> "we plan to open-source key parts of our code as well later on." Great!
Is this a unicorn or moon on a stick statement? [1]
11–20 of 51 posts
> "we plan to open-source key parts of our code as well later on." Great!
Is this a unicorn or moon on a stick statement? [1]
I'm always skeptical of browser/JS based crypto, but it is nice to see that they're at least upfront with the risks involved in doing such a thing.
They probably downplay the risk of a MITM attack a little much, but otherwise I'm glad to see they're realistic about possible weaknesses of the platform.
Running infrastructure in those DC's can't be cheap (compared to regular co-lo facilities). Thats on top of probably having to deploy more gear (or higher perf gear than a regular email provider) since the work load is probably CPU heavy.
The security details page[1] makes for interesting reading. Hopefully the new norm is 'E2E' encryption. It's actually starting to feel inevitable, and the hopelessness that followed in the wake of the 'Summer of Security' is perhaps evaporating bit by bit, through universal encryption, bit by bit. - [1] https://protonmail.ch/pages/security_details.php
> Messages are stored on ProtonMail servers in encrypted format. They are also transmitted in encrypted format between our server and users’ browsers . Messages between ProtonMail users are transmitted in encrypted form completely within our secured server network . Because they never leave our secured environment , there is no possibility to intercept the encrypted messages enroute. Emphasis mine. That doesn't sound…
This part is only noting that inter-user messages never even leave their 'secured environment'. By all accounts it does seem as well secured as any other provider I've looked into.
True end to end encryption would mean everything is transferred as an encypted thing, and only people with a key can open it. If any email you send out ultimately is unencrypted so that the other side can read it, we aren't much closer than where we started are we?
If an email ends up in an unencrypted IMAP mailbox on a server somewhere, how is that more secure than what happens now?
> https://protonmail.ch/blog/protonmail-threat-model/ I'm always skeptical of browser/JS based crypto, but it is nice to see that they're at least upfront with the risks involved in doing such a thing. They probably downplay the risk of a MITM attack a little much, but otherwise I'm glad to see they're realistic about possible weaknesses of the platform.
Plus, they're offering self-destructing e-mails, which is impossible to provide, so already there's a bit of snake oil there. If they said, "It's not possible to provide real self-destructing e-mails, but you can set it up so that (assuming you trust us), we'll delete the messages from our servers after a certain amount of time, which is the best anyone can do." Instead they say that they are "more ephemeral than SnapChat."
So, if I send an ecrypted protonmail to someone else's yahoo mail, what happens? Is it only encrypted in the protonmail ecosystem? True end to end encryption would mean everything is transferred as an encypted thing, and only people with a key can open it. If any email you send out ultimately is unencrypted so that the other side can read it, we aren't much closer than where we started are we? If an email ends up in…
Presumably they'll have some way to distribute the password in some ephemeral or slightly out-of-band way. It's probably less secure than messages within their environment, but it shouldn't ever hit another mailserver in plaintext (ideally ProtonMail wouldn't even have the plaintext anyway).