- tz - lists.iana.org

Re: Bug in yearistype.sh
by ado Sept. 1, 1995

Sept. 1, 1995

The file tzdata95g.tar.gz on in the ~ftp/pub directory on elsie.nci.nih.gov (165.112.73.1) has been updated to incorporate the "yearistype" correction. --ado

1 0

Bug in yearistype.sh
by Rickard Faith Sept. 1, 1995

Sept. 1, 1995

I'm not sure if this is fixed or not, but I'm passing it on just in case. The yearistype.sh version in question is 7.3. - ------- Forwarded Message Return-Path: koop(a)biene.u-strasbg.fr Date: Fri, 1 Sep 1995 08:52:50 -0400 (EDT) From: "Felix A. KOOP" <koop(a)biene.u-strasbg.fr> To: "Michael K. Johnson" <johnsonm(a)nigel.vnet.net> Subject: yearistype in util-linux-2.4 Message-ID: <Pine.LNX.3.91.950901084813.424A-100000(a)biene.u-strasbg.fr> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII - - -----BEGIN PGP SIGNED MESSAGE----- Hi, I found something bizarre in the yearistype script in util-linux-2.4. If you are no longer the maintainer, please can you forward it to the person who maintains it now. Thanks. The script reads like this: > 2-nonpres) case $1 in > *[02468][048]|*[13567][26]) exit 1 ;; ^^^ Should be 579 I think > *) exit 0 ;; > esac ;; > 2-odd) case $1 in > *[13579]) exit 0 ;; > *) exit 1 ;; > esac ;; > 2-uspres) case $1 in > *[02468][048]|*[13567][26]) exit 0 ;; ^^^ Here also 579 > *) exit 1 ;; > esac ;; If I'm wrong, can you explain me why? - - -----BEGIN PGP SIGNATURE----- Version: 2.6.2i Comment: PGP Signed with PineSign 2.0 iQCVAwUBMEcCGnKYpFAkbPv9AQF0dAQAli7pctlXzX6IsqX/o5kSgfYuO1aW1Es4 PL6CEJfF6Q96bn80+cprZbUwVC588mgzU8A5McG7CJKGQLV2yDJGdFXHrf+XLmof iiueJ/jvG0Ir/HRkvChAW/llx9UlkbF3Ijxk24FGlYk7WnW73kSEl0N8wCKzwqK8 EcSVY3tFkXU= =E/kA - - -----END PGP SIGNATURE----- Greetings Felix A. Koop (Felix.Koop(a)ensps.u-strasbg.fr) <A HREF="http://sneezy.u-strasbg.fr/~felix/">Felix A. Koop</A> Disclaimer: Any suggestions, comments, or ideas promulgated are absolutely and positively disclaimed to be claimed for the purposes herein, whereas, and therefore. Any resemblance to intelligent ideas are purely coincidental. - ------- End of Forwarded Message

1 0

Win 95 banned in India due to timezone map
by Monty Solomon Aug. 30, 1995

Aug. 30, 1995

FYA Begin forwarded message: Date: Tue, 29 Aug 1995 12:35:00 -0400 From: Rod.Holland(a)East.Sun.COM (Rod Holland - Sun BOS Software) Subject: fenestration X-Sun-Charset: US-ASCII > > Subject: Really interesting Win 95 banned in India > Content-Length: 1246 > X-Lines: 30 > > Wanted to update you guys on this. Kind of hard to > believe, but this is a real issue. > > The timezone map doesn't display the correct borders for > north region of Kashmir according to the Indian > government. > > This is an area they are at war with Pakistan over. The > map in Win95 is the map used by the UN, so we thought it > correct. The Indian Goverment has banned all sales of > Windows 95 in India and have also threatened to ban all > sales of all MS products in India. (Note that the > difference were talking about is a dozen pixels on the > map.) I've heard this is front page news there too. > > Looks like the solution of disabling the map is ok for > them, so that's what we'll do and re-release a special > English version for India. We'd change the bitmap to > please India, but who knows how long it'd stay correct. In > the future, we'll have a more politically correct timezone > picker that avoids this sort of thing so we don't have to > ship a special Indian version. > > If you have any special advise on how else we might deal > with this, please let me know.I don't have a schedule yet > on when we'll be able to re-release disks and CDs for > India. > > ---End of forwarded mail from Eric Johnson DAD ProjDev, Microsoft. > > Hi Stan, > > Apparently the Microsoft story is true ... > > Tom > > --------------------------------------------------------------------- > > Subject: Re: Windows 95 > Content-Length: 570 > X-Lines: 18 > > Unbelievable, but true. > Looks like the rectification is in the works already(according to the > Economic Times of India) > > Microsoft removes software feature with wrong Indian map (Eco. Times) > > Microsoft will omit the time zone feature in its Windows 95 software > package for the Indian market. This follows the Indian Customs > department's decision not to allow the import of software that has > maps showing Jammu and Kashmir in Pakistan and China. Microsoft has > also decided to rectify erroneous maps of India in several of its > other products. >

1 0

Re: Australia/Sydney patch for March 1996
by ado Aug. 29, 1995

Aug. 29, 1995

The file tzdata95f.tar.gz in the anonymous ftp directory on elsie.nci.nih.gov (165.112.73.1) incorporates Paul Eggert's change to the australasia file. --ado

