eap Mailing List: RE: [Issue 203] Comments on EAP-Peer state machine

[Nick Petroni (npetroni]

> [Joe] I guess what I am really compaining about is that there seems to
> be an overabundance of signals, some of which seem to be internal to the
> method.  I would like to see the minimal set of signals used between the
> method and the EAP state machine.  Perhaps we already have the minimal
> set of signals, but there seems to be a lot (1 boolean and two
> tri-states).  I think this can be simplified to 4 different signals from
> the method, IGNORE, COND_SUCCESS, SUCCESS, FAILURE.   However the
> machine is sufficiently complex that I could be missing something.

No, you understand it perfectly. I guess the difference in opinion is
whether or not there is any value in showing  more detail. I think the
first question would be if there is such value (and I believe Yoshihiro's
point was that there may be). The second question is whether the current
representation provides both the appropriate level of extra information
and the accuracy of that information.

It could very well be that the best state machine has the least amount of
variables and just more text describing what might happen in a method.

nick