- tz - lists.iana.org

did the Holiday Is observe DST after the rest of Queensland stopped?
by Paul Eggert Oct. 31, 1996

Oct. 31, 1996

According to the current tz database, Queensland stopped observing DST after summer 1991/1992. But the IATA SSIM (which I'm currently slogging through, thanks to Gwillim Law) says that the Holiday Islands (Hayman, Lindeman, Hamilton) observed DST through 1993/1994. Does anyone know the true story here? If the IATA is right, we'll need a new entry Australia/Hamilton to handle the anomaly. I know that those islands exist (they're mentioned in <URL:http://www.lonelyplanet.com/dest/aust/gbreef.htm>, which says Hamilton has the most people) but none of them are in my Rand McNally Goode's World Atlas or in Shanks's International Atlas, so I'll need further info to find Hamilton's latitude and longitude.

1 0

Re: Clock change orgy of 1996-10-27
by Rob Jones 4-6597 MR Sys Oct. 30, 1996

Oct. 30, 1996

Anyone willing to venture a guess as to the number of clocks involved ? (including uwaves, vcrs, etc...) 11 clocks at my home, but only 3 people. -- Rob Jones On Oct 30, 12:13am, Chris Carrier wrote: > Subject: Clock change orgy of 1996-10-27 > On 1996-10-28 05:04 GMT I posted: > > more people changed time today, 1996 October 27, than on any other date > in history. > > To which Paul Eggert commented: > > >Could be. 1991-09-15 is another candidate; it's the last time that > >China changed its clocks. > > Good point. The magnitude of the clock changing orgy, in terms of number of > people, would have been about the same. Speaking in very broad general terms, > the population of the US (ex. eastern IN, HI, and AZ) is about 250 million, > Mexico 100 million, EU 350 million, Russia 150 million, totalling 850 million, > plus or minus 100 M. The population of China is somewhere in the range of > 1100-1200 million, so it would take another 300 million people, in non-EU > Europe and the Central Asian republics of the CIS, changing time on 1996 > October 27 to exceed the Chinese 1991 record. Sounds like a close thing. > > I furthermore posted an: > > Editorial comment on the above: it should have been on the last Sunday in > September not October! >

1 0

Clock change orgy of 1996-10-27
by Chris Carrier Oct. 30, 1996

Oct. 30, 1996

On 1996-10-28 05:04 GMT I posted: more people changed time today, 1996 October 27, than on any other date in history. To which Paul Eggert commented: >Could be. 1991-09-15 is another candidate; it's the last time that >China changed its clocks. Good point. The magnitude of the clock changing orgy, in terms of number of people, would have been about the same. Speaking in very broad general terms, the population of the US (ex. eastern IN, HI, and AZ) is about 250 million, Mexico 100 million, EU 350 million, Russia 150 million, totalling 850 million, plus or minus 100 M. The population of China is somewhere in the range of 1100-1200 million, so it would take another 300 million people, in non-EU Europe and the Central Asian republics of the CIS, changing time on 1996 October 27 to exceed the Chinese 1991 record. Sounds like a close thing. I furthermore posted an: Editorial comment on the above: it should have been on the last Sunday in September not October! To which jhuggins(a)comsys-inc.com wanted to know: >I am a rather new subscriber to this newsgroup. Could you explain your >editorial remark about September versus October. I believe DST should end at the end of September because in October (in the North Temperate Zone, where most but not all of the people on this list live) there is simply not enough daylight in the morning when DST is used. The following examples for sunrise on the last day of DST this year (October 26) in certain parts of the United States are used for illustration: Dallas, TX = 7:45 Seattle, WA = 7:45 Oklahoma City = 7:46 Atlanta, GA = 7:51 Salt Lake City = 7:51 Russell, KS = 7:55 Cincinnati = 8:00 Midland, TX = 8:00 N. Platte, NE = 8:06 Louisville, KY = 8:02 Boise, ID = 8:12 Kalispell, MT = 8:13 Marquette, MI = 8:22 Williston, ND = 8:30 And since the Europeans and Russians have adopted our bad habit: Bonn.de = 8:12 Dublin.ie = 8:11 London.uk = 7:42 Madrid.es = 8:37 Moscow.ru = 8:22 Paris.fr = 8:27 St. Pete.ru = 9:04 Seville.es = 8:42 Going back on Standard Time on the last Sunday in September would make each of these sunrises an hour earlier; clearly a much better situation for public convenience and the safety of children on their way to school; in fact a DST ending of Labor Day (the American day, Mon => Sep 1) would serve schoolchildren even better. From: Garrett Wollman > INTERNET:wollman@lcs.mit.edu had a: >Comment on the editorial comment: who cares when the sun rises? Comment on the comment on the comment: Me, and anyone else who is sick and tired of getting up in the dark. Generally more DST is favored when it's not being used; I remember polls in December 1973 in the US in which a 3:1 majority was in favor of year-round DST. Then on 1974-01-06 we actually did go on DST and the 3:1 majority in favor quickly became a 3:1 majority against. The original advocates of DST did not favor DST in October. When Franklin semi-seriously suggested DST in the 18th century, he suggested that DST be used to advance the sunrise to when most people wake up. (He must have been thinking of city people; the majority lived on farms and woke up with the sun during his lifetime, whether sunrise was 3:00 or 9:00). When William Willett proposed it in 1905, he suggested advancing time 80m (20m on each of the first four Sundays) in April and ending it be turning the clock back 20m on each of the first four Sundays in SEPTEMBER, which would have meant reversion to standard time not later than September 28. My own personal view is that I don't really care (most of the time) if sunset is 14:00 or 23:59, I just want it light when I have to get up to go to work! Chris Carrier

2 1

Moscow time & public transit in CIS
by Chris Carrier Oct. 30, 1996

Oct. 30, 1996

According to a friend of mine who rode the Trans-Siberian Railroad from Moscow to Irkutsk in 1995, public air and rail transport in Russia after the USSR breakup still follows Moscow time, no matter where in Russia it is located; he mentioned getting off the train in Irkutsk at 03:00 railway time to see the sun up and people going to work at 08:00 local time, and when flying back having to wait an unexpected 5 hours in the airport for a plane which he thought was leaving on local time when it was really leaving on Moscow time. In those parts of the USSR which were in Europe and outside the Russian Federation, public transport runs on local time (which has a standard of GMT+2, moving back with Moscow in 1991 when they were still Soviet but staying on GMT+2 when Moscow standard time moved back to GMT on 1992-01-18, right after the breakup). Don't know what the Central Asian CIS republics are doing about transport time; anyone know? Chris Carrier

1 0

problem with tzcode96l
by Todd C. Miller Oct. 30, 1996

Oct. 30, 1996

Hi, we're trying to get OpenBSD to pass the NIST POSIX test suite and it tries to do a mktime() with TZ=dst0std,J057,J059 with a time of Tue Feb 26 01:59:59 1991 (which should return 667533600). mktime() from tzcode96l (and earlier versions) returns -1 with that TZ setting. There seems to be a problem with the end of daylight savings time specification. If you comment out the getrule() on line 751 it at least keeps mktime from returning -1 :-) I really don't understand the code in question. Any clues would be greatly appreciated. thanks, todd --- Todd C. Miller Sysadmin--University of Colorado Todd.Miller(a)cs.colorado.edu

