[PROPOSED] Be more consistent about UT vs UTC
* NEWS, africa, asia, australasia, europe, localtime.c: * newctime.3, newtzset.3, northamerica, southamerica, theory.html: * tz-link.htm, tzselect.ksh, zdump.c, zic.8, zic.c: Be more consistent about using "UT" instead of "UTC" when talking about timestamps that might predate the introduction of UTC. * tzselect.ksh: Use AEST instead of the (obsolescent) abbreviation GST when talking about UT offsets. --- NEWS | 29 +++++++++++++++++------------ africa | 2 +- asia | 14 +++++++------- australasia | 4 ++-- europe | 6 +++--- localtime.c | 2 +- newctime.3 | 2 +- newtzset.3 | 20 ++++++++++---------- northamerica | 2 +- southamerica | 4 ++-- theory.html | 6 +++--- tz-link.htm | 14 +++++++------- tzselect.ksh | 5 +++-- zdump.c | 14 +++++++------- zic.8 | 4 ++-- zic.c | 2 +- 16 files changed, 68 insertions(+), 62 deletions(-) diff --git a/NEWS b/NEWS index de81afc..2254650 100644 --- a/NEWS +++ b/NEWS @@ -33,6 +33,11 @@ Unreleased, experimental changes that have snprintf and use pre-C99 compilers. (Problem reported by Jon Skeet.) + Changes to code + + Diagnostis and commentary are now more careful about + distinguishing UT from UTC. + Release 2017c - 2017-10-20 14:49:34 -0700 @@ -929,8 +934,8 @@ Release 2015f - 2015-08-10 18:06:56 -0700 (Thanks to Jon Skeet and Arthur David Olson.) Constraints on simultaneity are now documented. - The two characters '%z' in a zone format now stand for the UTC - offset, e.g., '-07' for seven hours behind UTC and '+0530' for + The two characters '%z' in a zone format now stand for the UT + offset, e.g., '-07' for seven hours behind UT and '+0530' for five hours and thirty minutes ahead. This better supports time zone abbreviations conforming to POSIX.1-2001 and later. @@ -1053,13 +1058,13 @@ Release 2015c - 2015-04-11 08:55:55 -0700 The spring 1988 transition was 1988-10-09, not 1988-10-02. The fall 1990 transition was 1990-03-11, not 1990-03-18. - Assume no UTC offset change for Pacific/Easter on 1890-01-01, + Assume no UT offset change for Pacific/Easter on 1890-01-01, and omit all transitions on Pacific/Easter from 1942 through 1946 since we have no data suggesting that they existed. One more zone has been turned into a link, as it differed from an existing zone only for older time stamps. As usual, - this change affects UTC offsets in pre-1970 time stamps only. + this change affects UT offsets in pre-1970 time stamps only. The zone's old contents have been moved to the 'backzone' file. The affected zone is America/Montreal. @@ -1089,7 +1094,7 @@ Release 2015b - 2015-03-19 23:28:11 -0700 Some more zones have been turned into links, when they differed from existing zones only for older time stamps. As usual, - these changes affect UTC offsets in pre-1970 time stamps only. + these changes affect UT offsets in pre-1970 time stamps only. Their old contents have been moved to the 'backzone' file. The affected zones are: America/Antigua, America/Cayman, Pacific/Midway, and Pacific/Saipan. @@ -1141,7 +1146,7 @@ Release 2015a - 2015-01-29 22:35:20 -0800 Some more zones have been turned into links, when they differed from existing zones only for older time stamps. As usual, - these changes affect UTC offsets in pre-1970 time stamps only. + these changes affect UT offsets in pre-1970 time stamps only. Their old contents have been moved to the 'backzone' file. The affected zones are: Asia/Aden, Asia/Bahrain, Asia/Kuwait, and Asia/Muscat. @@ -1188,7 +1193,7 @@ Release 2014j - 2014-11-10 17:37:11 -0800 Some more zones have been turned into links, when they differed from existing zones only for older time stamps. As usual, - these changes affect UTC offsets in pre-1970 time stamps only. + these changes affect UT offsets in pre-1970 time stamps only. Their old contents have been moved to the 'backzone' file. The affected zones are: Africa/Addis_Ababa, Africa/Asmara, Africa/Dar_es_Salaam, Africa/Djibouti, Africa/Kampala, @@ -1278,7 +1283,7 @@ Release 