It looks like the boys at Microsoft Research implemented my idea: tiled vectors. Their demo makes one thing obvious, however. Douglas-Peucker just doesn’t cut it for topologies. This becomes a non-issue with a pixel-level of detail, but just looks bad in their demo. Their implementation doesn’t seem to preserve existing points between zooms; it’s probably just as efficient this way. Via Vish. (Check out Idevio, while you’re at it.)
Winners: Me! Virtual Earth stole my idea!! (LOL… “my” idea)
Losers: Me! Virtual Earth stole my idea!!
February 1, 2008 at 6:36 pm |
Can you blame them? It was a good idea.
I think I’m going to stake a claim on that triangle between Utah and Nevada.
February 1, 2008 at 10:28 pm |
It works well but I agree the demo leaves much to be desired. Maybe I missed a post, but what did you discover about the topology, or is it just an assume issue at this point?
February 2, 2008 at 6:12 pm |
NetTopologySuite has a Topology Preserving Simplifier but its implementation conflicted with some of my ideas on preserving data between zoom levels. If you used “dumb” tile sets like MS Research did, this wouldn’t be an issue. I’ll dig that code up and see if I can’t put something decent together worth posting.