3 ms·
It's amazing that we anticipate having to revoke malicious CAs as a crucial part of a security model, yet we have basically no plan to ensure that we don't acce
by pryce 8y ago
It's amazing that we anticipate having to revoke malicious CAs as a crucial part of a security model, yet we have basically no plan to ensure that we don't accept a competent-but-malicious CA into the fold in the first place.
- callalex 8y agoCompetency in this case can be objectively reinforced, but maliciousness requires one to device who is “bad” and who is “good” which is not a technical problem.