2 7

Re: Non-EU Europe, record set today
by peter＠aldie.co.uk Oct. 29, 1996

Oct. 29, 1996

> I would like to ask: given that the European Union has extended summer time > until October 27, and that Russia did the same, can we assume that all > European countries that are *not* in the EU and those former republics of the > ex-Soviet Union that observed DST also delayed their ending date this year > until the end of October? No. Although it won't affect the overall numbers much, it appears that Latvia at least didn't take part in this historic mass orgy of clock changing. I was checking with a relative in Estonia about the IATA manual, which gives a different time of change for Estonia to that which he had told me before, and he said unprompted that he knew Latvia had changed on the last Sunday in September. On the IATA manual and its claim that Estonia changes at 23:00 UTC he said: `I do not [know] exactly but there are some little different (confusing) rules for International Air and Railway Transport Schedules conversion in Sunday connected with end of summer time in Estonia. Anyway Estonia will change 03:00 local time.' This may be a more general problem. Does this IATA manual specifically say that it gives times of change for civil time, or does it describe some special IATA rules? Peter Ilieve peter(a)aldie.co.uk

2 1

Non-EU Europe, record set today
by Chris Carrier Oct. 28, 1996

Oct. 28, 1996

I would like to ask: given that the European Union has extended summer time until October 27, and that Russia did the same, can we assume that all European countries that are *not* in the EU and those former republics of the ex-Soviet Union that observed DST also delayed their ending date this year until the end of October? Another thought. Given that most of the USA, Mexico, the EU and Russia all changed back on the same date, I believe that some sort of record was set today: more people changed time today, 1996 October 27, than on any other date in history. (The combined population of the four nations/unions mentioned above being in the range 800-900 million). Editorial comment on the above: it should have been on the last Sunday in September not October! Chris Carrier

2 1

Re: multithreading and timezones
by Nathan Myers Oct. 22, 1996

Oct. 22, 1996

Valery Aronov <valery(a)ind.tansu.com.au> wrote: > How is it possible to manage current zone information in MT-safe fashion? From: Paul Eggert <eggert(a)twinsun.com> > Write and use a thread-safe version of the tz library. > > This requires changing the interfaces to localtime, gmtime, tzset, etc. > (localtime_r is not enough.) One must also make the locale a > parameter, for strftime at least. And I'm afraid that this will > require going into the guts of the C library, which will make it hard > to do portably. > > On my (long) list of things to do is to make a thread-safe version of > the glibc version of the tz library. Once you know you're dealing > with that particular library, the job becomes tractable. The non-re-entrancy, and the globalness, of the time functions has long bothered me. With the C locale itself global, there's not much to be done about that. The (draft) standard C++ locale is not (necessarily) global, and supports user extensions such as time zone handling. Now would be a good time to start defining a standard time zone facet for the C++ locale, so that it can be made official at the next round, in a few years. The glibc locale functions being worked on by Ulrich Drepper are designed to support a re-entrant (C) interface. We should co-ordinate with him to get the time zone code integrated into his architecture. Please, let's get the interface right this time. Nathan Myers ncm(a)cantrip.org

1 0

Solaris multithreading and timezones
by Valery Aronov Oct. 22, 1996

Oct. 22, 1996

We are moving our software towards multithreading (SunOS 5.5.1). One of the many concerns is the MT-unsafety of tzset() function. It is said to be MT-unsafe in man pages. It is partly due to the keeping references to TZ in the environment variable, I guess. Our software needs to deal with more than one timezone in one process. The only way I know to implement it is to change the value of environment variable TZ and call tzset() every time I want to switch to a new timezone. External variables tzname and others also makes it problematic to maintain access to them: other vendors' software we are using might use their own policy of access. How is it possible to manage current zone information in MT-safe fashion? I would appreciate any help. Valery

3 3

c-common.c
by ado Oct. 18, 1996

Oct. 18, 1996

The file ftp://elsie.nci.nih.gov/pub/c-common.c is now available; this is a replacement for c-common.c in gcc (either 2.7.2 or 2.7.2.1) that does strftime format checking (with special warning for formats that generate two digits worth of years and thus may be especially subject to year 2000 challenges). If anyone has feedback before I ship it off to the GNU folk, I'd love it! --ado

3 3