Slashdot Mirror

← Back to Stories (view on slashdot.org)

Google Open Sources Encrypted Email Extension For Chrome (onthewire.io)

Posted by EditorDavid on from the encrypting-emails dept.

Last week Google released E2EMail, "a Gmail client that exchanges OpenPGP mail." Google's documentation promises that "Any email sent from the app is also automatically signed and encrypted... The target is a simple user experience -- install app, approve permissions, start reading or send sending messages." Trailrunner7 quotes On The Wire: People have been trying to find a replacement for PGP almost since the day it was released, and with limited success. Encrypted email is still difficult to use and painful to implement in most cases, but Google has just released a Chrome plugin designed to address those problems. The new E2EMail extension doesn't turn a user's Gmail inbox into an encrypted mail client. Rather, it is a replacement that gives users a separate inbox for encrypted messages. The system is built on Google's end-to-end encryption library, and the company has released E2EMail as an open-source project.
Wired quotes a web security researcher who calls the open sourcing "a telltale sign the project isn't going anywhere. This is a way for them to get their work out there but to absolve themselves of future obligations." But Google's privacy and security product manager responds that they're tackling some very thorny issues like secure key handling, and "The reason we want to put this into the open source community is precisely because everyone cares about this so much. We don't want everyone waiting for Google to get something done."

44 comments

  1. privacy by mscommunity · · Score: 1

    encrypted email provide privacy! We have to work on this

    1. Re: privacy by Anonymous Coward · · Score: 0

      Crypto kinda has to be Open Source.
      The phrase "Google's crypto library" sent suspicious waves through my mind, out the back door, and all the way home to Virginia.

  2. Let me count the problems... by Entrope · · Score: 5, Insightful

    Having a plugin is nice, but it doesn't solve the PKI (key distribution and reputation) problem, and I am not very inclined to trust a plugin made by a company whose primary line of business is advertising by building user profiles.

    1. Re:Let me count the problems... by Anonymous Coward · · Score: 3, Insightful

      Having a plugin is nice, but it doesn't solve the PKI (key distribution and reputation) problem,

      RTFA. They've provided a keyserver based on OAuth and a "trust on first use / warn on change" local cache. It solves the problem better than traditional PGP, albeit with less (nonfunctional) kool aid.

      and I am not very inclined to trust a plugin made by a company whose primary line of business is advertising by building user profiles.

      Then read the source, or expect others to do so and destroy Google's reputation if there are backdoors. The alternatives to the plugin are written by anyone who can send a pull request, ie. NSA. The fact that Google has some reputation to lose puts the situation above average, same as with Chrome. Either way, trust in the adversarial scenario depends on someone auditing the source and on long-term author reputations.

    2. Re:Let me count the problems... by geek · · Score: 1

      Having a plugin is nice, but it doesn't solve the PKI (key distribution and reputation) problem, and I am not very inclined to trust a plugin made by a company whose primary line of business is advertising by building user profiles.

      What does PKI have to do with OpenPGP? Its fucking open source, why do you care who makes it if its open and you can see whether its spying on you?

    3. Re:Let me count the problems... by Anonymous Coward · · Score: 0

      I am not very inclined to trust

      So you don't trust OpenPGP? May I suggest IT is not the right industry for you.

    4. Re:Let me count the problems... by Anonymous Coward · · Score: 0

      This is a plugin that involves gmail. Google can already trawl through your email (like any other online email site can trawl through the emails they host) with or without this plugin to build 'user profiles'.

    5. Re:Let me count the problems... by Zemran · · Score: 0

      Despite the trolls, I agree with you. Google do not have a reputation to lose as it is already lost. They are world famous for reading your mail and selling the content to their advertisers. They are exactly the people I want a good encrypted mail system to avoid. At best this just encourages more people to use to Google who they believe will not use encryption for the majority of their mail and will therefore provide more mail for Google to read and sell. You are not the customer to Google, you are the product and they will continue to treat you as sheep.

      --
      I love stacking my barbecues in the shed at the end of summer - you can't beat a bit of grill on grill action.
    6. Re:Let me count the problems... by Anonymous Coward · · Score: 0

      Um. You are not actually counting the problems. Instead, you are listing one misunderstanding (it does reasonably solve the PKI problem) and one personal opinion of yours.

    7. Re:Let me count the problems... by Anonymous Coward · · Score: 0

      Having a plugin is nice, but it doesn't solve the PKI (key distribution and reputation) problem, and I am not very inclined to trust a plugin made by a company whose primary line of business is advertising by building user profiles.

