Hi Paul,
>> After reading the arguments of Bill: What is the need to place the
>> annotations on the implemented interfaces or superclasses, and not on
>> the implementing class?
> I asked myself the same question :-) I would like to better understand
> what is driving this use case.
first read, than write, Mr. Koops :-) :
https://jsr311.dev.java.net/issues/show_bug.cgi?id=16
According to this the ideas were
1. to isolate the published interface of the service. But IMO that's an
invalid argument: The published interface is HTTP (as you said on
JavaPolis 2007, see
http://www.parleys.com/display/PARLEYS/Home#slide=1;talk=9044049;title=JSR%20311%20-%20JAX-RS%20The%20Java%20API%20for%20RESTful%20Web%20Services)
2. The other argument was, that this "also allow the EJB specification
to adopt REST and be able to publish remote, local, and restful
interfaces to the outside world". That's an argument we could follow,
IMO. But I don't no much about EJB 3.x, so there are others who could
give more established statements than me.
best regards
Stephan