[Bug 70119] HTTP/2 guide: “its RFC 91 13”

https://bz.apache.org/bugzilla/show_bug.cgi?id=3D70119

Philippe Cloutier <[email protected]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEEDINFO                    |NEW

--- Comment #2 from Philippe Cloutier <[email protected]> ---
(In reply to Rich Bowen from comment #1)
> I don't understand the phrase "should precede something countable". One is
> countable, so I presume you're saying something different than that. 9113=
 is
> indeed its RFC - the RFC describing HTTP/2.
There are several RFCs specifying HTTP/2, including 7540 and 9113. Saying =
=E2=80=9Cits
RFC 9113=E2=80=9D is syntactically valid, but it suggests that refers to th=
e 9113th
HTTP RFC. In reality, the vast majority of RFCs are not about HTTP.

Some non-misleading ways to phrase this could be:
1. =E2=80=9CThe most normative are its RFCs (9113 and 7540)=E2=80=9D
2. =E2=80=9CThe most normative is its specification=E2=80=9D
3. =E2=80=9CThe most normative is RFC 9113=E2=80=9D

> Regarding the definition of h2c, surely "HTTP/2 over TCP" is far more
> specific and meaningful than "a protocol that does not use TLS". FTP is a
> protocol that does not us TLS. Gopher is a protocol that does not use TLS.
> That feels like a meaningless definition.
Of course. That quote is from
https://www.rfc-editor.org/info/rfc9113/#name-starting-http-2-for-https-u a=
nd
in the context of HTTP.

> What are you suggesting that we change here?
Well, the ideal would be to remove that item, since h2c was removed years a=
go.
But =E2=80=9Ch2c is an HTTP/2 protocol that does not use TLS.=E2=80=9D woul=
d be OK.

--=20
You are receiving this mail because:
You are the assignee for the bug.=