Slashdot Mirror


Blockchains Are Poised To End the Password Era (technologyreview.com)

schwit1 shares a report from MIT Technology Review: Blockchain technology can eliminate the need for companies and other organizations to maintain centralized repositories of identifying information, and users can gain permanent control over who can access their data (hence "self-sovereign"), says Drummond Reed, chief trust officer at Evernym, a startup that's developing a blockchain network specifically for managing digital identities. Self-sovereign identity systems rely on public-key cryptography, the same kind that blockchain networks use to validate transactions. Although it's been around for decades, the technology has thus far proved difficult to implement for consumer applications. But the popularity of cryptocurrencies has inspired fresh commercial interest in making it more user-friendly.

Public-key cryptography relies on pairs of keys, one public and one private, which are used to authenticate users and verify their encrypted transactions. Bitcoin users are represented on the blockchain by strings of characters called addresses, which are derived from their public keys. The "wallet" applications they use to hold and exchange digital coins are essentially management systems for their private keys. Just like a real wallet, they can also hold credentials that serve as proof of identification, says Reed. Using a smartphone or some other device, a person could use a wallet-like application to manage access to these credentials. But will regular consumers buy in? Technologists will need to create a form factor and user experience compelling enough to convince them to abandon their familiar usernames and passwords, says Meltem Demirors, development director at Digital Currency Group, an investment firm that funds blockchain companies. The task calls for reinforcements, she says: "The geeks are working on it right now, but we need the designers, we need the sociologists, and we need people who study ethics of technology to participate."

129 comments

  1. Sorry by Anonymous Coward · · Score: 2, Informative

    "The geeks are working on it right now, but we need the designers, we need the sociologists, and we need people who study ethics of technology to participate."

    Sorry, ethics died a year ago.

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

      she lost.. get over it. that "imwithher" tshirt's getting a bit ragged.

    2. Re:Sorry by NicknameUnavailable · · Score: 4, Insightful

      "The geeks are working on it right now, but we need the designers, we need the sociologists, and we need people who study ethics of technology to participate."

      We literally need none of those people, they're just the buddies of the people who write stupid articles like this.

    3. Re:Sorry by Anonymous Coward · · Score: 0

      I don't care that she lost.

      Its that HE won...

    4. Re:Sorry by gweihir · · Score: 4, Informative

      Quite to the contrary of the article, some experts (derided as "geeks" in the article) are currently exploring what the blockchain is all _not_ good for or at least not any better than traditional solutions. Unfortunately, that likely includes basically everything, with a slight chance that some variant of the failed "currency" angle ("Bitcoin") may be salvageable if there is a way to reliably curb speculation and create stability.

      In fact, I am currently supervising a BA thesis in this area and I expect that a mostly negative result will save the industry-partner a ton of money. I also already told the student that a negative result can certainly get a good grade if argued and justified well (same as a positive result, really). I have a second thesis lined up with a different student and industrial partner. Will be interesting to see what the outcomes are, but I do expect a "not now" or a "not ever unless special conditions are met".

      Incidentally, "ethics" in the commercial space these days means "how can we manage to not get caught and still get rich". And please keep out the sociologists and designers, they really are mostly useless with very few exceptions.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    5. Re:Sorry by networkBoy · · Score: 1

      NASDAQ is introducing a BTC futures market in 2018. Futures markets have the effect of smoothing thrash and adding longer term stability to a commodity.

      --
      whois gawk date unzip strip find touch finger mount join nice man top fsck grep eject more yes exit umount sleep dump
    6. Re:Sorry by gweihir · · Score: 1

      I know. That is why I think the "currency" angle may be salvageable. Will be interesting to see whether that works or not.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    7. Re:Sorry by SlaveToTheGrind · · Score: 0

      Yeah, I voted this swill down as "stupid" this afternoon -- apparently that convinced them it would get a lot of clicks and needed to run post haste. LOL

    8. Re: Sorry by Anonymous Coward · · Score: 0

      You sound like someone who does not understand the implications of decentralization on a global scale. It will have a huge impact and change many things. How in the hell can you be supervising theses and not understand this?

    9. Re: Sorry by gweihir · · Score: 1

      The problem is on your side: You are clueless. Decentralization must happen in order to have an impact. It will not happen here, at least not anytime soon. And I do understand that, while you are obviously a "tech cheerleader". Ask yourself how many past trends you perceived as game-changers and what actually became of them. And then maybe read up on the Dunning-Kruger effect. At the end of that process you may be able to understand why I am qualified and capable to supervise these theses. Or you may not.

