On Fri, Jan 20, 2012 at 10:20 AM, Yoav Nir <[email protected]> wrote:
> > On Jan 20, 2012, at 3:15 PM, Phillip Hallam-Baker wrote: > > Stating the problem in terms of PKIX is way too specific. The requirements > should apply equally to any new technology being proposed. > > Otherwise we end up with a set of requirements that are trivially > satisfied just by having something that is not called PKIX. > > > SOPA/PIPA attempt to force actions on the DNS, an infrastructure that the > US Congress appears to regard itself as being in control of. So the same > issues are raised for DNSSEC in spades. I would not be surprised if some > idiot attempts to 'fix' SOPA/PIPA by giving the plaintiffs the power to > order ICANN and/or the registry to insert fraudulent records. > > > I agree, except I'm not sure that DNS is within the scope of this mailing > list. > The scope of the list is the problem. If there is a BOF in Paris a possible outcome could be 'problem X can actually be handled in PKIX, or DANE or whatever.' The biggest problem is deployment. If the near term deployment incentive for whatever came out of this was that it also fixed the PIPA/SOPA hole in DNS, I am perfectly OK with that. As a matter of practicality, having an application make separate calls to a name service and a trust service is inefficient. It means two packets to two separate services, either of which could fail. So there is an availability issue to be considered. > As for the activities of intelligence services and co-operation therewith, > it seems rather unlikely that any intelligence service is going to attempt > to engage in a covert operation that leaves highly visible traces unless > the object is to be visible or they are very desperate. Fraudulent > certificates are rather visible. > > > Without pinning, fraudulent certificates are invisible, especially when > applied to specific individuals. The fraudulent diginotar cerrtificates in > Iran would have gone unnoticed had it not been for the pinning of Google > certificates. > The Google certs were not pinned. Pinning is a very specific mechanism. It is actually more than is required to detect a possible anomaly. Perspectives, the EFF Observatory and other mechanisms are going to be deployed for detecting anomalies regardless of whatever standards action might occur. What is necessary is to have an understanding of what 'normal' is. Pinning is an explicit assertion by a (presumably) authoritative party as to what normal is. -- Website: http://hallambaker.com/
_______________________________________________ therightkey mailing list [email protected] https://www.ietf.org/mailman/listinfo/therightkey
