The vote result was clear: drop 1.3 and move ASAP to Java 5 and
generics. However, I don't like mob rule which will leave a minority
out in the cold. Therefore I want to wait with closing 1.3 until after
1.3.3. This will give us a chance to triage the most blocking issues
for 1.3 and fix them. We will branch 1.3.x immediately after the 1.3.3
release and start work in Wicket With Generics (WWG) in trunk.

The process I think we should do is the following:

 * assign issues to 1.3.3 that are high priority and can be solved in
a timely manner
 * assign all other issues to 1.4M1
 * fix the 1.3.3 issues and release 1.3.3
 * when the release is approved, branch 1.3.x
 * trunk is now Wicket 1.4
 * apply generics, flatten directory hierarchy, fix build scripts
 * fix issues as we go along (if possible apply to 1.3.x)
 * release 1.4-M1
 * fix the generics stuff issues discovered
 * release 1.4-M2
 * go into release candidate mode
 * release 1.4-rc1 until 1.4-rcx ...
 * release 1.4 final

In the mean time we could release 1.3.4, 1.3.5 if necessary.

There is one thing I'd like to extend our 1.4 release with: replacing
wicket.util.concurrent with the JDK API. This removes the additional
license burden and these classes.

What goes into 1.4-M1?
 - generics
 - removal of wicket.util.concurrent
 - flattening of directory structure
 - all of Wicket compiled with Java 5
 - fixes to release script to accommodate for the flattening
 - bug fixes from 1.3

What goes into 1.4-M2?
 - fixes onto generics API (in case we didn't get it right the first time)
 - fixes for concurrent stuff (in case we didn't get it right the first time)
 - more bug fixes from 1.3 era

Martijn

-- 
Buy Wicket in Action: http://manning.com/dashorst
Apache Wicket 1.3.2 is released
Get it now: http://www.apache.org/dyn/closer.cgi/wicket/1.3.2

Reply via email to