On Fri, May 25, 2012 at 11:09 AM, Lester Caine <lester@lsces.co.uk> wrote:
Tobias Conradi wrote:
By using only [A-Za-z0-9] in name components and "/" as component separator, the IANA time zone database would allow easy transformation from IANA time zone names into class names or ids and backward.
Currently having increasing fun with an insane attempt to restrict names to some simple 'English' subset's of characters in a couple of other areas, I think the discussion should be on ALLOWING the full Unicode character set. This is the 21st century and many more users do not even speak English! And many more do every day. But it doesn't matter. What matters is if people can type. Many people cannot type the full range of UNICODE easily. But everyone writing HTML, C, Java, python can write [A-Za-z0-9._-]
And I am English and only speak English ... I don't speak only English, but I speak almost no Indonesian or Afrikaans, still I can type it easily, since both only use the ISO basic Latin alphabet AFAIK.
More at: http://en.wikipedia.org/wiki/ISO_basic_Latin_alphabet#Equivalent_alphabets
Personally I could make a case for using ISO 3166 country codes and perhaps augmenting that with an area code, so that in addition to English versions of zone names, local translation tables could be introduced in a standard way? But this probably a separate project to the time zone data :( CLDR uses UN/LOCODEs which have the ISO 3166-1 alpha-2 country code as the first two characters. But country codes can change, so this would lead either to divergence of the codes as CLDR does, or to zone name change caused by ISO 3166-1 alpha-2 changes
Some environments don't allow "/". Since "/" is part of the zone name, a replacement may be needed.
The use of "/" makes perfect sense since it groups zones into natural folders and sub-folders. It's up to you code to decide what to do with it.
That folders are natural is something I have never heard before. -- Tobias Conradi Rheinsberger Str. 18 10115 Berlin Germany http://tobiasconradi.com/