> I've always asked myself why we wouldn't try some quadtree algoritm for
> parsing/routing/filing our data...

As far as I understand quad-trees are useful if you want a fast query
to find you data.

However I wanted mainly a basic node-edges structure for those who
would like to play with routing algorithms (let's imagine a student
learning graph theory and who would like to try them on real data).
Personnaly, I just load all the data in memory so I have no access
speed problem (France is only 1M edges big).

If you want some spatial queries, postgis (using GIST-indexes) works
pretty well.
For a more portable app (let's say a desktop app), sqlite has a R-tree support.

Of course if you have a specific idea in mind (routing over very big
regions with little memory), then I don't know :)

> The day we heard that the IGN data would become "protected", and we
> weren't able to use it, we had a reason to be unhappy...

Well, now we have OSM ;)

_______________________________________________
Routing mailing list
[email protected]
http://lists.openstreetmap.org/listinfo/routing

Reply via email to