2014h - 2014-09-25 18:59:03 -0700 Some more zones have been turned into links, when they differed from existing zones only for older timestamps. As usual, - these changes affect UTC offsets in pre-1970 timestamps only. + these changes affect UT offsets in pre-1970 timestamps only. Their old contents have been moved to the 'backzone' file. The affected zones are: Africa/Blantyre, Africa/Bujumbura, Africa/Gaborone, Africa/Harare, Africa/Kigali, Africa/Lubumbashi, @@ -1363,7 +1368,7 @@ Release 2014g - 2014-08-28 12:31:23 -0700 Some more zones have been turned into links, when they differed from existing zones only for older timestamps. As usual, - these changes affect UTC offsets in pre-1970 timestamps only. + these changes affect UT offsets in pre-1970 timestamps only. Their old contents have been moved to the 'backzone' file. The affected zones are: Africa/Bangui, Africa/Brazzaville, Africa/Douala, Africa/Kinshasa, Africa/Libreville, Africa/Luanda, @@ -1513,7 +1518,7 @@ Release 2014f - 2014-08-05 17:42:36 -0700 standard and daylight saving time the abbreviations are AEST and AEDT instead of the former EST for both; similarly, ACST/ACDT, ACWST/ACWDT, and AWST/AWDT are now used instead of the former CST, CWST, and WST. - This change does not affect UTC offsets, only time zone abbreviations. + This change does not affect UT offsets, only time zone abbreviations. (Thanks to Rich Tibbett and many others.) Asia/Novokuznetsk shifts from NOVT to KRAT (remaining on UT +07) @@ -1550,8 +1555,8 @@ Release 2014f - 2014-08-05 17:42:36 -0700 Treindl sent helpful translations of two papers by Guo Qingsheng.) Some zones have been turned into links, when they differed from existing - zones only for older UTC offsets where data entries were likely invented. - These changes affect UTC offsets in pre-1970 timestamps only. This is + zones only for older UT offsets where data entries were likely invented. + These changes affect UT offsets in pre-1970 timestamps only. This is similar to the change in release 2013e, except this time for western Africa. The affected zones are: Africa/Bamako, Africa/Banjul, Africa/Conakry, Africa/Dakar, Africa/Freetown, Africa/Lome, diff --git a/africa b/africa index 3a60bc2..be44580 100644 --- a/africa +++ b/africa @@ -425,7 +425,7 @@ Link Africa/Nairobi Indian/Mayotte # # The Nautical Almanac for the Year 1970, p 264, is the source for -0:44:30. # -# In 1972 Liberia was the last country to switch from a UTC offset +# In 1972 Liberia was the last country to switch from a UT offset # that was not a multiple of 15 or 20 minutes. The 1972 change was on # 1972-01-07, according to an entry dated 1972-01-04 on p 330 of: # Presidential Papers: First year of the administration of diff --git a/asia b/asia index ac39af3..91076fd 100644 --- a/asia +++ b/asia @@ -50,7 +50,7 @@ # 9:00 KST KDT Korea when at +09 # 9:30 ACST Australian Central Standard Time # Otherwise, these tables typically use numeric abbreviations like +03 -# and +0330 for integer hour and minute UTC offsets. Although earlier +# and +0330 for integer hour and minute UT offsets. Although earlier # editions invented alphabetic time zone abbreviations for every # offset, this did not reflect common practice. # @@ -647,17 +647,17 @@ Zone Asia/Hong_Kong 7:36:42 - LMT 1904 Oct 30 # time", in which abolished the adoption of Western Standard Time in # western islands (listed above), which means the whole Japan # territory, including later occupations, adopt Japan Central Time -# (UTC+9). The adoption began on Oct 1, 1937. The original text can +# (UT+9). The adoption began on Oct 1, 1937. The original text can # be found on Wikisource: # https://ja.wikisource.org/wiki/明治二十八年勅令第百六十七號標準時ニ關スル件中改正ノ件 # -# That is, the time zone of Taipei switched to UTC+9 on Oct 1, 1937. +# That is, the time zone of Taipei switched to UT+9 on Oct 1, 1937. # From Yu-Cheng Chuang (2014-07-02): -# I've found more evidence about when the time zone was switched from UTC+9 -# back to UTC+8 after WW2. I believe it was on Sep 21, 1945. In a document +# I've found more evidence about when the time zone was switched from UT+9 +# back to UT+8 after WW2. I believe it was on Sep 21, 1945. In a document # during Japanese era [1] in which the officer told the staff to change time -# zone back to Western Standard Time (UTC+8) on Sep 21. And in another +# zone back to Western Standard Time (UT+8) on Sep 21. And in another # history page of National Cheng Kung University [2], on Sep 21 there is a # note "from today, switch back to Western Standard Time". From these two # materials, I believe that the time zone change happened on Sep 21. And @@ -1505,7 +1505,7 @@ Rule Japan 1950 1951 - May Sun>=1 2:00 1:00 D # # ...the Showa Emperor announced Ordinance No. 529 of Showa Year 12 ... which # means the whole Japan territory, including later occupations, adopt Japan -# Central Time (UTC+9). The adoption began on Oct 1, 1937. +# Central Time (UT+9). The adoption began on Oct 1, 1937. # https://ja.wikisource.org/wiki/明治二十八年勅令第百六十七號標準時ニ關スル件中改正ノ件 # Zone NAME GMTOFF RULES FORMAT [UNTIL] diff --git a/australasia b/australasia index 5f7c86d..9a968e4 100644 --- a/australasia +++ b/australasia @@ -683,8 +683,8 @@ Zone Pacific/Guadalcanal 10:39:48 - LMT 1912 Oct # Honiara # From Steffen Thorsen (2012-07-25) # ... we double checked by calling hotels and offices based in Tokelau asking # about the time there, and they all told a time that agrees with UTC+13.... -# Shanks says UTC-10 from 1901 [but] ... there is a good chance the change -# actually was to UTC-11 back then. +# Shanks says UT-10 from 1901 [but] ... there is a good chance the change +# actually was to UT-11 back then. # # From Paul Eggert (2012-07-25) # A Google Books snippet of Appendix to the Journals of the House of diff --git a/europe b/europe index cf6d7e4..7bca59b 100644 --- a/europe +++ b/europe @@ -2772,9 +2772,9 @@ Zone Asia/Omsk 4:53:30 - LMT 1919 Nov 14 # # https://regnum.ru/news/society/1957270.html # has some historical data for Altai Krai: -# before 1957: west part on UTC+6, east on UTC+7 -# after 1957: UTC+7 -# since 1995: UTC+6 +# before 1957: west part on UT+6, east on UT+7 +# after 1957: UT+7 +# since 1995: UT+6 # http://barnaul.rusplt.ru/index/pochemu_altajskij_kraj_okazalsja_v_neprivychn... # confirms that and provides more details including 1995-05-28 transition date. diff --git a/localtime.c b/localtime.c index 2e1d621..9b8266a 100644 --- a/localtime.c +++ b/localtime.c @@ -1982,7 +1982,7 @@ time2sub(struct tm *const tmp, && ((BIGGEST (INT_FAST32_MIN, LONG_MIN) + yourtm.TM_GMTOFF) <= mytm.TM_GMTOFF)))) { - /* MYTM matches YOURTM except with the wrong UTC offset. + /* MYTM matches YOURTM except with the wrong UT offset. YOURTM.TM_GMTOFF is plausible, so try it instead. It's OK if YOURTM.TM_GMTOFF contains uninitialized data, since the guess gets checked. */ diff --git a/newctime.3 b/newctime.3 index 8b89935..b376ae1 100644 --- a/newctime.3 +++ b/newctime.3 @@ -191,7 +191,7 @@ extra initial argument specifying the time zone to be used for conversion. If .B zone -is null, UTC is used; otherwise, +is null, UT is used; otherwise, .B zone should be have been allocated by .I tzalloc diff --git a/newtzset.3 b/newtzset.3 index c3e22dc..beee20a 100644 --- a/newtzset.3 +++ b/newtzset.3 @@ -61,7 +61,7 @@ If the implied call to .B tzalloc fails, .I tzset -falls back on UTC. +falls back on Universal Time (UT). .PP If .B TZ @@ -74,7 +74,7 @@ in the system time conversion information directory, is used. If .B TZ is the empty string, -Universal Time (UT) is used, with the abbreviation "UTC" +UT is used, with the abbreviation "UTC" and without leap second correction; please see .IR newctime (3) for more about UT, UTC, and leap seconds. If @@ -256,22 +256,22 @@ extensions to POSIX. .TP .B EST5 stands for US Eastern Standard -Time (EST), 5 hours behind UTC, without daylight saving. +Time (EST), 5 hours behind UT, without daylight saving. .TP .B FJT\*-12FJST,M11.1.0,M1.3.4/75 stands for Fiji Time (FJT) and Fiji Summer Time (FJST), 12 hours ahead -of UTC, springing forward on November's first Sunday at 02:00, and +of UT, springing forward on November's first Sunday at 02:00, and falling back on January's third Thursday at 75:00 (i.e., 03:00 on the first Sunday on or after January 18). .TP .B IST\*-2IDT,M3.4.4/26,M10.5.0 stands for Israel Standard Time (IST) and Israel Daylight Time (IDT), -2 hours ahead of UTC, springing forward on March's fourth +2 hours ahead of UT, springing forward on March's fourth Thursday at 26:00 (i.e., 02:00 on the first Friday on or after March 23), and falling back on October's last Sunday at 02:00. .TP .B WART4WARST,J1/0,J365/25 -stands for Western Argentina Summer Time (WARST), 3 hours behind UTC. +stands for Western Argentina Summer Time (WARST), 3 hours behind UT. There is a dummy fall-back transition on December 31 at 25:00 daylight saving time (i.e., 24:00 standard time, equivalent to January 1 at 00:00 standard time), and a simultaneous spring-forward transition on @@ -282,9 +282,9 @@ is a placeholder. .TP .B WGT3WGST,M3.5.0/\*-2,M10.5.0/\*-1 stands for Western Greenland Time (WGT) and Western Greenland Summer -Time (WGST), 3 hours behind UTC, where clocks follow the EU rules of -springing forward on March's last Sunday at 01:00 UTC (\-02:00 local -time) and falling back on October's last Sunday at 01:00 UTC +Time (WGST), 3 hours behind UT, where clocks follow the EU rules of +springing forward on March's last Sunday at 01:00 UT (\-02:00 local +time) and falling back on October's last Sunday at 01:00 UT (\-01:00 local time). .PP If no @@ -297,7 +297,7 @@ by the file .B posixrules in the system time conversion information directory are used, with the -standard and summer time offsets from UTC replaced by those specified by +standard and summer time offsets from UT replaced by those specified by the .I offset values in diff --git a/northamerica b/northamerica index e5d3eca..505300a 100644 --- a/northamerica +++ b/northamerica @@ -1896,7 +1896,7 @@ Zone America/Edmonton -7:33:52 - LMT 1906 Sep # manager of the Creston & District Museum. The article was written in May 2009. # http://www.ilovecreston.com/?p=articles&t=spec&ar=260 # According to the article, Creston has not changed its clocks since June 1918. -# i.e. Creston has been stuck on UTC-7 for 93 years. +# i.e. Creston has been stuck on UT-7 for 93 years. # Dawson Creek, on the other hand, changed its clocks as recently as April 1972. # Unfortunately the exact date for the time change in June 1918 remains diff --git a/southamerica b/southamerica index bbae226..e0b3265 100644 --- a/southamerica +++ b/southamerica @@ -25,7 +25,7 @@ # https://www.jstor.org/stable/1774359 # # These tables use numeric abbreviations like -03 and -0330 for -# integer hour and minute UTC offsets. Although earlier editions used +# integer hour and minute UT offsets. Although earlier editions used # alphabetic time zone abbreviations, these abbreviations were # invented and did not reflect common practice. @@ -1068,7 +1068,7 @@ Zone America/Rio_Branco -4:31:12 - LMT 1914 # From Paul Eggert (2015-04-03): # Shanks & Pottenger says America/Santiago introduced standard time in -# 1890 and rounds its UTC offset to 70W40; guess that in practice this +# 1890 and rounds its UT offset to 70W40; guess that in practice this # was the same offset as in 1916-1919. It also says Pacific/Easter # standardized on 109W22 in 1890; assume this didn't change the clocks. # diff --git a/theory.html b/theory.html index 965135d..3e686a0 100644 --- a/theory.html +++ b/theory.html @@ -598,7 +598,7 @@ POSIX has the following properties and limitations. and daylight saving time (DST) zone names. Starting with POSIX.1-2001, <var>std</var> and <var>dst</var> may also be - in a quoted form like '<code><UTC+10></code>'; this allows + in a quoted form like '<code><+09></code>'; this allows "<code>+</code>" and "<code>-</code>" in the names. </dd> <dt><var>offset</var></dt><dd> @@ -646,7 +646,7 @@ POSIX has the following properties and limitations. </dd> </dl> Here is an example POSIX TZ string for New Zealand after 2007. - It says that standard time (NZST) is 12 hours ahead of UTC, + It says that standard time (NZST) is 12 hours ahead of UT, and that daylight saving time (NZDT) is observed from September's last Sunday at 02:00 until April's first Sunday at 03:00: @@ -678,7 +678,7 @@ POSIX has the following properties and limitations. applications that an administrator wants used only at certain times – without regard to whether the user has fiddled the TZ environment - variable. While an administrator can "do everything in UTC" to get + variable. While an administrator can "do everything in UT" to get around the problem, doing so is inconvenient and precludes handling daylight saving time shifts - as might be required to limit phone calls to off-peak hours.) diff --git a/tz-link.htm b/tz-link.htm index 835df40..2a8a486 100644 --- a/tz-link.htm +++ b/tz-link.htm @@ -432,7 +432,7 @@ It is freely available under the <abbr>LGPL</abbr>.</li> implementation of a binary file reader. It is freely available under the Apache License.</li> <li><a href="https://github.com/google/cctz">CCTZ</a> is a simple C++ -library that translates between <abbr>UTC</abbr> and civil time and +library that translates between <abbr>UT</abbr> and civil time and can read binary files. It is freely available under the Apache License.</li> <li><a href="http://bmsi.com/java/#TZ">ZoneInfo.java</a> @@ -660,7 +660,7 @@ world time zone boundaries distributed under the href="https://en.wikipedia.org/wiki/Territorial_waters">territorial waters</a> of any nation uses that nation's time. In international waters, time zone boundaries are meridians 15° apart, except that -<abbr>UTC</abbr>−12 and <abbr>UTC</abbr>+12 are each 7.5° +<abbr>UT</abbr>−12 and <abbr>UT</abbr>+12 are each 7.5° wide and are separated by the 180° meridian (not by the International Date Line, which is for land and territorial waters only). A captain can change ship's @@ -884,10 +884,10 @@ protocols.</li> Formats on the Web</a> surveys web- and Internet-oriented date and time formats.</li> <li>Alphabetic time zone abbreviations should not be used as unique -identifiers for <abbr>UTC</abbr> offsets as they are ambiguous in +identifiers for <abbr>UT</abbr> offsets as they are ambiguous in practice. For example, in English-speaking North America -"<abbr>CST</abbr>" denotes 6 hours behind <abbr>UTC</abbr>, -but in China it denotes 8 hours ahead of <abbr>UTC</abbr>, +"<abbr>CST</abbr>" denotes 6 hours behind <abbr>UT</abbr>, +but in China it denotes 8 hours ahead of <abbr>UT</abbr>, and French-speaking North Americans prefer "<abbr title="Heure Normale du Centre">HNC</abbr>" to "<abbr>CST</abbr>". The <code><abbr>tz</abbr></code> @@ -895,7 +895,7 @@ database contains English abbreviations for many time stamps; unfortunately some of these abbreviations were merely the database maintainers' inventions, and are gradually being removed.</li> <li>Numeric time zone abbreviations typically count hours east of -<abbr>UTC</abbr>, e.g., +09 for Japan and +<abbr>UT</abbr>, e.g., +09 for Japan and −10 for Hawaii. However, the <abbr>POSIX</abbr> <code><abbr>TZ</abbr></code> environment variable uses the opposite convention. For example, one might use <code><abbr>TZ</abbr>="<abbr @@ -909,7 +909,7 @@ confusion, handle old time stamps better, and insulate you better from any future changes to the rules. One should never set <abbr>POSIX</abbr> <code><abbr>TZ</abbr></code> to a value like <code>"GMT-9"</code>, though, since this would incorrectly imply that -local time is nine hours ahead of <abbr>UTC</abbr> and the time zone +local time is nine hours ahead of <abbr>UT</abbr> and the time zone is called "<abbr>GMT</abbr>".</li> </ul> <h2 id="see-also">See also</h2> diff --git a/tzselect.ksh b/tzselect.ksh index 759ba5f..d71a95e 100644 --- a/tzselect.ksh +++ b/tzselect.ksh @@ -342,8 +342,9 @@ while while echo >&2 'Please enter the desired value' \ 'of the TZ environment variable.' - echo >&2 'For example, GST-10 is a zone named GST' \ - 'that is 10 hours ahead (east) of UTC.' + echo >&2 'For example, AEST-10 is a zone named AEST' \ + 'that is 10 hours ahead (east) of UT' + echo >&2 'and does not observe daylight saving time.' read TZ $AWK -v TZ="$TZ" 'BEGIN { tzname = "(<[[:alnum:]+-]{3,}>|[[:alpha:]]{3,})" diff --git a/zdump.c b/zdump.c index d4e6084..3cd4f42 100644 --- a/zdump.c +++ b/zdump.c @@ -265,7 +265,7 @@ tzfree(timezone_t env) } #endif /* ! USE_LOCALTIME_RZ */ -/* A UTC time zone, and its initializer. */ +/* A UT time zone, and its initializer. */ static timezone_t gmtz; static void gmtzinit(void) @@ -280,7 +280,7 @@ gmtzinit(void) } } -/* Convert *TP to UTC, storing the broken-down time into *TMP. +/* Convert *TP to UT, storing the broken-down time into *TMP. Return TMP if successful, NULL otherwise. This is like gmtime_r(TP, TMP), except typically faster if USE_LOCALTIME_RZ. */ static struct tm * @@ -730,8 +730,8 @@ adjusted_yday(struct tm const *a, struct tm const *b) } #endif -/* If A is the broken-down local time and B the broken-down UTC for - the same instant, return A's UTC offset in seconds, where positive +/* If A is the broken-down local time and B the broken-down UT for + the same instant, return A's UT offset in seconds, where positive offsets are east of Greenwich. On failure, return LONG_MIN. If T is nonnull, *T is the timestamp that corresponds to A; call @@ -847,10 +847,10 @@ format_local_time(char *buf, size_t size, struct tm const *tm) : my_snprintf(buf, size, "%02d", hh)); } -/* Store into BUF, of size SIZE, a formatted UTC offset for the +/* Store into BUF, of size SIZE, a formatted UT offset for the localtime *TM corresponding to time T. Use ISO 8601 format +HHMMSS, or -HHMMSS for timestamps west of Greenwich; use the - format -00 for unknown UTC offsets. If the hour needs more than + format -00 for unknown UT offsets. If the hour needs more than two digits to represent, extend the length of HH as needed. Otherwise, omit SS if SS is zero, and omit MM too if MM is also zero. @@ -925,7 +925,7 @@ format_quoted_string(char *buf, size_t size, char const *p) %f zone name %L local time as per format_local_time - %Q like "U\t%Z\tD" where U is the UTC offset as for format_utc_offset + %Q like "U\t%Z\tD" where U is the UT offset as for format_utc_offset and D is the isdst flag; except omit D if it is zero, omit %Z if it equals U, quote and escape %Z if it contains nonalphabetics, and omit any trailing tabs. */ diff --git a/zic.8 b/zic.8 index ab95b08..43ef2ce 100644 --- a/zic.8 +++ b/zic.8 @@ -366,7 +366,7 @@ is used to show where the of the time zone abbreviation goes. Alternately, a format can use the pair of characters .B %z -to stand for the UTC offset in the form +to stand for the UT offset in the form .RI \(+- hh , .RI \(+- hhmm , or @@ -376,7 +376,7 @@ using the shortest form that does not lose information, where .IR mm , and .I ss -are the hours, minutes, and seconds east (+) or west (\(mi) of UTC. +are the hours, minutes, and seconds east (+) or west (\(mi) of UT. Alternately, a slash (/) separates standard and daylight abbreviations. diff --git a/zic.c b/zic.c index a3d7383..82e653d 100644 --- a/zic.c +++ b/zic.c @@ -2055,7 +2055,7 @@ abbroffset(char *buf, zic_t offset) minutes = offset % MINSPERHOUR; offset /= MINSPERHOUR; if (100 <= offset) { - error(_("%%z UTC offset magnitude exceeds 99:59:59")); + error(_("%%z UT offset magnitude exceeds 99:59:59")); return "%z"; } else { char *p = buf; -- 2.7.4
On Tue, Nov 7, 2017 at 1:43 AM, Paul Eggert <eggert@cs.ucla.edu> wrote:
diff --git a/NEWS b/NEWS index de81afc..2254650 100644 --- a/NEWS +++ b/NEWS @@ -33,6 +33,11 @@ Unreleased, experimental changes that have snprintf and use pre-C99 compilers. (Problem reported by Jon Skeet.)
