On Apr 13, 2005, at 6:36 PM, Nadia Poulou wrote:While preparing this release, Nico Klasens and Andr� van Toly offered valuable help with input, reviews, comments etc. Since I am no developer/commitor and with their agreement, I name them as initiators of the vote.
Hi Nadia,
Naming committers is obviously not a valid way to call for a vote. You'll have to persuade one of them to call for a vote himself.
Still, you can contribute significantly to this process, doing the necessary groundwork, by submittng code and providing some introduction and motivation - you've made a good start here already.
This was exactly her question when she contacted us. She couldn't find any clear answer about who is allowed to call a vote. I searched for it too and could not find an answer either. We decided that she would make the vote and mention us as callers.
Some notes:
The dependency with the JGroups .jar is only in compile time, this causes thus no problems in runtime (for the case that the jgroups jar is not distributed).
I don't think this is true. The new classes do rely on the JGroups classes at runtime.
This is true as long as you don't load these classes into memory. When you don't use this implementation then you don't have problems in runtime when you don't have the jgroup jars. Distributing mmbase with these classes without the jars is possible. (Already point out by Michiel)
In a clumsy way the classnames have been appended with "JG" and the classes have moved to the *.change package, resulting in sources that cannot compile.
All implementations are in the *.change package. Easiest is to prefix the classes instead of postfix.
I don't know a lot about licenses, but I think Gerard has a point. We shouldn't introduce confusion about licenses, because that will only push people away.
Nico _______________________________________________ Developers mailing list [email protected] http://lists.mmbase.org/mailman/listinfo/developers
