[rfc-i] Summary: use case for 2119 markup

Nico Williams nico at cryptonector.com
Fri Jun 20 13:49:39 PDT 2014

On Fri, Jun 20, 2014 at 12:37 PM, Ted Lemon <mellon at fugue.com> wrote:
> I think you did not capture two points I was trying to make:
>  this is actually a process change, and hence should go through the IETF
> consensus process
>  what it would mean to use this markup is not at all clear

I don't see how this is a process change, at least so long as we're
only talking about marking up RFC2119 terms.

First, it might not be required for I-D authors (just as XML isn't now
either).  The RFC-Editor might add it on their behalf (reviewable in
AUTH48 and so on).

Second, RFC2119 isn't a requirement.  There are post-RFC2119
Standards-Track RFCs that are full of non-RFC2119 normative language
-- not many, I know, but there are some.

Third, even if it was required, it's just markup -- the semantic
requirement would have been there regardless.

Fourth, it wouldn't be IETF process: the IETF is not the only stream
feeding the RFC-Editor.  This would be an ISOC process matter, if it
were a process matter.

This is a tools thing, not a process thing.


More information about the rfc-interest mailing list