1 0

Australia/Sydney patch for March 1996
by Paul Eggert Aug. 14, 1995

Aug. 14, 1995

In May, Bradley White wrote that next year New South Wales will conform to the other DST-using states of Australia for the autumn change, but I see that this isn't reflected in the latest released tzdata. I assume the change is still on (though with Australian politics one never knows :-). Here is a patch. According to the tables, the last holdout is Tasmania's spring change: it switches the first Sunday in October instead of the last Sunday like the others. =================================================================== RCS file: RCS/australasia,v retrieving revision 1994.8 retrieving revision 1994.8.1.1 diff -c -r1994.8 -r1994.8.1.1 *** australasia 1994/12/10 16:26:05 1994.8 --- australasia 1995/08/14 20:28:15 1994.8.1.1 *************** *** 111,117 **** Rule AN 1986 1989 - Mar Sun>=15 3:00 0 - Rule AN 1986 only - Oct 19 2:00 1:00 - Rule AN 1987 max - Oct lastSun 2:00 1:00 - ! Rule AN 1990 max - Mar Sun>=1 3:00 0 - # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Australia/Sydney 10:04:52 - LMT 1895 Feb 10:00 - EST 1917 Jan 1 0:01 --- 111,118 ---- Rule AN 1986 1989 - Mar Sun>=15 3:00 0 - Rule AN 1986 only - Oct 19 2:00 1:00 - Rule AN 1987 max - Oct lastSun 2:00 1:00 - ! Rule AN 1990 1995 - Mar Sun>=1 3:00 0 - ! Rule AN 1996 max - Mar lastSun 3:00 0 - # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Australia/Sydney 10:04:52 - LMT 1895 Feb 10:00 - EST 1917 Jan 1 0:01 *************** *** 700,705 **** --- 701,710 ---- # From Arthur David Olson (March 8, 1992): # Sources differ on whether DST ended March 6 or March 20 in 1988; # March 20 (the "confirmed" date) is in the chosen rules. + + # From Bradley White (1995-05-20): + # Prem Bob Carr announced NSW will fall into line with other E states + # and SA and continue daylight savings to the last Sun in Mar. # Yancowinna

1 0

OpenVMS to support tz
by Paul Eggert Aug. 5, 1995

Aug. 5, 1995

In <news:3vipkg$u9n@nntpd.lkg.dec.com>, Ken Cowan of Digital Equipment Corporation <cowan(a)rtl.enet.dec.com> writes ``The DEC C library interface to timezones should be available in the first field test of OpenVMS V7.0''. DEC will supply zic, and users should be able to use the latest master tz tables, or tweak their own.

2 1

tzdata95e.tar.gz
by ado July 8, 1995

July 8, 1995

tzdat95e.tar.gz is now available for anonymous ftp from elsie.nci.nih.gov (165.112.73.1) in the ~ftp/pub directory. The only change is a leap second at the end of this year. Thanks to Bradley White for forwarding news on the leap second. --ado

1 0

timezone - data
by halanek＠ibm.net July 4, 1995

July 4, 1995

Hi, 1) I discovered your ftp server, where you have an excellent list of timezones. I am making an astronomy program, which I want to sell (when it's complete), and I want to use some of your data. Because I can't find any copyright, I want to ask you, if you allow me to do this, and if yes, if I must write your name,address,... into the copyright of my program ? 2) Do you know where I can get a list of as many cities as possible, with there name (german and or english), there geographical position and timezone (for the timezones, I would like to use your list (see above)). 3) Do you know where I can get an exact digital worldmap ? The map should not be saved as a picture, but the continets should be saved as points => I want to zoom to the map, and I want to show it as a globe, ... Thanks a lot. Bye/2 FROM Rainer WITH OS/2 3.00

1 0

tzdata95d.tar.gz
by ado July 3, 1995

July 3, 1995

Paul Eggert's changes have been incorporated into tzdata95d.tar.gz, available by anonymous ftp from elsie.nci.nih.gov (165.112.73.1) in the ~ftp/pub directory. I should obviously avoid doing time zone work on a holiday weekend :-) --ado

1 0

tzdata95c.tar.gz
by ado July 3, 1995

July 3, 1995

The file tzdata95c.tar.gz is now available for anonymous ftp from elsie.nci.nih.gov (165.112.73.1) in the ~ftp/pub directory. It incorporates changes to "asia", "backward", "europe", and "southamerica" (read: northamericacentrics need not apply). --ado

2 1

Argentinian time zones
by Hernan G. Otero June 26, 1995

June 26, 1995

Hi! > # This data is by no means authoritative; if you think you know better, > # go ahead and edit the file (and please send any changes to > # tz(a)elsie.nci.nih.gov for general use in the future). I am sending modifications to the Argentinian time zone table... please check comments below to find out the source used. We did three modifications: * The table was completed for the missing years 1989-1993. * The name of the time zone was changed from America/Buenos_Aires to Argentina/Buenos_Aires * The (short) names where changed from E%sT to AR%s (AR was chosen because they are the ISO letters that represent Argentina). --- 8< ----------------------------------------------------------- ############################################################################### # Argentina # From Bob Devine (January 28, 1988): # Argentina: first Sunday in October to first Sunday in April since 1976. # Double Summer time from 1969 to 1974. Switches at midnight. # From U. S. Naval Observatory (January 19, 19889): # ARGENTINA 3 H BEHIND UTC # Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S Rule Arg 1930 only - Dec 1 0:00 1:00 D Rule Arg 1931 only - Apr 1 0:00 0 S Rule Arg 1931 only - Oct 15 0:00 1:00 D Rule Arg 1932 1940 - Mar 1 0:00 0 S Rule Arg 1932 1939 - Nov 1 0:00 1:00 D Rule Arg 1940 only - Jul 1 0:00 1:00 D Rule Arg 1941 only - Jun 15 0:00 0 S Rule Arg 1941 only - Oct 15 0:00 1:00 D Rule Arg 1943 only - Aug 1 0:00 0 S Rule Arg 1943 only - Oct 15 0:00 1:00 D Rule Arg 1946 only - Mar 1 0:00 0 S Rule Arg 1946 only - Oct 1 0:00 1:00 D Rule Arg 1963 only - Oct 1 0:00 0 S Rule Arg 1963 only - Dec 15 0:00 1:00 D Rule Arg 1964 1966 - Mar 1 0:00 0 S Rule Arg 1964 1966 - Oct 15 0:00 1:00 D Rule Arg 1967 only - Apr 1 0:00 0 S Rule Arg 1967 1968 - Oct Sun<=7 0:00 1:00 D Rule Arg 1968 1969 - Apr Sun<=7 0:00 0 S Rule Arg 1974 only - Jan 23 0:00 1:00 D Rule Arg 1974 only - May 1 0:00 0 S Rule Arg 1974 1976 - Oct Sun<=7 0:00 1:00 D Rule Arg 1975 1977 - Apr Sun<=7 0:00 0 S Rule Arg 1985 only - Nov 2 0:00 1:00 D Rule Arg 1986 only - Mar 14 0:00 0 S Rule Arg 1986 1987 - Oct 25 0:00 1:00 D Rule Arg 1987 only - Feb 13 0:00 0 S Rule Arg 1988 only - Feb 7 0:00 0 S Rule Arg 1988 only - Dec 1 0:00 1:00 D # # These corrections were contributed by InterSoft Argentina S.A., # obtaining the data from the: # Talleres de Hidrografia Naval Argentina # (Argentinian Naval Hidrography Institute) # Rule Arg 1989 only - Mar 5 0:00 0 S Rule Arg 1989 only - Oct 15 0:00 1:00 D Rule Arg 1990 only - Mar 4 0:00 0 S Rule Arg 1990 only - Oct 21 0:00 1:00 D Rule Arg 1991 only - Mar 3 0:00 0 S Rule Arg 1991 only - Oct 20 0:00 1:00 D Rule Arg 1992 only - Mar 1 0:00 0 S Rule Arg 1992 only - Oct 18 0:00 1:00 D Rule Arg 1993 only - Mar 7 0:00 0 S # # From this moment on, the law that mandated the daylight saving # time corrections was derogated and no more modifications # to the time zones (for daylight saving) are now made. # # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Argentina/Buenos_Aires -3:53:48 - LMT 1894 Nov -4:17 - CMT 1920 May # Cordoba Mean Time -4:00 - AST 1930 Dec -4:00 Arg A%sT 1969 Oct 5 -3:00 Arg AR%s --- 8< ----------------------------------------------------------- -- ----------------------------------------------------------------- Hernan G. Otero InterSoft Argentina S.A. Senior Software Engineer Av. Cordoba 883 Piso 13 hernan(a)isoft.com.ar (1054) Buenos Aires - Argentina http://www.isoft.com.ar/~hernan Voice: +54-1-318-8900 ,,, Fax: +54-1-318-8999 0(o o)0 -------------------------------oOO--(_)--OOo---------------------

