- From: hedlund <hedlund@best.com>
- Date: Sat, 11 Oct 1997 20:23:03 -0700 (PDT)
- To: Yaron Goland <yarong@microsoft.com>
- Cc: http-state@lists.research.bell-labs.com, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
On Sat, 11 Oct 1997, Yaron Goland wrote: > I understand the concerns regarding unsigned cookies but at the same > time I do not believe we can create restrictions that are not arbitrary. > For example, the two hierarchy level restriction. I agree that the two-dot rule you refer to is a compromise that neither fully protects privacy nor consistently ascertains the need to do so. It was the best solution we found, and the other privacy restrictions in the RFC were intended to make up for this compromise. I do not agree that the privacy restrictions are arbitrary, though I would be foolish but to agree that they are controversial. > As such I believe the best we can do is state "You want security? Use a > signature." > > How many systems do you know that go out of there to specify security in > situations where the user intentionally chooses not to use any security? Security is not the same thing as privacy. There are plenty of systems that protect privacy even if the user does not explicitly request privacy protection. That is my understanding of the intent of the privacy restrictions in the cookie rfc. M. Hedlund <hedlund@best.com>
Received on Saturday, 11 October 1997 20:27:17 UTC