On Tue, Feb 9, 2016 at 12:01 AM, Yoav Nir <ynir.ietf@xxxxxxxxx> wrote: > >> On 9 Feb 2016, at 1:20 AM, Phillip Hallam-Baker <phill@xxxxxxxxxxxxxxx> wrote: >> >> On Mon, Feb 8, 2016 at 5:39 PM, Carsten Bormann <cabo@xxxxxxx> wrote: >>> Ronald Bonica wrote: >>>> The words "many" and "some" don't do justice to the conversation. https://tools.ietf.org/html/draft-ietf-v6ops-ipv6-ehs-in-real-world-02 provides more concrete numbers from real-world observation. >>> >>> Ah, but the result is much simpler. >>> >>> Some other real world data (Google QUIC experiments) already tell us >>> that a sizable part of the Internet (was it 7 %?) is not reachable via >>> UDP at all. This just ups that number slightly for IPv6 and UDP >>> protocols that don't have their own segmentation. >>> >>> UDP, it was nice to have known you. >> >> Maybe what we needed all along was a better TCP that allowed data to >> be sent on the first packet. > > And could be stateless on the server > > And you could get the same size data as the other side sent you without adding your own message layer. The stateless server is the real win for UDP and the reason I am very skeptical of the idea of using TCP in DPRIV. What rather puzzles me is that people don't seem to accept that a server can still be stateless if it receives one UDP request packet and returns ten (say). Obviously unreliable transport means that there is a limit to the number of packets that can be returned without some sort of flow/retransmit scheme. But it is a lot more than one.