Consider the function $$M_p(s,t)=\left(\frac{s^p+t^p}{2}\right)^{\frac{1}{p}}$$ where $s,t,p\in\mathbb{R}$.
The limit for $p$ going to zero is $M_0(s,t)=\sqrt{st}$.
To obtain it, I take log, apply l'hôpital's rule and then take exponential. Is this way of deriving the limit correct? Are there better (less clumsy) proofs?
Yes, deriving this limit using L'Hôpital's rule as you described is valid.