Skip to main content
10 events
when toggle format what by license comment
Oct 15, 2010 at 12:14 comment added Denis Serre David, if $n$ is not among the integers $1+\binom{m+1}{2}$, we may at least choose $m$ such that the dimension of the level set $s_1(M)=\cdots=s_m(M)=1$ is smaller than $1+2\binom{n}{2}$. Then it makes sense to speak of linking. $SO_n({\mathbb R})$ and $O_n^-({\mathbb R})$ are linked if every sub-manifold $X$ of dimension $1+\binom{n}{2}$, with boundary $SO_n({\mathbb R})$, intersects $O_n^-({\mathbb R})$. Is there any problem with this definition ?
Oct 14, 2010 at 14:23 comment added David E Speyer That also works.
Oct 14, 2010 at 13:57 comment added Denis Serre At first glance, the question makes sense if $n=1+\binom{m+1}{2}$ (that is $n=2,4,7,11,16,...$) and $V_n$ is defined by $s_1(M)=\cdots=s_m(M)=1$.
Oct 14, 2010 at 13:48 comment added David E Speyer But when $r=0$, we have $1-kr/(n-1)=1$. Doesn't that fix the question?
Oct 14, 2010 at 13:43 comment added Denis Serre @David. Actually, the choice of smaller $s_2,\ldots$ does not work, because the corresponding set does not contain $O_n({\mathbb R})$. Orthogonal matrices have $s_1=\cdots=s_n=1$, and are characterized by this property. I'll think to how fix the question.
Oct 14, 2010 at 13:36 comment added Denis Serre Thank you ! I agree with your calculation and your suggestion. After all, it was important to impose $s_1=1$, but for the choice of $s_2,\ldots,s_{n-1}$, I was lazy. I change my question accordingly.
Oct 14, 2010 at 13:34 history edited Denis Serre CC BY-SA 2.5
$V_n$ instead of $M_n(R)$ in the first sentence
Oct 14, 2010 at 13:25 history edited David E Speyer CC BY-SA 2.5
deleted 82 characters in body
Oct 14, 2010 at 13:08 history edited David E Speyer CC BY-SA 2.5
edited body
Oct 14, 2010 at 12:48 history answered David E Speyer CC BY-SA 2.5