If a facade approach is used to support the legacy branch, that development could occur in many other places than the log4j_1_2 branch. It could be done within slf4j.org like the JDK 1.4 facade, it could be part of a log4j 1.x release but could also support some earlier log4j.jar, or it could be done as part of the CVS HEAD and back ported as necessary.
The CVS HEAD currently contains an integrated implementation of slf4j which could be used to establish the performance cost of a facade implementation. If a 1.x minor release were to include an integrated implementation, it would likely be back ported from the then current implementation for the CVS HEAD.
Since the cost to reestablish a 1.2_slf4j branch if ever needed is small, I don't see any compelling reason to have the branch until active development is imminent.
On May 4, 2005, at 12:33 PM, Ceki G�lc� wrote:
I am less optimistic than you. In my experience, once cast, many committers albeit not all, would first drink lethal poison before reverting their -1 vote. Whatever the reasons given today, those who cast -1 votes will continue to do so under various pretexts, which may be justified, unjustified, honest, disguised or otherwise. In short, it is practically impossible for a rejected proposal to get subsequently accepted. In any case, I won't be presenting a proposal. Now if a heroic soul wishes to try, I wish them all the luck in the world.
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
