Is it possible to use $$\frac{\sin \pi s}{\pi s}=\prod _{n\ge 1}\left(1-\frac{s^2}{n^2}\right)$$to get a closed form for $\zeta(2n)$?
I am told that it is possible by using Taylor expansion. I can see that expanding the LHS gives $\zeta(2)$ as the coefficient of $s^2$ which makes that case easy to handle. But the other coefficients are not so straightforward. For example, is it at least possible to get the value of $\zeta(4)$ from this approach?