Carl Banks wrote: > Come on, I don't think anyone's under the impression we're being > indiscriminate here. Ok, but I don't think that in the case at hand we should recommend a metaclass solution. Michele Simionato