Theory and pragmatics of the tz code and data

Scope of the tz database

The tz database attempts to record the history and predicted future of civil time scales. It organizes time zone and daylight saving time data by partitioning the world into timezones whose clocks all agree about timestamps that occur after the POSIX Epoch (1970-01-01 00:00:00 UTC). Although 1970 is a somewhat-arbitrary cutoff, there are significant challenges to moving the cutoff earlier even by a decade or two, due to the wide variety of local practices before computer timekeeping became prevalent. Most timezones correspond to a notable location and the database records all known clock transitions for that location; some timezones correspond instead to a fixed UTC offset.

Each timezone typically corresponds to a geographical region that is smaller than a traditional time zone, because clocks in a timezone all agree after 1970 whereas a traditional time zone merely specifies current standard time. For example, applications that deal with current and future timestamps in the traditional North American mountain time zone can choose from the timezones America/Denver which observes US-style daylight saving time (DST), and America/Phoenix which does not observe DST. Applications that also deal with past timestamps in the mountain time zone can choose from over a dozen timezones, such as America/Boise, America/Edmonton, and America/Hermosillo, each of which currently uses mountain time but differs from other timezones for some timestamps after 1970.

Clock transitions before 1970 are recorded for location-based timezones, because most systems support timestamps before 1970 and could misbehave if data entries were omitted for pre-1970 transitions. However, the database is not designed for and does not suffice for applications requiring accurate handling of all past times everywhere, as it would take far too much effort and guesswork to record all details of pre-1970 civil timekeeping. Although some information outside the scope of the database is collected in a file backzone that is distributed along with the database proper, this file is less reliable and does not necessarily follow database guidelines.

As described below, reference source code for using the tz database is also available. The tz code is upwards compatible with POSIX, an international standard for UNIX-like systems. As of this writing, the current edition of POSIX is: The Open Group Base Specifications Issue 7, IEEE Std 1003.1-2017, 2018 Edition. Because the database's scope encompasses real-world changes to civil timekeeping, its model for describing time is more complex than the standard and daylight saving times supported by POSIX.1-2017. A tz timezone corresponds to a ruleset that can have more than two changes per year, these changes need not merely flip back and forth between two alternatives, and the rules themselves can change at times. Whether and when a timezone changes its clock, and even the timezone's notional base offset from UTC, are variable. It does not always make sense to talk about a timezone's "base offset", which is not necessarily a single number.

Timezone identifiers

Each timezone has a name that uniquely identifies the timezone. Inexperienced users are not expected to select these names unaided. Distributors should provide documentation and/or a simple selection interface that explains each name via a map or via descriptive text like "Czech Republic" instead of the timezone name "Europe/Prague". If geolocation information is available, a selection interface can locate the user on a timezone map or prioritize names that are geographically close. For an example selection interface, see the tzselect program in the tz code. The Unicode Common Locale Data Repository contains data that may be useful for other selection interfaces; it maps timezone names like Europe/Prague to locale-dependent strings like "Prague", "Praha", "Прага", and "布拉格".

The naming conventions attempt to strike a balance among the following goals:

• Uniquely identify every timezone where clocks have agreed since 1970. This is essential for the intended use: static clocks keeping local civil time.
• Indicate to experts where the timezone's clocks typically are.
• Be robust in the presence of political changes. For example, names are typically not tied to countries, to avoid incompatibilities when countries change their name (e.g., Swaziland→Eswatini) or when locations change countries (e.g., Hong Kong from UK colony to China). There is no requirement that every country or national capital must have a timezone name.
• Be portable to a wide variety of implementations.
• Use a consistent naming conventions over the entire world.

Names normally have the form AREA/LOCATION, where AREA is a continent or ocean, and LOCATION is a specific location within the area. North and South America share the same area, 'America'. Typical names are 'Africa/Cairo', 'America/New_York', and 'Pacific/Honolulu'. Some names are further qualified to help avoid confusion; for example, 'America/Indiana/Petersburg' distinguishes Petersburg, Indiana from other Petersburgs in America.

Here are the general guidelines used for choosing timezone names, in decreasing order of importance:

