In message <[EMAIL PROTECTED]> Russ White writes: > > Geoff Huston wrote: > > I'm against this exact match requirement - I think its a case of > > semantic overload in mixing basic authorities (permission to originate) > > with routing policy (explicit nomination of what prefixes to route). > > Agreed.... > > I could see someone wanting to say: "nothing longer than /x will be > advertised in this space," but I don't see any reason to make this an > exact match only. > > :-) > > Russ
Seems like an argument for the following expressiveness: prefix/len1[-len2] Geoff would have to use prefix/len1-24 (or 28?, or ??) for most prefixes, assuming that most prefixes would want some limit. If not, using prefix/len1-32 notation could be used to do what Jeff wants. Or some tool could display this as prefix/len1* to indicate any more specific being allowed. In this notation prefix/len would imply exact match. Obviously there is no need to store information in ascii string format. We want this expressiveness and maybe for display a common notation would be helpful but not manditory. It might be used in an implimentation's XML representation or it might not be (an alternative would be explicit prefix and len tags). Curtis _______________________________________________ Sidr mailing list [email protected] https://www1.ietf.org/mailman/listinfo/sidr
