Hello, i am writing to you about an issue with Nicosia the capital of Cyprus. Cyprus is an EU member since May 1st 2004 which makes it a European country but you have it as an Asia Continent Country. Before May 1st 2004 Cyprus had Asia Timezone but since then we have Eastern European Time and we have also adopted the Eastern European Summer Time. Basically you representation of Nicosia as UTC+2 and UTC+3 (in summer) time is correct but in the wrong Continent. Cyprus as a European Union Member: https://europa.eu/european-union/about-eu/countries/member-countries/cyprus_... Map of Europe Union including Cyprus: https://europa.eu/european-union/about-eu/countries_en?country=CY#map Waiting for you reply Thank you GsaTitanium
If you prefer to think of Nicosia as being in Europe you can set TZ="Europe/Nicosia", which is equivalent to setting TZ="Asia/Nicosia". This is covered in the 'asia' file, which says: # Classically, Cyprus belongs to Asia; e.g. see Herodotus, Histories, I.72. # However, for various reasons many users expect to find it under Europe. Link Asia/Nicosia Europe/Nicosia
Hello mr.Eggert Thank you for replying, i appreciate it. I have read both your Asia/Nicosia and Europe/Nicosia files. Europe file it just say to refer to the Asia one. Now, I don’t think of Nicosia and general Cyprus as being in Europe because i like it. It is the law. The links i sent you proves what i am saying to you and you can find all the informations you need. The “classically, Cyprus belongs to Asia; e.g. see Herodotus, Histories, I.72.” is technically wrong. For the time it was written Cyprus was considered in Asia. But as the 21st century, Cyprus is in the EU and recognized by the UN. Politically i am correct. Eastern Europe Time and Eastern Europe Summer time are the standard times of the Republic of Cyprus. Thank you GsaTitanium Sent from my iPhone
On 23 May 2018, at 23:33, Paul Eggert <eggert@cs.ucla.edu> wrote:
If you prefer to think of Nicosia as being in Europe you can set TZ="Europe/Nicosia", which is equivalent to setting TZ="Asia/Nicosia". This is covered in the 'asia' file, which says:
# Classically, Cyprus belongs to Asia; e.g. see Herodotus, Histories, I.72. # However, for various reasons many users expect to find it under Europe. Link Asia/Nicosia Europe/Nicosia
On 05/23/2018 04:43 PM, George Savva wrote:
I don’t think of Nicosia and general Cyprus as being in Europe because i like it. It is the law.
That's fine, as tzdb is not intended to take or endorse any position on legal issues. If tzdb had authoritative legal effect in Cyprus or Crimea or Jerusalem or any of dozens of other areas of conflict, we would never be able to get our work done. Luckily for us, tzdb is merely a low-level tool to help programs set civil-time clocks. Names like "Europe/Nicosia" and "Europe/Istanbul" are not supposed to be visible to any but experienced users, and applications are expected to instead display names that users are likely to recognize and understand. For more on this subject, please see: https://data.iana.org/time-zones/theory.html#naming which points to the Unicode Common Locale Data Repository (CLDR) here: http://cldr.unicode.org/ The CLDR gives the strings "Λευκωσία" and "Κωνσταντινούπολη" as the suggested way to display the abovementioned time zones for Greek-language user interfaces. Errors in these strings can be reported as described here: https://www.unicode.org/contacts.html
Hello mr.Eggert Thank you for replying. i have checked CLDR. Under their Time Zone section say and i quote: "The time zone IDs (tzid) are language-independent, and follow the TZ time zone database [Olson <https://unicode.org/reports/tr35/tr35.html#Olson>] and naming conventions.” Link: https://unicode.org/reports/tr35/tr35-dates.html#Time_Zone_Names <https://unicode.org/reports/tr35/tr35-dates.html#Time_Zone_Names> Where Olson is referring to iana tz database. Link: https://unicode.org/reports/tr35/tr35.html#Olson <https://unicode.org/reports/tr35/tr35.html#Olson> CLDR say that by locale display name may very. Basically what you said to me. But their core data are getting it from iana. So basically iana data: Asia/Nicosia. must change to Europe/Nicosia. I understand that core data must not be disabled to any user and that the locale names must be showed, but developers are skipping the process to include the locale data, resulting to display Asia/Nicosia in some programs. Are they wrong for not including locale data? Yes they are. But i am trying to resolve the issue from the root, in this case is iana. I undestand that you do not like to change data and that your data are based in geological locations and not political. But in this case your data are misleading. i know that you are not willing to change your data or the way you collect them so i will end this conversation here. Again thank you for replying to me, i appreciate it. I hope in the near future you change you way of doing business before a new organisation come and become the new standard for tz database. GsaTitanium
On 24 May 2018, at 3:34 AM, Paul Eggert <eggert@cs.ucla.edu> wrote:
On 05/23/2018 04:43 PM, George Savva wrote:
I don’t think of Nicosia and general Cyprus as being in Europe because i like it. It is the law.
