Joe Touch wrote: > > I think it is clear that there are several hard questions in this > space, > > and we simply do not know whether, e.g., some of the mapping ideas > > actually work well in real life. My belief is that finding out > requires > > not just implementation, but also well planned trials, experiments, > > simulation, or measurements. Otherwise its too easy to focus on > > relatively minor protocol details and miss the big picture. > > Absolutely. THAT step - setting up the experiments, trials, etc., seems > outside the scope of this WG, however. It's hard to talk about those > trials and experiments absent a documented protocol, however, which is > why this seems like a very useful step to everyone.
I disagree that it's out of scope. Specifically, if a doc is put forward as experimental, it is very helpful to state things like what the purpose of the experiment is, what the success criteria are, how long the experiment is expected to run for, etc. These sorts of things are appropriate in experimental documents (e.g., RFC 2770), especially when it is known what the potential issues are that one wants to experiment with. -Dave _______________________________________________ Int-area mailing list [email protected] https://www.ietf.org/mailman/listinfo/int-area
