💾 Archived View for rawtext.club › ~sloum › geminilist › 007621.gmi captured on 2024-03-21 at 15:43:09. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2021-11-30)
-=-=-=-=-=-=-
Stephane Bortzmeyer stephane at sources.org
Sat Nov 20 08:05:53 GMT 2021
- - - - - - - - - - - - - - - - - - -
On Sat, Nov 20, 2021 at 02:36:37AM -0500, Matthias Portzel <matthiasportzel at gmail.com> wrote a message of 85 lines which said:
I believe Molly Brown, Agate, Jetforce, and others, do this simple
case-sensitive hostname check.
The intuitive behavior here is definitely that case shouldn’t effect
domain names.
It is not just intuition, it is the standard. RFC 1034, on DNS, inits section 3.1 "By convention, domain names can be stored witharbitrary case, but domain name comparisons for all present domainfunctions are done in a case-insensitive manner, assuming an ASCIIcharacter set, and a high order zero bit." And section 3.5 "Note thatwhile upper and lower case letters are allowed in domain names, nosignificance is attached to the case. That is, two names with thesame spelling but different case are to be treated as if identical."
Also, in RFC 3986 on URI, section 3.2.2 "The host subcomponent iscase-insensitive."
(IDN complicate things a bit and, unfortunately, Gemini standard issilent about IDN.)
So I would guess that these servers need to be patched.
Indeed, there are clearly wrong and deserve a bug report.
Unless there are objections, it might be worth adding a line to the
specification to that effect.
Here, I disagree. The standard is already clear. A standard is notsupposed to mention all the ways the readers can be wrong. (Acompanion document "Things to keep in mind when implementing thestandard" could be useful but not the specifictaion itself.)