      Incidentally, the students came to me with this topic because I was recommended to them for my expertise. There was zero pressure on them, and they defined the topic themselves before they contacted me.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    10. Re:Sorry by vtcodger · · Score: 1

      I know next to nothing about blockchains, and cryptocurrencies have always looked to me like the dumbest financial "innovation" since the South Seas Company. But I sort of vaguely thought that blockchains MIGHT be useful as sort of the digital equivalent of Notarization of paper documents. For example, imagine a trust set up by lawyers for any of the thousand usual nefarious purposes or maybe even some legitimate purpose. The trustee is incarcerated for unrelated acts having to do with failing to touch the proper bases while impoverishing widows and children. If there is only a digital copy, how is the document modified in a controlled manner to reflect a new trustee? I sort of thought blockchain might somehow be able to maintain a chain of continuity acceptable to courts.. Not so?

      --
      You can't see ANYTHING from a car, You've got to get out of the goddamned contraption and walk...Edward Abbey
    11. Re:Sorry by phantomfive · · Score: 1

      I would argue that bitcoin works brilliantly as a currency (though not much else: I can't imagine how it would work to keep track of your private info). It will never work as a day-to-day currency: you won't commonly go to buy pizza or laundry detergent with it, but for large transactions it can work great. Note also that this has been a problem historically with gold: people didn't mostly do day-to-day transactions with gold.

      --
      "First they came for the slanderers and i said nothing."
    12. Re:Sorry by Anonymous Coward · · Score: 0

      The fundemental engineering problem is centralizing trust relationships between multiple parties; effectively you have to figure out how to get a group of people to have power, and be able to charge money for a service people want, but not be able to do anything but provide that service. Trust between people backs these systems, and when you have a federal government that encourages millions of people to stream in from foreign countries and literally steal citizen's identities by forge identity documents in order to take jobs. This is the same government that then gets their underwear in a bunch over Russian Cyberwarfare targeting the US banking industry (they've been at it for a decade), you know there's a trust issue.

      Trust was meant to be local, and the technologies that back it were also meant to be local. Once the system grows too large, it becomes too unruly, and its ability to ensure what it was originally meant to will fail. This is one of the concepts that keeps a government in check and one that would be truely terrifying if ever solved.

    13. Re:Sorry by gweihir · · Score: 1

      Usefulness as a currency and usefulness as a speculation object are mutually exclusive. Bitcoin is currently a complete fail as currency.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    14. Re:Sorry by Anonymous Coward · · Score: 0

      Complete fail? Bitcoin worked pretty well for the Silk Road before it got busted. I suppose the question is whether bitcoin offers any advantage besides anonymity, which is great when you're buying/selling illegal stuff, but not so crucial elsewhere.

    15. Re:Sorry by Anonymous Coward · · Score: 0

      get over it anyway. there's no turning back the clock. gotta stop moping and figure out how to repair the damage, starting with the poor misguided MAGA dumb-shits who voted for him, and soon will be paying for it with their jobs sent overseas, unaffordable healthcare, a new tax-cut-fueled market bubble crash, Russians spreading FUD everywhere and laughing about it, and maybe even a hot nuke war with the Rocket Man. Hillary and the dems ignored the "deplorables" and gave the election away. Time to own it and move on.

    16. Re:Sorry by phantomfive · · Score: 0

      Bitcoin is currently a complete fail as currency.

      This is either ignorant or an outright lie: it fails to acknowledge the actual uses of bitcoin as a transaction medium. In brief many of the transactions e-gold was used for, bitcoin has taken over.

      I don't any bitcoin at all (I wish I did), but come on, if you're advising BA theses, you better get your head screwed on straight.

      --
      "First they came for the slanderers and i said nothing."
    17. Re:Sorry by Anonymous Coward · · Score: 0

      This blind arrogance is why Silicon Valley keeps being blindsided..
      - Facebook being used for personalised propaganda by Russia.
      - The information society actually means 'nobody knows what's true anymore'
      - A rise in self-cencorship as data increasingly influences chances to get a job, insurance, etc.
      - Large scale ICT projects failing to model or get adopted by reality.
      - Building a surveillance infrastructure of immense proportions.
      - We have machine learning algorithms that amplify discrimination.

      If we don't bridge these worlds we'll keep getting "solutionism".

