[ since my last message on this subject was not approved for posting to this list, i am inflicting myself directly on you. apologies. ]
i am told that multiply signed roas are reappearing like whack-a-mole. they are as operationally useful as gears on a fish, and serve only to make it harder for the poor thing to swim. the fantasy that receiving two aggregatable prefixes from two different parents is as O(likely) as brownian motion causing all the molecules of air in the room to rush into one corner and stay there, and all the complexity it would produce in relying implementations would save the announcement of one prefix in the global routing table. wow! just announce the two prefixes. sheesh! if it would save the over 50% pollution of the routing table by idiocy, then i would leap to advocating it. but one bleeping prefix, and with a 0.00001% probability at that? you must be kidding. is life not hard enough already? notice how rirs' promises of delivery have already slipped by years. this is not easy stuff. do everything you can to simplify it, not complicate it. and surely we can find better ways to amuse ourselves by complicating things. like how about we propose a second and less efficient protocol for fetching repositories? randy _______________________________________________ sidr mailing list [email protected] https://www.ietf.org/mailman/listinfo/sidr
