7 ms·

The surprises never cease: > Further below, in § 4.2.2, it's noted that the ordinal day number of the week starts at 1, which is Monday. 2000-01-01 is defined

The surprises never cease: > Further below, in § 4.2.2, it's noted that the ordinal day number of the week starts at 1, which is Monday. 2000-01-01 is defined to be Saturday and the week calendar continues as a series of contiguous calendar weeks. This leads to e.g. the first day of 2019-W1 (a Monday) being 2018-12-31.

jorams 6y ago
Monday is considered the first day of the week in many countries[1]. Week years are based on the year in which most of the week occurs / whichever year contains the Thursday. [1]: https://en.wikipedia.org/wiki/Names_of_the_days_of_the_week#Numbered_days_of_the_week https://en.wikipedia.org/wiki/Names_of_the_days_of_the_week#...
masklinn 6y ago
I assume you are american? That's the definition of the ISO week, which is very much the standard and in common use throughout europe. Monday being the first day of the week is also standard in europe.
- npongratz 6y ago
  The concept of Monday being the first day of the week isn't what surprises me. It was interesting to see the first day of the first week of 2019 is a date in 2018. EDIT: And I'm not in any way rendering judgment, merely uttering surprise. (The country that claims me as its citizen is irrelevant, and obviously we are indeed discussing ISO's definition of a "week". )
  - randallsquared 6y ago
    If weeks are to be considered as a unit that exists "in" a year, then there is no non-surprising solution to this notion of the first and last week of the year.
    - Dylan16807 6y ago
      Depends on your threshold of surprise. I would say that the end of a week going into the next year is much less surprising or even non-surprising.
      - gowld 6y ago
        It's surprising that a day in 2019 is not part of a week in 2019.
        Dylan16807 6y ago
        Out of context yes. In context the direction matters. If I say that 3am on Saturday is part of "Friday night", that's not very surprising. If I say that 10pm on Monday is part of "Tuesday morning", that's quite surprising.
  - masklinn 6y ago
    > (The country that claims me as its citizen is irrelevant) The country is indeed irrelevant. The culture however is not: as I noted, this week ordinal definition is standard in europe, and the only people I've seen 1. not know about this definition of the week (and the concept of week-year) and 2. care enough to express any sort of surprise, have been american. > we are indeed discussing ISO's definition of a "week". The information in my comment was that this is the standard week in europe, any time somebody talks in weeks (which is common in many, many businesses, people'll tell you they'll do a job W23, or will ship your order W36), that's the one they're talking about.
  - mvolfik 6y ago
    I think there was some bug in a big service (Twitter? very unsure) a few years ago related to this conflict of ISO and 'real' year
  - OJFord 6y ago
    Well, there are I think only two reasonable definitions. One leads to the first day of the week being in the previous year; the other to the first day of the first week being six days into the year. Each is independently 'surprising' in that it seems 'obviously wrong' without thinking about it.
pimlottc 6y ago
A common date formatting error is to accidentally use the “iso week year” instead of “calendar year”, which is the same 99% of the time except sometimes at the end of the year, as your example shows. https://ericasadun.com/2018/12/25/iso-8601-yyyy-yyyy-and-why-your-year-may-be-wrong/ https://ericasadun.com/2018/12/25/iso-8601-yyyy-yyyy-and-why...

jorams 6y ago

Monday is considered the first day of the week in many countries[1]. Week years are based on the year in which most of the week occurs / whichever year contains the Thursday. [1]: https://en.wikipedia.org/wiki/Names_of_the_days_of_the_week#Numbered_days_of_the_week https://en.wikipedia.org/wiki/Names_of_the_days_of_the_week#...

masklinn 6y ago

I assume you are american? That's the definition of the ISO week, which is very much the standard and in common use throughout europe. Monday being the first day of the week is also standard in europe.

npongratz 6y ago
The concept of Monday being the first day of the week isn't what surprises me. It was interesting to see the first day of the first week of 2019 is a date in 2018. EDIT: And I'm not in any way rendering judgment, merely uttering surprise. (The country that claims me as its citizen is irrelevant, and obviously we are indeed discussing ISO's definition of a "week". )
- randallsquared 6y ago
  If weeks are to be considered as a unit that exists "in" a year, then there is no non-surprising solution to this notion of the first and last week of the year.
  - Dylan16807 6y ago
    Depends on your threshold of surprise. I would say that the end of a week going into the next year is much less surprising or even non-surprising.
    - gowld 6y ago
      It's surprising that a day in 2019 is not part of a week in 2019.
      - Dylan16807 6y ago
        Out of context yes. In context the direction matters. If I say that 3am on Saturday is part of "Friday night", that's not very surprising. If I say that 10pm on Monday is part of "Tuesday morning", that's quite surprising.
- masklinn 6y ago
  > (The country that claims me as its citizen is irrelevant) The country is indeed irrelevant. The culture however is not: as I noted, this week ordinal definition is standard in europe, and the only people I've seen 1. not know about this definition of the week (and the concept of week-year) and 2. care enough to express any sort of surprise, have been american. > we are indeed discussing ISO's definition of a "week". The information in my comment was that this is the standard week in europe, any time somebody talks in weeks (which is common in many, many businesses, people'll tell you they'll do a job W23, or will ship your order W36), that's the one they're talking about.
- mvolfik 6y ago
  I think there was some bug in a big service (Twitter? very unsure) a few years ago related to this conflict of ISO and 'real' year
- OJFord 6y ago
  Well, there are I think only two reasonable definitions. One leads to the first day of the week being in the previous year; the other to the first day of the first week being six days into the year. Each is independently 'surprising' in that it seems 'obviously wrong' without thinking about it.

pimlottc 6y ago

A common date formatting error is to accidentally use the “iso week year” instead of “calendar year”, which is the same 99% of the time except sometimes at the end of the year, as your example shows. https://ericasadun.com/2018/12/25/iso-8601-yyyy-yyyy-and-why-your-year-may-be-wrong/ https://ericasadun.com/2018/12/25/iso-8601-yyyy-yyyy-and-why...