understood as outlined, cannot be expected to cause any problems, I continue to think season intervals should be included in the upcoming standard.

I did challenge the choice of MM values 21 through 24 for 3-month / quarter-year seasons, because they do cause problems.