|
|
|
|
|
|
|
One might wish that ISO usage had designated and consistently used different terms for calendar month and calculational month. As it stands, the distinction must be made with regard to context. |
|
|
|
|
|
|
|
|
A second point in this regard is that ISO durations may assume strange forms, as a consequence of the definitions. A duration of 364 days, for example, would be described in ISO parlance as one year and four days, or P1Y0M4D. This usage seems counterintuitive, since the period is less than a calendar year. |
|
|
|
|
|
|
|
|
The time of day notations are inconsistent from one section of the document to another. Section 5.3 states that, "hours are represented by two digits from [01] to [24], whereas minutes and seconds are represented by two digits from [01] to [60]." Errata (1) in the Preface corrects this obvious oversight to read, "hours are represented by two digits from [00] to [24], whereas minutes and seconds are represented by two digits from [00] to [59]." This correction eliminates the obvious problem of a lack of zero but introduces more problems. |
|
|
|
|
|
|
|
|
First, it contradicts the carry-over points designated in section 5.5.3.2.1 as maximum values for the standard units. These values are probably in error. |
|
|
|
|
|
|
|
|
Second, section 5.3.3 expressly permits the expression of UTC, which implies the necessity of permitting a leap second, but the range allowed in section 5.3 does not permit it. |
|
|
|
|
|
|
|
|
The intention of the ISO standard is to address current commercial needs, not date/time issues in general. As a consequence, the range of expression is too restricted for general calendric usage. In general, reference is made to the Gregorian calendar, specifically to a four-digit year. No provision is made for the astronomical representation of year, and the use of the hyphen, as described in section 5.2.3.3 would be confused with negative year numbers, if such a provision were made. |
|
|
|
|
|
|
|
|
Section 5.3.2 expressly permits the representation of midnight as either 0 or 24 hours and therefore does not prescribe a single standard. As a consequence, a single date/time does not have a single unequivocal representation. The date/time of 1999 December 31 24:00:00 is the same as the date/time of 2000 January 1 00:00:00. The equivocation is unnecessary. |
|
|
|
|
|
|
|
|
This section discusses the SCDTL data structures used to represent the formats presented in the ISO 8601 document. |
|
|
|
|
|
|
|
|
8.2.1
The ISO_YWD Structure |
|
|
|
|
|
|
|
|
The ISO_YWD structure contains the data items needed to store a date in the year-week-day format prescribed by document ISO 8601. |
|
|
|
|
|