• Use only valid POSIX file name components (i.e., the parts of names other than '/'). Do not use the file name components '.' and '..'. Within a file name component, use only ASCII letters, '.', '-' and '_'. Do not use digits, as that might create an ambiguity with POSIX.1-2017 TZ strings. A file name component must not exceed 14 characters or start with '-'. E.g., prefer America/Noronha to America/Fernando_de_Noronha. Exceptions: see the discussion of legacy names below.
• A name must not be empty, or contain '//', or start or end with '/'.
• Do not use names that differ only in case. Although the reference implementation is case-sensitive, some other implementations are not, and they would mishandle names differing only in case.
• If one name A is an initial prefix of another name AB (ignoring case), then B must not start with '/', as a regular file cannot have the same name as a directory in POSIX. For example, America/New_York precludes America/New_York/Bronx.
• Uninhabited regions like the North Pole and Bouvet Island do not need locations, since local time is not defined there.
• If all the clocks in a timezone have agreed since 1970, do not bother to include more than one timezone even if some of the clocks disagreed before 1970. Otherwise these tables would become annoyingly large.
• If boundaries between regions are fluid, such as during a war or insurrection, do not bother to create a new timezone merely because of yet another boundary change. This helps prevent table bloat and simplifies maintenance.
• If a name is ambiguous, use a less ambiguous alternative; e.g., many cities are named San José and Georgetown, so prefer America/Costa_Rica to America/San_Jose and America/Guyana to America/Georgetown.
• Keep locations compact. Use cities or small islands, not countries or regions, so that any future changes do not split individual locations into different timezones. E.g., prefer Europe/Paris to Europe/France, since France has had multiple time zones.
• Use mainstream English spelling, e.g., prefer Europe/Rome to Europa/Roma, and prefer Europe/Athens to the Greek Ευρώπη/Αθήνα or the Romanized Evrópi/Athína. The POSIX file name restrictions encourage this guideline.
• Use the most populous among locations in a region, e.g., prefer Asia/Shanghai to Asia/Beijing. Among locations with similar populations, pick the best-known location, e.g., prefer Europe/Rome to Europe/Milan.
• Use the singular form, e.g., prefer Atlantic/Canary to Atlantic/Canaries.
• Omit common suffixes like '_Islands' and '_City', unless that would lead to ambiguity. E.g., prefer America/Cayman to America/Cayman_Islands and America/Guatemala to America/Guatemala_City, but prefer America/Mexico_City to America/Mexico because the country of Mexico has several time zones.
• Use '_' to represent a space.
• Omit '.' from abbreviations in names. E.g., prefer Atlantic/St_Helena to Atlantic/St._Helena.
• Do not change established names if they only marginally violate the above guidelines. For example, do not change the existing name Europe/Rome to Europe/Milan merely because Milan's population has grown to be somewhat greater than Rome's.
• If a name is changed, put its old spelling in the 'backward' file as a link to the new spelling. This means old spellings will continue to work. Ordinarily a name change should occur only in the rare case when a location's consensus English-language spelling changes; for example, in 2008 Asia/Calcutta was renamed to Asia/Kolkata due to long-time widespread use of the new city name instead of the old.

Guidelines have evolved with time, and names following old versions of these guidelines might not follow the current version. When guidelines have changed, old names continue to be supported. Guideline changes have included the following:

• Older versions of this package used a different naming scheme. See the file 'backward' for most of these older names (e.g., 'US/Eastern' instead of 'America/New_York'). The other old-fashioned names still supported are 'WET', 'CET', 'MET', and 'EET' (see the file 'europe').
• Older versions of this package defined legacy names that are incompatible with the first guideline of location names, but which are still supported. These legacy names are mostly defined in the file 'etcetera'. Also, the file 'backward' defines the legacy names 'Etc/GMT0', 'Etc/GMT-0', 'Etc/GMT+0', 'GMT0', 'GMT-0' and 'GMT+0', and the file 'northamerica' defines the legacy names 'EST5EDT', 'CST6CDT', 'MST7MDT', and 'PST8PDT'.
• Older versions of these guidelines said that there should typically be at least one name for each ISO 3166-1 officially assigned two-letter code for an inhabited country or territory. This old guideline has been dropped, as it was not needed to handle timestamps correctly and it increased maintenance burden.

The file zone1970.tab lists geographical locations used to name timezones. It is intended to be an exhaustive list of names for geographic regions as described above; this is a subset of the timezones in the data. Although a zone1970.tab location's longitude corresponds to its local mean time (LMT) offset with one hour for every 15° east longitude, this relationship is not exact. The backward-compatibility file zone.tab is similar but conforms to the older-version guidelines related to ISO 3166-1; it lists only one country code per entry and unlike zone1970.tab it can list names defined in backward. Applications that process only timestamps from now on can instead use the file zonenow.tab, which partitions the world more coarsely, into regions where clocks agree now and in the predicted future; this file is smaller and simpler than zone1970.tab and zone.tab.

The database defines each timezone name to be a zone, or a link to a zone. The source file backward defines links for backward compatibility; it does not define zones. Although backward was originally designed to be optional, nowadays distributions typically use it and no great weight should be attached to whether a link is defined in backward or in some other file. The source file etcetera defines names that may be useful on platforms that do not support POSIX.1-2017-style TZ strings; no other source file other than backward contains links to its zones. One of etcetera's names is Etc/UTC, used by functions like gmtime to obtain leap second information on platforms that support leap seconds. Another etcetera name, GMT, is used by older code releases.

