If a function is analytic there are two ways to represent its differential. In the twenty-first century we would like to have an animation that goes along with our description: not for that we think our readers are stupid enough to confuse an effect with proof, but because looking away from a surface in order to think of a plane is a trick we're hard-pressed to explain the necessity of.
Any routine that builds a picture must have an orientation. If we 'trust the math' to the degree that we say 'transformation' without being quite sure how to explain a non-linear function, or use something that then is as explicable to us as the weather in order to show that an explanation is available in our mother tongue, we're unlikely to know how to count bytes. So that we're not only relying on things people thought about before they had computers to play with, we're also relying on what only became available when computer circuits became a thousand times smaller than the smallest virus (according to my sources).
Credulity is not a sin; but it is one to call a man a fool. As an example we look at something any fool can solve: assume we have a struct containing coordinates, and a routine which moves a set of structs to the right of a gui without making them overlap. Referring to coordinates by their ordinal brings us to looking at linear spaces before we've drawn anything; and anyhow, an index parameter introduces inefficiency that can't be justified. Switching coordinates, then, if we don't want to have two languages--which is truly what the preprocessor gives us--which is all we need to do in order to move the set of structs to the bottom of the gui, requires duplication that seems to us to be unnecessary. We do know that without an ordinal, or index, parameter, the object-native code will need to be different for each case.
Now let's talk about the legibility of code.
I don't know about you, but I reckon we aught to get it on the screen, first.
Note that the introductory statement is not quite correct, but those who understand what I'm saying are unlikely to miss my point.