a path forward on DPD & DPV

[Stephen Kent <kent@xxxxxxx>]

During this week's IETF, Tim and I held  several meetings with the authors of the drafts for SCVP and for the OCSP v2 extensions that deal with delegated path discovery and validation. We have arrived at the following procedure and time line for selecting a single solution from these competing protocols.

1. I will send a message examining the rationales for developing facilities to support delegated path discovery (DPD) and validation (DPV). This message will propose requirements and criteria for evaluating proposed solutions that offer these services, and will be issued before the end of this month. Many of the issues discussed in this message were presented at the WG meeting on Tuesday afternoon. 

2. The WG will discuss the proposed requirements and evaluation criteria during January and we will reach closure by the end of the month, with a final set of requirements and criteria. During this time, advocates for each solution may refine their proposals to track the evolving requirements and criteria.

3. During February, the proponents of each approach will have an opportunity to revise their proposals to better match the finalized requirements and criteria, and to present arguments demonstrating how their proposals meet the requirements and evaluation criteria.

4. By the first (full) week of March the WG will select a single protocol as the standard. If the WG chairs do not feel that (rough) consensus has been achieved, a straw poll will be conducted. This should constitute WG last call, since the selected protocol document should be stable and thoroughly reviewed.

This schedule permits a final decision to be reached prior to the next IETF meeting, which takes place 3/19-23, in Minneapolis. However, we probably will not manage to reach (or complete) IETF last call prior to this meeting, so we should not expect IESG approval until late April, at the earliest.