      This problem was pointed out 50 years ago:
      https://en.wikipedia.org/wiki/The_Two_Cultures

    18. Re:Sorry by gweihir · · Score: 1

      You seriously do not get it. Fascinating. It is fortunately I do not supervise your thesis, you would fail completely. Your analytical skills suck.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    19. Re:Sorry by gweihir · · Score: 1

      You seem to have overlooked a little, but critical component of my statement: The word "currently". Incidentally, Bitcoin is not really anonymous either. Unless you are very careful and add extra measures, you can often be identified after a few transactions to several others. It just takes some effort.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    20. Re:Sorry by NicknameUnavailable · · Score: 1

      Are you joking? Silicon Valley isn't a tech hub, they're a .com hub. They are a part of the problem, but they most certainly aren't innovating anything. They're marketing and PR hacks, lots of writers and social justice types, who focus on presenting themselves as innovative. Nobody actually working in tech buys that nonsense.

    21. Re:Sorry by painandgreed · · Score: 1

      Incidentally, "ethics" in the commercial space these days means "how can we manage to not get caught and still get rich".

      As described by my ethics people where I work, ethics literally means following company policy. If you are following written company policy, then it's ethical, and if not, it isn't. Whether it's legal or moral has no bearing on if it's ethical.

    22. Re:Sorry by phantomfive · · Score: 1

      Your analytical skills suck.

      And you had none. Your post is pure insults.

      But I can insult you too: if you think the only use for bitcoin is a speculation object, you're just ignorant. Fortunately ignorance can be cured, stupidity is forever.

      --
      "First they came for the slanderers and i said nothing."
  2. These are experts ... by CaptainDork · · Score: 3, Informative

    ... with, apparently, no experience:

    ... a startup that's developing a blockchain network specifically for managing digital identities.

    --
    It little behooves the best of us to comment on the rest of us.
    1. Re:These are experts ... by Luthair · · Score: 1

      Based on past stories that made it to the front page its the same sort of fine journalism one can expect from the MIT Technology Review.

  3. The last thing we need by Anonymous Coward · · Score: 0

    is more psychological manipulation in society.

  4. Ad by Anonymous Coward · · Score: 2, Informative

    See subject

  5. Admin hell by fluffernutter · · Score: 1

    As someone who needs to sign in with a password to a server possibly a hundred times a day, this sounds like it could possibly be hell.

    --
    Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    1. Re:Admin hell by Anonymous Coward · · Score: 0

      Erp? Use ssh, like everybody else. Disable password logins. Use key based auth.

      There's no need to type in passwords a hundred times a day to administer systems, unless you are doing it very, VERY wrong.

    2. Re:Admin hell by gweihir · · Score: 2

      And as somebody that uses certificate-based logins (ssh) regularly, I wonder what problem they are trying to solve....

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    3. Re:Admin hell by sexconker · · Score: 3, Insightful

      Erp? How does that help anything? If you're not providing your password each time you authenticate, then somewhere it exists in an accessible form to some automated system. Using keys? You need to encrypt them (with a password). There's no getting around it. A password is at the heart of all proper authentication schemes because it is the only method that is even possible to be secure. It is the secret, the "something you know", that exists only in your head. Nothing changes if you use it to encrypt a key or a keychain or a password database or whatever else.

    4. Re:Admin hell by Dorianny · · Score: 2

      The problem of how to squander the hundreds of millions being poured into blockchain startups by VC's that mostly don't even understand what it is

    5. Re:Admin hell by Anonymous Coward · · Score: 0

      It is the secret, the "something you know", that exists only in your head.

      I've never encountered a system used by actual people where what you said is true.

    6. Re:Admin hell by gweihir · · Score: 1

      Makes sense. So they are trying to scam VC's basically.

      --
      Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
    7. Re:Admin hell by Anonymous Coward · · Score: 0

      > Using keys? You need to encrypt them (with a password). There's no getting around it.

      Sure there is. Generate the keys on and keep the keys in a smartcard like a YubiKey.

    8. Re:Admin hell by fluffernutter · · Score: 1

      How often do you change your certificates? If an admin that worked for you a year ago got your private key and then got canned in a very nasty manner, would it still be valid or do you change it every 90 days or so?

      --
      Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
    9. Re:Admin hell by Anonymous Coward · · Score: 0

      I wonder what problem they are trying to solve....

      They want to track you... more big brother crap. Blockchains track user activities, so big brother has info about when and where you logged into what system.

    10. Re:Admin hell by u801e · · Score: 1

      And as somebody that uses certificate-based logins (ssh) regularly, I wonder what problem they are trying to solve....

      A better problem to solve would be to have a more intuitive UI for managing client side TLS certificates. It would be nice if I could just use a client side TLS certificate to authenticate myself on Slashdot instead of (or in addition to) a username and password.

    11. Re:Admin hell by Anonymous Coward · · Score: 0

      The problem they are trying to solve is having to use those pesky browser fingerprints and ultrasonic sounds to track you across the interwebs.

      Sincerely,
      MBA

  6. Sounds idiotic by Anonymous Coward · · Score: 2, Interesting

    Blockchain wallets have to be secured, else anyone can impersonate the user and do what they will with the contents. So what would a blockchain credential system be? An online password wallet, in effect, exactly as secure as the protection on the wallet... which is either going to be what you have (an app on your device) or what you know (a password).
     

    1. Re:Sounds idiotic by Anonymous Coward · · Score: 0

      Exactly, we would just be trading one type of lack of security for another. Nothing is going to change the fact that there really is no such thing as security. There will always be a 14 y/o kid that cracks your most secure server, for the fun of it, using some technique that no one before him/her thought of.

    2. Re:Sounds idiotic by Desler · · Score: 1

      It is idiotic, but since it uses “blockchain” in it that means they’ll easily score heaps of VC money to burn for years.

  7. Um... that's exactly when Private Keys are best! by Anonymous Coward · · Score: 0

    If you're always signing into something like that, then you should have already setup a public/private key solution for yourself, fool.

    Seriously. This is why our computing experience sucks; we've got fools like "fluffernutter" running things.

    Work smart, not hard.

  8. Why is "Blockchain" even necessary, here? by Anonymous Coward · · Score: 0

    This is just public/private key authentication (and possibly encryption).

    That's how MIT already requires its students to access web-based resources, anyway.

    Here's the real reason passwords haven't yet gone away: They have been good enough.

    Of course, now that scammers/"hackers" are so sophisticated, and computing systems so complex as to be stupidly designed, passwords might no longer be good enough.

    1. Re:Why is "Blockchain" even necessary, here? by freeze128 · · Score: 1

      Offline access will need to be protected with passwords anyway.

    2. Re:Why is "Blockchain" even necessary, here? by arglebargle_xiv · · Score: 1
      In fact the headline is wrong, it should be:

      Are Blockchains Poised To End the Password Era?

    3. Re:Why is "Blockchain" even necessary, here? by hoggoth · · Score: 1

      Exactly. Yubikey does this now. You have a secret private key that never leaves the security hardened device. The device will accept any token and encrypt it with the private key ("sign it"). The website or resource can decrypt with your public key to authenticate you.

      No blockchain necessary here. Blockchain would add nothing.

      --
      - For the complete works of Shakespeare: cat /dev/random (may take some time)
  9. bottleneck vunerability? by PhantomHarlock · · Score: 5, Insightful

    If everyone uses one private/public key set for everything, then if that is compromised then the third party gets access to absolutely everything and can impersonate the user?

    For those of us who use different usernames/emails/passwords from server to server that seems like a downgrade in security.

    Tell me I'm wrong and I'm missing something. I've used PGP in the past and use keys for SSH logins but I've never used blockchain related stuff.

    1. Re:bottleneck vunerability? by Anonymous Coward · · Score: 0

      Why, then you create a new wallet that contains your ID, ask for a hard fork of the blockchain, and voila!

    2. Re: bottleneck vunerability? by Monster_user · · Score: 1

      Which is essentially certificate revocation by the upstream certificate authority.

    3. Re: bottleneck vunerability? by Monster_user · · Score: 3, Insightful

      It is simply a password manager for more complex passwords.

      A downgrade in security most definitely, but it should have the same pros and cons of a password manager.

    4. Re:bottleneck vunerability? by DarkOx · · Score: 1

      As an individual quite possibly as an organization not so much. Right now attackers go after the authentication/authorization server / infrastructure (very often AD but not always). Its the first and primary target because if they can compromise that odds are good some of the following become true:

      1) They can authorize some existing account they have access to already
      2) They can change the authentication information for an account they want access to
      3) They can get the authentication information in bulk, maybe they have to crack it maybe not.
      3a) If they have to crack it for a large organization there will be a number of accounts with weak credentials that can be recovered using a dictionary attack of some kind, with some number of GPUs dedicated to the effort.

      This really addresses 2,3,3a and could help address one by making authorization changes transnational, and easier to audit..

      --
      Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
    5. Re: bottleneck vunerability? by Monster_user · · Score: 1

      The blockchain essentially becomes the"root certificate. If an individual account is compromised but the "root certificate" (block chain) is not, the key for that account (password) can be "revoked" or "changed".
      However, there will most likely be a window of opportunity before the "certificate" gets revoked, that is there is a window before the block chain is forked by a trusted authority, or otherwise before the integrity of the block chain is restored and the fraudulent access chain is no longer trusted. It is perhaps an automated mitigation strategy.

    6. Re:bottleneck vunerability? by Anonymous Coward · · Score: 0

      There shouldn't be any reason to limit yourself to a single "wallet". However, this would essentially be like having a password manager if you did.

