On 01/18/13 09:54, Shawn Steele wrote:
What has been the past policy for determining these abbreviations? Particularly for non-Latin scripts
The tz database uses English and ASCII, so (for example) even if the most common name for the time in mainland China is "北京时间", we can't use that in the tz database. Instead, we use "CST", which is short for the English phrase "China Standard Time". Similarly, for Bangladesh we can't use "বাংলাদেশ মান সময়" or an abbreviation, since that's not ASCII, so we need to use some abbreviation for "Bangladesh Standard Time". We try to look for the real-world consensus, which often involves looking at both official sources as well as unofficial but reliable sources such as newspapers that publish at least partly in English. There's no hard and fast rule here, but the basic idea is that we're trying to reflect common practice, not prescribe it.