On OCT17@11:53, Paul Wouters wrote: > On Mon, 17 Oct 2022, Eliot Lear wrote: > > > Let's please leave Internet lawyering to lawyers. If people want a > > legal opinion on this draft, the IETF does have resources for that. > > But it is to the core of the ICANN / IETF divide, so IETF shouldn't wade > into ICANN territory. > > > We cannot assume that DNS will forever be the only Good approach and > > that all others will forever be Bad. Given that, we as a community are > > obligated to search for better, and to try new things. > > Sure. The IETF method is to start a BoF, form a WG, do your thing. See > Speedy/QUIC. See SSL/TLS, See PGP/OpenPGP. >
I am a bit confused now. Let's say there is BoF and, eventually, a WG. How is that WG supposed to "do it's thing" with respect to alternative name spaces if there is literally no way to safely use and test it's results alongside DNS? Wouldn't such a WG need the .alt as a prerequisite to any actual work or at least have .alt as its first work item? Would especially dnsop not care if another WG starts defining a protocol that is able to resolve names that look like/are DNS names because it is designed to possibly replace it in the future? BR > > There exist many registries for things the IETF doesn't recommend. One > > need look no further than TLS 1.3 crypto-suites as an example. > > These are not equivalent. For TLS to interop with non-IETF stuff, they > need codepoints for within the IETF defined TLS protocol. They are not > replacing TLS with something else on port 443. (and on top, non-TLS WG > entries are marked as NOT RECOMMENDED) > > > No matter what we say in the ALT draft, someone could burden the IETF > > with a new draft. People do so every day. If it gains sufficient > > support, it would have to be at least considered, no matter the topic. > > Sure, but "replacing DNS with something else", would definitely not be > in dnsop, or the ISE, but via BoF and a new WG. I dout any of the > alternatives would follow that approach, especially as I cannot see us > starting a new (replacement or not) naming scheme where there is a new > landrush for domain names with all of its ICANNlike associated problems. > > I can see DNS 2.0 that replaces all of DNS, but would still hook into > the existing EPP and RRR ICANN model though. Again, that would not go > via the ISE path. > > Paul > > _______________________________________________ > DNSOP mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/dnsop _______________________________________________ DNSOP mailing list [email protected] https://www.ietf.org/mailman/listinfo/dnsop
