Question

Attempting to decide on how to implement our RPC on the server side.
From the documentation for the google RequestFactory it tells me there is 2 ways to implement an RPC service. http://code.google.com/webtoolkit/doc/latest/DevGuideRequestFactory.html

“Services can be implemented on the server in one of two ways: as static methods in a type or as instance methods in a service class accompanied by a ServiceLocator”

I did a simple service using the Static methods way and worked fine, job done. Using the service locator requires a lot more complexity. But is it that simple, anyone know what the down side, if any, to using static methods is ? To me the simplest way is usually the best. Might there be scalability issues or anything?

Thanks …

Answer 1

If you take a look at the employee example in the RequestFactory documentation, from the architectural point of view, does it make sense to have a class Employee, effectively a domain object, being responsible for persisting itself?

If you want to split those responsibilities, then the second approach is preferred, and as you said it is a bit more complex (but not that much).