Lemma 99.16.5. In Situation 99.16.3. Denote $\mathcal{X} = \mathcal{C}\! \mathit{omplexes}_{X/B}$. Then $\Delta : \mathcal{X} \to \mathcal{X} \times \mathcal{X}$ is representable by algebraic spaces.
Proof. Consider two objects $x = (T, g, E)$ and $y = (T, g', E')$ of $\mathcal{X}$ over a scheme $T$. We have to show that $\mathit{Isom}_\mathcal {X}(x, y)$ is an algebraic space over $T$, see Algebraic Stacks, Lemma 94.10.11. If for $h : T' \to T$ the restrictions $x|_{T'}$ and $y|_{T'}$ are isomorphic in the fibre category $\mathcal{X}_{T'}$, then $g \circ h = g' \circ h$. Hence there is a transformation of presheaves
Since the diagonal of $B$ is representable (by schemes) this equalizer is a scheme. Thus we may replace $T$ by this equalizer and $E$ and $E'$ by their pullbacks. Thus we may assume $g = g'$.
Assume $g = g'$. After replacing $B$ by $T$ and $X$ by $X_ T$ we arrive at the following problem. Given $E, E' \in D(\mathcal{O}_ X)$ satisfying conditions (1), (2) of Lemma 99.16.2 we have to show that $\mathit{Isom}(E, E')$ is an algebraic space. Here $\mathit{Isom}(E, E')$ is the functor
where $E_ T$ and $E'_ T$ are the derived pullbacks of $E$ and $E'$ to $X_ T$. Now, let $W \subset B$, resp. $W' \subset B$ be the open subspace of $B$ associated to $E, E'$, resp. to $E', E$ by Lemma 99.16.1. Clearly, if there exists an isomorphism $E_ T \to E'_ T$ as in the definition of $\mathit{Isom}(E, E')$, then we see that $T \to B$ factors into both $W$ and $W'$ (because we have condition (1) for $E$ and $E'$ and we'll obviously have $E_ t \cong E'_ t$ so no nonzero maps $E_ t[i] \to E_ t$ or $E'_ t[i] \to E_ t$ over the fibre $X_ t$ for $i > 0$. Thus we may replace $B$ by the open $W \cap W'$. In this case the functor $H = \mathop{\mathcal{H}\! \mathit{om}}\nolimits (E, E')$
is an algebraic space affine and of finite presentation over $B$ by Lemma 99.16.1. The same is true for $H' = \mathop{\mathcal{H}\! \mathit{om}}\nolimits (E', E)$, $I = \mathop{\mathcal{H}\! \mathit{om}}\nolimits (E, E)$, and $I' = \mathop{\mathcal{H}\! \mathit{om}}\nolimits (E', E')$. Therefore we can repeat the argument of the proof of Proposition 99.4.3 to see that
for some morphisms $c$ and $\sigma $. Thus $\mathit{Isom}(E, E')$ is an algebraic space. $\square$
Post a comment
Your email address will not be published.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)
There are also: