3 ms·

> Even with open source software you will never know what is actually running on the servers. If the clients are open-source and properly implement end-to-end

> Even with open source software you will never know what is actually running on the servers. If the clients are open-source and properly implement end-to-end encryption, and you verify that they are not sending your keys to the servers, then what is running on the servers is irrelevant.

DangerousPie 6y ago
... if you have the technical expertise to audit the full source code, and run and audit your own build (on both ends).
- _-___________-_ 6y ago
  Sure, but that's a different argument than what parent was making.
- hajile 6y ago
  I'm qualified to audit loads of software I don't have the time to write myself.
intricatedetail 6y ago
But they may run modified software e.g. with added backdoors and you wouldn't know as you cannot check what is actually running on servers.
- _-___________-_ 6y ago
  Yes, but the servers only transfer encrypted payloads for which the servers do not have the decryption keys, and you can verify that just by looking at the clients (which are open source in this scenario). That is the entire point of end-to-end encryption.
  - intricatedetail 6y ago
    Are you saying that MITM is not possible? For example your client will receive a key prepared by rogue server and it will decrypt and encrypt conversations on the fly. You wouldn't be able to tell unless you find a way to verify the person on the other side tried to exchange different keys.
    - _-___________-_ 6y ago
      Resisting MITM is the entire point of end-to-end encryption. Verification can be made with the security code that WhatsApp uses, and the safety number that Signal uses (same thing, different name). Other systems have other, similar methods. You can verify that they match in order to verify that you're not communicating with a man-in-the-middle, and if the key changes then both apps show a prominent warning. Granted, a lot of people may not actually bother to verify.
- emiliobumachar 6y ago
  End-to-end encryption properly implemented on clients is resistant to any malicious software that may run on servers. The only relevant vulnerability is stealth updates infecting the client, but the client could disallow it as well.

DangerousPie 6y ago

... if you have the technical expertise to audit the full source code, and run and audit your own build (on both ends).

_-___________-_ 6y ago
Sure, but that's a different argument than what parent was making.
hajile 6y ago
I'm qualified to audit loads of software I don't have the time to write myself.

intricatedetail 6y ago

But they may run modified software e.g. with added backdoors and you wouldn't know as you cannot check what is actually running on servers.

_-___________-_ 6y ago
Yes, but the servers only transfer encrypted payloads for which the servers do not have the decryption keys, and you can verify that just by looking at the clients (which are open source in this scenario). That is the entire point of end-to-end encryption.
- intricatedetail 6y ago
  Are you saying that MITM is not possible? For example your client will receive a key prepared by rogue server and it will decrypt and encrypt conversations on the fly. You wouldn't be able to tell unless you find a way to verify the person on the other side tried to exchange different keys.
  - _-___________-_ 6y ago
    Resisting MITM is the entire point of end-to-end encryption. Verification can be made with the security code that WhatsApp uses, and the safety number that Signal uses (same thing, different name). Other systems have other, similar methods. You can verify that they match in order to verify that you're not communicating with a man-in-the-middle, and if the key changes then both apps show a prominent warning. Granted, a lot of people may not actually bother to verify.
emiliobumachar 6y ago
End-to-end encryption properly implemented on clients is resistant to any malicious software that may run on servers. The only relevant vulnerability is stealth updates infecting the client, but the client could disallow it as well.