I am reading journal of number theory and the paper quites a result.
$$ \text{Ext}^1(\mathbb{Q}, \mathbb{Z}) = \hat{\mathbb{Z}}/\mathbb{Z} = \mathbb{A}_f / \mathbb{Q} $$
Unfortunately, before I prove this result, this statement contains a number of concepts. What is is the Ext group measuring in this context? How do we construct elements of $\hat{\mathbb{Z}}$ which is the profinite completion of $\mathbb{Z}$.
The paper also mentions a group action of $\text{PSL}_2(\mathbb{Q})$ on $\mathbb{A}_f/ \mathbb{Q}$ which looks like the quotient of the finite adeles by rationals under multiplication: $$ z \cdot \left( \begin{array}{cc} a & b \\ c & d\end{array} \right) = \frac{b + dz}{a + cz} \in \mathbb{A}_f / \mathbb{Q} $$
There is a note (and blog) where they walk us though the proof of $(*)$. I still have a lot of questions. Here are a random few observations from it:
$\mathbb{Q} \otimes \mathbb{Q} \simeq \mathbb{Q}$
$\mathbb{Z}/n \otimes \mathbb{Q} = 0$
$G \otimes H \simeq H \otimes G$ this seems trivially true.
$\mathbb{Q}/\mathbb{Z} \simeq \bigoplus_p \mathbb{Z}/p^\infty$ where $\mathbb{Z}/p^\infty := \mathbb{Z}[p^{-1}]/\mathbb{Z}$
There are some discussion of exact sequences:
- $\text{Hom}(\mathbb{Q}, - )$ is a functor.
- $ 0 \to \mathbb{Z} \to \mathbb{Q} \to \mathbb{Q}/\mathbb{Z} \to 0 $ this exact sequence (of additive groups, $(\mathbb{Z},+)$ and $(\mathbb{Q},+)$) seems.
That these objects exists at all is quite interesting. Looks on the surface quite elementary (e.g. unique factorization).
Maybe related: