> On 1 Mar 2021, at 20:08, Brian E Carpenter <[email protected]> > wrote: > > But it would take many years for it to be widely implemented and deployed, > during which time the Internet would be opaque to such addresses.
Answering this separately. You are correct that it would take years for it to be deployed on the big I internet, and it is a rollout of staggering proportions, However more and more traffic is going to flow between the client and a server embedded close to the edge. Furthermore there is a world developing where IP/UDP is used to tunnel from client application to server, and where there is no fixed constraint on what actually runs over that UDP/IP tunnel - it is a private matter for the application designer. If and it is a big if, there was a compelling competitive advantage in say mobile networks in using a new packet design to get from the application to the MEC server, then you could see a new design rolling out in a series of well controlled restricted domains. The point being that in such domains the overwhelming scale issues associated with the big I internet may not apply. The driver for this would need to be an application that had an overwhelming advantage if it used such a change. I am not yet sure I know what that application is, but given the huge network and application transitions we see I would not bet that such an application would never emerge, - Stewart
_______________________________________________ Int-area mailing list [email protected] https://www.ietf.org/mailman/listinfo/int-area
