On Fri, Jul 07, 2017 at 08:09:30AM -0700, Paul Vixie wrote: > Nico Williams wrote: > >... > > > >I'm well aware that as to clients and servers, deploying new RR types is > >easy. The hard part is the management backend and UIs. Not all of them > >allow you to enter raw RDATA (hex-encoded or whatever). > > > >We've struggled with this in KITTEN WG. Deploying the URI RR type when > >you're using a hosting service can be anywhere from annoying (must enter > >raw RDATA) to impossible (the hosting service doesn't give a damn). I > >suppose it's just a matter of time; perhaps things have improved since > >we last looked. > > that would be a prior restraint, and bad for innovation. just like middle > boxes or the nearly universal assumption that ip means tcp or udp. i know > this kind of thinking is common. but there is a stark choice for all of us: > do all of our future protocol work on tcp/80 and make everything fit in > JSON, so that we can get our work done; or keep doing what the internet used > to mean, and eventually put enough stress on middle box makers, isp's, and > in your case dns hosting services, that they have to learn about the > internet rather than just the web. > > i know which future i'd rather live in. i also feel in-year pressure to get > my work done. i vacillate as to who gets to receive which burdens. I'm with you on this. I badly want to be able to use new RR types without further ado. I would prefer to push on the middleboxes and middlepeople -- but we need to do a better job of this altogether so that we don't have to wait a decade. We don't have a spec police, so all we can do is nag. Nico --