Are mail providers retarded?

Your problem that you don't create a new Simple security questions thread. Look at this soykaf. What the fuarrrk is wrong with this? https://proton.me/support/proton-mail-encryption-explained >All messages in your Proton Mail mailbox are stored with zero-access encryption. This means we cannot read any of your messages or hand them over to third parties. How is this possible? Where the fuarrrk do they store the key? If locally, then how can I possibly log in from another device? But I can, then either they have copy of the key, or the password itself is the key or masterkey for the actual key. In this case they CAN fuarrrking read all of your mail if they feel like it. What the fuarrrk is wrong with this gaslighting? Where am I wrong?

You're not wrong. Password likely unlocks the encrypted private key that Proton stores for you, while the public key is used to encrypt your incoming plaintext emails before storage. Yes, Proton can push malicious software whenever they feel like it to get hold of that secret. https://www.devever.net/~hl/webcrypto >A cryptosystem is incoherent if its implementation is distributed by the same entity which it purports to secure against. The upside of Proton is not the trust-based snakeoil encryption, it's that at least they claim not to steal your data and metadata, unlike Google. Want real encryption? Use protocols like PGP, TLS, OMEMO, Tox directly, don't trust some company to push unsupervised binary updates of the only client.

Did a quick look and I think I have a way it might be done. Hashing of passwords there is dats loss as part of the process so it's not possible to decrypt anything hashed. It's a one way street. The only way around this is if they use their hashing until they find a hash that matches through a sort of brute force attack. It may be a key is generated relating to the password if not the password itself as a way of always getting large values that would be more difficult to copy again through brute forcing hashes. This would mean the password is the only possible weak point since it would be easier to brute force than a hash if the hash is something made to be longer and more complex than the password.

>>25174 Do a few deep inhales and make this comprehensive. Hashing is needed because unique password yields unique hash, but the process is not reversible, so storing hash instead of password is good way to avoid leaking the password itself. It is only useful with salt and everyone who has two brain cells these days uses salt. But it does not in any way prevent Proton from decrypting your mail and reading it. Considering their services are essentially retarded and don't even allow you to have free IMAP, I see no reasons why they wouldn't violate their privacy policy. Big tech like that can afford. They're also advertised everywhere so they're a honeypot. I hate them.

They're probably fine but way too expensive : ( Things aren't honeypots because they're big. It's just that big things are going to get compromised if they can be, same with small things if they posses value to compromise, and can be so. Protonmail also encrypts emails which were not so as they come in, which again is relying on trust. I would assume if they have a warrant they would stop doing so.

>>25200 >They're probably fine >Things aren't honeypots because they're big Totally not glowing.

I'll just leave this here. Proton's architecture is the same as Tuta's, i.e. user trusting the company fully, closer to Gmail than it is to proper end-to-end encrypted communications. https://archive.org/details/nov3am http://gangstersout.com/nov3am.pdf A. So, I was briefed on a storefront that was being created or had been created in order to attract targets – criminal targets to this online encrypted service that was being created, in order for them to – the criminals or the targets to use this service in order for intelligence to be collected by the agency that created the storefront. Q. And the name of that storefront was what? A. It’s an online end-to-end encryption service called Tutanota. Q. And is that spelled T-U-T-A-N-O-T-A? A. That’s correct. Q. So, if people begin to use that service, how will that assist, ultimately, Canada? A. So, if targets begin to use that service, the agency that’s collecting that information would be able to feed it back – that information into the 5EYES system, and then back into the RCMP. Q. Okay. And when you say that’s – that’s – accessing that information would the storefront, the agency that is supporting the storefront be like, intercepting e-mails, or something like that? A. That’s correct. Q. And would that – the intercepted material then become part of high-side intelligence that the RCMP would be able to rely on? A. That’s correct. https://archive.org/details/nov6pm http://gangstersout.com/nov6pm.pdf Q. So, just continuing where we left off before the lunch, you’d agree with me, sir, that Mr. Ramos was a 5EYES law enforcement target? A. At various points, correct. Q. And you say the goal of this e-mail exchange with Mr. Ramos was to get him to move to Tutanota. Is that right, sir? A. That’s correct. Q. And that Tutanota was a government-run storefront that was being used to intercept intelligence or evidence of alleged criminals? A. At the time, that’s what I was briefed on. ... Q. And you still want us to believe that Tutanota was run by a foreign agency? A. That’s correct. Q. And they would allow there to be security vulnerabilities and ask for donations from users? A. It seems like ordinary course of business for online e-mail services. Q. Okay. A. They all do... Q. Run... A. ...this. Q. ...run by foreign agencies? A. If they wanna keep up the pretext... Q. Mm-hmm? A. ...this would be something that they would do. Absolutely. This is from Cameron Ortis' auditions after he was arrested for corruption. He was the director general of the National Intelligence Coordination Center, a branch of the Royal Canadian Mounted Police that deals with online threats. https://www.bbc.com/news/world-us-canada-68228941

>>25204 From this perspective do you think it would be more reasonable to, for example, pay mailbox.org (found on wikipedia just now, so don't think of it as of an indepth take into the issue) or a similar service with cash, if you want your identity to go under the radar? If they do notice illegal activity they will probably get back to you without much trouble, but do they keep accumulating profiles continuously and everywhere or is there a chance that they only unleash their tooling when they need to? Is a paid company less likely to be honeypot, especially since it (mailbox.org) brags about snake oil encryption just the same as Proton or Tuta? Is there actually an option to prevent government from spying on you without any single fuarrrking reason to do so?

>>25206 Accept that e-mail is a plaintext protocol, both at the sending and receiving end. Therefore there can't be a guarantee of privacy with a remote e-mail service. Same as raw TCP (HTTP, FTP, Telnet) which was good enough in the early days of the Wired, but then they tacked on SSL encryption for online payments. With e-mail we could manually tack on PGP (or something better), but people are too dumb to set it up, or they'll Re: you back the decrypted plaintext with their reply. As a society, out of laziness, to maximize convenience, we decided managed plaintext e-mail is preferable to a proper automatically E2EE default communication protocol. Like we decided illegal (now legalized) mass surveillance in the U.S. is fine. Mailbox could be false flagging like Tuta does. That said, I'd rather use that or Proton, Tuta than Google Mail, because at least they don't double dip, they probably don't commercialize your data on top of forwarding it to the glowies. Use anything you like, e-mail surveillance can't be helped, blend in. You can host your e-mail server on your own hardware, but the other end will 99% be some compromised cloud service. An extreme anti-privacy take: https://www.tastyfish.cz/lrs/privacy.html

>>25211 >Use anything you like, e-mail surveillance can't be helped, blend in. All I want is for them to be unable to track my identity without effort, so I can actually do stuff on internet in peace. Don't get me wrong I had a bunch of unpleasant experience with certain agencies, even though I didn't do anything illegal, and I am both fuarrrking paranoid and hate them. They don't need to know I post on lainchan and soykafpost on certain mailing lists.

>>25212 Can try Tor against dragnet surveillance, these have hidden services and there are more on the Hidden Wiki, hard to post Onion links on this site: cock.li (when they open up registrations, >>>/Ω/81483, >>>/Ω/64846): Alt Address (disposable) Proton's Tor hidden service maybe These can receive clearnet e-mail while accessed anonymously through Tor. Provider can snoop on your mail, but doesn't easily find out who you are from browser info. If a service rejects these e-mail addresses, then go full tinfoil and fuarrrk the service, or go full cattle and fuarrrk privacy.

yeah the OBVIOUS security loophole there is that you're 'sold' the idea of muh switzerland so neutral so encrymtims the reality is that 5-and-now-9-eyes gets to take a peek at any encrypted traffic that crosses borders so this, like tor, ooh so many destinations, how could they ever trace 3 whole steps?? is actually the thing that's fuarrrking you by Ensuring!! your soykaf gets picked up, then decrypted later by the Massive data crunching centers at their disposal also anything you type in the browser gets autosaved, no option to turn that off

>>25217 AES, Ed25519, SHA3 aren't breakable with all the computers in the world Side-channel attacks is how they get people, for example they conventionally hack the hidden php website, traffic analyze the guy using an outdated version of ricochet chat, corrupt cryptographic random generators, distribute backdoored phones (anom), social engineering, snitching Hidden services aren't 3 hops, that's for clearnet destinations which is open to traffic analysis, yes. Hidden services have more relay hops after the 3rd node, and unlike clearnet exit nodes their traffic isn't decrypted by the last node. If there are a few honest nodes in the chain it's fine. If 90% of nodes glow, then Tor would be compromised of course, that's the threat model

>>25217 > the reality is that 5-and-now-9-eyes gets to take a peek at any encrypted traffic that crosses borders What law is this? Something like the CLOUD Act in the US? Or a different law in mind?

>>25211 >With e-mail we could manually tack on PGP (or something better), but people are too dumb to set it up Is thing like Delta Chat a solution? Dumb or lazy people can use PGP with only email like it's messaging app (with only chat of course and no calls)

>>25173 It's still better, because if they were grabbing keys en masse, someone would have noticed. If they do this soykaf, it's only against very few customers. It's not too great and you shouldn't trust it, but on average you reduce your exposure a bit. Is it worth giving up a mail client though? Probably not. >>25174 When you create your account, your password (which is not stored in their servers in plaintext) is used to encrypt the mail database clientside. When you log in, you use your password to decrypt the database like you would decrypt a password protected 7zip archive.

>Bruh its SSL its secure and private Thats what everyone says on Matrix and Discord and Telegram so we are safe everything else is FUD.

>>25223 >>SSL >>Surreptitious Entry >>TEMPEST >>Durov arrest = Cover for action, source/method sanitation. Pwnd.

>>25224 Good thing Tails and Whonix use TEMPEST Fonts to make it harder for interceptors.

>>25225 Trolling? Should I go to the Guardian or not? Probably not. EU has been Christmas market terror attack free for a few years. I think we'd all like to keep it like that right? And too those Iranian terrorists that keep getting nabbed. Probably best not to talk about Tempest. fonts lol...

>>25221 >someone would have noticed What makes you think so? >it's only against very few customers What makes you think so? Save user's password, store it elsewhere. That's all. Mail received? Decrypt, analyze, decide what to do. Mail sent? Decrypt, analyze, decide what to do. Maybe not as bad as google but since Proton requires doxxing too, they can get on your case whenever they want and for whatever purposes they want. They're most likely just a honeypot but it doesn't make things better for you. You're most likely spied upon as much as with google. And the point is, if I wanted to blend in, I'd do it with google. I've been thinking about cash lately. Cash to crypto to monero to mail. But few accept monero. Maybe pay cash directly to the mail provider. No idea. Cock.li has the most sane approach to it. No bullshit.

>>25220 There are tons of E2EE messaging solutions: >>>/Ω/50332. People are intellectually lazy or dumb to discover and use them. It's different, confusing, they want an app on their iPhone that works automatically, that their contacts already use, they are averse to anything new or slightly difficult. People offload all effort, predatory companies jump on the opportunity, easy money. Tox is as simple as launching an .exe and copy-pasting 56A1ADE4B65B86BCD51CC73E2CD4E542179F47959FE3E0E21B4B0ACDADE51855D34D34D37CB5 to your contact. No 50 page EULA and privacy policy, account, password, recovery email, phone and email verification, 2FA, security questions, premium plan up-sell, donation begging, Tor/VPN block, questionable blobs, contact list upload, device fingerprinting, telemetry. Pure P2P E2EE messaging, nothing more. Is this not a more satisfying experience for texts and calls, a breath of fresh air? Is it not fun to try new things for free? While lack of history sync and multi-device in Tox's case could be rational roadblocks, I believe the cause of why people don't use these apps is they cannot imagine what happens beyond the screen, immediate convenience is the only factor. If they received their dragnet dataset every week, with AI roasts based on their consolidated "private" activity, or even worse if everyone got that package of all their friends, they'd go full E2EE at lightning speed. Mass surveillance industries would know this, that's why they STFU about the troves of data and analyses they most likely hoard. Ask anyone how their favorite messenger works. Even most professional developers won't be able to confidently state e.g. for Telegram that "by default it's TLS from client to centralized server, stored as plaintext on server". Many will assume it's somewhat encrypted because the word "encrypted" is displayed in the app store. Few have ever used the bare-bones E2EE feature in Telegram, and nobody will notice the cryptographically weak emoji pubkey hash or consider whether a home-brew encryption scheme and centralized app distributor are suitable. >>25221 >if they were grabbing keys en masse, someone would have noticed They don't have to grab keys. All incoming and outgoing mail is plaintext, metadata is rich in information, the from, to, subject fields, plaintext mail body which 99% of emails use. The at-rest encryption they tout happens after the plaintext transfer and is only useful if some hacker breaches their storage servers. It doesn't prevent them from copying all incoming, outgoing mail unencrypted, and it's exactly how they comply with government data requests. For the few people who are targets and actually use the built-in PGP feature, Proton can deploy seamless malicious JS or app updates to get the key, and since PGP has no forward secrecy, the key decrypts all past recorded encrypted messages. Whether they do or don't is beside the point, it's Proton's choice, not the user's. >>25223 Thank you, finally a simple answer. I don't know why they make it so complicated. :-) >>25224 >>25225 >>25226 >>25227 >too many redpills >quick saturate the thread with meaningless implications kek

>>25232 >People are intellectually lazy or dumb to discover and use them This statement is weird to me, because, if that is the case, what the hell happened to teamspeak/skype in 2017~18 when discord entered the game? Everyone, including me, hopped from skype to discord. I didn't know what the hell that was until all my friends were telling me they were on there now. The effort to make the switch, to be an early adopter, for anyone would have been too much. The only thing I remember from that era was hearing how the quality of the services they offered back then were just better. Is that all a service needs to be the new standard? Or was it all just an elaborate marketing technique? Well. Considering how little marketing these E2EE services (matrix, tox, etc...) do, maybe just doing any kind of marketing push for privacy would convince alot of folks to make the effort to make the switch.

>>25234 I was talking about E2EE software, Discord is the complete opposite, it's the perfect example of "convenience is the only factor". Skype to Discord, from one abusive centralized plaintext provider to another because it's more convenient, heavily marketed, has that precious network effect going. Discord doesn't allow anonymous registration and use. Real life identity is associated with pseudonymous Discord accounts. Discord stores all messages, private or public, in a single database without pretense of privacy: https://discord.com/blog/how-discord-stores-trillions-of-messages Discord client transmits a lot of metadata that a user isn't informed about and isn't shown. Discord can arbitrarily make you choose between uploading photo ID and getting your account deleted. Discord forbids third party clients. Discord has ongoing expenses and doesn't make enough money to cover them. Skim through the TOS and Privacy Policy which you agreed to. Request and review your GDPR data package, if Discord allows it. Undoubtedly Discord is more convenient and accessible than any E2EE software. As long as people, at the tiniest hurdle, shift from fully private, non-spying E2EE to the siren call of exploitative user-hostile software, the trend of surveillance will not end. Most people will choose convenience. If Matrix somehow got a critical mass going and became the default, people wouldn't host small servers and join them, but flock to emerging centralized Discord-like abusive "Matrix provider" companies.

>Discord Triggers my retard hatred. Absolute piece of soykaf. Voice chat is ugly, performance is like kicking dead whales down the beach and the sound is terrible. Self hosting mumble, probably hidden by a vpn (so it's harder for 3rd party to kick in), is by far a better solution for chatting/playing games. Completely agree with >>25232. People are essentially retarded and don't give a fuarrrk. Why would they? Spying doesn't hurt so we'll simply pretend it's not there. Talk about IQ. fuarrrk them.

>>25232 >Proton can deploy seamless malicious JS or app updates to get the key, and since PGP has no forward secrecy, the key decrypts all past recorded encrypted messages. Whether they do or don't is beside the point, it's Proton's choice, not the user's. This guy gets it. Never put your keys on an internet connected device. But for general secure use at the confidential level where risks aren't death or life imprisonment, what do you think about simplex chat? Also to anyone else if you're absolutely serious about making sure not even the CI-GAyyy can get your keys: https://github.com/maqp/tfc

>>25238 Not that lainon, but it's worth noting that pgp can be much safer than every mainstream messenger, as long as you handle the encryption/decryption of messages in a separate non-netwotked sandbox (e.g. a qube).

>>25204 >Q. And you still want us to believe that Tutanota was run by a foreign agency? How reliable is this Ortis guy, is he just making soykaf up to save himself? Why does he call it a "storefront"?

>>25240 > Not that lainon, but it's worth noting that pgp can be much safer than every mainstream messenger, as long as you handle the encryption/decryption of messages in a separate non-netwotked sandbox (e.g. a qube). Seems futile if you have to blindly trust whoever you communicate with is doing the same or is there something im missing?

If you want to exchange private, secret, or sensitive information with someone, the first thing you have to do is find someone trustworthy to communicate this information with. That part is unavoidable, No one can be trusted absolutely, but some people can be trusted more than others. So you find someone you can trust, say, 99% Assuming you want to do your communication via computer, wire, electromagnetically, you then need the same encryption key. It needs to be generated on an air-gapped computing device protected by a Faraday cage and protected from surveillance via sound, light, vibration. It shgould be a battery operated device so no power cable extends outside the Faraday cage. The encryption key should be so long and random that no human could possible remember it, especially you and your communications partner. No one can use duress to get information out of someone who doesn't even know the information. The encryption key needs to be saved on a device that can be instantly destroyed by intention, but not accidentally destroyed. This physical device has to be hand delivered to the communications partner, and that person must know how to store safely (preferably on his body 24/7) and use it safely and how to destroy it if need be. Once all of that is accomplished You have a means of communication relatively privately using very keep-it-simple-stupid methods instead of unknown-black-box software packages, none of which can really be trusted at all.