[OSM-talk] Mapnik has busstops

[Schuyler Erle]

* On  3-May-2007 at  3:01PM PDT, Martin Spott said:
> 
> The current case with OSM is a bit different: There's an unambiguous
> way for two-way transformation of OSM's lat/lon numbers into WKT
> representation of an OGC-style database - what you call "spatial
> columns". This means the basic data model can remain unchanged and
> you'll still profit from the WKT representation - including the
> indexing that is attached to it. Only small adjustment to the SELECT
> and INSERT/UPDATE statemens buys you a big advantage here without being
> exposed to the risk of redundant data storage.

FWIW this is exactly what I am attempting with the OSM-on-Rails-
on-PostGIS experiment. All this speculation about triggers-good /
triggers-bad is interesting, but the only way we'll find out is to try
it and see. ;-)

SDE