On 2004-08-03, JINMEI Tatuya / [EMAIL PROTECTED]@C#:H wrote: > > If the unsolicited NAs as a signal did not cause any bad effect on > existing implementation, I'd also call that "interoperable" solution. > But they actually do cause a bad effect of increasing the odds of > disruption
... in a case so rare that users should be more concerned about being struck by lightning and/or asteroids. > On the other hand, the new ND option is really an "interoperable" > solution, since existing unmodified routers simply ignore the new > option (see, e.g., Section 6.1.1 of RFC2461) and this does not have > any bad effect on existing implementations. I guess there's "interoperable" and "interoperable". OptiDAD is interoperable with unmodified routers in that it actually works with them. This isn't the case for a new ND option. I actually quite like the new ND option alternative, it has the potential to get things done with less signalling overhead. But the aim of OptiDAD is to work with unmodified equipment, so that your OptiDAD capable MN will work with every Starbucks AR. Anyway, perhaps we can meet and discuss before the session. I'm presenting in MobOpts right now, I don't know what you look like so if you see me around (Monash t-shirt) say hello! I'd also be interested in other opinions ... somehow we've got to reach a rough consensus on this (we've already got running code ...) -----Nick -------------------------------------------------------------------- IETF IPv6 working group mailing list [EMAIL PROTECTED] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
