> > This approach has, I believe, some significant advantages, especially if one > can do a good job determining the best fit, as it can significantly reduce > the number of terms that need to be iterated and it also collapses lat/lon > into a single field which also helps improve search times. The downside is > it often takes up more index space due to the many zoom levels. > > Does this match up with other people's understanding? >
Yes, this is a good summary of the approach (as I understand it) However, 2^n tiles is not a requirement, just the approach taken. MGRS breaks the world based on UTM zones with sizes 100M, 10M 1M etc. and everything you describe would work with this encoding. ryan --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
