DISPATCH-Style Working Groups and the SIP-Change Process

described processes for evolving and maintaining the Session Initiation Protocol (SIP) and related technologies in the former Real-time Application and Infrastructure (RAI) area. These processes are collectively known as the "SIP-Change Process". While areas do not normally have "charters" per se, RFC 5727 effectively served as a charter for RAI. The language in RFC 5727 was tightly bound to the RAI area and to the the DISPATCH and SIPCORE working groups. in 2015, The RAI area merged with the Applications (APP) area to form the Applications and Real-Time (ART) area. This document updates RFC 5727 to remove its dependency on RAI and its working group structure. The updates in this document do not depend on the names of the new area, or any specific working group. Rather, the authors seek to future-proof the SIP-Change Process against future reorganizations. RFC 5727 specified that the DISPATCH working group assesses potential new work for the area, and determines where such work should occur. DISPATCH does not itself take on such new work. The SIPCORE working group is responsible for maintenance of SIP. Other historically RAI area working groups develop extensions to SIP that do not change the core protocol, new applications of SIP, and other technologies for interactive communication among humans. This document further generalizes the processes of the DISPATCH working group so that they can be applied to other areas, or to clusters of technologies within an area. This document does not change any other aspect of RFC 5727. While areas and working groups may change over time, the rules and procedures for changing SIP and other historically RAI protocols remain the same, until such time that they are updated by future documents.

The DISPATCH working group has proven successful at managing new work for the RAI and ART areas. Areas may choose to adopt DISPATCH-like procedures, either for an entire area, or for technology clusters in an area or across areas. A "Dispatch-Style" working group operates according to procedures similar to those used for DISPATCH. This document is not intended to recommend dispatch-style groups for any specific IETF area other than ART. Different areas have different needs, and those needs may change over time. It is up to the community and respective Area Directors to determine if a dispatch-style group is appropriate for any given situation. The "Dispatch Style" includes the following essential elements: The working group evaluates proposals for new work for an area, or for a well-defined technology cluster. It acts as a filter for the area or cluster to determine whether a proposal is a reasonable use of or addition to associated technologies. This determination may depend upon established criteria (for example, the SIP-Change Process), the experience and expertise of the participants, or a combination of the two. The dispatch-style working group determines an appropriate venue for the work. The venue could be an existing working group. If no appropriate group exists, it may develop a charter for a BoF, a new working group, or an exploratory group. The group might also recommend that a proposal progress as an AD-sponsored individual draft, or even that a proposal should not be acted upon at the time. The dispatch-style working group does not complete the proposed work. It may, however, adopt milestones needed to properly dispatch the work. For example, it may produce charter text for a BoF or a new working group, an initial problem statement, or documentation about why certain work was not pursued. Nothing in this list prevents existing working groups from directly adopting new work that reasonably fits their charters, nor does it prevent new-work proposals from going directly to BoF meetings when appropriate. For borderline cases, the decision whether new work should start in a dispatch-style group or elsewhere is made by the responsible Area Directors and chairs. Likewise, in cases where an area has multiple dispatch-style groups for different purposes or technology clusters, deciding which group will handle a particular proposal is up to the responsible Area Directors and relevant chairs. The charter of a dispatch-style group should make that fact clear, either by referencing this document, or by directly describing similar procedures.

This document clarifies that the SIP-Change Process is not bound to any particular area or working group structure. All references to the RAI area in RFC 5727 should be interpreted as "the cluster of SIP and closely related application and infrastructure technologies, as well as other technologies designed primarily for interactive communication, historically among humans." While the DISPATCH and SIPCORE working groups are expected to continue in their current capacities, nothing in the SIP-Change Process prevents their responsibilities from being assigned to other working groups in the future. All other aspects of the SIP-Change process are to continue as described in RFC 5727.

This document makes no requests to IANA.

This document discusses the roles and responsibilities of areas and working groups. It does not create new security considerations in the conventional sense. However, organizational structures come with their own security considerations. A dispatch-stye working group has the potential to concentrate the control of work for an area or cluster in the hands of a much smaller set of people than those in the whole area or cluster. This could effectively create bottlenecks or roadblocks for new work in an area or cluster. Likewise, such a concentration could reduce the quality of decisions about new work. Care must be taken to avoid this risk. The best mitigation is active participation in the group by as many people in the area or cluster as possible.

The authors would like to thank all the previous authors of the SIP-Change Process for their contributions. Jon Peterson, Cullen Jennings, and Robert Sparks authored RFC 5727. That RFC obsoleted , which was in turn written by Allison Mankin, Scott Bradner, Rohan Mahy, Dean Willis, Brian Rosen, and Joerg Ott. The authors additionally thank the present and past chairs of DISPATCH and SIPCORE, as well as all the participants in the former RAI area since its inception.