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 want to ask the definition of a concept I see around but has not been able to find the definition for. Given a finite dimensional vector bundle $\pi : E \rightarrow K$ over a normed convex vector space K (with isomorphic fibers), what would be the definition of a continuous family of norms {$|.|_{x\in K}$}? I have two guesses but I don't know which one is correct, or if any of it is correct,

1- It is continuous as a map, $|.|: E \rightarrow R$ in the sense that for any sequence $v_n \rightarrow v$ in $E$ with $\pi(v_n)=x_n$, we have $|(v_n)_{x_n}|_{x_n}\rightarrow |(v)_{x}|_{x}$.

2- Or, first we choose a basis for each fiber $E_x$ write an isomorphism $T_{xy} :E_x \rightarrow E_y$, between each fibers. And then given a vector $v_x$ in the fiber $E_x$ and a sequence of points $x_k$ in K converging to x, $|T_{x_nx}(v_x)|_{x_n} \rightarrow |v_x|_{x} $

The difference between the two is whether if you fix vector component $v_x$ or now. I could ask the same question about continuous family of operators $T_x :E \rightarrow E$, is it continuity as a map from E to E or should we somehow again identify each fiber like in the second case above.


share|cite|improve this question
up vote 2 down vote accepted

One would be far more likely to define such a notion with something like your first definition, though the property people would really care about is that for any continuous section $\xi$ of your vector bundle $E \to X$, $x \mapsto |\xi(x)|$ defines a continuous map $X \to \mathbb{R}$. In particular, for continuity to make sense, all you need is for $X$ to be a topological space.

As for what actually tends to appear in practice, one usually doesn't just want a continuous family of norms but rather a metric, a continuous family $g$ of inner products in the sense that for any continuous sections $\xi$ and $\eta$ of $E \to X$, $x \mapsto g(x)(\xi(x),\eta(x))$ defines a continuous map $X \to \mathbb{R}$ (or $\mathbb{C}$, if $E \to X$ is complex). Again, you simply need $X$ to be a topological space. An example, of course, is the Riemannian metric on a Riemannian manifold $X$, which is a metric on the tangent bundle $TX \to X$, though more generally, in the case of smooth vector bundles over smooth manifolds, one usually wants metrics to be smooth in the same sense.

share|cite|improve this answer
On behalf of Finsler metrics, I raise an objection to "but rather a metric, a continuous family $g$ of inner products". After all, what is now known as a Finsler metric was the original definition of a metric by Riemann. – user53153 Feb 2 '13 at 3:59
@5PM I stand corrected, though I find it rather interesting that the notion of Finsler metric one wants involves a family of asymmetric norms, and not full-blown norms. – Branimir Ćaćić Feb 2 '13 at 23:55

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.