1 0

tz{code,data}
by Rickard Faith June 4, 1995

June 4, 1995

I've been distributing your tz* packages in a utility package for the Linux community (util-linux). Although I distribute the files freely (some are under the GNU GPL, so they must be distributed with sources), sometimes the package is included on a commercial CDROM (I don't have any control over this). I was going over the copyright notices of the packages in util-linux, and I could not find any notices in the tz* packages which says anything about redistribution restrictions: some files are under a freely distributable UCB license, but others are not. I wanted to check with you to make sure that redistribution (even commercial redistribution) is ok. Thanks for your help in this matter. Let me know if I can provide more information about anything or if I can clarify any of my questions.

1 0

australasia:Rule AN
by Bradley White May 21, 1995

May 21, 1995

I don't know if this is official or proposed, but ... > Apr 29 Sydney. > Prem Bob Carr announced NSW will fall into line with other E states > and SA and continue daylight savings to the last Sun in Mar.

3 3

Re: timezone magic number
by ado May 5, 1995

May 5, 1995

> > I have one comment regarding the timezone file format. > > > > Why not use a few of 20 remaining bytes in tzh_reserved field for a > > magic number that would be unique to timezone data files? 4 bytes at > > the beginning of the file would be great, but if you're concerned > > about space then 2 bytes would do the job if selected carefully. > > I would like to add that we need it to be PERSISTENT: even if timezone > would not exist anymore its magic number should be preserved. It > makes possible to use them for database storage. I'm unclear here. One possibility would be to drop a string such as try man tzfile (or something shorter) into each timezone file at a fixed place, so that the UNIX "file" command (in conjunction with an appropriate "magic" entry) could identify it. (A string such as that above might have the benefit of making the file self-documenting.) But the suggestion above seems to involve generating particular tags for particular files. Can anyone make things clearer? --ado

