[Zope3-dev] Strawman Proposal: Timezone as a location

[Guido van Rossum]

> In an effort to help simplify the requirements for datetime, I am putting
> forth a strawman proposal that some notion of timezone be treated as
> equivalent to a mapped location on the world. In other words, there is a
> one-to-one correspondence between a timezone code and a place on the earth.
> Likewise, no place on the earth belongs in more than one timezone at any
> given point in time.
> 
> I would argue that this notion of timezone location should be stored with a
> datetimetz object, in addition to tzoffset and tzname. (Perhaps tzlocation?)
> There should be 25 valid values ranging from -12 to +12. Simple date
> calculations would not alter the timezone location value.
> 
> In support of this argument, here are links to various timezone maps.
> 
> Free (as in GPL) World and US Time Zones:
> http://www.manifold.net/download/freemaps.html
> 
> Standard Time Zones Of The World, according to the United States Central
> Intelligence Agency (CIA):
> http://www.odci.gov/cia/publications/factbook/ref/pdf/802801.pdf
> 
> Feel free to shoot holes in this. :-)

I already have and I'm not going to repeat it.  Please try not to
limit how the tzinfo object can represent its step function from UTC
to tzoffset and tzname values.

Read http://webexhibits.org/daylightsaving/g.html if you don't believe
me.

--Guido van Rossum (home page: http://www.python.org/~guido/)