Google Closure Templates vs JSP performances

Question

Recently I am playing with Google soy templates. I wonder if anyone has measured the performances of just using soys as templates (in the same way jsp or freemarker is used).

• Are the soy templates in any way faster then eg from JSP? I am interested in real-life usage of soy, like one where you also pass many parameters to soy context needed for the template.
• Can we see somehow the generated java classes for soy templates?

Answer 1

The primary reason to use Soy is to allow you to easily render either Server side or client side using the same templates. You don't have that flexibility with JSPs.
Also Soy has good support for proper escaping based on context, which results in a more secure site