Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have read a chapter about differentiability in two variables. I now have two questions:

  1. Why do we need the constraint that $|\vec{u}|=1$ when we calculate the directional derivative?
  2. Definition of gradient: $\nabla f = \frac{\partial f}{\partial x_1 }\mathbf{e}_1 + \cdots + \frac{\partial f}{\partial x_n }\mathbf{e}_n$ but why do we need f to be differetiable? Is it simply becuase we "take the partial derivatives"? I have learned the pure defn of differetiable, but maybe I don't have the "gut feeling" of what it really is, in simple language.
share|cite|improve this question
up vote 1 down vote accepted

1) You are free to choose $u$ not unitary. But you have to be careful. For example, if $|u|=2$ you get double the amount you get with $|u|=1$, as it means that you are "moving twice as fast".

For example, consider a function $f(x,y)$, and consider the "curve" $(x=t, y=0)$, which is a straight line along the $x$ axis.

How do we "see" $f$ from "inside" the curve? We get: $$ f(x(t), y(t)) = f(t,0). $$

The derivative of $f(t,0)$ with respect to $t$ is exactly the directional derivative in the direction of $x$.

Now pick the "curve" $(x=2t, y=0)$, which is a straight line like before, but now the "speed" is double. What happens now if you calculate: $$ f(x(t), y(t)) = f(2t,0), $$

and derive it with respect to $t$?

2) "Differentiable" intuitively means that "its partial derivatives exist and behave well". Which translates to: "if $f$ is not differentiable, it makes no sense to talk about its gradient."

share|cite|improve this answer
can you elaborate on 1) ? – jacob Apr 19 '14 at 15:13
@jacob Edited the question, adding an example...Let me know if it is clear! – geodude Apr 19 '14 at 15:22

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.