Time zone abbreviations

When this package is installed, it generates time zone abbreviations like 'EST' to be compatible with human tradition and POSIX. Here are the general guidelines used for choosing time zone abbreviations, in decreasing order of importance:

• Use three to six characters that are ASCII alphanumerics or '+' or '-'. Previous editions of this database also used characters like space and '?', but these characters have a special meaning to the UNIX shell and cause commands like 'set `date`' to have unexpected effects. Previous editions of this guideline required upper-case letters, but the Congressman who introduced Chamorro Standard Time preferred "ChST", so lower-case letters are now allowed. Also, POSIX from 2001 on relaxed the rule to allow '-', '+', and alphanumeric characters from the portable character set in the current locale. In practice ASCII alphanumerics and '+' and '-' are safe in all locales.

  In other words, in the C locale the POSIX extended regular expression [-+[:alnum:]]{3,6} should match the abbreviation. This guarantees that all abbreviations could have been specified by a POSIX.1-2017 TZ string.

• Use abbreviations that are in common use among English-speakers, e.g., 'EST' for Eastern Standard Time in North America. We assume that applications translate them to other languages as part of the normal localization process; for example, a French application might translate 'EST' to 'HNE'.

  These abbreviations (for standard/daylight/etc. time) are: ACST/ACDT Australian Central, AST/ADT/APT/AWT/ADDT Atlantic, AEST/AEDT Australian Eastern, AHST/AHDT Alaska-Hawaii, AKST/AKDT Alaska, AWST/AWDT Australian Western, BST/BDT Bering, CAT/CAST Central Africa, CET/CEST/CEMT Central European, ChST Chamorro, CST/CDT/CWT/CPT Central [North America], CST/CDT China, GMT/BST/IST/BDST Greenwich, EAT East Africa, EST/EDT/EWT/EPT Eastern [North America], EET/EEST Eastern European, GST/GDT Guam, HST/HDT/HWT/HPT Hawaii, HKT/HKST/HKWT Hong Kong, IST India, IST/GMT Irish, IST/IDT/IDDT Israel, JST/JDT Japan, KST/KDT Korea, MET/MEST Middle European (a backward-compatibility alias for Central European), MSK/MSD Moscow, MST/MDT/MWT/MPT Mountain, NST/NDT/NWT/NPT/NDDT Newfoundland, NST/NDT/NWT/NPT Nome, NZMT/NZST New Zealand through 1945, NZST/NZDT New Zealand 1946–present, PKT/PKST Pakistan, PST/PDT/PWT/PPT Pacific, PST/PDT Philippine, SAST South Africa, SST Samoa, UTC Universal, WAT/WAST West Africa, WET/WEST/WEMT Western European, WIB Waktu Indonesia Barat, WIT Waktu Indonesia Timur, WITA Waktu Indonesia Tengah, YST/YDT/YWT/YPT/YDDT Yukon.

• For times taken from a city's longitude, use the traditional xMT notation. The only abbreviation like this in current use is 'GMT'. The others are for timestamps before 1960, except that Monrovia Mean Time persisted until 1972. Typically, numeric abbreviations (e.g., '-004430' for MMT) would cause trouble here, as the numeric strings would exceed the POSIX length limit.

  These abbreviations are: AMT Asunción, Athens; BMT Baghdad, Bangkok, Batavia, Bermuda, Bern, Bogotá, Brussels, Bucharest; CMT Calamarca, Caracas, Chisinau, Colón, Córdoba; DMT Dublin/Dunsink; EMT Easter; FFMT Fort-de-France; FMT Funchal; GMT Greenwich; HMT Havana, Helsinki, Horta, Howrah; IMT Irkutsk, Istanbul; JMT Jerusalem; KMT Kaunas, Kyiv, Kingston; LMT Lima, Lisbon, local; MMT Macassar, Madras, Malé, Managua, Minsk, Monrovia, Montevideo, Moratuwa, Moscow; PLMT Phù Liễn; PMT Paramaribo, Paris, Perm, Pontianak, Prague; PMMT Port Moresby; PPMT Port-au-Prince; QMT Quito; RMT Rangoon, Riga, Rome; SDMT Santo Domingo; SJMT San José; SMT Santiago, Simferopol, Singapore, Stanley; TBMT Tbilisi; TMT Tallinn, Tehran; WMT Warsaw.

  A few abbreviations also follow the pattern that GMT/BST established for time in the UK. They are: BMT/BST for Bermuda 1890–1930, CMT/BST for Calamarca Mean Time and Bolivian Summer Time 1890–1932, DMT/IST for Dublin/Dunsink Mean Time and Irish Summer Time 1880–1916, MMT/MST/MDST for Moscow 1880–1919, and RMT/LST for Riga Mean Time and Latvian Summer time 1880–1926.