1 0

Floating-point time
by Monty Solomon May 5, 1995

May 5, 1995

FYI. Excerpts from RISKS DIGEST 17.08, 17.09, 17.10, 17.11 ----------------------------- Date: Sun, 23 Apr 1995 22:02:42 +0059 (EDT) From: Robert J Horn <rjh(a)world.std.com> Subject: Floating-Point Time The opponents of floating-point representation for time have done an insufficient analysis. About twenty years ago I was part of a research group doing extensive time series analysis of weather and related data. We needed a good way to represent time. Fortunately we had a few astronomers on the team, so time was reasonably well understood. We chose "second of century", using a double precision floating point representation. Analysis showed that this would preserve millisecond accuracy for the span of interest. (Actually for all of recorded history and more.) Since we usually were satisfied with one minute accuracy this seemed sufficient. There was a brief debate about using a better time base, but 12:00:01 AM GMT, 1 January, 1901 was easy to explain to everyone. There are a few applications that need better than millisecond precision, but for most of the worlds applications double precision floating point will provide enough precision for the next few millenia. (A simple test for those who are unsure about their needs. Do you compensate for the variations in the rate of the Earth's rotation? If not, you probably don't need millisecond accuracy.) This notation had some interesting side effects. At the time, floating point turned out to be somewhat faster than 64-bit integers due to a quirk of hardware. It also led to excellent compatibility with the other time series processing. Time was just another well behaved variable. This notation eliminated a lot of the mistakes made by the typical programmer who is ignorant of traditional time notations and their problems. There could have been some round-off issues, but we rarely did any arithmetic other than addition or subtraction of two times, where millisecond accuracy is maintained. It even led to a simple notation for interval time span data, e.g. "0.01 inches of rain fell between 1633 and 1647 on ...", which is how many meteorological measurements are made. The difficult problems were in translation to and from local. The most severe problem was the inherent ambiguity of local time in recent decades. There are two true times corresponding to each time in the one hour of overlap when Daylight Savings shifts back to Standard. Correctly resolving this ambiguity was always a headache. Fortunately most professional measurements have been recorded in UTC, or GMT before UTC was defined. A word of caution, double precision floating point is suitable for an internal representation of UTC, or "absolute" time. You have to do your own analysis if you are interested in timing relative to some event. Rob Horn rjh(a)world.std.com P.S. The turn of century problem has made The NY Times. It may be so widely hyped that almost all the problems are fixed by the time it comes. [Hmm! According to you, it comes at 1/1/01 rather than 1/1/00. I wonder who agrees with that! PGN] ----------------------------- ----------------------------- Date: Tue, 25 Apr 1995 10:54:01 -0400 From: rjh(a)world.std.com (Robert J Horn) Subject: Re: Floating-Point Time In comp.risks you write: > [Hmm! According to you, it comes at 1/1/01 rather than 1/1/00. > I wonder who agrees with that! PGN] Who am I to argue with astronomers? They work in mathematics and FORTRAN, so counting starts with one. Therefore the first day of the first century is 1/1/00000001. C was invented later, and never really accepted by astronomers. We just moved things closer by starting at 1901. Rob Horn rjh(a)world.std.com [The customary convention was also commented on by jcs(a)zoo.bt.co.uk (John Sager), pbm1001(a)thor.cam.ac.uk (Paul Menage), harper(a)kauri.vuw.ac.nz (John Harper), ag325(a)detroit.freenet.org (William M. Bickham), "david (d.p.) woodman" <woodman(a)bnr.ca>, rjwells(a)undergrad.math.uwaterloo.ca (justin wells), and Greg Lindahl <gl8f(a)fermi.clas.virginia.edu>, an astronomer who noted "Astronomers get to go to 2 sets of turn-of-the-century parties... you nay-sayers only get to go to one." I'll be at the former, when most of the computer-related risks are likely to begin. PGN] ----------------------------- Date: Tue, 25 Apr 95 21:02:17 PDT From: "Peter G. Neumann" <neumann(a)csl.sri.com> Subject: Re: Floating-Point Time Oh, yes, I certainly agree that the astronomers and ephemeris/almanac folks like 1/1/01 as the century start. However, because there was no year ZERO, that does not scale backwards. The first century BC clearly began on 1/1/-100, and the first millennium BC on 1/1/-1000. The only SANE way to handle this is to provide a 99-year first century; just as we have leap-years and leap-seconds, we could have a backwards leap-century. Indeed, it is just as well there were no computers in virtual-year 0000, or the religious wars would have been proven recursively unsolvable. PGN ----------------------------- Date: Tue, 25 Apr 95 13:53:39 PDT From: geoff(a)ficus.CS.UCLA.EDU (Geoff Kuenning) Subject: Re: Floating-Point Time (Robert J Horn, RISKS-17.08) > We chose "second of century", using a double precision floating point > representation. Analysis showed that this would preserve millisecond > accuracy for the span of interest. Sigh. There is more to time calculations than just understanding time, and there is more to numerical analysis than just the range of the mantissa. I'll try this once more, keeping it brief. This project wouldn't have used IEEE format 20 years ago, but let's proceed with the analysis under modern assumptions. The IEEE double-precision floating-point representation provides 53 bits of significance in the mantissa. Using the approximation that 2^10 = 10^3, this can be see to allow a range of about 8x10^15 in the mantissa, before bits start getting dropped. Since there are about 3x10^7 seconds in a year, or about 10^8 every 3 years, one can represent about 8x16x3 = 384 years to millisecond precision without violating that range, right? Wrong. This is true only if you represent time as integer milliseconds. Since the representation used seconds, the milliseconds were represented fractionally. There are only four millisecond values that can be represented accurately in a binary floating-point system: 0, 250, 500, and 750. ANYTHING ELSE is an approximation. This is especially true if your calculations involve any addition and subtraction of times. > Since we usually were satisfied with one minute accuracy this > seemed sufficient. It sounds like this project involved some sort of modeling of the physical world. In such cases, the loss of a few bits of accuracy may not matter, although I still think programmers ought to understand numerical analysis better than most do. But remember that the original discussion of time representation arose in the context of measuring time in computer and financial applications, where these bits can matter. > There are a few applications that need better than millisecond precision, > but for most of the worlds applications double precision floating point > will provide enough precision for the next few millennia. (A simple test > for those who are unsure about their needs. Do you compensate for the > variations in the rate of the Earth's rotation? If not, you probably don't > need millisecond accuracy.) This is both a broad and a biased statement. Here's a simple response to the simple test: I'm not dealing with astronomical phenomena, and so I couldn't care less about variations in the Earth's rotation. What I care about is whether I can query the system time, do something, query the time again, and subtract to get an accurate elapsed time. I care about not only milliseconds, but microseconds (and in a few years, nanoseconds). Now it's true that I generally deal with time in a small range. Unfortunately, the system time is represented in terms of a relatively long base interval. So I have to be able to take two times measured in years, subtract them, and get microsecond precision in my results. If the times are represented as integer microseconds, I can be sure that everything will "balance," as the accountants say. If a sloppy floating-point representation insists on giving it to me as integer and fractional seconds, things won't add up, and my papers may not get published. > There could have been some round-off issues, but we > rarely did any arithmetic other than addition or subtraction of two > times, where millisecond accuracy is maintained. But addition and subtraction are precisely the places where numerical errors are introduced! Millisecond accuracy was *not* maintained in this situation. Since you only cared about minutes, this hardly mattered, but it's still an incorrect statement. > A word of caution, double precision floating point is suitable for an > internal representation of UTC, or "absolute" time. It's only appropriate when you're doing modeling of the physical world, where you don't care about losing a few bits of accuracy in the fraction because your measurements aren't that good anyway. Double precision is a *terrible* way to model elapsed time inside a computer system. It has nothing to do with the nature of time, and everything to do with numerical analysis. Geoff Kuenning g.kuenning(a)ieee.org geoff(a)ITcorp.com http://www.cs.ucla.edu/ficus-members/geoff/ ----------------------------- ----------------------------- Date: Sat, 29 Apr 1995 19:49:08 GMT From: dcline(a)netcom.com (David Cline) Subject: Re: Floating-Point Time (Kuenning, RISKS-17.09) > ... Since there are about 3x10^7seconds in a year, or about 10^8 every > 3 years, one can represent about 8x16x3 = 384 years to millisecond > precision without violating that range, right? Wrong. This confuses milliseconds and microseconds; You can represent 285 years to *microsecond* accuracy in 53 bits. If you only care about millisecond accuracy, you can represent about 285,000 years. There are also ways of using the sign bit to double the effective range. Dave Cline Spring Valley Software dcline(a)netcom.com [Your moderator is dismayed that this is dragging on so long! PGN] ----------------------------- Date: Fri, 28 Apr 95 11:12:30 EDT From: hopkins(a)VFL.Paramax.COM Subject: Re: Floating-Point Time On the year-zero and religious wars: PGN suggests [RISKS-17.09] that first-century dates (which were, after all, not invented until well after the fact) would have created religious wars had there been computers to suggest that there should be a year zero. Any self-respecting computer, however, would have balked at attempts to divide the factions by zero. Bill Hopkins hopkins(a)VFL.Paramax.Com Unisys Corporation (Soon to be Loral, they say) 610-648-2854 or 363-7464 Valley Forge Eng'g Ctr, POB 517, Paoli PA 19301 ----------------------------- ----------------------------- Date: Tue, 2 May 1995 19:49:51 -0400 From: "Andrew D. Fernandes" <adfernan(a)cnd.mcgill.ca> Subject: Re: Floating-point time It seems to me that the real issue as to whether or not floating point representations are appropriate for timestamps is not "can I get microsecond resolution?" but "is it going to work unconditionally?" Floating point numbers bedevil numerical programmers for several reasons; we need to know the radix, number of mantissa and exponent digits, behaviour of rounding, etc. to guarantee "good" results---there is a reason that textbooks have been written about numerical analysis. Modern computers generally stick to the IEEE-754 standard for floating point arithmetic, and thus code is usually fairly interchangeable, but anyone who has ever done intensive number crunching on old IBMs, VAXen, or CRAYs can tell horror stories about interchanging floating point programs between systems. Even going between a SPARCstation and my '486 at home, I can see a few digits change. "Unconditional portability" is ludicrous under these conditions. A 64 bit integer can represent 1.84e19 values, which implies about 2 nanosecond resolution of a thousand year interval. Integer math is very portable across computer systems, or at least is very easy to describe fully, and 1 + 1 always evaluates as 2, and not 2.0001. -Andrew D. Fernandes (adfernan(a)cnd.mcgill.ca) ----------------------------- Date: Thu, 4 May 95 11:42 PDT From: ludemann(a)netcom.com (Peter Ludemann) Subject: Re: Floating-point time In my original note I mentioned something really simple: handling clock ticks. Not handling time, which is a topic for PhD theses (Stanford is about to grant a PhD for a thesis on temporal representation in databases). My problem was simple: I needed to do 48-bit arithmetic on a machine which had 32-bit integers. I had been given an assembler routine which did the 48-bit arithmetic wrong; and I was fortunate to test it in December when its accumulated error was about 3 minutes in converting clock ticks to a date (in January, it had almost no error). So, what to do? Obviously, writing 48-bit arithmetic in assembler was error-prone and tedious. Doing it in PL/I (this was before C was generally available) wasn't much more fun nor less error-prone (anyone who has read the PL/I manual's description of handling overflow and precisions will agree that it's not a pretty sight; and I later learned that the PL/I implementors had got it wrong anyway). And I didn't have access to LISP's rational-number packages. In a flash of inspiration, I realized that double-precision floating point, with 53-bit mantissas, would handle my 48-bit integers with no round-off errors. Converting between 48-bit integers and floating point is simple (it's a "well-known assembler idiom" that takes about 3 instructions). And everything would then proceed perfectly (even a Pentium can get floating-point addition and subtraction right). At one stroke I had removed a few pages of difficult assembler and replaced them by a few lines of PL/I. Less code, less chance of error. [Inspired by this, I wrote a disk accounting package that eschewed packed decimal arithmetic and instead used floating point (of course, I calculated everything in cents because 1.01 doesn't represent exactly but 101 does. Another success.)] One of my jobs as a programmer is to reduce the risk of error in my system. I can only test so much. Testing 48-bit integer arithmetic is not fun, at least for me; and proving the code right is difficult (Knuth allegedly wrote "Beware of bugs in the above code; I have only proved it correct, not tried it."). PL/I's implementation of 15-digit packed decimal was buggy. And I wasn't being paid by the number of lines of code (au contraire: if I finished early, I took time off). The lessons for RISKS: before you try to do something complicated, maybe there's something simple and reliable lying around that you can pervert slightly into the form you want. [My original RISKS lesson was to be sure to test time/date conversion routines at many times of the year; if they work in January, they might not work in December.] - Peter Ludemann ----------------------------- Date: Tue, 2 May 1995 18:26:06 GMT0BST From: Phil Brady <P.R.Brady(a)swansea.ac.uk> Subject: Re: Floating-point time [Your moderator is dismayed that this is dragging on so long! PGN] Agreed - the debate has been interesting, but hasn't it run it's course yet? The fact is that every programmer needs to decide whether to use floating, decimal, integer, string, etc and the appropriate precision for every variable in every program depending on the application. If they don't know the appropriate form for time for the problem in hand, then they aren't programmers! Phil Brady, University of Wales, Swansea 01792 295160 ----------------------------- ----------------------------- Date: 24 March 1995 (LAST-MODIFIED) From: RISKS-request(a)csl.sri.com Subject: Info on RISKS (comp.risks), contributions, subscriptions, FTP, etc. The RISKS Forum is a moderated digest. Its USENET equivalent is comp.risks. Undigestifiers are available throughout the Internet, but not from RISKS. SUBSCRIPTIONS: PLEASE read RISKS as a newsgroup (comp.risks or equivalent) on your system, if possible and convenient for you. BITNET folks may use a LISTSERV (e.g., LISTSERV@UGA): SUBSCRIBE RISKS or UNSUBSCRIBE RISKS. U.S. users on .mil or .gov domains should contact <risks-request(a)pica.army.mil> (Dennis Rears <drears(a)pica.army.mil>). UK subscribers please contact <Lindsay.Marshall(a)newcastle.ac.uk>. Local redistribution services are provided at many other sites as well. Check FIRST with your local system or netnews wizards. If that does not work, THEN please send requests to <risks-request(a)csl.sri.com> (which is not yet automated). SUBJECT: SUBSCRIBE or UNSUBSCRIBE; text line (UN)SUBscribe RISKS [address to which RISKS is sent] CONTRIBUTIONS: to risks(a)csl.sri.com, with appropriate, substantive Subject: line, otherwise they may be ignored. Must be relevant, sound, in good taste, objective, cogent, coherent, concise, and nonrepetitious. Diversity is welcome, but not personal attacks. PLEASE DO NOT INCLUDE ENTIRE PREVIOUS MESSAGES in responses to them. Contributions will not be ACKed; the load is too great. **PLEASE** include your name & legitimate Internet FROM: address, especially from .UUCP and .BITNET folks. Anonymized mail is not accepted. ALL CONTRIBUTIONS CONSIDERED AS PERSONAL COMMENTS; USUAL DISCLAIMERS APPLY. Relevant contributions may appear in the RISKS section of regular issues of ACM SIGSOFT's SOFTWARE ENGINEERING NOTES, unless you state otherwise. All other reuses of RISKS material should respect stated copyright notices, and should cite the sources explicitly; as a courtesy, publications using RISKS material should obtain permission from the contributors. RISKS can also be read on the web at URL http://catless.ncl.ac.uk/Risks Individual issues can be accessed using a URL of the form http://catless.ncl.ac.uk/Risks/VL.IS.html (Please report any format errors to Lindsay.Marshall(a)newcastle.ac.uk) RISKS ARCHIVES: "ftp unix.sri.com<CR>login anonymous<CR>[YourNetAddress]<CR> cd risks<CR> or cwd risks<CR>, depending on your particular FTP. Issue J of volume 17 is in that directory: "get risks-17.J<CR>". For issues of earlier volumes, "get I/risks-I.J<CR>" (where I=1 to 16, J always TWO digits) for Vol I Issue j. Vol I summaries in J=00, in both main directory and I subdirectory; "bye<CR>" I and J are dummy variables here. REMEMBER, Unix is case sensitive; file names are lower-case only. <CR>=CarriageReturn; UNIX.SRI.COM = [128.18.30.66]; FTPs may differ; Unix prompts for username and password. Also ftp bitftp(a)pucc.Princeton.EDU. WAIS repository exists at server.wais.com [192.216.46.98], with DB=RISK (E-mail info(a)wais.com for info) or visit the web wais URL http://www.wais.com/ . Management Analytics Searcher Services (1st item) under http://all.net:8080/ also contains RISKS search services, courtesy of Fred Cohen. Use wisely. -----------------------------

