It is long since time we made email secure. Or replaced it with something else that would allow us to send messages to people securely (in a decentralized way).
Having to log in to a half-maintained, slow web portal with terrible UI that is down 25% of the time is a really terrible way to get your sensitive and often important documents.
email can be read by any server in the chain between the sender & recipient. It's not secure. PGP doesn't fully fix this, it still leaks message content (subject) and metadata. So does S/MIME. That doesn't mean attachments are leaked, but it does mean email isn't compliant with any of the standards which require communicating securely.
I agree. The problem is that anything which solves this won't be compatible with email, since these issues are inherent to the protocol. Fixing email isn't possible, so the solution has to be replacing it. Unfortunately that's very difficult.
It would be interesting if there was a DNS record of some kind to announce a PKI registry for a particular domain's email addresses. This would allow for some kind of decentralized discovery of suitable public keys for individual addressees.
But, a potential sender with important messages still needs do some pre-flight email verification over this path, e.g. sending a verification code encrypted the same way, to validate that the intended user can read it and send it back via another authenticated channel. This ensures that the sender-specific user identity actually possesses control of that email address and the discovered private key. It also reduces the impact of a malicious email provider to only denial-of-service rather than content interception.
So, is that discovery protocol worth it, versus just having an opt-in setting where the user supplies their desired public key and email address in the same form?
Comments
It is long since time we made email secure. Or replaced it with something else that would allow us to send messages to people securely (in a decentralized way).
Having to log in to a half-maintained, slow web portal with terrible UI that is down 25% of the time is a really terrible way to get your sensitive and often important documents.
email can be read by any server in the chain between the sender & recipient. It's not secure. PGP doesn't fully fix this, it still leaks message content (subject) and metadata. So does S/MIME. That doesn't mean attachments are leaked, but it does mean email isn't compliant with any of the standards which require communicating securely.
Right. I'm saying this state of affairs is unacceptable for the dominant digital messaging system in 2026.
I agree. The problem is that anything which solves this won't be compatible with email, since these issues are inherent to the protocol. Fixing email isn't possible, so the solution has to be replacing it. Unfortunately that's very difficult.
It would be interesting if there was a DNS record of some kind to announce a PKI registry for a particular domain's email addresses. This would allow for some kind of decentralized discovery of suitable public keys for individual addressees.
But, a potential sender with important messages still needs do some pre-flight email verification over this path, e.g. sending a verification code encrypted the same way, to validate that the intended user can read it and send it back via another authenticated channel. This ensures that the sender-specific user identity actually possesses control of that email address and the discovered private key. It also reduces the impact of a malicious email provider to only denial-of-service rather than content interception.
So, is that discovery protocol worth it, versus just having an opt-in setting where the user supplies their desired public key and email address in the same form?