Re: Registration of text/calendar MIME property GEO_BOUNDS

[Doug Royer <doug@xxxxxxxxxxxxxx>]

Bruce_Kahn@xxxxxxxx wrote:
> 
> Doug in part wrote:
> >> Most of the time, yes.  Are there any weird cases out there, with
> >> overlapping time zones? For example, a country where some
> >> religious minority doesn't observe DST; then which time zone you
> >> want will depend on the user's religion.
> >
> >Correct time zones can overlap. If you get > 1, choose.
> 
> I was with ya up until the "If you get > 1, choose" bit.  This is a
> recipe for interop disaster/nightmare.  If I use CUA 1 I get 1 set of
> date/times and if I use CUA 2 to view the same entry I get another
> because it guessed differently?!?  Argh!  Dont do it!!

As you recall from RFC2445-7, a VTIMEZONE MUST be included
in a VCALENDAR when any component uses a TZID.

Adding GEO_BOUNDRY simply specifies (if included) in addition to
what alredy MUST be sent, the physical boundry of the TZID
supplied.

The only time a user must choose is when a CUA would query a CS
to get a TZID or list of TZID's to use. If they get > 1, then
the CU or CUA chooses which TZID MUST be sent
with the object. One VTIMEZONE for each unique TZID.

I have not proposed eliminating a VTIMEZONE as a MUST
in a VCALENDAR.