[OSM-dev] Current Multipolygon definition of OSM considered harmfull!

[Frederik Ramm]

Hi,

  ah, computational geometry raises its head again.

>    I just stumbled over this page
>    http://wiki.openstreetmap.org/index.php/Relations/Multipolygon

Yes multipolygons are a kind of ugly hack and a misnomer. This has been
discussed countless times on this list. I think the person to bring
it up first was chrschmidt, perhaps even before they were introduced
I cant't remember.

Here are some other discussions about the same topic:
http://lists.openstreetmap.org/pipermail/dev/2008-February/008856.html
http://lists.openstreetmap.org/pipermail/talk/2008-March/023876.html

We are aware of the fact that our definition of multipolygon is
different from other people's definition and many of us think that
re-introducing an extra type for areas wouldn't be too bad.

But on the face of it, all this is just cosmetics. With our current
clumsy, ugly, badly-named method, you can express everything we
need to express. It's not that mappers all over the world throw their
hands in the air and say "oh dear we can't map this because we don't
have proper polygons!"

There's a huge number of things we can work at that really improve
life for mappers (and data users). Giving them proper polygons is 
*somewhere* on this list - nobody says it would be a bad thing - 
but it's not on page 1.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"