[Tagging] Mapping time zones as geometries (relations)
kevin.b.kenny+osm at gmail.com
Tue Mar 7 01:06:28 UTC 2017
The timezone database does not map the geographic extents of timezones.
Rather, it lists them by name and gives the rules for what time they keep.
The geography is explicitly out of scope, so the tzdata maintainers really
shouldn't have anything to say about OSM mapping time zones!
There are reasonable use cases for wanting to take a timezone name and get
back a multipolygon, or to get a list of name and multipolygon for all the
timezones (or all the timezones on a given map). That's hard to do unless
there are relations for the timezones. Not impossible, we could query
administrative boundaries for the tag, but that's likely to be SLOW since
the tag will at best be in hstore and not indexed. Grouping disjoint items,
such as "all the administrative regions in a time zone" seems tailor-made
It seems that having a 'group' relation for all the administrative regions
that use a given timezone would be useful. The timezone data itself, of
course, belongs separate, but the group relation would, to my thinking, be
more useful than merely tagging each administrative region with the
timezone it keeps.
As someone who imports tzdata into a third-party system on a very regular
basis, I can verify that while the rules for the timezones undergo constant
churn - there are 20-30 releases of the database in a typical year - the
boundaries change hardly at all, except for splits when one administrative
region goes one way and another region goes another. In those cases, we
already very likely have the administrative boundaries, and merely need to
move bits from one relation to another.
The mapping between tzdata names and geography really isn't available
anywhere else, and it's surely something that I'd use if OSM had it.
On Mon, Mar 6, 2017 at 6:30 PM, Aun Johnsen <lists at gimnechiske.org> wrote:
> > On 3/6/17 11:21 AM, Paul Johnson wrote:
> >> On Mar 5, 2017 18:30, "Frederik Ramm" <frederik at remote.org
> >> <mailto:frederik at remote.org>> wrote:
> >> Hi,
> >> I would like to start a discussion about the mapping of time
> >> What do you think?
> >> I'm generally opposed to mapping timezones in OpenStreetMap unless the
> >> tzdata maintainers are 100% on board. Since timezones are a royal
> >> pain to keep track of, often changing 100+ times a year, on as little
> >> as a few hours notice in some cases.
> > i agree. this a perfect example of something that belongs in its own
> > database or
> > shape file, available to be overlaid on the map when it's wanted.
> > richard
> > --
> > rwelty at averillpark.net
> > Averill Park Networking - GIS & IT Consulting
> > OpenStreetMap - PostgreSQL - Linux
> > Java - Web Applications - Search
> Actually, timezones are to great extent already mapped, you only have to
> find a better way to extract the data, instead of creating new relations.
> There might still be administrative relations that doesn’t have timezone
> tag set making the timezone data incomplete, but this is not a reason to
> create new relations, but rather add the missing data in existing relations.
> Aun Johnsen
> Tagging mailing list
> Tagging at openstreetmap.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Tagging