--On Monday, August 29, 2022 08:38 -0400 Timothy Mcsweeney <tim@xxxxxxxxxxxxxx> wrote: > >> On 08/29/2022 8:30 AM EDT Olaf Kolkman <kolkman@xxxxxxxx> >> wrote: >> >> >> On 29 Aug 2022, at 14:15, Timothy Mcsweeney wrote: >> >> >> On 8/29/22 07:33, Timothy Mcsweeney wrote: >> >> >> >>> Do you agree with the title? >> >> >> >> No. And I'm not sure what point you're trying to make. >> > >> > The point I'm trying to make is that HTTP is a domain name >> > with no specific location and I wanted to see what the >> > people on this list thought of that premise. >> >> What is the conclusion you want to base on the premise? >>... > That the characters of a URI are in fact domain names and need > to resolved before making sense of anything else. If that is the issue, then the answer is to a slightly different set of questions: Q: Could the string of characters, "HTTP", be used as a domain name label? A: Yes, of course, because that question is probably about the syntax rules for domain names. Q: Is "HTTP" valid as a TLD label? A: That is an administrative question, not a technical one. The answer, as of this instant, is "no". The attempts to use DNS or registry database lookup tools to answer your question appear to me to be attempts to demonstrate that answer but maybe those who showed them in messages had other ideas in mind. Q: Can one determine whether a string is a "domain name" (and should be treated as one, e.g., looking it up in the DNS with or without "completion") by looking at in isolation, without lexical or other context? A: Of course not. If the answer were "yes", it would lead directly to questions like "What is a 'tim'?" or, more extremely and with apologies to anyone given that name very early in life, "What is a 'karen'?" Other examples occur to me that would probably get me in trouble with the Moderators. Q: If "HTTP" were somehow intrinsically a domain name, what relatively short strings consisting entirely of ASCII letters and digits would not be? A: Null set best, john