      And I'm not too inclined to trust a company whose primary line of business is handing over everything-under-the-sun to vindictive ex-wives.

    8. Re: Let me count the problems... by Entrope · · Score: 2

      Adam wants to send a message to Betty without anyone being able to snoop on it. Eve wants to snoop, for example by tricking Adam into thinking Eve's key belongs to Betty, or keeping Betty from reporting that get key changed due to a compromise. PKI is how you keep Eve from being able to fool with keys.

    9. Re: Let me count the problems... by geek · · Score: 1

      Thats not PKI. PKI is Public Key Infrastructure, as in certificate managers and issuers. You're point is juvenile and underlays your ignorance on the topic.

    10. Re: Let me count the problems... by Entrope · · Score: 1

      Wikipedia agrees with me that a WoT is one form of PKI, and the published verification and trust statements that make up the WoT work as certificates of the associated public keys.

      You have an unreasonably constrained view of what qualifies as a PKI. A PKI is merely something that helps users reliably identify the public keys that are used by particular other users. Google's system here does not solve the PKI problem because it really only associates the public keys with an account, not with the end user, but people usually want to know who the person on the other end is.

  3. SMIME and DANE ? by johnjones · · Score: 3, Insightful

    How about support for SMIME ?

    It would be nice if they supported DANE so that all the keys where looked up automatically!

    Why not ?

    John

    1. Re: SMIME and DANE ? by corychristison · · Score: 2

      I long for the day that we can universally use DANE with SSL/TLS Certificates, and cut out the Certificate Authorities.

    2. Re:SMIME and DANE ? by Anonymous Coward · · Score: 0

      They're a big fish now and everybody trusts their CA, so there's no interest in DANE. DANE only takes control away from them.

  4. Lets get pragmatic here for a moment by JustNiz · · Score: 1, Offtopic

    I hate DRM as much as anyone but lets face it, if he did not ratify it into the standard, DRM isn't just gonna magically go away.
    The only effect not ratifying it would actually have is to ensure the continued existence of a fragmented mess of multiple different actual implementations across different sites.

    1. Re:Lets get pragmatic here for a moment by The+Raven · · Score: 1

      You replied to the wrong article.

      --
      "I will trust Google to 'do no evil' until the founders no longer run it." Hello Alphabet.
    2. Re:Lets get pragmatic here for a moment by JustNiz · · Score: 1

      Oops TY :-)

      By now I should know better than to get on the computer before I get my first coffee.

  5. Stop the FUD already. by Anonymous Coward · · Score: 0

    > Encrypted email is still difficult to use and painful to implement

    No, it isnt.

    I mean, I commend Google for doing that, but this stupid FUD that PGP/GPG is somehow "difficult" really gets on my nerves. With a functioning mail client it's not very difficult. Of course, if you're doing webmail... but you shouldn't be doing that in the first place.

    1. Re: Stop the FUD already. by Entrope · · Score: 1

      It's not hard to do PGP. It's only hard to do it properly, so that the Web of Trust works like it is supposed to.

    2. Re: Stop the FUD already. by Anonymous Coward · · Score: 0

      This is just a corollary of "security is hard".

    3. Re:Stop the FUD already. by Anonymous Coward · · Score: 0

      So true. I just taught my students to install and configure Thunderbird+Enigmail, it takes less than 15 minutes to have properly encrypted email.

      Also, a real mail cliente is much better tan any webmail.

    4. Re:Stop the FUD already. by Anonymous Coward · · Score: 1

      > this stupid FUD that PGP/GPG is somehow "difficult" really gets on my nerves.
      > With a functioning mail client it's not very difficult.

      Well, there's the crux, ain't there, buddy. There is zero to none 'functioning email client' in that respect, as far as I see it!

      Opportunistic encrption? Nope. Automatic key exchange? Nope. Hell...built-in functionality by default, nevermind it being turned on in the first place? Nope, nope, nope.

      The MUA's are the primary reason, why PGP IS 'difficult' to use. Instead of install and forget we get to talk to Grandma about 'per-recipient rules' and other nonsense.

      The WoT is highly overrated, though a technically clever idea. Except it don't work at scale and mostly not at all.
      Ditto keyservers. The reliance on the has got to be the stupidest thing ever. MUA's should directly and automatically exchange keys. End of story. And go jump out a window before you reply with any nonsense about key validity, that can't be ascertained in 90%+ of cases anyway!

      There are some, who have indeed used PGP since it came out and had high hopes for it. It's a cool idea, it can be done. But it's way, WAY more unreasonably involved and difficult than it needs to be. So wake me up, when Thunderbird et all has it actually activated by default and I can install it like HTTPS-Everywhere for Grandma and it just works out-of-the-box! Perhaps with a lower security level (naturally), but works nevertheless!

  6. Google? by aglider · · Score: 1

    So why is e2email by e2email-org and not by Google?

    --
    Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
  7. Yeah, but no by Anonymous Coward · · Score: 0

    the first problem, which is hard to fix, is: Google.

    While their core business is advertisement, their reach and influence has in parallel turned Google into a political mission. It's not in their interest to give people 100% private communications. Private from other people and petty criminals? Sure. Private from Google and the government they have to answer to? No, not a chance, never, ever.

  8. Greetings from the alternate universe! by TheRealHocusLocus · · Score: 4, Insightful

    People have been trying to find a replacement for PGP almost since the day it was released

    I've been around since PGP first popularized public key email and while there have been various problems with Zimmerman's implementation from time to time (as with S/MIME since)... I do not recall any broad opposition to it or GnuPG... besides intelligence agencies who would be satisfied with nothing less than outlawing non-escrow encryption. We were in fact excited and intrigued by it, and it was fun to use even if you weren't paranoid. This must be a dispatch from the Millennial Alternate Universe where or any project emitted by Microsoft or promised by Google or announced in a press release is considered to be a vast improvement on what came before it.

    End-To-End Encryption implemented solely in Javascript which is served up by the company that's not supposed to be spying on you is not worth the paper it's printed on. And Key Transparency is a fancy way of saying, use our single point of failure Internet Gizmo 'solution' to handle key management so you don't have to think about insurmountable issues of trust, as were directly addressed in Zimmerman's day (key signing parties, etc.).

    --
    <blink>down the rabbit hole</blink>
    1. Re: Greetings from the alternate universe! by Entrope · · Score: 1

      Sure, if you ignore X.509 and all the other PKI standards, no one has been trying to replace PGP's key distribution and verification schemes.

      But when you look at what has actually been going on, is pretty clear that -- whether their reasons are good or bad -- lots of groups have rejected the PGP approach to public key crypto.

    2. Re:Greetings from the alternate universe! by Anonymous Coward · · Score: 0

      as were directly addressed in Zimmerman's day (key signing parties, etc.).

      Key-signing parties are ridiculous.
        - let's gather a bunch of activists who want to meet without the government's knowledge, . . . then upload their social graph to Geocities? WHAT WERE YOU THINKING?
        - let's get a bunch of people who ought to communicate pseudonymously, who don't allow themselves to be photographed, who shun peers for "doxing", . . . then retrain their etiquette to demand government ID and put in-the-clear wallet name envelopes on all the encrypted messages they send? WHAT WERE YOU THINKING?

      The continued popularity of key-signing parties shows that nerds are sometimes incapable of thinking for themselves. Not only are they useless, all their premises are the exact opposite of what you should be doing within the practice (or actual) threat model.

    3. Re:Greetings from the alternate universe! by Anonymous Coward · · Score: 0

      Indeed, tools like GnuPG don't have to be difficult to use - I've got a plugin for Apple Mail that works well. The real "problem" with encrypted mail is that few people care about encrypting their mail.

      (Responding anonymously so I don't undo mods)

    4. Re:Greetings from the alternate universe! by Anonymous Coward · · Score: 0

      It works really well.... except when a new version of macOS comes out, then it breaks for around 6 months.

    5. Re:Greetings from the alternate universe! by TheRealHocusLocus · · Score: 1

      There is nothing ridiculous there. Key signing parties are the ONLY solution to the trust problem. Everything else compromises the idea through implicit (or unintended) centralized trust, misleading obfuscation or outright snake oil. The problem itself is ridiculous, not the only real solution.

      Which of my public keys is the right one? The first one you see in an unencrypted email to you or DNS-steered web page? The one that comes to you armored within SSL or S/MIME signed through a CA chain to Symantec whose subsidiary Thawte had gone a little rogue? It's that first key exchange between users that holds the greatest danger... and in the real world people suddenly feel the need for encryption only sometimes, such as a submission to WikiLeaks, I imagine that among the world's CAs the pressure to sign rogue www.wikileaks.com certs is intense. It is even illogical to assume it has never happened.

      My ability to create a 'fake' WikiLeaks PGP key pair is very useful because it reveals underlying truth. You know you could do it too and therefore, the state of being vulnerable is known to both of us. To solve the problem of how key signing parties might become practical, is also to solve the people-trust problem. They are the same.

