Closed markafoltz closed 11 months ago
@tidoust : I wonder if these terms should go into the section at the end under "Terms Defined by Reference"?
These terms were locally re-defined because ReSpec did not have good support for cross-referencing sections in other specs that need to be interpreted as actual terms. ReSpec now has better support for that. Two options:
Keep a local definition.
<p>The term <dfn data-cite="rfc9110#rfc.section.12.5.4">Accept-Language</dfn>
is used as defined in EcmaScript. [[!ECMASCRIPT]].</p>
<p>I can then reference the term directly with <a>Accept-Language</a>.</p>
Only use references:
<p>The term can be referenced directly through
<a data-cite="rfc9110#rfc.section.12.5.4">Accept-Language</a>.</p>
<p>But all occurrences need to have the same data-cite attribute,
as in <a data-cite="rfc9110#rfc.section.12.5.4">Accept-Language</a>.
<a>Accept-Language</a> won't work in particular.</p>
Both options will correctly put the term in the list of terms defined by reference. I personally prefer the second option that avoids creating an otherwise useless local definition, but it requires using the data-cite
attribute in all references to the term.
The same approach would work with most of the remaining terms in the Terminology section. Do you want me to update this pull request to apply the second option and extend it to other terms as well?
I actually prefer the first version; the second version creates duplicate cross references that will be hard to update if needed.
I can update this PR to use local definitions for the HTTP terms, and followup with another PR to update remaining terms.
New commit should reflect your suggestion, PTAL @tidoust
RFC 9110 (HTTP Semantics) has obsoleted the RFC for HTTP 1.1.
Update term references to point to RFC 9110.
Preview | Diff