Alan DeKok <al...@deployingradius.com> writes: > Simon Josefsson wrote: >> I think the text proposed is mostly fine. I would say "normalization >> and/or comparison" instead of just "normalization" to allow for >> mechanisms that just specify comparison-based rules rather than >> normalization-based rules. So: > > The text looks OK to me. > > My only concern is that standardizing i18n in *each* possible EAP > document is duplicating work. If we end up with only one tunnel method > that is standardized, then there's less of a problem.
Standardizing multiple tunnel-based methods seems like a bad idea for many reasons. I agree that standardizing i18n in every document will lead to duplication of work (and duplication of discussions and the potential for mistakes), but placing too specific requirements on all future EAP documents today is similar to placing too specific crypto requirements on all future EAP documents today. The algorithms are evolving and there needs to be some element of agility in the requirements. If there is, any mistakes in this area can be fixed later on. /Simon _______________________________________________ Emu mailing list Emu@ietf.org https://www.ietf.org/mailman/listinfo/emu
