> give me some more time,
Yes, I think time is needed here. No rush, as these are important things.
> here is what I was looking at yesterday ... a one level delegation with a
> "module/app cluster of CL" at the 2nd level. You still keep a valid
> "initiating CL" (per VM spec) but you can cycle the "defining loaders" as
> nodes under it....
Ah, ok, I see, you want an inverse delegation model where the top CL can ask
any of its subloaders for a class and simply picks the first one. That would
make the semantics clear, although it is still somewhat non-deterministic if
two apps have the same class and they're of different versions.
What you get is essentially a set model instead of a tree model. (of course,
since there's only one level, not counting the top one which doesn't really
do anything other than delegate).
Interesting. As a coincidence I think JMX has the same approach.
Yes, interesting.
The pro is that it is fairly simple, even to implement, and the main con is
that you can't have several versions of the same class within an
application. Which will probably won't be necessary for most cases buuut...
anyway..
> still research (so I am not saying I will find something) but give me some
> time...
Go ahead, you have the entire holidays to explore this :-) If I'm guessing
right with the above it would be an interesting concept.
You still need to have something to bind the EAR's together, but it would
basically be just a "I'm part of the logical app Foo" thingy.
> PS: also I strongly recommend that ACM paper I posted yesterday it really
> helps
Will do.
/Rickard