That's fine, as tzdb is not intended to take or endorse any position on legal issues. If tzdb had authoritative legal effect in Cyprus or Crimea or Jerusalem or any of dozens of other areas of conflict, we would never be able to get our work done. Luckily for us, tzdb is merely a low-level tool to help programs set civil-time clocks. Names like "Europe/Nicosia" and "Europe/Istanbul" are not supposed to be visible to any but experienced users, and applications are expected to instead display names that users are likely to recognize and understand. For more on this subject, please see:
https://data.iana.org/time-zones/theory.html#naming
which points to the Unicode Common Locale Data Repository (CLDR) here:
The CLDR gives the strings "Λευκωσία" and "Κωνσταντινούπολη" as the suggested way to display the abovementioned time zones for Greek-language user interfaces. Errors in these strings can be reported as described here:
These sorts of name complaints are among the most common messages this list gets other than legitimate updates to time zones. There's a whole FAQ dedicated to it, and a large part of the point of doing it this way is to avoid endless squabbling about unimportant details or conflicts between laws and cultures about what things are called, etc. The fact that a lot of them are "incorrect" by some standard is unimportant - what matters is that *a* decision was made. I suspect it would be greatly appreciated by the members of the TZ database project if a second organization wanted to step in and maintain a second, complementary time zone database competing to be the standard. That sounds sarcastic mostly because it's so unlikely, but don't take it with any sort of bitterness. A second group of maintainers also working on time zone standards would be a great check on this project and, if they were aiming to be a "user friendly" version of the tz database might draw some of these discussions onto their mailing list. In any case, you may want to read some of the endless discussions in the archives of this mailing list to see the arguments that despite this policy we have at least partially failed to avoid. The one I recall most often is the Kiev/Kyiv debate, so you can maybe use that as a search term. Best, Paul (G, not E) On 05/24/2018 01:45 PM, GsaTitanium wrote:
Hello mr.Eggert
Thank you for replying.
i have checked CLDR. Under their Time Zone section say and i quote: "The time zone IDs (tzid) are language-independent, and follow the TZ time zone database [Olson <https://unicode.org/reports/tr35/tr35.html#Olson>] and naming conventions.” Link: https://unicode.org/reports/tr35/tr35-dates.html#Time_Zone_Names <https://unicode.org/reports/tr35/tr35-dates.html#Time_Zone_Names>
Where Olson is referring to iana tz database. Link: https://unicode.org/reports/tr35/tr35.html#Olson <https://unicode.org/reports/tr35/tr35.html#Olson>
CLDR say that by locale display name may very. Basically what you said to me. But their core data are getting it from iana. So basically iana data: Asia/Nicosia. must change to Europe/Nicosia.
I understand that core data must not be disabled to any user and that the locale names must be showed, but developers are skipping the process to include the locale data, resulting to display Asia/Nicosia in some programs.
Are they wrong for not including locale data? Yes they are. But i am trying to resolve the issue from the root, in this case is iana.
I undestand that you do not like to change data and that your data are based in geological locations and not political. But in this case your data are misleading.
i know that you are not willing to change your data or the way you collect them so i will end this conversation here.
Again thank you for replying to me, i appreciate it. I hope in the near future you change you way of doing business before a new organisation come and become the new standard for tz database.
