On 6/7/07, Eric Newcomer <[EMAIL PROTECTED]> wrote: > > > > Can we *please* stop this? > > Superior in one way does not mean superior in all ways, and in fact implies > strengths and weaknesses, which we all know by now is the case.
Of course, Eric. As Pete said, statelessness has a performance cost because it means larger messages. But we're just talking about reliability here, and specifically the impact of stateless communication on it in the general case. If anything I've said was construed as suggesting that statelessness should be used in all cases, then I apologize, because it clearly isn't. There's nothing religious or dogmatic about this AFAICT. It's just software architecture. Mark.
