Have enfolding cartographically input and render mainstream 'geographic information' within transformed spaces

[geoglrb]

Enfolding presently depends on the user to input 2d images--which are often maps, but not necessarily so. Enfolding then allows the user to depart from pre-rendered absolute space into more general types of spaces. But this can result in some strange outcomes. Textures, text, and linear features, among others, can end up looking quite distorted. This can be a feature or a bug.

What if enfolding allowed for the inputting of classical absolute-spatial data (e.g., GeoJSON or a Shapefile) that were subsequently transformed before any sort of cartographic rendering occurred?

In that case, enfolding could be given styling information along with the transformed data to input into some sort of renderer (Mapbox GL, Mapnik, etc.). This would yield visual outputs more proper to the transformed spaces users were generating in enfolding.

Using Mapnik to do the above would result in a one-off rasterization that three.js could view. Mapbox GL could be used to do that too, though its strength is in interactive vector output. Mapbox GL doesn't seem to really think in the generalized 3d terms we do, though, conceivably, someday, one could imagine having it interface with or replace three.js.

This is quite a departure and won't happen anytime soon.

[geoglrb]

A step in this would be implementing code that transforms a spatial data file from one coordinate system to another via a particular warping. For the way that a GImaginS might approach that, it's closely related to what @DOSull and I have been talking about under the rubric of 'hyperproj', an imagined modification of proj.4 to handle a more general set of analytical-cartographic transformations.