On 6/27/22 22:05, John Levine wrote:
But there is a great deal of software that expects the names it uses to look like hostnames, and won't work with anything else.
The software for new applications which would use a _foo pseudo-TLD namespace is not yet written. It is for future applications, for which we can hope to push TLD-like use of things like "onion" into namespaces like "_onion". Of course, for Tor, it is not feasible because that software is written, and (more importantly) those .onion names are already out there. I see no reason why, if Tor was started today, the software written for it should not be able to support _onion, if that was the BCP for doing it. Tor software would be written for that purpose at the time. Or am I missing something here?
The argument for *.alt is that if ICANN sells another round of vanity TLDs, as seems depressingly likely, here's a hostname we promise won't have new name collisions.
IMO the promise of "this name will never be delegated, thus have no DNS collisions" delivers a thing quite different from an as-close-as-possible alternative to a TLD namespace. In other words, it's not as tailored a solution for people who are currently "squatting" TLDs. Really, if they're not using the DNS, what's going on is not squatting. It's just a semantic collision of dotted strings, that's it. This is why I wrote that DNS is not necessarily the primary naming system, depending on perspective. When thinking about such other naming schemes, we should give up the DNS-centric view of "squattable namespace property". Pushing alternative naming schemes down one level, reserving the top level for DNS stuff, seems like a rather cocky attitude. Still, a no-collision name at the top level may be useful, namely in situations where a collision-free letter-only top label is needed. That may be a worthwhile purpose, but it does not cater to the same needs. I'm not implying how common the needs for _*-style names would be. It's just that the claimed need for "alt" to be used for alternative naming schemes mounted into the second level is not as close to a first-class citizen naming scheme as it could be. (The "alt" draft is exclusively about such non-DNS contexts; the first sentence of the abstract already says so.) I think it is better addressed with a closer-to-TLD approach. If alternative naming scheme communities don't accept such a TLD-reminiscent proposal, then perhaps there is no solution at all.
But I don't recall ever seeing anyone squatting on a name that isn't a hostname. This should give us a hint.
It may well be that it just tells us that this hasn't been thought of before. Think of QUIC -- it wouldn't have been fair 10 years ago to say that nobody has done it, so it's not a good idea. Thanks, Peter -- https://desec.io/ _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
