[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: VTIMEZONE and signular TZID



Frank_Dawson@xxxxxxxxx wrote:
> 
> Doug (and David):
> 
> I don't see that there is a problem with having numerous definitions
> for time zones with a different TZID or TZNAME but with the same time
> zone information.

I agree with you "over the wire". I was thinking of IANA registration
overhead. If one changes - how do the IANA people know they
are "THE SAME" as another? Unless we have some out of band
information? 

> 
> Time zones, by nature, are under local governmental control. If there
> are two  or more locals at the same UTC offset, then they would fall
> under this case.

I agree - Not the issue.  US/Pacific is simply an old name
for America/Los_Angeles. So it is NOT that I was asking for
all gmt-offset values that have the same offset to be the same name.
I was asking what do we do with the historical names?

> A given time zone may have had its definition changed over time. This
> would account for multiple definitions for the same time zone, also.

Yes - those are easy. If the name is different - we register its
different name. Each with there own unique data.

> Likewise, the duplicate definitions could result from a change in the
> name or identifier of one or more of these time zone definitions.

Exactly the question!

So how do we specify that XX == YY and only the name has changed?
Using VTIMEZONE restrictions as defined by the IETF docs?

> In any case, our role is merely to capture all the definitions in a common
>  format. Not to become time zone tsars and redefine the time zone
> definitions, names or whatever. Right?

We do NOT want to be the czar of timezones!

We do want clients to be able to use existing and historical names to
be able to understand and communicate iCalendar/VTIMEZONE data.

-Doug

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature