quinncnl
commented
6 months ago Hi @atorger , thanks for your feedback. This falls into the country specific rendering domain. We already have some special treatment for Nordic countries and Australia. See comparison https://mc.bbbike.org/mc/?lon=11.855571&lat=61.196854&zoom=6&num=2&mt0=mapnik&mt1=tracestrack-lang-default But it's only limited to place names at the moment. We can do some investigation on showing even more details on lower zooms for sparsely populated areas.
A common problem with OSM-based maps compared to traditional maps is that when zooming out you don't get a good overview and it's hard to orientate. An example is shown below at zoom level 12.
The area in the example is not 100% fully mapped, but in all places where the forest cover is, all the same details and tags is there as is in the map from the Swedish land survey authority.
Note the difference in names, roads, streams, buildings etc.
Position is: https://www.openstreetmap.org/#map=12/65.3843/16.1348&layers=P
This is just a broad suggestion, and I don't know if it even is possible with current pipeline with mapnik etc. I know that we cannot expect to get cartographic generalization (simpler/clearer geometries when zoomed out), but I don't think that is a big issue, the big problem is that most names are not seen until zoomed in very far, small roads, paths and streams are removed far too early etc. This might be good for New York or other dense areas, but in outdoor maps when there are sparsely placed small hamlets, the map becomes pretty bad in terms of being useful in zoom levels less than 14. Buildings not showing up until zoom level 17 also feels a bit odd.
One key(?) problem is that if we look at 1 km in Sweden and then 1 km in central India, the distance is about 2,2 times longer in Sweden due to the projection. That is a static map style that works with metropolitan areas near the equator won't work that well in rural areas nearer the poles. I'm not sure if it's possible with the current tools used to adapt the style depending on actual data density in the map, but that is probably what is required to make overviews that work everywhere in the world.