[bobbylarrybobby]

North being up doesn't make left west because left is relative to front, not up, and front can easily change directions whereas up cannot (at least not relative to the direction of gravity). It happens that when you're looking at a map mounted on a wall with north up, west is left, but if you were to turn 90° left yourself, then west would be straight (front). This is all consistent with north being up.

[matrss]

But if you define "up" as the direction opposite to where gravity pulls things (which very much makes sense) then there is no possible situation in which north being up would be correct either. A compass needle never points up. You are mounting a map of something horizontal onto a vertical surface, it is just as correct - or depending on the alignment of the wall arguably even more correct - to put east up. You can only ever get at most two directions correct, after all.

[ndriscoll]

Obviously "up" is the direction of of the normal vector for Earth's orbit around the Sun. North is then tilted but mostly up.

[matrss]

/s, right? Since the path of earth's orbit around the sun lies on a plane your decision to make the normal vector point in the same general direction of "north" seems arbitrary too. You could just as well call the negative of that the normal vector, and then south would be mostly up instead.

In any case, within the reference frame of earth that seems to be a bad definition. Contrary to popular belief, I am pretty sure that Australian's look up at the sky, not down.

[ndriscoll]

Mostly sarcasm (you already get North from Earth's rotation on its axis and don't need the Sun), but fixing "up" on a map (so that it doesn't rotate as you move around) makes sense, and agreeing with general mathematical convention is convenient so that we don't need to teach more exceptions. If we're going to flip the sign of something it should be electric charge.