> | forcing apps to deal with scopes at all is a serious problem. > > This is one view. Another is that an address is just a number, > without some kind of scope to define its use, it is meaningless. > Even "global" needs something to say which globe is being considered. > > It can be argued that all apps should always deal with address scoping > and always should have.
yes, it *can* be argued. OTOH, I haven't seen a convincing argument for this. because if you want apps to work reliably under these conditions then you are essentially asking hosts to do routing in the absence of routing information - so ideally the hosts should listen to routing updates and do their own route computation. which sort of implies that they're attached to the network for long periods of time rather than being nomadic. and if you designed the network to work in this way then you'd want your higher-layer protocols to be able to tolerate multiple destination addresses and/or changes to the destination address (with all of the security issues that this implies). so no, the argument doesn't stand up to scrutiny. at least, not in anything resembling the current Internet. Keith -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
