[OSM-talk] Topology (was: OSM the mediocre alternative)

[Lars Aronsson]

SteveC wrote:

> And yes you're right, topological is really useful since OSM is 
> a wiki and we track changes in nodes. Otherwise moving an 
> intersection of many roads would mean updating many linestrings 
> not one node.

Actually, we could still have topology at intersections and 
non-topology for the curvy linear roads between intersections. I 
don't know if this would bring any benefit in performance, but it 
certainly is an option worth considering.  It would make editing 
more complicated, though, as moving a "knee" on a linear curvy 
road would be different from moving an intersection "node".

However, the great lack in OSM is not whether our current 
implementation is this way or that way, but that most people here 
(it's a community issue, not a technical one) haven't heard about 
"topological" before they entered this project.  It's a word that 
brings up a great "huh?" and the FAQ "where do I read up on these 
GIS things?" doesn't even have a clear answer. In contrast, most 
who walk in through the developer door *have* heard about 
programming, operating systems, TCP/IP, sockets, XML, RPC, 
databases, SQL, ODBC and other things that we also need to know.


-- 
  Lars Aronsson (lars at aronsson.se)
  Aronsson Datateknik - http://aronsson.se