0
$\begingroup$

$\DeclareMathOperator\span{span}$Denote two $p$-blades $\nu=v_1\wedge \dots \wedge v_p$ and $\omega=w_1\wedge \dots \wedge w_p$ $\in \bigwedge^p X$, where $X$ is an inner product space. How to define the inner product of $p$-blades $\langle \nu, \omega\rangle$ from the inner product of $X$ so that $$(1)\qquad \frac{\langle \nu, \omega\rangle}{\|\nu\|\|\omega\|}=1 \iff \span(v_1,\dots, v_p)=\span(w_1,\dots, w_p);$$ $$(2)\qquad\langle \nu, \omega\rangle=0 \iff \span(v_1,\dots, v_p)\perp \span(w_1,\dots, w_p).$$ I know one may define $\langle \nu, \omega\rangle=\det(\langle v_i,w_j\rangle)$, but this defintion doesn't satisfy (2). See the arXiv paper Grassmann angle formulas and identities by Mandolesi.

$\endgroup$

1 Answer 1

2
$\begingroup$

It is not possible. Let (i,j,k) be the canonical basis of $\mathrm{R}^3$, $\nu=i\wedge j$ and $\omega_t = (\cos (t) i + \sin(t) k)\wedge j$. For $t=0$ your condition (1) requires $\langle\nu,\omega_0\rangle=1$, and as $\omega_\pi = -\omega_0$ we have $\langle\nu,\omega_\pi\rangle=-1$. By continuity $\langle\nu,\omega_t\rangle=0$ for some $0<t<\pi$, but $\omega_t$ is never completely orthogonal to $\nu$.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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