• Use 'LMT' for local mean time of locations before the introduction of standard time; see "Scope of the tz database".
• If there is no common English abbreviation, use numeric offsets like -05 and +0530 that are generated by zic's %z notation.
• Use current abbreviations for older timestamps to avoid confusion. For example, in 1910 a common English abbreviation for time in central Europe was 'MEZ' (short for both "Middle European Zone" and for "Mitteleuropäische Zeit" in German). Nowadays 'CET' ("Central European Time") is more common in English, and the database uses 'CET' even for circa-1910 timestamps as this is less confusing for modern users and avoids the need for determining when 'CET' supplanted 'MEZ' in common usage.
• Use a consistent style in a timezone's history. For example, if a history tends to use numeric abbreviations and a particular entry could go either way, use a numeric abbreviation.
• Use Universal Time (UT) (with time zone abbreviation '-00') for locations while uninhabited. The leading '-' is a flag that the UT offset is in some sense undefined; this notation is derived from Internet RFC 3339.

Application writers should note that these abbreviations are ambiguous in practice: e.g., 'CST' means one thing in China and something else in North America, and 'IST' can refer to time in India, Ireland or Israel. To avoid ambiguity, use numeric UT offsets like '-0600' instead of time zone abbreviations like 'CST'.

Accuracy of the tz database

The tz database is not authoritative, and it surely has errors. Corrections are welcome and encouraged; see the file CONTRIBUTING. Users requiring authoritative data should consult national standards bodies and the references cited in the database's comments.

Errors in the tz database arise from many sources:

• The tz database predicts future timestamps, and current predictions will be incorrect after future governments change the rules. For example, if today someone schedules a meeting for 13:00 next October 1, Casablanca time, and tomorrow Morocco changes its daylight saving rules, software can mess up after the rule change if it blithely relies on conversions made before the change.
• The pre-1970 entries in this database cover only a tiny sliver of how clocks actually behaved; the vast majority of the necessary information was lost or never recorded. Thousands more timezones would be needed if the tz database's scope were extended to cover even just the known or guessed history of standard time; for example, the current single entry for France would need to split into dozens of entries, perhaps hundreds. And in most of the world even this approach would be misleading due to widespread disagreement or indifference about what times should be observed. In her 2015 book The Global Transformation of Time, 1870–1950, Vanessa Ogle writes "Outside of Europe and North America there was no system of time zones at all, often not even a stable landscape of mean times, prior to the middle decades of the twentieth century". See: Timothy Shenk, Booked: A Global History of Time. Dissent 2015-12-17.
• Most of the pre-1970 data entries come from unreliable sources, often astrology books that lack citations and whose compilers evidently invented entries when the true facts were unknown, without reporting which entries were known and which were invented. These books often contradict each other or give implausible entries, and on the rare occasions when they are checked they are typically found to be incorrect.
• For the UK the tz database relies on years of first-class work done by Joseph Myers and others; see "History of legal time in Britain". Other countries are not done nearly as well.
• Sometimes, different people in the same city maintain clocks that differ significantly. Historically, railway time was used by railroad companies (which did not always agree with each other), church-clock time was used for birth certificates, etc. More recently, competing political groups might disagree about clock settings. Often this is merely common practice, but sometimes it is set by law. For example, from 1891 to 1911 the UT offset in France was legally UT +00:09:21 outside train stations and UT +00:04:21 inside. Other examples include Chillicothe in 1920, Palm Springs in 1946/7, and Jerusalem and Ürümqi to this day.
• Although a named location in the tz database stands for the containing region, its pre-1970 data entries are often accurate for only a small subset of that region. For example, Europe/London stands for the United Kingdom, but its pre-1847 times are valid only for locations that have London's exact meridian, and its 1847 transition to GMT is known to be valid only for the L&NW and the Caledonian railways.
• The tz database does not record the earliest time for which a timezone's data entries are thereafter valid for every location in the region. For example, Europe/London is valid for all locations in its region after GMT was made the standard time, but the date of standardization (1880-08-02) is not in the tz database, other than in commentary. For many timezones the earliest time of validity is unknown.
• The tz database does not record a region's boundaries, and in many cases the boundaries are not known. For example, the timezone America/Kentucky/Louisville represents a region around the city of Louisville, the boundaries of which are unclear.
• Changes that are modeled as instantaneous transitions in the tz database were often spread out over hours, days, or even decades.
• Even if the time is specified by law, locations sometimes deliberately flout the law.
• Early timekeeping practices, even assuming perfect clocks, were often not specified to the accuracy that the tz database requires.
• The tz database cannot represent stopped clocks. However, on 1911-03-11 at 00:00, some public-facing French clocks were changed by stopping them for a few minutes to effect a transition. The tz database models this via a backward transition; the relevant French legislation does not specify exactly how the transition was to occur.
• Sometimes historical timekeeping was specified more precisely than what the tz code can handle. For example, from 1880 to 1916 clocks in Ireland observed Dublin Mean Time (estimated to be UT −00:25:21.1); although the tz source data can represent the .1 second, TZif files and the code cannot. In practice these old specifications were rarely if ever implemented to subsecond precision.
• Even when all the timestamp transitions recorded by the tz database are correct, the tz rules that generate them may not faithfully reflect the historical rules. For example, from 1922 until World War II the UK moved clocks forward the day following the third Saturday in April unless that was Easter, in which case it moved clocks forward the previous Sunday. Because the tz database has no way to specify Easter, these exceptional years are entered as separate tz Rule lines, even though the legal rules did not change. When transitions are known but the historical rules behind them are not, the database contains Zone and Rule entries that are intended to represent only the generated transitions, not any underlying historical rules; however, this intent is recorded at best only in commentary.
• The tz database models time using the proleptic Gregorian calendar with days containing 24 equal-length hours numbered 00 through 23, except when clock transitions occur. Pre-standard time is modeled as local mean time. However, historically many people used other calendars and other timescales. For example, the Roman Empire used the Julian calendar, and Roman timekeeping had twelve varying-length daytime hours with a non-hour-based system at night. And even today, some local practices diverge from the Gregorian calendar with 24-hour days. These divergences range from relatively minor, such as Japanese bars giving times like "24:30" for the wee hours of the morning, to more-significant differences such as the east African practice of starting the day at dawn, renumbering the Western 06:00 to be 12:00. These practices are largely outside the scope of the tz code and data, which provide only limited support for date and time localization such as that required by POSIX.1-2017. If DST is not used a different time zone can often do the trick; for example, in Kenya a TZ setting like <-03>3 or America/Cayenne starts the day six hours later than Africa/Nairobi does.
• Early clocks were less reliable, and data entries do not represent clock error.
• The tz database assumes Universal Time (UT) as an origin, even though UT is not standardized for older timestamps. In the tz database commentary, UT denotes a family of time standards that includes Coordinated Universal Time (UTC) along with other variants such as UT1 and GMT, with days starting at midnight. Although UT equals UTC for modern timestamps, UTC was not defined until 1960, so commentary uses the more general abbreviation UT for timestamps that might predate 1960. Since UT, UT1, etc. disagree slightly, and since pre-1972 UTC seconds varied in length, interpretation of older timestamps can be problematic when subsecond accuracy is needed.
• Civil time was not based on atomic time before 1972, and we do not know the history of earth's rotation accurately enough to map SI seconds to historical solar time to more than about one-hour accuracy. See: Stephenson FR, Morrison LV, Hohenkerk CY. Measurement of the Earth's rotation: 720 BC to AD 2015. Proc Royal Soc A. 2016;472:20160404. Also see: Espenak F. Uncertainty in Delta T (ΔT).
• The relationship between POSIX time (that is, UTC but ignoring leap seconds) and UTC is not agreed upon. This affects time stamps during the leap second era (1972–2035). Although the POSIX clock officially stops during an inserted leap second, at least one proposed standard has it jumping back a second instead; and in practice POSIX clocks more typically either progress glacially during a leap second, or are slightly slowed while near a leap second.
• The tz database does not represent how uncertain its information is. Ideally it would contain information about when data entries are incomplete or dicey. Partial temporal knowledge is a field of active research, though, and it is not clear how to apply it here.

