> just to provide a "simple" answer to a simple question: > > NO - I do not agree with Keith's assertion as quoted. > YES - I agree with what I *believe* Keith wanted to say, which is that > "introduction of *site-scoped* addresses causes far more problems than it > solves". > > I do NOT believe that introduction of link-local addresses causes far more > problems than it solves. I see scoped addressing as a general problem, of which v6 site-locals and v4 link-locals are both examples. That said, I think v6 link-locals as "mostly harmless" provided that we use them only for things like router discovery. If we started expecting ordinary applications to be able to use them, I'd move them back to the "harmful" category - they'd have most of the problems of site-locals.