[dns-operations] A dns-proxy for DNS over HTTP(s)

[Mark Delany]

On 27Aug15, Matthew Ghali allegedly wrote:
> On Aug 26, 2015, at 9:26 PM, Mark Delany <x2t at foxtrot.emu.st> wrote:
> >> 
> > My thesis is that there has been little incentive to reduce the cost
> > of TCP state but that DNS-over-TCP/HTTP presents such an incentive.
> 
> I suspect that anyone working on QUIC at Google would take issue with that.

Heh. I doubt that QUIC is going to reduce state per connection
compared to TCP. I would expect quite the opposite given multiple
independent streams per session, complicated compression semantics,
built-in encryption and more. QUIC semantics are vastly more
complicated than TCP which suggests lots of state.


Mark.