Date: Fri, 16 Feb 2001 15:47:48 -0500 From: "Olson, Arthur David (NCI)" <olsona@dc37a.nci.nih.gov> Message-ID: <4E0F9DDCA724D411B2120008C75DB71DD0F593@nihexchange2.nih.gov> I agree with the generality of what you say, but unfortunately ... | topmost geographic location to phone system area code | middle phone system area code to time zone identifier that particular layering won't work. As well as regularly altering its time zone rules, Australia also just recently altered all its phone system area codes. Now we have just 4 area codes for the country. Anyone following this discussion will know that we have more than 4 timezones (using the tzdata definition of a timezone anyway). Eg: South Aus, NT, and West Aus are all the same phone area code, but are three different timezones. Vic and Tas the same. NSW and (I think) Lord Howe Island the same (I'm not certain what area code LHI uses, but I think it will be the same as NSW). If Qld ever decides to do the rational thing and introduce summer time in its south east corner (one part of Aus, along with north east NSW of course) where it really would be useful, then that would be another. Now if you extended the phone system identifier to include the exchange prefix, things have a better chance at working, but now you have a nightmare of administration (as those things spring up new almost daily). And even then, there's no guarantees - phone numbers are allocated to suit the requirements of the phone system (and to a lesser degree, its users...) any relationship with timezones is purely accidental. The relationship really needs to be directly from gegraphical location (perhaps postal area code, or zip code - those are usually small enough so they will probably be within a timezone - though even there, they are assigned to suit the post office, not the sun) to timezone, and from the same thing to phone area code (if that is needed). kre