Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Does metric determines derivatives?

  1. Apr 16, 2003 #1
    If we have a metric could we define a derivative?..in fact the derivative would be Lim D(f(x+h),f(x))
  2. jcsd
  3. Apr 16, 2003 #2


    User Avatar
    Gold Member

    Covariant derivative, metric compatibility, Levi-Civita connections, etc.
  4. Apr 18, 2003 #3
    I am not sure what you are exactly meant by "metric" here.
    If a vector space is equipped with "length" (I am using more intuitive notion.) it is called metric space. If you consider a set of nice functions over some domain they usually form a vector space. If this space has notion of norm, (and being complete) then it is called Banach space. With this norm you can define a derivative by usual calculus text book way. ( In usual book way absolute value of real number plays the roll of norm). In metric space you can use this metric to define a norm.

    Metric on Riemannian manifold is slightly different concept even though they are of course related. You can define a covariant derivative with any affine connections on Riemannian manifold. Fumdamental theorem of Riemannian geometry says, then, there is a unique covariant derivative that is compatible with metric. i.e. you can use metric to fix your affine connection to define covariant derivative.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook