Could anyone just clear up a question regarding openstreetmap data. Are there different OSM data sets for the various set scales, or do the data get generalized on-the-fly when the map is zoomed out to a smaller scale? I understand that the many objects are hidden in smaller scales, but are there also different generalized datasets for the smaller scales as well?




What you're seeing is primarily different styles for the map, with scale-dependent rules in Mapnik - the OSM project itself does not maintain any scale-separated data. It's all the same roads, but some of them appear earlier and later in the map style.

The only minor exception to this is coastlines, which are shown at lower-detail levels with simplified boundaries, and at high levels of detail more complex, detailed coastlines are used.


What you see in openstreetmap.org is the images that have been made from OSM data. The images are rendered by a tile map server. the images in the server are made for each zoom level, according to the generalization rules, i.e. at smaller scale, objects are hidden. so when you go to smaller zoom levels, new images are requested and not generalized. there is only a single dataset for whole world.

p.s. sometimes the images are stretched when you zoom in, but never combined(generalized) on zooming out


There are some generalisations: In lower zoom levels, two different simpified coastline datasets are used, as well as border lines. These are read from shapefiles, to speed up rendering compared to large-area postgis queries.

Furthermore, roads are generalized and stored in a seperate table planet_roads for low zooms levels. For higher zoom levels, planet_line is used.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.