The meeting started slightly after 17:40. Blake Ramsdell, the chair, reviewed the agenda and there were no objections. About 20 people were in attendance.

The chair reviewed the status of the drafts currently active in the working group.

* The certificate capabilities draft was published as RFC 4262.

* The GOST algorithm draft is the the RFC Editor's queue.

* The symkeydist draft is still blocked in the RFC Editor's queue on a reference to the CMC draft draft-ietf-pkix-2797-bis in the PKIX working group. However, the CMC draft is progressing through the IESG and will hopefully be completed shortly.

The chair reviewed the milestones, which break down into two different areas:

* Draft standard progression for existing S/MIME RFCs.

* KEM last call and proposed standard.

The chair explained that he has serious concerns about the progression of the S/MIME RFCs to draft standard, due to the ultimate dependence on the PKIX working group completing an interoperability profile on their certificate RFCs.

The KEM draft was revised since the last working group meeting. It is attempting to track the current state in the ANSI X9.44 working group. There are some concerns about the availability of the normative references that will need to be resolved before final publication in the IETF.

There was some review of the action items that were proposed at the last meeting in Vancouver. These were:

* gost to WG last call. This was completed.

* Continued pressure to get CMC released to get symkeydist moving again. As reported earlier, CMC is moving through the final IESG review.

* Continue discussion about SHA-256 transition. This work is still in progress.

The chair then reviewed the issues with digest migration. He gave a quick overview of the reasons for the transition (weaknesses and suspected weaknesses in the existing algorithms), and explained that the primary goal is to come up with a sustainable plan for digest algorithm replacement. He pointed out that one complicating factor is the transition period where multiple algorithms might be in use, and how to best cope with that.

The general plan that was explained was that:

* An alternative digest profile would be developed. It was explained that this work was already started in draft-ramsdell-smime-sha256.

* Guidance would be created about how to deal with the transition period.

* The format of these would be along the lines of a separate document to modify the relevant existing RFCs.

Jim Schaad then made a presentation about the changes to revise ESSCertID which is currently hardcoded to use SHA-1 and needs to be modified to support alternative digest algorithms. He explained that his proposal is already largely completed, and would be an update for RFC 2634 (Enhanced Security Services for S/MIME). There was a small amount of discussion between Russ Housley and Jim regarding the use of SHA-1 as a default digest algorithm with this new syntax.

Jim then went on to discuss multiple signature layers. Himself, Russ Housley and Sean Turner will be undertaking this work as a clarification of the signature processing guidance for CMS. The intent is to clarify what to do with multiple signatures at the same level, as well as what to do with nested signatures.

Jim proposed handling issues at four different levels -- strict error, possible error, warning, and no problem. There would need to be some discussion about what types of issues to report at what level, as well as how to combine the information found at each level.

Jim then explained his analysis of existing working group documents where digest algorithms are used, and whether they were fixed or parameterized uses of a digest. He reviewed the types of signature attacks that we are trying to defend against, ultimately leading to the proposal of a multiple signature attribute to maintain the integrity of any applied signatures. The intent would be to prevent the removal of a "better" signature.

The chair then reviewed the status of the IBE draft forwarded by Voltage Security in draft-martin-bfibecms. The chair and Russ Housley had some discussion about whether the work should be brought in as a draft of the working group instead of an individual draft. The chair was concerned about intellectual property issues surrounding the technology, and what would happen if Voltage and the IETF could not come to terms. Russ indicated that the worst case would be that the draft would simply not be able to become standards track, and that it could still be informational as a product of the working group.

With that confusion out of the way, the chair took a straw poll of the members present, just to make sure that there were no strong objections to taking this work into the group. There were not, and the members in attendance were supportive of allowing the work. This is not finalized, and the chair will take it to the mailing list to see if there is consensus. But as it stands right now, the chair and the area director are supportive of undertaking this work in the group.

The meeting adjourned promptly at 18:40.