Shepherd writeup

draft-aranda-dispatch-q4s has been presented to the ISE for publication as an Informational RFC in the Independent Submissions steam.

The document describes an application level protocol for the communication of end-to-end QoS compliance information based on the Hypertext Transfer Protocol (HTTP) and the Session Description Protocol (SDP). It provides a mechanism to negotiate and monitor latency, jitter, bandwidth, and packet, and to alert whenever one of the negotiated conditions is violated.

The work is chiefly experimental in nature, but the authors do not propose any ongoing experimentation. Instead, they believe that recording this work in the permanent archive will offer a foundation for future similar work.

The Abstract carefully notes that:

   This protocol specification is the product of research conducted
   over a number of years, and is presented here as a permanent record
   and to offer a foundation for future similar work.  It does not
   represent a standard protocol and does not have IETF consensus.

This is, of course, in addition to the standard boilerplate that will be added to the RFC before publication.

The draft was originally taken to the Dispatch WG at IETF-80. I contacted Mary Barnes and Ben Campbell about the history, and they told me that the idea received no support, but not that it was rejected for technical reasons.

Further discussions, including with Mirja Kuehlewind, pointed out that the world has moved on across the last 25 IETF meetings, and that this document needed to make reference to OWAMP/TWAMP and to QUIC. The authors duly added section 1.4 and some discussions in 1.1.

Along the way, this draft received reviews Tommy Pauly and from me. It was updated three times before reaching this stage.

A final note on code points...

This document originally requested code points from the IANA, but after discussions with the authors we have decided to present all code points from new registries in line, and section 11 in terms of the future allocation that would be needed were this work to progress in the IETF at some time.