1 0

Re: timezone magic number
by Valery Aronov April 27, 1995

April 27, 1995

Daniel Quillan wrote: > I have one comment regarding the timezone file format. > > Why not use a few of 20 remaining bytes in tzh_reserved field for a > magic number that would be unique to timezone data files? 4 bytes at > the beginning of the file would be great, but if you're concerned > about space then 2 bytes would do the job if selected carefully. I would like to add that we need it to be PERSISTENT: even if timezone would not exist anymore its magic number should be preserved. It makes possible to use them for database storage. Valery -- Valery Aronov ,-_;\ Telstra Australia ph (61-2) 395 33 59 / \ Telecom Plaza fax (61-2) 395 32 25 \_,-._* <-- 320 Pitt Street valery(a)ind.tansu.com.au v Sydney South NSW 2000 Australia

1 0

timezone magic number
by quinlan＠yggdrasil.com April 27, 1995

April 27, 1995

I have one comment regarding the timezone file format. Why not use a few of 20 remaining bytes in tzh_reserved field for a magic number that would be unique to timezone data files? 4 bytes at the beginning of the file would be great, but if you're concerned about space then 2 bytes would do the job if selected carefully. Thanks. Dan -- Daniel Quinlan Member of the League for Programming Freedom quinlan(a)yggdrasil.com

1 0

Japan to use DST. Indiana changed? Changes for Israel, W-SU, etc.
by eggert＠twinsun.com April 4, 1995

April 4, 1995

The big news is that Japan will most likely start using daylight saving next year. It isn't confirmed yet, but it seems probable. See below for details. A lot of software in Japan will have to be rewritten, since they haven't used daylight saving since the dawn of the computer era. This looks like a growth opportunity for tz.... Also, while I was in Japan last month I read a newspaper article talking about daylight saving; it mentioned something about the Indiana legislature changing the time zone rules again in the last few weeks. No details. Anybody else heard anything about this rumor? Other than that, here are proposed changes to the current tz rules: Proposed changes that affect behavior: * Ephraim Silverberg's correction for this autumn's transition in Israel. * Make `W-SU' a link to Europe/Moscow instead of something with ???? in it. Proposed comment and/or cleanup changes: * Remove unused GB-Eire rules starting in 1996, and comment on the relationship between GB-Eire and EC. * Comments about the proposed daylight saving time in Japan in 1996. * Comments about DST in Japan between 1948 and 1951. * Comment Serbia/Yugoslavia's name more accurately. PS. I'll be on vacation starting Thursday so won't answer mail for a while. =================================================================== RCS file: RCS/asia,v retrieving revision 1994.8 retrieving revision 1994.8.1.1 diff -c -r1994.8 -r1994.8.1.1 *** asia 1994/12/10 16:26:05 1994.8 --- asia 1995/04/04 18:04:39 1994.8.1.1 *************** *** 48,54 **** # 8:00 PST PDT Philippines* # 8:00 SGT Singapore # 8:00 UST UDT Ulan Bator* ! # 9:00 JST Japan # 9:00 KST KDT Korea # 9:00 MLT Moluccas* # 9:30 CST Australian Central Standard Time --- 48,54 ---- # 8:00 PST PDT Philippines* # 8:00 SGT Singapore # 8:00 UST UDT Ulan Bator* ! # 9:00 JST JDT Japan # 9:00 KST KDT Korea # 9:00 MLT Moluccas* # 9:30 CST Australian Central Standard Time *************** *** 461,467 **** Rule Zion 1994 only - Apr 1 0:00 1:00 D Rule Zion 1994 only - Aug 28 0:00 0:00 S Rule Zion 1995 only - Mar 31 0:00 1:00 D ! Rule Zion 1995 only - Aug 27 0:00 0:00 S # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Asia/Tel_Aviv 2:19:04 - LMT 1880 --- 461,474 ---- Rule Zion 1994 only - Apr 1 0:00 1:00 D Rule Zion 1994 only - Aug 28 0:00 0:00 S Rule Zion 1995 only - Mar 31 0:00 1:00 D ! # On 1995-03-13 Ephraim Silverberg corrected the next line from Aug 27 to Sep 3. ! Rule Zion 1995 only - Sep 3 0:00 0:00 S ! ! # From Ephraim Silverberg (1995-03-13): ! # The Spokeswoman's office confirmed that there are not yet any dates ! # for the timezone conversion for the years 1996 and beyond yet ! # and this is one of the things the newly-appointed Minister of Interior ! # will have to decide sometime this year. # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Asia/Tel_Aviv 2:19:04 - LMT 1880 *************** *** 475,494 **** # `9:00' and `JST' is from Guy Harris. ! # From Paul Eggert <eggert(a)twinsun.com> (November 18, 1993): ! # Shanks says that the far southern Ryukyu Is (Nansei-Shoto) are 8:00, ! # but we don't have a good location name for them; ! # we don't even know the name of the principal town. ! # There is no information for Marcus. ! # Other Japanese possessions are probably like Asia/Tokyo. # From Shanks (1991): # Japan switched from the Japanese calendar on 1893 Jan 1. # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Asia/Tokyo 9:19:04 - LMT 1896 9:00 - JST #Zone Asia/South_Ryukyu 8:14:44 - LMT 1896 # Amitori # 8:00 - CST # Jordan # From Paul Eggert <eggert(a)twinsun.com> (November 18, 1993): --- 482,525 ---- # `9:00' and `JST' is from Guy Harris. ! # From Paul Eggert <eggert(a)twinsun.com> (1995-03-06): ! # Today's _Asahi Evening News_ (page 4) reports that Japan had ! # daylight saving between 1948 and 1951, but ``the system was discontinued ! # because the public believed it would lead to longer working hours.'' ! # Shanks writes that daylight saving in Japan during those years was as follows: ! # Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S ! #Rule Japan 1948 only - May Sun>=1 2:00 1:00 D ! #Rule Japan 1948 1951 - Sep Sat>=8 2:00 0 S ! #Rule Japan 1949 only - Apr Sun>=1 2:00 1:00 D ! #Rule Japan 1950 1951 - May Sun>=1 2:00 1:00 D ! # but the only locations using it were US military bases. ! # We go with Shanks and omit daylight saving in those years for Asia/Tokyo. ! # ! # The same news article also reports that Japan is likely to go on DST in 1996 ! # as follows, where I've guessed the AT and LETTER/S columns: ! # Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S ! #Rule Japan 1996 max - Apr Sun>=1 2:00 1:00 D ! #Rule Japan 1996 max - Oct lastSun 2:00 0 S # From Shanks (1991): # Japan switched from the Japanese calendar on 1893 Jan 1. # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Asia/Tokyo 9:19:04 - LMT 1896 9:00 - JST + # If Japan adopts DST in 1996, replace the above line with the following lines: + # 9:00 - JST 1996 + # 9:00 Japan J%sT + # and uncomment the two `Rule' lines commented out above. + + # From Paul Eggert <eggert(a)twinsun.com> (November 18, 1993): + # Shanks says that the far southern Ryukyu Is (Nansei-Shoto) are 8:00, + # but we don't have a good location name for them; + # we don't even know the name of the principal town. + # Zone NAME GMTOFF RULES FORMAT [UNTIL] #Zone Asia/South_Ryukyu 8:14:44 - LMT 1896 # Amitori # 8:00 - CST + # There is no information for Marcus. + # Other Japanese possessions are probably like Asia/Tokyo. # Jordan # From Paul Eggert <eggert(a)twinsun.com> (November 18, 1993): =================================================================== RCS file: RCS/backward,v retrieving revision 1995.2 retrieving revision 1995.2.1.1 diff -c -r1995.2 -r1995.2.1.1 *** backward 1995/02/21 15:43:22 1995.2 --- backward 1995/03/23 01:42:23 1995.2.1.1 *************** *** 73,76 **** --- 73,77 ---- Link Pacific/Samoa US/Samoa Link Etc/UTC UTC Link Etc/Universal Universal + Link Europe/Moscow W-SU Link Etc/Zulu Zulu =================================================================== RCS file: RCS/europe,v retrieving revision 1995.2 retrieving revision 1995.2.1.1 diff -c -r1995.2 -r1995.2.1.1 *** europe 1995/02/21 15:43:19 1995.2 --- europe 1995/04/04 10:39:24 1995.2.1.1 *************** *** 714,724 **** Rule GB-Eire 1971 only - Oct 31 3:00 0 GMT Rule GB-Eire 1972 1980 - Oct Sun>=23 2:00s 0 GMT # 1981 on ! Rule GB-Eire 1981 max - Mar lastSun 1:00u 1:00 BST Rule GB-Eire 1981 1989 - Oct Sun>=23 1:00u 0 GMT Rule GB-Eire 1990 1995 - Oct Sun>=22 1:00u 0 GMT ! Rule GB-Eire 1996 max - Oct lastSun 1:00u 0 GMT ! # Also see EC, which (starting 1996) differs only in LETTER/S. # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Europe/London -0:01:15 - LMT 1847 Sep 22 --- 714,723 ---- Rule GB-Eire 1971 only - Oct 31 3:00 0 GMT Rule GB-Eire 1972 1980 - Oct Sun>=23 2:00s 0 GMT # 1981 on ! Rule GB-Eire 1981 1995 - Mar lastSun 1:00u 1:00 BST Rule GB-Eire 1981 1989 - Oct Sun>=23 1:00u 0 GMT Rule GB-Eire 1990 1995 - Oct Sun>=22 1:00u 0 GMT ! # See EC for rules starting in 1996. # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Europe/London -0:01:15 - LMT 1847 Sep 22 *************** *** 760,766 **** Rule EC 1979 1995 - Sep lastSun 1:00u 0 - Rule EC 1981 max - Mar lastSun 1:00u 1:00 " DST" Rule EC 1996 max - Oct lastSun 1:00u 0 - - # Also see GB-Eire, which (starting 1996) differs only in LETTER/S. # W-Eur differs from EC only in that W-Eur uses standard time. Rule W-Eur 1800 only - Jan 1 0:00 0 - --- 759,764 ---- *************** *** 824,830 **** Zone WET 0:00 EC WET%s Zone MET 1:00 M-Eur MET%s Zone EET 2:00 EC EET%s - Zone W-SU 3:00 M-Eur ???? # Tom Hoffman says that MET is also known as Central European Time --- 822,827 ---- *************** *** 1789,1795 **** 12:00 - ASK 1992 Jan 19 2:00s 13:00 Russia AS%s ! # Serbia # They switched from the Julian to the Gregorian calendar on 1918 Mar 18. # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Europe/Belgrade 1:22:00 - LMT 1884 --- 1786,1792 ---- 12:00 - ASK 1992 Jan 19 2:00s 13:00 Russia AS%s ! # Yugoslavia (what's left of it -- mostly Serbia) # They switched from the Julian to the Gregorian calendar on 1918 Mar 18. # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Europe/Belgrade 1:22:00 - LMT 1884