+ Changes to code + + Diagnostis and commentary are now more careful about + distinguishing UT from UTC. +
Since the difference between UT and UTC isn't widely know outside certain communities, would it be worth having a reference? Maybe just "See theory.html for more information." --Matthew Donadio (matt@mxd120.com)
On 2017-11-06 22:43:33 (-0800), Paul Eggert wrote:
* NEWS, africa, asia, australasia, europe, localtime.c: * newctime.3, newtzset.3, northamerica, southamerica, theory.html: * tz-link.htm, tzselect.ksh, zdump.c, zic.8, zic.c: Be more consistent about using "UT" instead of "UTC" when talking about timestamps that might predate the introduction of UTC.
I wonder if it wouldn't be easier to simply note "where we write UTC, we actually mean UT when referring to timestamps before UTC was introduced". While UTC is comparatively poorly understood, UT is simply unknown to anyone who hasn't been reading tzdata diffs or hanging out with time nerds for a while. Also worth considering, where 'output' is concerned, people expect "timezones" to be in the form of "UTC+-XXXXXX" and "UT" may be confusing. Philip -- Philip Paeps Senior Reality Engineer Ministry of Information
When I read the patch, I found it confusing and it seemed like many of the changes would become harder to understand. I went back and reread Paul's front matter explaining the concern was timestamps prior to the existance of UTC. Then I threw away my composed message expressing concern. So I think Philip's point is probably correct. This change seems like it creates more confusion, both in documentation and in strings printed by the code. --jhawk@mit.edu John Hawkinson Philip Paeps <philip@trouble.is> wrote on Wed, 8 Nov 2017 at 01:08:34 +0800 in <20171107170834.GF13707@rincewind.trouble.is>:
I wonder if it wouldn't be easier to simply note "where we write UTC, we actually mean UT when referring to timestamps before UTC was introduced".
While UTC is comparatively poorly understood, UT is simply unknown to anyone who hasn't been reading tzdata diffs or hanging out with time nerds for a while.
On Tue 2017-11-07T12:14:43-0500 John Hawkinson hath writ:
So I think Philip's point is probably correct. This change seems like it creates more confusion, both in documentation and in strings printed by the code.
Confusion is inevitable when interpreting historic use. The best that can be done is to explain and apologize for our ancestors. -- Steve Allen <sla@ucolick.org> WGS-84 (GPS) UCO/Lick Observatory--ISB 260 Natural Sciences II, Room 165 Lat +36.99855 1156 High Street Voice: +1 831 459 3046 Lng -122.06015 Santa Cruz, CA 95064 http://www.ucolick.org/~sla/ Hgt +250 m
On 2017-11-07 09:32:41 (-0800), Steve Allen wrote:
On Tue 2017-11-07T12:14:43-0500 John Hawkinson hath writ:
So I think Philip's point is probably correct. This change seems like it creates more confusion, both in documentation and in strings printed by the code.
Confusion is inevitable when interpreting historic use. The best that can be done is to explain and apologize for our ancestors.
After reading through the diff more carefully (and with more context in view), I agree that it makes sense to differentiate more carefully between UTC and UT. Explaining / apologising for our ancestors seems wise though. Perhaps a more prominent note somewhere for the uninitiated? Possibly just copying large chunks of your (Steve's) previous email in this thread. Philip -- Philip Paeps Senior Reality Engineer Ministry of Information
On 11/07/2017 10:01 AM, Philip Paeps wrote:
Explaining / apologising for our ancestors seems wise though. Perhaps a more prominent note somewhere for the uninitiated?
I'm hoping that the bullet point in theory.html suffices; I mean the one that says "In the tz database commentary, UT denotes a family of time standards that includes Coordinated Universal Time (UTC)...." If this wording could be improved or if featuring it more prominently would help, specific suggestions are welcome. Before the recent change, tzdb was inconsistent; although quite a bit already respected the distinction between UT and UTC, quite a bit didn't. The idea of the patch was to change tzdb to follow the bullet point more systematically.