GsaTitanium
On 24 May 2018, at 3:34 AM, Paul Eggert <eggert@cs.ucla.edu> wrote:
On 05/23/2018 04:43 PM, George Savva wrote:
I don’t think of Nicosia and general Cyprus as being in Europe because i like it. It is the law.
That's fine, as tzdb is not intended to take or endorse any position on legal issues. If tzdb had authoritative legal effect in Cyprus or Crimea or Jerusalem or any of dozens of other areas of conflict, we would never be able to get our work done. Luckily for us, tzdb is merely a low-level tool to help programs set civil-time clocks. Names like "Europe/Nicosia" and "Europe/Istanbul" are not supposed to be visible to any but experienced users, and applications are expected to instead display names that users are likely to recognize and understand. For more on this subject, please see:
https://data.iana.org/time-zones/theory.html#naming
which points to the Unicode Common Locale Data Repository (CLDR) here:
The CLDR gives the strings "Λευκωσία" and "Κωνσταντινούπολη" as the suggested way to display the abovementioned time zones for Greek-language user interfaces. Errors in these strings can be reported as described here:
On May 24, 2018, at 10:45 AM, GsaTitanium <gsatitanium@gmail.com> wrote:
i have checked CLDR. Under their Time Zone section say and i quote: "The time zone IDs (tzid) are language-independent, and follow the TZ time zone database [Olson] and naming conventions.” Link: https://unicode.org/reports/tr35/tr35-dates.html#Time_Zone_Names
Where Olson is referring to iana tz database. Link: https://unicode.org/reports/tr35/tr35.html#Olson
CLDR say that by locale display name may very. Basically what you said to me. But their core data are getting it from iana. So basically iana data: Asia/Nicosia. must change to Europe/Nicosia.
"Asia/Nicosia" and "Europe/Nicosia" are tzids. As Paul Eggert explained, they are not intended to be shown to end users. The CLDR gives the display names; *those* are what should be shown to end users. CLDR does *not* get the display names from IANA, so the IANA names don't need to change in order to make the CLDR display names correct.
I understand that core data must not be disabled to any user and that the locale names must be showed, but developers are skipping the process to include the locale data, resulting to display Asia/Nicosia in some programs.
This is why I've suggested switching to using some form of very-definitely-not-intended-for-humans-to-be-able-to-read IDs for tzdb zones, so that any developer who "[skips] the process" ends up with an app offering choices such as "67382" or "SKB78" to end users, the end users complain, and they're forced to stop offering those choices (or, if it's some small embedded device that can't support the CLDR database, to have a printed manual giving tzids for locations, or whatever).
Are they wrong for not including locale data? Yes they are. But i am trying to resolve the issue from the root, in this case is iana.
If "the issue" is end users being forced to say "Shanghai" when they're in Beijing, or to say "Kiev" when they're in Київ, or to say "Asia/Nicosia" when they're in an EU country, one can quite validly argue that the root of the problem is in the software being, as you say, wrong in using the tzid rather than the CLDR display name (or, even better, offering a list of cities and towns, in the locale's language and script, so they don't have to choose a tzid region, they can choose a nearby city or town), rather than being the IDs that the IANA TZDB arbitrarily chooses not being convenient for end users.
I undestand that you do not like to change data and that your data are based in geological locations and not political. But in this case your data are misleading.
Misleading to end users forced to use them. Again, I view this as being the fault of the software developers, not the TZDB maintainers. (As far as I'm concerned, *everybody* developing tzdb region selecting software should be taken to a nearby Apple Store and shown macOS's selector, in detail, and told "OK, your job is now to do something that works as well as this, if not better.)
Again thank you for replying to me, i appreciate it. I hope in the near future you change you way of doing business before a new organisation come and become the new standard for tz database.
Perhaps many of the people here would enjoy watching some new organization try to do that, especially if it means *they* would no longer have to spend time and energy maintaining the database and dealing with zone naming complaints.
On Thu, May 24, 2018 at 1:43 AM, George Savva <gsatitanium@gmail.com> wrote:
Now, I don’t think of Nicosia and general Cyprus as being in Europe because i like it. It is the law. The links i sent you proves what i am saying to you and you can find all the informations you need.
You are making confusion between Europe and the European Union. They are different things, even if related. Cyprus is an example of a country which is in the European Union, but not in Europe. Switzerland is an example of a country which is not in the European Union, but it is in Europe. BTW, laws do not have the power to change geographycal facts. Or do we have to move UK from Europe to some other continent shortly? which one? PB
And staying close to Cyprus, Turkey is not a member of the EU, but is a member of NATO. North_Atlantic/Istanbul? On Fri, 25 May 2018, 7:30 am Pierpaolo Bernardi, <olopierpa@gmail.com> wrote:
On Thu, May 24, 2018 at 1:43 AM, George Savva <gsatitanium@gmail.com> wrote:
Now, I don’t think of Nicosia and general Cyprus as being in Europe because i like it. It is the law. The links i sent you proves what i am saying to you and you can find all the informations you need.
You are making confusion between Europe and the European Union. They are different things, even if related.
Cyprus is an example of a country which is in the European Union, but not in Europe.
Switzerland is an example of a country which is not in the European Union, but it is in Europe.
BTW, laws do not have the power to change geographycal facts. Or do we have to move UK from Europe to some other continent shortly? which one?
PB
On 2018-05-24 18:12, Sanjeev Gupta wrote:
On Fri, 25 May 2018, 7:30 am Pierpaolo Bernardi wrote:
On Thu, May 24, 2018 at 1:43 AM, George Savva wrote:
Now, I don’t think of Nicosia and general Cyprus as being in Europe because i like it. It is the law. The links i sent you proves what i am saying to you and you can find all the informations you need. You are making confusion between Europe and the European Union. They are different things, even if related. Cyprus is an example of a country which is in the European Union, but not in Europe. Switzerland is an example of a country which is not in the European Union, but it is in Europe. BTW, laws do not have the power to change geographycal facts. Or do we have to move UK from Europe to some other continent shortly? which one? And staying close to Cyprus, Turkey is not a member of the EU, but is a member of NATO. North_Atlantic/Istanbul?
Turkey and Istanbul are in both eastern Europe and western Asia i.e. $ ls /usr/share/zoneinfo/*/Istanbul /usr/share/zoneinfo/Asia/Istanbul /usr/share/zoneinfo/Europe/Istanbul Even the division and number of continents is cultural; see: https://en.wikipedia.org/wiki/Continent#Number for another issue over which some may not agree. -- Take care. Thanks, Brian Inglis, Calgary, Alberta, Canada
Brian Inglis wrote:
Turkey and Istanbul are in both eastern Europe and western Asia i.e.
$ ls /usr/share/zoneinfo/*/Istanbul /usr/share/zoneinfo/Asia/Istanbul /usr/share/zoneinfo/Europe/Istanbul
A historical note for tzdb nerds. In 1993 when I originally wrote the entry for Istanbul, I naively thought that the city was primarily in Asia. But after I looked it up, I found that geographically the city was more European, e.g., its European side was larger and its ancient heart is on the Europe side. To avoid similar confusion by others, when I renamed my private draft's zone to Europe/Istanbul I added a link to the Asia name. I figured that Istanbul would present a unique problem in Zone naming as it is the only major city that straddles continental boundaries. Little did I know, eh?
On May 26, 2018, at 13:35, Paul Eggert <eggert@cs.ucla.edu> wrote:
I figured that Istanbul would present a unique problem in Zone naming as it is the only major city that straddles continental boundaries. Little did I know, eh?
Perhaps a ‘Constantinople’ entry? <ducking> Cheers! |----------------------------------------------------------------------| | Frederick F. Gleason, Jr. | Chief Developer | | | Paravel Systems | |----------------------------------------------------------------------| | But in our enthusiasm, we could not resist a radical overhaul of the | | system, in which all of its major weaknesses have been exposed, | | analyzed, and replaced with new weaknesses. | | -- Bruce Leverett | | "Register Allocation in Optimizing Compilers" | |----------------------------------------------------------------------|
participants (9)
-
Brian Inglis -
Fred Gleason -
George Savva -
GsaTitanium
-
Guy Harris -
Paul Eggert -
Paul G -
Pierpaolo Bernardi -
Sanjeev Gupta