1
$\begingroup$

Let $V$ be a projective variety defined over $\mathbb{F}_p$. For any extension of $\mathbb{F}_p$ let $V(k)$ be the $k$-points of $V$, that is, if $V$ is defined by the homogeneous polynomials $f_1,\ldots,f_r\in\mathbb{F}_p[X_0,\ldots,X_n]$, then $V(k)=\{(x_0:\ldots :x_n)\in k^n:f_i(x_0,\ldots,x_n)=0\ \forall i\}$. Then we can talk about dimension of $V(k)$ for any $k$ because dimension is just a concept of topological spaces and all $V(k)$ are topological spaces with the Zariski topology.

1-Are all the numbers $\dim V(k)$ equal? (assume $k$ runs over subfields of $\bar{\mathbb{F}_p}$). If not, when we read the expression "Let $V$ be a projective variety defined over $\mathbb{F}_p$ of dimension $r$", do we have to read $r=\dim V(\bar{\mathbb{F}_p})$?

2-Similarly I have read the expression "Let $V$ be a smooth projective variety defined over $\mathbb{F}_p$". The concept of smoothness I know is in term of dimension, that is, $V$ is smooth if the rank of the matrix $[\partial f_i/\partial x_j]$ in each $(a_0:\ldots:a_n)\in V$ is $n-\dim V$. So my question is which is the definition of that expression?.

$\endgroup$
  • $\begingroup$ If you take polynomials with no non-trivial roots over $\mathbb{F}_p$, then $V(\mathbb{F}_p)=\emptyset$, but there will always be some field extension $k$ such that $V(k)\neq \emptyset$. $\endgroup$ – Matt Feb 11 '13 at 23:43
2
$\begingroup$

If $k$ is finite, then $V(k)$ is finite, hence of dimension $\le 0$. The dimension of $V$ as an algebraic variety is the dimension of $V(\bar{\mathbb F}_p)$ with its Zariski topology. In the smoothness criterion, the dimension of $V$ is also defined this way.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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