Re: [External] Request for information - Challenges in routing related to semantic addressing

[Robert Raszuk <[e-mail]>]

>
> Similarly, I'm aware of some (many?) ways to use the 64 bit interface
> identifier, some of which embed one-off tokens in the field in an extreme
> case of privacy addressing.


Indeed and I am also aware of such use cases.

See I am actually not sure what this discussion is all about :)  Orthodox
v6-ers claim that you only can assign IPv6 addresses to the interfaces. And
frankly this is ok. Except since early moments router's interfaces were
both physical and logical. And in the latter case logical interface is
nothing else then instantiation of some semantics. It can be simple
loopback, it can be virtual tunnel termination, it can be dial-up, it also
can be service chain anchor. In modern slide decks it is called SR
endpoint.

So last 64 bits give you great place to express to the destination which
attracts traffic to such address on how you want this packet to be handled.

Such indication can be implicit or explicit - I think some people do not
like to see it explicit. But honestly it should be in best interest of the
IETF to make it as explicit as folks want to disclose it for better
interoperability across vendors, wider adoption, solicitation of more
comments etc ... .

Happy Easter Holidays to all,
Robert