Re: [rtcweb] I-D Action: draft-ietf-rtcweb-ip-handling-01.txt

["Drage, Keith (Nokia - GB)" <[e-mail]>]

Assuming it does not contain normative language in its own right, then it might well be a BCP, but then it needs to tighten up on some of its internal language – if we don’t mean RFC 2119 language then we should try and avoid the lower case forms as well, in order to avoid confusion.

I’d note that in the past there has been some discussion of what a BCP really is so that might also require some more debate.

My understanding of an applicability document is that it is a PROFILE as defined by ISO 9646. That means that it converts optional capabilities into mandatory ones and so on. Those in themselves are normative statements that would require RFC 2119 language. This document does not appear to be such a document.

Regards

Keith

From: EXT Ted Hardie [mailto:[e-mail]]
Sent: 22 March 2016 23:10
To: Drage, Keith (Nokia - GB)
Cc: [e-mail]
Subject: Re: [rtcweb] I-D Action: draft-ietf-rtcweb-ip-handling-01.txt

Speaking as an individual, rather than chair:

On Tue, Mar 22, 2016 at 3:31 AM, Drage, Keith (Nokia - GB) <[e-mail]<mailto:[e-mail]>> wrote:
The header of this document says "standards track".

The title of the document says "recommendations" which would equate to a normative document of SHOULD strength

The abstract says "best practices"

Throughout the document I detected lots of lower case usage of RFC 2119 terms but did not see any upper case usage (but may have missed it).

All of which leaves me confused.

Would the chairs like to clarify the intent of this document in terms of document type?

Regards

Keith Drage


I don't think this document describes any new protocol mechanisms, so the lack of upper case RFC 2119 terms seems to me personally fairly natural.  What it does instead is lay out the results of doing IP address handling in specific modes, along with a recommendation on which would make a reasonable default.

Clearly it could have been part of JSEP, which would have made it part of a standards track doc.  The value to me of it being separate is that the advice can be revised independently of the core protocol specification.  To that extent, it is somewhere between "Applicability statement" and "BCP" in our formal terminology, with a thumb very slightly on the scale toward BCP. Ultimately, though, I am willing to be guided by our ADs or the community about the exact status.  The high order bits are the advice and its decomposability.
regards,
Ted