Patches for faster interface dispatch, type checking

[Godmar Back]

> 
> Are there any other arguments in favour of this "lazy" method-existance
> checking?
> 

Actually, I don't want to remain on the books as being in favor of 
this either.  I agree with Per and you.

The only concern I have, and again this is something we hit in Kaffe,
is that you won't be able to run pieces of software because you're
throwing exception too eagerly for missing pieces you don't actually 
invoke/need.  In this case (incompatible interface implementations), 
this shouldn't be the case.

	- Godmar