In short, many, perhaps most, of the tz database's pre-1970 and future timestamps are either wrong or misleading. Any attempt to pass the tz database off as the definition of time should be unacceptable to anybody who cares about the facts. In particular, the tz database's LMT offsets should not be considered meaningful, and should not prompt creation of timezones merely because two locations differ in LMT or transitioned to standard time at different dates.

Time and date functions

The tz code contains time and date functions that are upwards compatible with those of POSIX. Code compatible with this package is already part of many platforms, where the primary use of this package is to update obsolete time-related files. To do this, you may need to compile the time zone compiler 'zic' supplied with this package instead of using the system 'zic', since the format of zic's input is occasionally extended, and a platform may still be shipping an older zic.

POSIX.1-2017 properties and limitations

• In POSIX.1-2017, time display in a process is controlled by the environment variable TZ. Unfortunately, the POSIX.1-2017 TZ string takes a form that is hard to describe and is error-prone in practice. Also, POSIX.1-2017 TZ strings cannot deal with daylight saving time rules not based on the Gregorian calendar (as in Morocco), or with situations where more than two time zone abbreviations or UT offsets are used in an area.

  The POSIX.1-2017 TZ string takes the following form:

  stdoffset[dst[offset][,date[/time],date[/time]]]

  where:

  std and dst

  are 3 or more characters specifying the standard and daylight saving time (DST) zone abbreviations. Starting with POSIX.1-2001, std and dst may also be in a quoted form like '<+09>'; this allows "+" and "-" in the names.

  offset

  is of the form '[±]hh:[mm[:ss]]' and specifies the offset west of UT. 'hh' may be a single digit; 0≤hh≤24. The default DST offset is one hour ahead of standard time.

  date[/time],date[/time]

  specifies the beginning and end of DST. If this is absent, the system supplies its own ruleset for DST, typically current US DST rules.

  time

  takes the form 'hh:[mm[:ss]]' and defaults to 02:00. This is the same format as the offset, except that a leading '+' or '-' is not allowed.

  date

  takes one of the following forms:

  Jn (1≤n≤365)

  origin-1 day number not counting February 29

  n (0≤n≤365)

  origin-0 day number counting February 29 if present

  Mm.n.d (0[Sunday]≤d≤6[Saturday], 1≤n≤5, 1≤m≤12)

  for the dth day of week n of month m of the year, where week 1 is the first week in which day d appears, and '5' stands for the last week in which day d appears (which may be either the 4th or 5th week). Typically, this is the only useful form; the n and Jn forms are rarely used.

  Here is an example POSIX.1-2017 TZ string for New Zealand after 2007. 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:

  TZ='NZST-12NZDT,M9.5.0,M4.1.0/3'

  This POSIX.1-2017 TZ string is hard to remember, and mishandles some timestamps before 2008. With this package you can use this instead:

  TZ='Pacific/Auckland'

• POSIX does not define the DST transitions for TZ values like "EST5EDT". Traditionally the current US DST rules were used to interpret such values, but this meant that the US DST rules were compiled into each time conversion package, and when US time conversion rules changed (as in the United States in 1987 and again in 2007), all packages that interpreted TZ values had to be updated to ensure proper results.
• The TZ environment variable is process-global, which makes it hard to write efficient, thread-safe applications that need access to multiple timezones.
• In POSIX, there is no tamper-proof way for a process to learn the system's best idea of local (wall clock) time. This is important for 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 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.
• POSIX.1-2017 provides no convenient and efficient way to determine the UT offset and time zone abbreviation of arbitrary timestamps, particularly for timezones that do not fit into the POSIX model.
• POSIX requires that time_t clock counts exclude leap seconds.
• The tz code attempts to support all the time_t implementations allowed by POSIX. The time_t type represents a nonnegative count of seconds since 1970-01-01 00:00:00 UTC, ignoring leap seconds. In practice, time_t is usually a signed 64- or 32-bit integer; 32-bit signed time_t values stop working after 2038-01-19 03:14:07 UTC, so new implementations these days typically use a signed 64-bit integer. Unsigned 32-bit integers are used on one or two platforms, and 36-bit and 40-bit integers are also used occasionally. Although earlier POSIX versions allowed time_t to be a floating-point type, this was not supported by any practical system, and POSIX.1-2013 and the tz code both require time_t to be an integer type.

Extensions to POSIX.1-2017 in the tz code

• The TZ environment variable is used in generating the name of a file from which time-related information is read (or is interpreted à la POSIX.1-2017); TZ is no longer constrained to be a string containing abbreviations and numeric data as described above. The file's format is TZif, a timezone information format that contains binary data; see Internet RFC 8536. The daylight saving time rules to be used for a particular timezone are encoded in the TZif file; the format of the file allows US, Australian, and other rules to be encoded, and allows for situations where more than two time zone abbreviations are used.

  It was recognized that allowing the TZ environment variable to take on values such as 'America/New_York' might cause "old" programs (that expect TZ to have a certain form) to operate incorrectly; consideration was given to using some other environment variable (for example, TIMEZONE) to hold the string used to generate the TZif file's name. In the end, however, it was decided to continue using TZ: it is widely used for time zone purposes; separately maintaining both TZ and TIMEZONE seemed a nuisance; and systems where "new" forms of TZ might cause problems can simply use legacy TZ values such as "EST5EDT" which can be used by "new" programs as well as by "old" programs that assume pre-POSIX TZ values.

• The code supports platforms with a UT offset member in struct tm, e.g., tm_gmtoff, or with a time zone abbreviation member in struct tm, e.g., tm_zone. As noted in Austin Group defect 1533, a future version of POSIX is planned to require tm_gmtoff and tm_zone.
• Functions tzalloc, tzfree, localtime_rz, and mktime_z for more-efficient thread-safe applications that need to use multiple timezones. The tzalloc and tzfree functions allocate and free objects of type timezone_t, and localtime_rz and mktime_z are like localtime_r and mktime with an extra timezone_t argument. The functions were inspired by NetBSD.
• Negative time_t values are supported, on systems where time_t is signed.
• These functions can account for leap seconds; see Leap seconds below.

POSIX features no longer needed

POSIX and ISO C define some APIs that are vestigial: they are not needed, and are relics of a too-simple model that does not suffice to handle many real-world timestamps. Although the tz code supports these vestigial APIs for backwards compatibility, they should be avoided in portable applications. The vestigial APIs are:

• The POSIX tzname variable does not suffice and is no longer needed. To get a timestamp's time zone abbreviation, consult the tm_zone member if available; otherwise, use strftime's "%Z" conversion specification.
• The POSIX daylight and timezone variables do not suffice and are no longer needed. To get a timestamp's UT offset, consult the tm_gmtoff member if available; otherwise, subtract values returned by localtime and gmtime using the rules of the Gregorian calendar, or use strftime's "%z" conversion specification if a string like "+0900" suffices.
• The tm_isdst member is almost never needed and most of its uses should be discouraged in favor of the abovementioned APIs. Although it can still be used in arguments to mktime to disambiguate timestamps near a DST transition when the clock jumps back on platforms lacking tm_gmtoff, this disambiguation does not work when standard time itself jumps back, which can occur when a location changes to a time zone with a lesser UT offset.

Other portability notes

• The 7th Edition UNIX timezone function is not present in this package; it is impossible to reliably map timezone's arguments (a "minutes west of GMT" value and a "daylight saving time in effect" flag) to a time zone abbreviation, and we refuse to guess. Programs that in the past used the timezone function may now examine localtime(&clock)->tm_zone (if TM_ZONE is defined) or tzname[localtime(&clock)->tm_isdst] (if HAVE_TZNAME is nonzero) to learn the correct time zone abbreviation to use.
• The 4.2BSD gettimeofday function is not used in this package. This formerly let users obtain the current UTC offset and DST flag, but this functionality was removed in later versions of BSD.
• In SVR2, time conversion fails for near-minimum or near-maximum time_t values when doing conversions for places that do not use UT. This package takes care to do these conversions correctly. A comment in the source code tells how to get compatibly wrong results.
• The functions that are conditionally compiled if STD_INSPIRED is nonzero should, at this point, be looked on primarily as food for thought. They are not in any sense "standard compatible" – some are not, in fact, specified in any standard. They do, however, represent responses of various authors to standardization proposals.
• Other time conversion proposals, in particular those supported by the Time Zone Database Parser, offer a wider selection of functions that provide capabilities beyond those provided here. The absence of such functions from this package is not meant to discourage the development, standardization, or use of such functions. Rather, their absence reflects the decision to make this package contain valid extensions to POSIX, to ensure its broad acceptability. If more powerful time conversion functions can be standardized, so much the better.

Interface stability

The tz code and data supply the following interfaces:

• A set of timezone names as per "Timezone identifiers" above.
• Library functions described in "Time and date functions" above.
• The programs tzselect, zdump, and zic, documented in their man pages.
• The format of zic input files, documented in the zic man page.
• The format of zic output files, documented in the tzfile man page.
• The format of zone table files, documented in zone1970.tab.
• The format of the country code file, documented in iso3166.tab.
• The version number of the code and data, as the first line of the text file 'version' in each release.

Interface changes in a release attempt to preserve compatibility with recent releases. For example, tz data files typically do not rely on recently added zic features, so that users can run older zic versions to process newer data files. Downloading the tz database describes how releases are tagged and distributed.

Interfaces not listed above are less stable. For example, users should not rely on particular UT offsets or abbreviations for timestamps, as data entries are often based on guesswork and these guesses may be corrected or improved.

Timezone boundaries are not part of the stable interface. For example, even though the Asia/Bangkok timezone currently includes Chang Mai, Hanoi, and Phnom Penh, this is not part of the stable interface and the timezone can split at any time. If a calendar application records a future event in some location other than Bangkok by putting "Asia/Bangkok" in the event's record, the application should be robust in the presence of timezone splits between now and the future time.

Leap seconds

Leap seconds were introduced in 1972 to accommodate the difference between atomic time and the less regular rotation of the earth. Unfortunately they caused so many problems with civil timekeeping that they are planned to be discontinued by 2035, with some as-yet-undetermined mechanism replacing them, perhaps after the year 2135. Despite their impending obsolescence, a record of leap seconds is still needed to resolve timestamps from 1972 through 2035.

The tz code and data can account for leap seconds, thanks to code contributed by Bradley White. However, the leap second support of this package is rarely used directly because POSIX requires leap seconds to be excluded and many software packages would mishandle leap seconds if they were present. Instead, leap seconds are more commonly handled by occasionally adjusting the operating system kernel clock as described in Precision timekeeping, and this package by default installs a leapseconds file commonly used by NTP software that adjusts the kernel clock. However, kernel-clock twiddling approximates UTC only roughly, and systems needing more precise UTC can use this package's leap second support directly.

