Definition

"The second layer encodes that number as a sequence of bits or decimal digits" - What is this? — Preceding unsigned comment added by 86.239.239.108 (talk) 00:56, 17 November 2018 (UTC)

script issues

As with many videos and etc. that Iv'e seen, i think a section about the flaws should be added. For example, Iphone Ios 7&8 will permanently fail if the date is manually set to January 1, 1970. There are probably other similar issues out there, and i think it would be nice to add a section about them. The garmine (talk) 05:07, 24 April 2018 (UTC)

Got a citation for the above claim about iphones? Everything on Wikipedia needs to fowwlo the rules at WP:V and WP:RS. --Guy Macon (talk) 08:31, 17 November 2018 (UTC)
Well, there's an article in the Grauniad and a 9 to 5 Mac article, both found with a Google search for
ios 7 failure setting date
That's a flaw, but in the firmware or OS software for the iPhone, not in the Unix date/time representation. I'm not sure whether code mishandling particular Unix time values is worth noting here or not. That one might belong in the iOS 7 and iOS 8 pages' "Problems" sections rather than here. Guy Harris (talk) 18:29, 17 November 2018 (UTC)

Does any system really keep a time a a pure linear count of seconds elapsed since 1970-01-01T00:00:00 *TAI*?

The Olson code (IANA tzdb reference implementation) can, if the leap seconds files are being used, convert a pure linear count of seconds elapsed since 1970-01-01T00:00:00 UTC to a UTC label such as 2015-06-30T23:59:60 UTC. It's not expecting a different-by-23-seconds pure linear count of seconds since 1970-01-01T00:00:00 TAI.

If a system were to keep a pure linear count of seconds since 1970-01-01T00:00:00 TAI, then a gmtime() implementation not taking leap seconds into account (such as the Olson code without the leap seconds files) would convert that to a TAI label. Guy Harris (talk) 19:51, 23 April 2016 (UTC)

Okay AsadNTR (talk) 14:20, 12 December 2018 (UTC)