[OSM-talk] House numbers... One more suggestion

Frederik Ramm frederik at remote.org
Tue Jul 29 11:02:34 BST 2008


> I once had a look at the Karlsruhe Schema and i found it "ugly" from a
> technical point of view. In my world one wants explicit links between
> two objects but the proposal simply draws points or polygons "next to"
> something which is not a explicit link.

That's correct. Many people seem to think that house numbers should  
somehow be tagged to ways. But our thinking was vey practical: You  
see a house. The house as a number. You tag it. The house is there.  
The house will not move, or be removed, should the road be  
demolished. In the real world, the house does exist independently of  
the road where it is located.

Now if you think of our nodes not representing "house numbers" but  
"houses" (or more precisely, "house entrances"), then it might not  
look too ugly to you.

As far as I am aware, nobody ever proposed to put, say, a pharmacy  
node into to the way representing the road the pharmacy is on. Think  
of a house number node as a placeholder for a house, and of a house  
number interpolation line as a placeholder for a string of houses.  
Both will be replaced by proper building outlines in due time.

Tagging house numbers onto intersections or ways is just a lame  
workaround employed by many existing data sets because they can't  
model every single house. I have absolutely zero doubt that sooner or  
later, OSM *will* model every single house. At that time, every  
single house will simply be tagged with its address and that's it.  
The Karlsruhe Schema is intended as a workable interim solution until  
we reach that. With this schema, you can freely decide whether you  
want to map the full detail (every house mapped and tagged as  
individual building with address), or just put a node for each house,  
or just but nodes at the start and end of the road and connect them.  
If you opt for one of the coarser methods, someone else can easily  
refine them later.

But I'm just explaining why we do what we're doing; I'm not trying to  
talk you into doing the same if you don't see the merit for yourself.


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

More information about the talk mailing list