[LEAPSECS] Crunching Bulletin B numbers (POSIX time)

[Ian Batten]

On 21 Feb 11, at 1333, Clive D.W. Feather wrote:


> Tony Finch said:

>>> There's no reason why you couldn't extend numeric tz offsites to include

>>> leapseconds, as in "Mon, 21 Feb 2011 11:17:16 +000021".

>> Yes there is: timezone offsets are relative to UTC not TAI.

> 

> That would be a matter of defining the new notation (offsets are currently

> sign plus 4 digits, after all).


But it's not as though any of it is consistent.   The main user of timestamps that insert +0100 or whatever seems to be RFC2822 mail, which permits (...) comments.  So as one example from your mail's header when I got it is postfix's "Mon, 21 Feb 2011 08:33:37 -0500 (EST)", it would be legal to append "(leap:14)" or something and parsers would correctly ignore it if they didn't know to pay attention to it.

And anyway, both you and I are playing fast and loose, as when you sent the mail your mailer added:

Received: (from clive at localhost)
	by nyder.davros.org (8.14.4/8.14.4/Submit) id p1LDXcj4020766
	for leapsecs at leapsecond.com; Mon, 21 Feb 2011 13:33:38 GMT
	(envelope-from clive)

and when mine received it it added: 

Received: from bham.ac.uk (mailer3.bham.ac.uk [147.188.128.54])
	by offsite2.batten.eu.org (8.14.3/8.14.3) with ESMTP id p1LDYJQ8026147
	for <igb at batten.eu.org>; Mon, 21 Feb 2011 13:34:19 GMT

and whatever we think our respective computers are ticking, it certainly isn't GMT, and we certainly aren't paying attention to DUT1.  We're casually assuming that everyone understands GMT to be "sort of" UTC, and not defining our terms any more accurately than that...

ian