Define PKI structure and policies #32
Labels
protocol research
Issues for tracking protocol research and choices
security
Potential and confirmed security issues
During the protocol development, we did not focus on the PKI part as there are a lot of state-of-the-art options that are readily available, and well researched and tested. As @mmaker reported, if the PKI part is implemented, it has to be modern and provide eventually canaries, transparency, revocation and more. Rather than doing actual crypto research again, we should investigate what suits our use-case best, and adopt them (looking at key transparency for newsrooms, transparency for FPF, canaries for FPF, requiring multiple n-of-m signatures, and revocation especially at the journalists level, as newsrooms must be able to securely offboard journalists).
An extension of that is adding proper rotation policies, for the longer term keys such as the fetching keys and expiration policies for ephemeral keys, so that we reduce the attack surface of key reuse (from for example a dishonest server).
The text was updated successfully, but these errors were encountered: