ercadio @ercadio@social.cleverlibre.org

I've had so many nasty issues with `std::thread`. At this point I don't even think static analysis could keep up with the amount of ways you can mess it up. Even `std::jthread` which I thought would fix some issues, is probably even worse.

@marathon Let's try it
Let $r_1,\dots,r_m$ be the row vectors of $A$. If $c$ is in the row space of $A$, then there exists coefficients $\alpha_1,\dots,\alpha_m$
such that $c=\sum_{i=1}^m\alpha_i r_i$. Let $x$ be a feasible solution. Then
$$\begin{align*}
c^\top x &= \sum_{i=1}^m\alpha_i r_i^\top x\\
&=\begin{bmatrix}\alpha_1 & \dots & \alpha_m\end{bmatrix}Ax\\
&=\begin{bmatrix}\alpha_1 & \dots & \alpha_m\end{bmatrix}b
\end{align*}
$$
Since all feasible solutions have the same value, they are all optimal

@freemo I'll probably just generate pictures if needed then. I know tusky doesn't support either anyway

@freemo afaik it supports both server-side and client-side rendering
```javascript
katex.render("c = \\pm\\sqrt{a^2 + b^2}", element, {
throwOnError: false
});
```

@freemo Thanks! It's already a really cool feature. I recommend looking at KaTeX since it's apparently faster than mathjax

lol seems like it's a bit broken
\[
\mathrm{Let\ }G=\left\lt V,E\right\gt \ \mathrm{be\ a\ digraph}, c\in\mathbb{R}^{|V|}_{+}, r\in V\\ \begin{align*}\mathrm{min}\ &c^\top x\\ \mathrm{subject\ to}\\&\sum_{uv\ \in\ E}x_{uv}=1\ \forall\ v\in V\setminus\left\{r\right\}\\&x\geq0\ \mathrm{integer}\end{align*}
\]

never thought I'd want a free twitter alternative with math support
\[x^2\]

edmond's algorithm is my biggest brainfuck since combinatorics