ISO 8601-1:2019 Revelation, with consequences

[tucotuco]

Please see tdwg/dwc#558.

[Tasilee]

It is interesting that this aspect of ISO 8601 date ranges has just been picked up.

As far as I can figure, these are the Core Tests that may be affected in some way (possibly only in implementation?):

#36 TG2-VALIDATION_EVENTDATE_INRANGE
#76 TG2-VALIDATION_DATEIDENTIFIED_INRANGE
#86 TG2-AMENDMENT_EVENTDATE_FROM_VERBATIM
#140 TG2-MEASURE_EVENTDATE_DURATIONINSECONDS

Do we document this range interpretation issue in Notes of these Tests? The Expected Responses/Specifications will be ok as read.

I will check the Test Data to see if there are any issues.

What else?

[chicoreus]

This should not be relevant to BDQ, see, in particular: "Time of day is treated as out of scope for BDQ Use Cases. In cases where time zone data are important, dates within a Multi Record from multiple sources may have multiple plus or minus one day errors introduced. " https://github.com/tdwg/bdq/blob/master/tg2/_review/docs/supplement/index.md#42-time-non-normative in particular with the reference to https://xkcd.com/2867/

[ArthurChapman]

It is not only Time issue, @chicoreus

1900-1909 is not 1 Jan 1900 - 31 Dec 1909, but 1 Jan 1900 - 31 Dec 1908.
Because "The start of the interval is icluded, and the end of the interval is excluded.