The directly supported mechanism assumes that time_t counts of seconds since the POSIX epoch normally include leap seconds, as opposed to POSIX time_t counts which exclude leap seconds. This modified timescale is converted to UTC at the same point that time zone and DST adjustments are applied – namely, at calls to localtime and analogous functions – and the process is driven by leap second information stored in alternate versions of the TZif files. Because a leap second adjustment may be needed even if no time zone correction is desired, calls to gmtime-like functions also need to consult a TZif file, conventionally named Etc/UTC (GMT in previous versions), to see whether leap second corrections are needed. To convert an application's time_t timestamps to or from POSIX time_t timestamps (for use when, say, embedding or interpreting timestamps in portable tar files), the application can call the utility functions time2posix and posix2time included with this package.

If the POSIX-compatible TZif file set is installed in a directory whose basename is zoneinfo, the leap-second-aware file set is by default installed in a separate directory zoneinfo-leaps. Although each process can have its own time zone by setting its TZ environment variable, there is no support for some processes being leap-second aware while other processes are POSIX-compatible; the leap-second choice is system-wide. So if you configure your kernel to count leap seconds, you should also discard zoneinfo and rename zoneinfo-leaps to zoneinfo. Alternatively, you can install just one set of TZif files in the first place; see the REDO variable in this package's makefile.

Calendrical issues

Calendrical issues are a bit out of scope for a time zone database, but they indicate the sort of problems that we would run into if we extended the time zone database further into the past. An excellent resource in this area is Edward M. Reingold and Nachum Dershowitz, Calendrical Calculations: The Ultimate Edition, Cambridge University Press (2018). Other information and sources are given in the file 'calendars' in the tz distribution. They sometimes disagree.

Time and time zones off Earth

The European Space Agency is considering the establishment of a reference timescale for the Moon, which has days roughly equivalent to 29.5 Earth days, and where relativistic effects cause clocks to tick slightly faster than on Earth.

Some people's work schedules have used Mars time. Jet Propulsion Laboratory (JPL) coordinators kept Mars time on and off during the Mars Pathfinder mission (1997). Some of their family members also adapted to Mars time. Dozens of special Mars watches were built for JPL workers who kept Mars time during the Mars Exploration Rovers (MER) mission (2004–2018). These timepieces looked like normal Seikos and Citizens but were adjusted to use Mars seconds rather than terrestrial seconds, although unfortunately the adjusted watches were unreliable and appear to have had only limited use.

A Mars solar day is called a "sol" and has a mean period equal to about 24 hours 39 minutes 35.244 seconds in terrestrial time. It is divided into a conventional 24-hour clock, so each Mars second equals about 1.02749125 terrestrial seconds. (One MER worker noted, "If I am working Mars hours, and Mars hours are 2.5% more than Earth hours, shouldn't I get an extra 2.5% pay raise?")

The prime meridian of Mars goes through the center of the crater Airy-0, named in honor of the British astronomer who built the Greenwich telescope that defines Earth's prime meridian. Mean solar time on the Mars prime meridian is called Mars Coordinated Time (MTC).

Each landed mission on Mars has adopted a different reference for solar timekeeping, so there is no real standard for Mars time zones. For example, the MER mission defined two time zones "Local Solar Time A" and "Local Solar Time B" for its two missions, each zone designed so that its time equals local true solar time at approximately the middle of the nominal mission. The A and B zones differ enough so that an MER worker assigned to the A zone might suffer "Mars lag" when switching to work in the B zone. Such a "time zone" is not particularly suited for any application other than the mission itself.

Many calendars have been proposed for Mars, but none have achieved wide acceptance. Astronomers often use Mars Sol Date (MSD) which is a sequential count of Mars solar days elapsed since about 1873-12-29 12:00 GMT.

In our solar system, Mars is the planet with time and calendar most like Earth's. On other planets, Sun-based time and calendars would work quite differently. For example, although Mercury's sidereal rotation period is 58.646 Earth days, Mercury revolves around the Sun so rapidly that an observer on Mercury's equator would see a sunrise only every 175.97 Earth days, i.e., a Mercury year is 0.5 of a Mercury day. Venus is more complicated, partly because its rotation is slightly retrograde: its year is 1.92 of its days. Gas giants like Jupiter are trickier still, as their polar and equatorial regions rotate at different rates, so that the length of a day depends on latitude. This effect is most pronounced on Neptune, where the day is about 12 hours at the poles and 18 hours at the equator.

Although the tz database does not support time on other planets, it is documented here in the hopes that support will be added eventually.

Sources for time on other planets:

• Michael Allison and Robert Schmunk, "Technical Notes on Mars Solar Time as Adopted by the Mars24 Sunclock" (2020-03-08).
• Zara Mirmalek, Making Time on Mars, MIT Press (March 2020), ISBN 978-0262043854.
• Jia-Rui Chong, "Workdays Fit for a Martian", Los Angeles Times (2004-01-14), pp A1, A20–A21.
• Tom Chmielewski, "Jet Lag Is Worse on Mars", The Atlantic (2015-02-26)
• Matt Williams, "How long is a day on the other planets of the solar system?" (2016-01-20).