On Nov 14, 2003, at 9:28 AM, Geir Magnusson Jr. wrote:
I'm staring at this, and now have to go back and read the letter from the lawyers, because the code for
org.apache.geronimo.core.service.Invocation is a riff on java.util.Map, namely :
public interface Invocation {
Object get(InvocationKey key); void put(InvocationKey key, Object value); }
Where the jboss version is a fairly rich concrete class. I'm not going to waste bits posting it, but here's the link :
http://cvs.sourceforge.net/viewcvs.py/jboss/jboss/src/main/org/jboss/ invocation/Invocation.java?rev=1.19&view=markup
So this can't be what they are upset about.
Can someone explain what JBoss is talking about?
I'm not sure... It's my understanding that the 3 named payloads
are "highly JBoss-specific". These no longer exist in Geronimo.
Also, there was the implied impression that Invocation itself
(the concept as well as the implementation) was somehow "unique"
to JBoss, that it's origin was totally within JBoss and that no one
other than current JBoss developers have touched it. Indications are
that this is almost certainly not the case; it is instead a collaborative
work derived from Richard's MethodInvocation class. It is curious that
one of the 1st changes to that class was the addition of Marc as
an author (presumably by Marc himself).
