On Oct 6, 2010, at 1:22 PM, Fernando Gont wrote: > On 06/10/2010 01:43 p.m., Keith Moore wrote: > >> Honestly, I don't think we can tell. In the short term, it certainly >> doesn't look good for end-to-end transparency. But unlike 10 years >> ago, today there's a widespread understanding of the problems caused >> by lack of transparency, and much less denial about it. > > It's not clear to me what you mean by "end to end transparency". If you > mean "end to end connectivity", then I'd say that quite a few people are > actually *concerned* about going back to end-to-end connectivity.
I mean having the sender's packets delivered to the receiver, completely intact except for ordinary TTL and IP option processing, with "best effort" or better reliability, delay, and jitter, except when prohibited by explicit end-user-specified policy. >> The central problem with the Internet seems to be that nearly >> everybody who routes traffic thinks it's okay to violate the >> architecture and alter the traffic to optimize for his/her specific >> circumstances - and the end users and their wide variety of >> applications just have to cope with the resulting brain damage. > > When applications that e.g. include point of attachment addresses in the > app protocol break in the presence of NATs, one should probably ask > whether the NAT is breaking the app, or whether the NAT is making it > clear that the app was actually already broken. It's perfectly reasonable for applications to include IP addresses and port numbers in their payloads, as this is the only way that the Internet Architecture defines to allow applications to make contact with particular processes at particular hosts. Some might see this as a deficiency in the Internet Architecture, but that's the best that we have to work with for now. DNS has never been, and never will be, suitable as a general endpoint naming mechanism. And so far nobody has managed to implement and deploy a better system for endpoint naming. If and when someone manages to do this, there will still be a need for old applications to use IP addresses. Meanwhile, those who insist on corrupting other parties' traffic and harming their applications are very good examples of the kind of short-term, self-serving harm to which I was referring. Keith _______________________________________________ Ietf mailing list [email protected] https://www.ietf.org/mailman/listinfo/ietf
