UTC offset for Asia/Choibalsan

[Olson, Arthur David (NIH/NCI) [E]]

> I'm still not hearing any closure on this issue. Do people agree this

> is a change we should make, or is the feeling that there isn't enough

> evidence yet?

> Do I need to propose a patch?

Our best information comes from Steffen Thorsen's 2008-03-31 message:
> eznis airways, which operates several domestic flights, has a flight
schedule here:
> http://www.eznis.com/Container.jsp?id=112 (click the English flag for
English)

> There it appears that flights between Choibalsan and Ulaanbatar arrive
about 1:35 - 1:50 hours later in local clock
> time, no matter the direction, while Ulaanbaatar-Khvod takes 2 hours
in the Eastern direction and 3:35 back, which
> indicates that Ulaanbatar and Khvod are in different time zones (like
we know about), while Choibalsan and
> Ulaanbatar are in the same time zone (correction needed).

This is good evidence that Choibalsan and Ulaanbaatar are now in the
same time zone. What we don't yet know is when Choibalsan shifted its
clocks to agree with those in Ulaanbaatar. Ideally, we'd use that
instant in the time zone file. If the instant can't be learned, we'd put
in some arbitrary cutover date with a note that it was arbitrary, a
la...

	# Choibalsan, a.k.a. Bajan Tuemen, Bajan Tumen, Chojbalsan,
	# Choybalsan, Sanbejse, Tchoibalsan
	# XXX actual instant when Choibalsan switched from GMTOFF of 9
to 8 unknown.
	Zone	Asia/Choibalsan	7:38:00 -	LMT	1905 Aug
					7:00	-	ULAT	1978
					8:00	-	ULAT	1983 Apr
					9:00	Mongol	CHO%sT	2008 Mar
30 # Choibalsan Time
					8:00	Mongol	CHO%sT

My Spanish was good enough to let me do research on the Cuban situation
earlier this year; I'm not so blessed when it comes to the Mongolian
situation. So: does anyone know when Choibalsan shifted clocks? If not,
does anyone have insights into the best phony cutover date to use?

				--ado