Yeah I wanted to find something more equivalentish, I tried: - geodesic https://observablehq.com/d/d867e0687ae44ee5 - bucky https://observablehq.com/d/b4b375fcc73c6c16 - H3 https://observablehq.com/d/3e6237e97329c719 but none of these is as elegant as the spherical phyllotaxis (bravo!)

I changed the point distribution a bit so the poles would look less awkward, see https://observablehq.com/@fil/spherical-phyllotaxis and https://github.com/d3/d3-force/pull/129

my other trick i use sometimes is to abs(cos(latitude)) as many points in each "row" rounded to a power of 2. it's fun to use the bitwise operations to do that one :-D - I'll put together a quick example of that grid too. the benefit is it's a strict subset of the {lat,lon} lattice which can have computational benefits - but nowhere near as pretty as the phyllotaxis thanks for the great suggestion to separate the geometry from the data - merged!

Cf. http://extremelearning.com.au/how-to-evenly-distribute-points-on-a-sphere-more-effectively-than-the-canonical-fibonacci-lattice/