With 3.0 getting really close, I do realize this is kind of last minute, but 
wanted to throw these out real quick.

Would anyone object to removing all the SOAP over TCP stuff that is currently 
in the SOAP binding?  It was a project started by a GSoC student several years 
ago.  However, due to the complexity and poor documentation, it was not 
“finished” in time.   No one has ever picked it up to finish it.  Thus, it’s 
incomplete, it doesn’t work correctly, won’t actually interoperate with 
anything, etc…  It pretty much just results in a bunch of extra classes in the 
soap binding, a few extra “provided” deps in pom, etc….    Plus, it never 
caught on.    If someone DOES want to pick it up in the future, the code could 
be resurrected from GIT.   I just don’t see that happening.  (in addition, 
there is the SOAP over Websockets thing from Microsoft which encompasses  much 
of the same thing, but using Websockets and would work with recent .NET things)

Any thoughts about the Java6/Java7 support level?   This *IS* a “.0” release 
which could be a good time to consider this.   I really don’t care either way 
at this point, but I kind of expect that by 3.1 or 3.2, we’ll want to drop 
Java6 anyway due to dependencies starting to require it.  (example: Jetty 9 
requires Java7)    Anyway, something to think about.    I’d be OK sticking with 
Java6 and saying we’ll go Java7 for one of the later releases.      I suppose 
one thought is to keep Java6 for 3.0 so we have one version of CXF that support 
JAX-RS 2.0 and runs on Java6.   

Related to that, what about CXF 2.6?   Once 3.0 is release, do we want to do a 
“final” 2.6.x and stop doing regular releases on that branch?   Doing so would 
allow removing all the Java5 JDK’s which is certainly something I’m keen on.   
Never got Java5 working on my Mac.  :-)     That said, it’s also the only 
branch we currently have that support JAX-RS 1.1.   


Thoughts on the above?

-- 
Daniel Kulp
[email protected] - http://dankulp.com/blog
Talend Community Coder - http://coders.talend.com

Reply via email to