Informal Minutes of the P1363 Editorial Teleconference. 30 January 1998, 10 a.m. Pacific Time Present: David Jablon Don Johnson Burt Kaliski Mike Markowitz Aram Perez Leo Reyzin Tim Skorick Jerry Solinas Roger Schlafly Yiqun Lisa Yin Rob Zuccherato 1. Lisa reported that she worked on the draft of the main body after the meeting and posted our first official draft on the Web site. She also posted sample test vectors recommending the format to be used by other test vectors. She also went to the Microprocessor Standards Committee meeting in January and told us that the balloting process seems fairly straightforward by following the MSC guidelines as soon as we have the final draft. 2. Jerry and Leo were working on Annex A, which was nearing its final state. Leo was working on converting the Annex into IEEE style and was planning to post something on the Web site within a few weeks. Don reported that X9.F1 meeting recently decided to incorporate the complex multiplication (CM) method of curve generation from P1363 into its documents. 3. David Kravitz was no longer working for Certco, but would probably continue working on P1363 to some extent. 4. Burt reported that he was working on conformance and was thinking about making some changes to the main body that would make conformance more readable from the engineering point of view. He expected to post something for discussion by mid February. 5. We discussed a recent message from Paulo Barreto in which he suggested that we provide Schoof's point counting algorithm for generating elliptic curves. Don said that X9.F1 decided to include it into a technical guideline on elliptic curve arithmetic, to be issued later, because Schoof's algorithm is an area of current research and to try to include it into standards under development would significantly delay them. Most of us thought that in the interests of time we did not want to include it into P1363 right now, but could include it later or point to the X9.F1 technical guideline. Don pointed out that it is easy to check if the number of points on the curve is correct once you know what it is, so you don't have to trust a particular Schoof's algorithm implementation because you can always verify it. 6. We then discussed whether we should recommend verifiably random curve selection and point counting as a curve generation method, and whether it had any security advantages. We did not come to a conclusion and agreed to continue the discussion on the mailing list and at the March meeting. 7. We discussed the message Burt had posted on the mailing list about modifying the cofactor multiplication in the discrete log and elliptic curve Diffie-Hellman primitives. Burt agreed to clarify his proposal and requested that people voice their opinions on the mailing list. 8. Burt asked for comments on the new patent / trademark solicitation letter. He said that he would approach patent holders individually on behalf of the group if anyone in the group wanted to generate a test vector for a patented technique in the standard and didn't have a license to do so. 9. Burt asked that people be ready to discuss security considerations at the March meeting. Lisa said that RSA Labs will provide security considerations for the IF section; she also said that Lily Chen would provide security considerations for the DL section (over prime fields) shortly. We adjourned at about 10:50 Pacific Time.