Talk:ISO 8601

Standard Date

You can use quite a couple templates, in the YYYY-MM-DD date format

{{date|2=ISO}} or {{ISO date}} could be used. — Preceding unsigned comment added by 98.31.29.4 (talk • contribs) 00:47, 10 November 2018 UTC (UTC)

Incorrect dash-like character changes

On 17 January 2024 Alexeyevitch (talk · contribs) made an edit which changed many dash-like character changes, quite a few of which were incorrect or inadvisable. For example, the article presented ""2004-05" as valid ISO 8601 notation for May 2004, which it is. The edit changed it to "2004–05" (that is, using an n-dash character) which has an entirely different meaning, 2004-2005. In another spot "−05:00 was changed to "−05:00" which isn't incorrect, but presents difficulties for editors, because in edit mode it is just about impossible to distinguish the various dash-like characters. Jc3s5h (talk) 15:17, 17 January 2024 (UTC)

Contradictory statements about prefixing times with 'T'

This (in Times):

ISO 8601-1:2019 allows the T to be omitted in the extended format, as in "13:47:30", but only allows the T to be omitted in the basic format when there is no risk of confusion with date expressions.

contradicts this (in Combined date and time representations):

In ISO 8601:2004 it was permitted to omit the "T" character by mutual agreement as in "200704051430",[37] but this provision was removed in ISO 8601-1:2019.

Lacking access to the published standard, I've no way of ascertaining which is correct. Roger Rohrbach (talk) 01:44, 11 March 2024 (UTC)

What's meant by the 1st statement is that "194705" is not allowed as a time (except when context implies it has to be time), because it could mean May 1947. It would have to be "T194705" or "19:47:05" to be unambiguously a time and not a date. The 2nd statement applies only when a date and a time are combined. There is no contradiction. Indefatigable (talk) 17:45, 11 March 2024 (UTC)

Fix an event to define the calendar time to a second

"ISO 8601:2004 fixes a reference calendar date to the Gregorian calendar of 20 May 1875 as the date the Convention du Mètre was signed in Paris": This is good. This gives each day an absolute date, for example the day after this event is called 21 May 1875, and the day before it 19 May 1875. Similarly, to prove that today is 2024/8/26 is to prove that exactly 54520 days, or about 4710528000 seconds, have elapsed since the event.

But we should have fixed an event for a time to the accuracy of a second, since this affects when we switch the date! For example, some particular moment during the signature could have been defined to have happened at, say, exactly 20 May 1875 12:00:00 UTC. Unfortunately, I don't think the signature is possible to be fixed within a second ... 14.52.231.91 (talk) 00:36, 26 August 2024 (UTC)

The standard is intentionally vague about time keeping. For example, time can be local time in an unspecified time zone, or even a time before time zones were invented. Jc3s5h (talk) 03:22, 26 August 2024 (UTC)

Well if I understood it correctly ... So we switch from a day to the next day just because all clocks in the world agree to do so :) 14.52.231.91 (talk) 04:02, 26 August 2024 (UTC)

Duration

Does ISO 8601 clarify Year duration, for example if a duration is 3 years, was one of them a leap year? Even more important for Months, if a duration is 3 months, was it Jan, Feb, March — 90 or 91 days, depending on leap year, or July, August, September —92 days? GeorgeHarnish (talk) 15:16, 23 September 2024 (UTC)