Squared distance is useful for both performance, and guaranteeing whole numbers on grid-aligned sample points. Need to figure out a nice way to encode this.
May be best to introduce a squared equivalent of the Point field, then parametrize fields that use it as a basis. This would also open up the possibility of compositions like a 'chebyshev line' or 'taxicab capsule'.
Raises the question of standardization; do any of the existing fields implement manual length measurement, i.e. in a different vector space or such? Has the potential to violate user assumptions if it lacks a standardized interface.
Squared distance is useful for both performance, and guaranteeing whole numbers on grid-aligned sample points. Need to figure out a nice way to encode this.
May be best to introduce a squared equivalent of the Point field, then parametrize fields that use it as a basis. This would also open up the possibility of compositions like a 'chebyshev line' or 'taxicab capsule'.
Raises the question of standardization; do any of the existing fields implement manual length measurement, i.e. in a different vector space or such? Has the potential to violate user assumptions if it lacks a standardized interface.