Hi Paul, RFC5114 doesn't say they're new, it just gives a definition for them. And the difference is in the curve equations. For group 19 RFC4753 says that the equation for the curve is y^2 = x^3 - 3x + b and RFC5114 say the equation for the curve is y^2 = x^3 + ax + b.
The equation for the curve is what defines whether a particular (x,y) combination is a valid point on the curve-- i.e. that it's in the group. That's why I find it interesting that the order (the number of points that satisfy the equation for the curve) is the same for both definitions. Of course, since p-3=a and the generator is the same it may not matter. But it still seems wrong to have two different documents defining the same curve differently, even if they are uncorrelated Informational RFCs. Can you elaborate on why you don't want to "ask the 4753bis authors to significantly expand their document in a way that they didn't intend in the original"? It seems like a perfectly reasonable request. And "no, we don't want to do that" is an acceptable response provided it has a reason. regards, Dan. On Sun, July 12, 2009 8:32 am, Paul Hoffman wrote: > At 6:27 PM -0700 7/9/09, Dan Harkins wrote: >> RFC 5114 claims it defines new ECP groups 19, 20, and 21 for IKE but >>so does RFC 4753. > > To be fair, I don't see where RFC 5114 claims that they are new. In fact, > it says "Three of these groups were previously specified for use with IKE > [RFC4753]". > >> Interestingly the curve definitions are different but >>the orders are the same (maybe it's just interesting because I don't >>understand why). > > Where do you think they differ? The values look the same to me, but I > could have missed something. > >> If there is no cryptographic difference between ECP curve 19 (20 and >> 21) >>as defined in RFC 5114 and RFC 4753 then can some mention be made in this >>draft to that effect? > > RFC 4753 and RFC 5114 are uncorrelated Informational RFCs written by > different groups of authors. Maybe it is better to keep them uncorrelated, > and only link them if one of the other goes on Standards Track. > >> And can it then obsolete RFC 5114 as well as RFC >>4753? > > As you point out, 5114 does things that 4753 does not (new MODP groups, > coverage of more protocols). I don't think we should ask the 4753bis > authors to significantly expand their document in a way that they didn't > intend in the original. > > --Paul Hoffman, Director > --VPN Consortium > _______________________________________________ > IPsec mailing list > IPsec@ietf.org > https://www.ietf.org/mailman/listinfo/ipsec > _______________________________________________ IPsec mailing list IPsec@ietf.org https://www.ietf.org/mailman/listinfo/ipsec