      In all these years since practical RSA, there has been plenty easier this or less attackable that, but in my view there has been only ONE true lightning-strike moment. As Perfect Forward Secrecy is implemented, at least now when private server keys are compromised we will no longer have previously captured encrypted intercepts, perhaps even years of traffic, suddenly readable.

      In the realm of key trust between strangers, no progress. Be wary of anyone who offers it to you. They're probably just asking you to trust them.

      --
      <blink>down the rabbit hole</blink>
    6. Re:Greetings from the alternate universe! by Anonymous Coward · · Score: 0

      Today's key-signing parties should be done in the seediest corners of the Second Life, with clothing optional avatars so everybody can see you've got nothing to hide.

    7. Re: Greetings from the alternate universe! by Anonymous Coward · · Score: 0

      Key signing parties were the distraction, the sleight of hand. They ensured that there would be encrypted traffic that wasn't necessarily covert-the haystack. The covert PGP use, the needles, were unsigned certs used on the sly between people who had made arrangements out of band to exchange keys. Without the bombast of key signing and public haystack-building, the actual traffic, the needles, would have been easier to find.

    8. Re:Greetings from the alternate universe! by grumpy_old_grandpa · · Score: 2

      When Snowden wanted to initiate communication with Greenwald, would it really have been a good idea to use keys which were linked to their real names? And either way, using existing keys or newly minted ones, wouldn't they have to confirm the key fingerprints off-channel anyway? In that scenario, you really want to make sure you got the right one.

      For other types of communication, the threat model is different: When I send a message to my family, the content of the message is probably enough to establish that it was genuine. It would still have been nice if all governments and spies along its route would have a harder time reading it, though.

      The scenario I could see signed keys being helpful in, is valuable communication between two strangers. E.g. if the two us wanted to make a trade, and you'd send me your Bitcoin address, I'd trust you more if the message was signed with a signed key. However, if you were selling me illegal goods, we're back to square one. Neither of us would communicate with real names.

    9. Re:Greetings from the alternate universe! by Anonymous Coward · · Score: 0

      Yeah, didn't the snowden leaks say google (and microsoft and apple) was in bed with the NSA?

      A more user friendly end to end encryption is desperately needed, so that everyone's Aunt Tillie can get in on it too...

      But not from those companies!

    10. Re:Greetings from the alternate universe! by Anonymous Coward · · Score: 0

      > A more user friendly end to end encryption is desperately needed, so that everyone's Aunt Tillie can get in on it too...
      > But not from those companies!

      What's E2Email do differently than, say, Mailvelope ( https://www.mailvelope.com ) anyway?

  9. Why don't Android mail apps not support encryption by Anonymous Coward · · Score: 0

    I would suggest, that Google should just add encryption in an easy and good way to their mail app on Android. This should really be a problem, so why not do it?

  10. it's google by Anonymous Coward · · Score: 0

    the nsa hooks are implemented pre point a and post point z - by passing the end to end.

  11. E2E anathema to Google core biz by Anonymous Coward · · Score: 0

    My assessment is that Google has done the calculus and figured it's better for their biz to spend $ rigorously defending warrants than promoting e2e. E2e is anathema to their ~sole source of revenue, funding all their other play things.

    Kinda like Bell Labs was before divestiture (look it up, kids). The National phone company put off so much $ the research guys, like Kernighan & Ritchie could go off and invent amazing, amazing stuff, and not worry about a path to $.

  12. PGP does n scale by bib1620 · · Score: 1

    It does not scale. You can create a key for any domain with no verification - remember people tend to blindly click accept. You can create key with no expiration date, or long expiration dates - I'm sorry, but that is stupid.

  13. If they intend on dropping it, at least its open by The_Revelation · · Score: 1

    Unlike useful applications like Picasa that Google just dumped unceremoniously without patching significant bugs like its inability to scale previews to different resolutions.

  14. One word: Darkmail by Anonymous Coward · · Score: 0

    Look it up!

  15. ProtonMail by Anonymous Coward · · Score: 0

    Of course we can trust someone in bed with the NSA to "encrypt" our emails.
    It's not like they have a backdoor or anything

    Encrypted email is not hard to use, it's called ProtonMail, it is open source, completely encrypted email, created by people who aren't NSA agents.
    And the best part is it isn't intrusive like most encrypted emails and the design looks really professional.

    *I'm not a shill, I promise :D*