Rule Russia incorrect
Time Zone Data v. 2011m (Released 24 October 2011) file contains a line ---------------------------------------------------------------------------------- Rule Russia 1993 2010 - Mar lastSun 2:00s 1:00 S ---------------------------------------------------------------------------------- which I believe means that last change to summer time was in 2010. This is incorrect, because this (2011) year we actually have changed time on last Sunday of March (by moving clock forward) I believe that for historical purposes the rule must looks as follows ---------------------------------------------------------------------------------- Rule Russia 1993 2011 - Mar lastSun 2:00s 1:00 S ---------------------------------------------------------------------------------- On the other hand Rule --------------------------------------------------------------------------------------- Rule Russia 1996 2010 - Oct lastSun 2:00s 0 - ----------------------------------------------------------------------------------------- is correct, because Russia stopped using Daylight Savings Time switching because of a federal law "On calculation of time" signed by President Dmitry Medvedev on June 9, 2011 -- Sincerely yours, Konstantin Kivi <konst@sirena2000.ru> IT Director, Comtech-N JSC Moscow, Russia, +7-495-941-60-02
Date: Mon, 31 Oct 2011 15:39:45 +0300 From: Konstantin Kivi <konst@sirena2000.ru> Message-ID: <4EAE9711.50408@sirena2000.ru> | Time Zone Data v. 2011m (Released 24 October 2011) | file contains a line | which I believe means that last change to summer time was in 2010. Yes, that is the way we have it encoded. | This is incorrect, because this (2011) year we actually have changed | time on last Sunday of March (by moving clock forward) The question is whether that change was to summer time, or a change of timezone, altering standard time. The latter is the way the timezone data has encoded it. Note the entry in the Europe/Moscow zone data ... 3:00 Russia MSK/MSD 2011 Mar 27 2:00s which says that as from Mar 27, 2011, the timezone is UTC+3 (whereas it had been UTC+2 earlier). This results in the transitions being ... Europe/Moscow Sat Mar 27 22:59:59 2010 UTC = Sun Mar 28 01:59:59 2010 MSK isdst=0 Europe/Moscow Sat Mar 27 23:00:00 2010 UTC = Sun Mar 28 03:00:00 2010 MSD isdst=1 Europe/Moscow Sat Oct 30 22:59:59 2010 UTC = Sun Oct 31 02:59:59 2010 MSD isdst=1 Europe/Moscow Sat Oct 30 23:00:00 2010 UTC = Sun Oct 31 02:00:00 2010 MSK isdst=0 Europe/Moscow Sat Mar 26 22:59:59 2011 UTC = Sun Mar 27 01:59:59 2011 MSK isdst=0 Europe/Moscow Sat Mar 26 23:00:00 2011 UTC = Sun Mar 27 03:00:00 2011 MSK isdst=0 Note that in 2010, when the time (in March) moved from UTC+2 to UTC+3, isdst becomes 1. On the other hand, when the similar change occurred in March 2011, from UTC+2 to UTC+3, isdst remains at 0. Now it is certainly true that this could be coded either way, and if the common opinion in Russia is that you now have summer time (or daylight saving) turned on all year around, then perhaps we should switch the way it is done. On the other hand, if the view is that you now have a different standard time than before, and no more summer time (daylight saving), then the way it is now seems better. Either way, the timestamps are (I think) correct, the only difference is whether we consider summer time is turned on or not. kre
Konstantin Kivi <konst@sirena2000.ru> writes:
Time Zone Data v. 2011m (Released 24 October 2011) file contains a line
---------------------------------------------------------------------------------- Rule Russia 1993 2010 - Mar lastSun 2:00s 1:00 S ---------------------------------------------------------------------------------- which I believe means that last change to summer time was in 2010. This is incorrect, because this (2011) year we actually have changed time on last Sunday of March (by moving clock forward) I believe that for historical purposes the rule must looks as follows ---------------------------------------------------------------------------------- Rule Russia 1993 2011 - Mar lastSun 2:00s 1:00 S ----------------------------------------------------------------------------------
You have to look a bit lower, where the Zone description takes place. For example, Moscow contains the following: Zone Europe/Moscow 2:30:20 - LMT 1880 [...] 3:00 Russia MSK/MSD 2011 Mar 27 2:00s 4:00 - MSK So on Mar 27, Moscow stopped observing ruleset Russia, and instead changed to permanent offset of +4:00. It's done this way, because permanent daylight saving is not really _saving_, it's simply a new zone. PM
Date: Mon, 31 Oct 2011 16:27:37 +0100 From: Petr Machata <pmachata@redhat.com> Message-ID: <m28vo1tc46.fsf@redhat.com> | So on Mar 27, Moscow stopped observing ruleset Russia, and instead | changed to permanent offset of +4:00. Yes, that description is much better than my version with the offsets all incorrect... kre
Is IANA picking up Robert's uploads automatically or is this done manually? Seems like the last update is a bit late.
On Mon, 31 Oct 2011, Petr Machata wrote:
So on Mar 27, Moscow stopped observing ruleset Russia, and instead changed to permanent offset of +4:00. It's done this way, because permanent daylight saving is not really _saving_, it's simply a new zone.
Yes, that what the tzdata files say. But is that an accurate reflection of what happened on the ground? The relevant legislation hadn't yet been passed on 27 March 2011. Wouldn't it be more accurate to say that there was a transition to daylight savings time on 27 March 2011, and then another transition from {base offset = +03:00, extra DST offset = 01:00} to {base offset = +04:00, no DST} on some later date when the new legislation took effect? (And for some timezones, several back and forth transitions as the legislators changed their minds, each transition affecting our notion of whether or not DST was in effect, but not affecting the setting of clocks that lack isdst indicators and timezone abbreviation indicators.) --apb (Alan Barrett)
It certainly could be more accurate, but definitely would be more confusing and would be unlikely to reflect the common thought "on the street" about exactly what changes were made and when. Although the actual legislation for Russia hadn't been passed as of 27 March 2011, there had been a presidential decree to similar effect as early as February (http://rt.com/news/daylight-saving-time-abolished/), so I feel that keeping the logical switch on that date makes sense. Just because a law was decreed on a given date, it doesn't mean that a transition *happened* on that date. -- Tim Parenti On Mon, Oct 31, 2011 at 13:22, Alan Barrett <apb@cequrux.com> wrote:
Yes, that what the tzdata files say. But is that an accurate reflection of what happened on the ground? The relevant legislation hadn't yet been passed on 27 March 2011.
Wouldn't it be more accurate to say that there was a transition to daylight savings time on 27 March 2011, and then another transition from {base offset = +03:00, extra DST offset = 01:00} to {base offset = +04:00, no DST} on some later date when the new legislation took effect?
(And for some timezones, several back and forth transitions as the legislators changed their minds, each transition affecting our notion of whether or not DST was in effect, but not affecting the setting of clocks that lack isdst indicators and timezone abbreviation indicators.)
--apb (Alan Barrett)
participants (6)
-
Alan Barrett -
Konstantin Kivi -
Petr Machata -
Robert Elz -
Tim Parenti -
Zoidiasoft Tech