This is related to #3. While Web Mercator is a widely used projection, it is recognised as a very low-quality CRS because of the poorly chosen parameters It would be ideal to permit (some) different projections, so as to at least have the chance of getting participation in this initiative from government mapping authorities. Where the profusion of different map projections reduces interoperability, we should be able to support a globe as the base perspective, perhaps.
It seems to me that globes and planar projections are two 'rendering modes'. When the perspective is that of globe, coordinates have to be transformed onto the sphere and then rendered; when viewing a planar projection map, the coordinates are rendered using a simpler transformation.
Not sure if this is a use case, but use cases arise from it, e.g. mashups require the data to share coordinate systems, either beforehand or by client processing. The latter gets into performance issues and is why GIS is going to remain important.
Support tile sets defined in different coordinate systems, and the ability to draw vector features defined as lat/lon points onto maps defined with those coordinate systems.
Support re-projecting maps defined in one projection into another projection or a 3D globe view.
Let's focus on discussing (1) here, and focus on (2) in #3.
This issue is for discussion of the map viewer capability "Display map tiles defined in various common coordinate systems".
This is related to #3. While Web Mercator is a widely used projection, it is recognised as a very low-quality CRS because of the poorly chosen parameters It would be ideal to permit (some) different projections, so as to at least have the chance of getting participation in this initiative from government mapping authorities. Where the profusion of different map projections reduces interoperability, we should be able to support a globe as the base perspective, perhaps.
It seems to me that globes and planar projections are two 'rendering modes'. When the perspective is that of globe, coordinates have to be transformed onto the sphere and then rendered; when viewing a planar projection map, the coordinates are rendered using a simpler transformation.
Not sure if this is a use case, but use cases arise from it, e.g. mashups require the data to share coordinate systems, either beforehand or by client processing. The latter gets into performance issues and is why GIS is going to remain important.