Two comments. First, as she alludes to in her edit, the "rotated arrow" picture that RobotRollCall uses here is subtly backward. It does suggest the right thing

Two comments. First, as she alludes to in her edit, the "rotated arrow" picture that RobotRollCall uses here is subtly backward. It does suggest the right things and I used to think of it that way, but it eventually gets you in trouble. (In actuality, as you start to move through space, your motion through time speeds up... but this still leaves your arrow the same length because the geometry of space-time is hyperbolic: the Pythagorean theorem reads "x^2 - t^2 = c^2" instead of "x^2 + t^2 = c^2". This ends up avoiding LOTS of issues, some of which were stumbling blocks for people in the comments to her post. But it's certainly harder to visualize!) Second, several people have complained that RRC avoided the underlying question by saying "the arrow is always the same length". I think they may not be giving that answer the credit it deserves. Her claim isn't that the speed of light is the longest possible arrow (which I agree wouldn't help at all), but rather that every object's arrow has exactly this same length. That shifts the speed of light from being an arbitrary constraint to being simply a label for this universal fact. The question "Why is every arrow the same length?" is still valid, but there's much less reason to worry about it: no known process can change that length, just as no known process can change the rest mass of an electron.