Lemma 78.18.2. Let $B \to S$ as in Section 78.3. Let $(U, R, s, t, c)$ be a groupoid in algebraic spaces over $B$.
If $s$ and $t$ are open, then for every open $W \subset U$ the open $s(t^{-1}(W))$ is $R$-invariant.
If $s$ and $t$ are open and quasi-compact, then $U$ has an open covering consisting of $R$-invariant quasi-compact open subspaces.
Proof. Assume $s$ and $t$ open and $W \subset U$ open. Since $s$ is open we see that $W' = s(t^{-1}(W))$ is an open subspace of $U$. Now it is quite easy to using the functorial point of view that this is an $R$-invariant open subset of $U$, but we are going to argue this directly by some diagrams, since we think it is instructive. Note that $t^{-1}(W')$ is the image of the morphism
and that $s^{-1}(W')$ is the image of the morphism
The algebraic spaces $A$, $B$ on the left of the arrows above are open subspaces of $R \times _{s, U, t} R$ and $R \times _{s, U, s} R$ respectively. By Lemma 78.11.4 the diagram
is commutative, and the horizontal arrow is an isomorphism. Moreover, it is clear that $(\text{pr}_1, c)(A) = B$. Hence we conclude $s^{-1}(W') = t^{-1}(W')$, and $W'$ is $R$-invariant. This proves (1).
Assume now that $s$, $t$ are both open and quasi-compact. Then, if $W \subset U$ is a quasi-compact open, then also $W' = s(t^{-1}(W))$ is a quasi-compact open, and invariant by the discussion above. Letting $W$ range over images of affines étale over $U$ we see (2). $\square$
Your email address will not be published. Required fields are marked.
In your comment you can use Markdown and LaTeX style mathematics (enclose it like $\pi$). A preview option is available if you wish to see how it works out (just click on the eye in the toolbar).
All contributions are licensed under the GNU Free Documentation License.
Comments (0)