I wonder if it wouldn't be easier to simply note "where we write UTC, we actually mean UT when referring to timestamps before UTC was introduced".
Something like that was proposed more generally, even for applications outside tzdb, here: Sauter J. Extending Coordinated Universal Time to dates before 1972. 2017-07-08. https://commons.wikimedia.org/wiki/File:Extending_Coordinated_Universal_Time... Sauter also suggested that we drop the use of time_t for telling time: Sauter J. Avoid using POSIX time_t for telling time. 2017-10-08. https://commons.wikimedia.org/wiki/File:Avoid_Using_POSIX_time_t_for_Telling... If these recent proposals catch on, I suppose we could change tzdb eventually. Dropping time_t would be a big change, though. Even extending UTC for old timestamps is a stretch.
On Tue 2017-11-07T19:16:33+0000 Zefram hath writ:
To not make this change would perpetuate the existing confusion. The way forward is to use the correct terms and to include adequate explanation (in theory.html) to resolve the confusion.
During those intervals the folks who were tasked with providing time for their governments were getting new instructions as often as three times in a calendar year from assemblies of the IAU, CCIR, and URSI. Given that I am holding in my hands the issues of Bulletin Horaire from the relevant period of time, I am continuing to insert references into https://www.ucolick.org/~sla/leapsecs/timescales.html#UTC -- Steve Allen <sla@ucolick.org> WGS-84 (GPS) UCO/Lick Observatory--ISB 260 Natural Sciences II, Room 165 Lat +36.99855 1156 High Street Voice: +1 831 459 3046 Lng -122.06015 Santa Cruz, CA 95064 http://www.ucolick.org/~sla/ Hgt +250 m
On Wed 2017-11-08T01:08:34+0800 Philip Paeps hath writ:
I wonder if it wouldn't be easier to simply note "where we write UTC, we actually mean UT when referring to timestamps before UTC was introduced".
This makes it tricky to include contemporary documents which originally used one or the other. In many of those cases the usage was uninformed and often the term used was inappropriate for the actual meaning. The inception of all the forms of UT (UT0, UT1, UT2), and their usage, is available in issues of BIH's Bulletin Horaire from 1955. See https://www.ucolick.org/~sla/leapsecs/BH1955.html So definitions for terms related to UT are easy to pinpoint. It is much harder pinpoint UTC. The inception of UTC in print is found in Bulletin Horaire, Series J, No. 1 (1964 Jan/Feb) which was not actually prepared for printing until after 1965-05-01. That first use in print does not declaratively define UTC, but rather employs a term which had evidently been in informal use since some point after 1961 January. Nevertheless, most publications and radio broadcasts continued to use the term GMT or UT (and often local legal civil time) into the 1970s even when they were broadcasting signals that would better have been called UTC. The first documented use of the term UTC in radio broadcast time signals is by WWV and WWVH starting 1974-01-01. There is no formal definition of the term UTC until the 13th Plenary Assembly of the CCIR approved Recommendation 460-1 during 1974 July. Nevertheless, the distinction between UT,UT0,UT1,UT2 and UTC is clear from the proceedings during the decades from the 1950s through the 1970s. UT is a time scale derived directly from observation of the rotation of the earth in a way that subdivides days of the calendar. UTC was always a time scale derived from atomic chronometers and adjusted to remain within specified close agreement to UT. -- Steve Allen <sla@ucolick.org> WGS-84 (GPS) UCO/Lick Observatory--ISB 260 Natural Sciences II, Room 165 Lat +36.99855 1156 High Street Voice: +1 831 459 3046 Lng -122.06015 Santa Cruz, CA 95064 http://www.ucolick.org/~sla/ Hgt +250 m
On 7 November 2017 at 18:08, Philip Paeps <philip@trouble.is> wrote:
Also worth considering, where 'output' is concerned, people expect "timezones" to be in the form of "UTC+-XXXXXX" and "UT" may be confusing.
Is that so? Perhaps it depends on who these "people" are, but I wouldn't be surprised if more people expected "GMT+-XX" than "UTC+-XXXXXX".
participants (7)
-
John Hawkinson
-
Matthew Donadio
-
Paul Eggert
-
Philip Newton
-
Philip Paeps
-
Steve Allen
-
Zefram