How to 'normalize' Sphere-Coordinates? ( [0° long., 110° lat.] == [180° long., 70° lat.] != [0° long., 70° lat.] )

[Spammed]

The Blog-Post Range-Validated Quantity Points got me thinking.

I am reluctant to regard a latitude greater than 90° as physically meaningless.
Presumably because I do not view latitude in isolation, but implicitly as part of a coordinate pair used to describe a location on a sphere.

I welcome the ‘1-Dimensional Range’ validated (not to say normalized) quantity points.

Unfortunately, it seems to me that for 2 (or more) dimensions, one apparently needs correspondingly multidimensional ‘normalisation’.

[mpusz]

That is great feedback, @Spammed. I totally agree with you that for many use cases, the approach proposed in the article may not be enough. I was looking for the best use cases for each policy to describe it in the article, but maybe we can find even better ones. If so, please do let me know.

Regarding multidimensional coordinates, I think the burden of normalization should fall on users or their representation type. Since multidimensional coordinates are generally not ordered, the library can't perform any overflow checks on them. For more advanced use cases, we can try to design an even more advanced geo_position type that will do some smart normalization, but I think this is outside of the scope of the article and maybe even the library, as this type, for now, resides in the examples directory as a purely illustrative way of using the library.