An excellent post by Benjamin Carlyle (on the REST-discuss group):
REST pushes unformity of verbs and message structure at the network level, and in fact forces us to rethink the application level significantly. If the uniform version works, why waste time inventing a more specific or type-safe version? Why validate the whole data structure down in some network stack when you can transport the raw data to where it is used and only process what is needed when it arrives? Why insist on converting a message into an intricate structure of data when you are just going to reencode it again for transmission or display it to a user? Why not just retain it as strings until you know what you will do with it?
I value the passage where at zero state RPC and REST start converging. The REST representation however looks better. I need to handle the error condition anyway, regardless if it is thrown as an exception or represented as 404. I belief that the zero-state services are going to be introduced and also to be used. The fact that Googles translation service is not RESTFull does not mean that it is not USEfull.