[LEAPSECS] Bulletin C and all that
brooks at edlmax.com
Sun Jan 25 23:28:51 EST 2015
On 2015-01-25 10:04 PM, G Ashton wrote:
> Brooks Harris suggested ISO 8601:2004(E), 3.2.1 "The Gregorian calendar" as
> a source about the Gregorian calendar. Thanks for the suggestion, but I
> consider ISO 8601 to be garbage; it's so bad it makes me dislike the entire
I have my difficulties with 8601 too. In particular, it is famous for
its character representation of date and time. This convention is
followed in many places as if it where gospel. But that design is flawed
- it does not deal with time zones and daylight comprehensively, and
this leads to confusion in date-time representation implementations.
On the other hand, 8601's terms and definitions of time is quite
thorough, basing its terms in turn on IEC 60050-111 and IEC 60050-713.
As they say in the introduction "ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of
electrotechnical standardization." You can't really talk about time
standards without referring to 8601. Yes, I know organizations like OMG
have attempted to side-step it, but for purposes of precision
timekeeping you can't really ignore it. Its a normative reference in
many standards, POSIX, for example, and used in IEEE 1588 PTP.
8601 is the one place where the Gregorian calendar is standardized in
modern terms. I've looked at that in some detail, and I'd say it is "air
tight". It matches the POSIX definition of mktime() and the behavior of
(good) implementations of gmtime(). I've investigated those in detail.
8601, 3.2.1 The Gregorian calendar, says:
"This International Standard uses the Gregorian calendar for the
identification of calendar days."
Earlier, it says:
2.2.6 calendar day - "time interval starting at midnight and ending at
the next midnight, the latter being also the starting instant of the
next calendar day"
8601 is more carefully constructed than you give it credit for, and its
an international standard. "Difficult" yes. "Garbage", certainly not.
> Gerard Ashton
> LEAPSECS mailing list
> LEAPSECS at leapsecond.com
More information about the LEAPSECS