[tz] User time zones

[Alan Barrett]

On Tue, 06 Dec 2011, Steven Abner wrote:
> I could use some input. I have a different database, lots of 
> reasons for it, which has the attached file as part of the 
> database.

You didn't explain what you were trying to do, so I had to figure 
it out from the file attachment.  Let me try to explain it for 
others.

If I have correctly interpreted your intention, you want to 
make it easy for software to interpret time zone abbreviations 
in user input, and to that end you suggest making a lot of 
abbreviation-based time zones to supplement the existing 
location-based time zones.

Essentially, for each abbreviation that appears in the existing 
location-based time zone data, you would create another time zone 
whose name is based on the abbreviation and whose definition is 
a link to the location, and you would have some sort of method 
for resolving conflicts where the same abbreviation is used by 
multiple time zones.  You would also add some other abbreviations 
with other simple definitions, such as the "military" time zones A 
to Z.

For example, given the existing definition of Pacific/Chuuk:

# Zone  NAME            GMTOFF  RULES   FORMAT  [UNTIL]
Zone Pacific/Chuuk      10:07:08 -      LMT     1901
                         10:00   -       CHUT           # Chuuk Time

and given that the abbreviation "CHUT" is unique,
you would add this link:

Link Pacific/Chuuk      Abbr/CHUT

Once you have this, software that receives the string "CHUT" in a 
context where a time zone name would be appropriate, can try using 
the "Abbr/CHUT" time zone to make sense of it.

This seems fairly useful to me, despite the obvious problems with 
non-unique abbreviations.

--apb (Alan Barrett)