2 1

tzcode95c.tar.gz
by ado March 13, 1995

March 13, 1995

tzcode95c.tar.gz includes the one-line fix for sign extension problems in detzcode; it is available from elsie.nci.nih.gov by anonymous ftp. --ado

1 0

tzcode95b.tar.gz and tzdata95b.tar.gz
by ado March 11, 1995

March 11, 1995

The files tzcode95b.tar.gz and tzdata95b.tar.gz are now available by anonymous ftp from elsie.nci.nih.gov. Minor changes in both: the "code" file contains a workaround for the lack of "unistd.h" in Microsoft C++ version 7; the "data" file contains a fixed "Link" for America/Shiprock. --ado

2 1

Re: problem with detzcode on machines w/64 bit longs
by Chris Torek March 11, 1995

March 11, 1995

>Gawd - Chris, really, wouldn't just [perfectly simple portable code] >work rather more easily ... Yep, but I am not embarrassed ... as I said, it is these soma pills that are doing this to me. :-) (Seriously, this `muscle relaxant' seems to let one's mind leak out the ears. They do help out with the muscle spasm, but they make it like trying to program while drunk out of one's mind...) Chris

1 0

Re: problem with detzcode on machines w/64 bit longs
by Chris Torek March 11, 1995

March 11, 1995

>However, not all operating systems have [int32_t] ... Probably a case >for #ifdef'ing the code. Hm, good point, I was thinking of BSD systems specifically. (It is these `soma' muscle relaxant pills! I swear! :-) ) If one is to resort to #ifdef's, the original suggestion of testing the top bit and OR'ing in ~0L << 32 should work, using something like: #if LONG_MAX > 2147483647 /* long > 32 bits */ if (result & (1L << 31)) result |= ~0L << 32; #endif return (result); The (long)(int32_t) method may produce better code on an Alpha, Chris

2 1

Re: problem with detzcode on machines w/64 bit longs
by Chris Torek March 10, 1995

March 10, 1995

>How about ... (result >> 31) ? (result | (~0L << 32)) : result The minimum number of bits in a `long' is 32 (LONG_MIN and LONG_MAX must be at most -2147483647 and +2147483647 respectively; the former allows for ones-complement). The result of a shift operator is defined if and only if the shift count does not equal or exceed the number of bits in the type. Thus, on a machine with 32-bit ints, `~0L << 32' is undefined. More practically, on (say) a VAX, the shift instruction only looks at the low 5 bits of the shift count, so ~0L << 32 will be ~0L => -1. Since there is now a new ugly set of `intNN_t' typedefs in <machine/types.h>, your best bet is probably: return (time_t)(int32_t)result; Chris

2 1

Re: problem with detzcode on machines w/64 bit longs
by Tom Peterson March 10, 1995

March 10, 1995

>> yeah, that should work as desired, though i forget if "0L" is an >> ansi-C-ism, and don't have a book around to check... > >Yep, it's in there. Section 3.1.3.2 of the spec I have dated 1990. >I think this has been around for a while, probably in K&R too. Just checked & it's definitely in K&R too. Appendix A section 2.4.2. - Tom

1 0

Re: problem with detzcode on machines w/64 bit longs
by Bradley White March 10, 1995

March 10, 1995

> Chris Demetriou, who maintains the NetBSD DEC Alpha port, reported a > problem in localtime.c's detzcode() function that occurs because the > value being extracted from the string in that function is not sign > extended properly. > > He says: > The problem is that if you do what the current version does, > then you can end up with numbers like 0xffffyyyy being > returned from detzcode(), i.e. no sign extension into the > upper 32 bits of the long. Because of this, among other > things, 'date' will say that it's some time in the year > 2025... > > Using the following detzcode() implementation (which unrolls the loop > and doesn't mask the first character) works, but it suffers from a > small defect in that it assumes signed characters. > > static long > detzcode(codep) > const char * const codep; > { > register long result; > > result = (codep[0] << 24) \ > | (codep[1] & 0xff) << 16 \ > | (codep[2] & 0xff) << 8 > | (codep[3] & 0xff); > return result; > } > > casting codep[0] to signed char would fix that, but would require a > ANSI compiler. How about this simple change? *** tz95b/localtime.c Sat Mar 4 11:22:26 1995 --- localtime.c Fri Mar 10 08:51:53 1995 *************** *** 186,200 **** static long detzcode(codep) const char * const codep; { register long result; register int i; ! result = 0; for (i = 0; i < 4; ++i) result = (result << 8) | (codep[i] & 0xff); return result; } static void settzname P((void)) --- 186,200 ---- static long detzcode(codep) const char * const codep; { register long result; register int i; ! result = ~0; for (i = 0; i < 4; ++i) result = (result << 8) | (codep[i] & 0xff); return result; } static void settzname P((void))

3 4