      Like with a PW manager, if you lose the primary PW and the data base, it is game over.

      The advantage is that you never have to give any site your PW, it would be handled with a challenge/response. However, the trick is keeping your private key private.

      This is a good idea for how to handle public key exchange, however.

    7. Re:bottleneck vunerability? by Orgasmatron · · Score: 1

      There is no third party that needs to be trusted.

      Basically, you would open an account with your bank, and tell them that you have already registered id/PhantomHarlock. They would give you specs for a public key pair and a subidentifier. You'd create a private key to their requirements and publish the corresponding public key under the subidentifier they will be looking for.

      When you go to log in, you type in your username and their system consults the global distributed database. It finds the subidentifier they are looking for and makes a note of the public key you list.

      Then, it creates a random challenge, signs it with the bank's key, encrypts it with your pubkey and sends it to you. You decrypt it, verify the signature and now have a one-time password you can use to log in to the site. You know that it came from the bank because it was signed with their key, and they know that you are using it because you had to decrypt it with your key.

      Most of this can be automated. A browser plugin could pop up a box that you paste the challenge into, or it could be marked on the page in some way so that your browser can find it without your help. Much like SSL now, it could just pop up a box that says "You have received a challenge from XYZ corp. One of their published domain names matches the website you are on right now. Please enter the password for your private key store if you wish to log in." For the user it could easily be made to act very much like any other password safe plugins.

      --
      See that "Preview" button?
    8. Re:bottleneck vunerability? by sexconker · · Score: 1

      If you trust the bank's published public key, then you're not getting anything beyond the current method of trusting their SSL cert.
      Having a one time challenge presented is pointless within the context of an SSL-encrypted session.

      There's simply no benefit to such a scheme over creating a username and password for each site and using a password manager.

      The most impact you could have is the authenticating system's database not having to keep a copy of your hashed password for someone to eventually steal and try to crack. But the core problem there is someone using shitty passwords (and sites not using salts or key stretching), which is fixed by using a password manager.

    9. Re:bottleneck vunerability? by Orgasmatron · · Score: 1

      SSL involves trusting a few thousand third parties. The system I described would have zero. That would be worth the price of admission by itself.

      Second benefit: you can change your key just by updating your public record. No need to contact the other party and let them know.

      Third benefit: no one but you can change your key.

      --
      See that "Preview" button?
    10. Re:bottleneck vunerability? by ceoyoyo · · Score: 1

      As far as I can tell, the blockchain bit is clickbait. They're talking about public key encryption and signing.

  10. No by Anonymous Coward · · Score: 0

    No, they are not. This isn't going anywhere.

  11. Re: Um... that's exactly when Private Keys are bes by Monster_user · · Score: 1

    Your talking about public and private usage, when the real discussion seems to be about "Single Sign-On", or SSO.

    Otherwise what advantage does public and private keys for an individual offer over Kerberos NTLM authetication against a domain controller?

    If you're talking about multiple servers on different domains, then you're actually talking about implementing a SSO configuration for multiple domains using pre-shared keys in place of pre-shared passwords.

    Pre-shared keys require less typing, but are not always the ideal solution. How do you enable a more fine grained security approach to minimize the damage when a system is physically vulnerable? Say you turned your back and somebody tried to log into one of those server while you were working on another machine in the cubicle or office?

  12. LOL by EvilSS · · Score: 4, Funny

    users can gain permanent control over who can access their data

    So yea that's definitely not going to happen.

    --
    I browse on +1 so AC's need not respond, I won't see it.
    1. Re:LOL by Anonymous Coward · · Score: 0

      Discretionary Access Control was abandoned after the Windows 3.11 Workgroup era for a reason.

      DAC assumes every employee is trustworthy to the employer and is well-versed on information security to manage their own.

  13. "CryptoCurrency" Buzzword gets VC money by Anonymous Coward · · Score: 1

    Translation:
    Lets take Public Key/Private Key (ie PGP) methods, combine it with "BlockChain CryptoCurrency" words, and then get suckers that have more money than brains (or tech knowledge) to fund a startup that goes nowhere.

    1. Re: "CryptoCurrency" Buzzword gets VC money by Monster_user · · Score: 1

      Its reinventing the wheel, but it is a different technique/implementation.

      Does the blockchain mindset or methodology provide a superior visualization or understanding of the beat use implementation of the solution?

      Can people understand it better, so they can use it better?

    2. Re: "CryptoCurrency" Buzzword gets VC money by 110010001000 · · Score: 1

      People don't understand a blockchain or public/private keys or anything. The OP is right, just another guy trying to make a buck from VCs, and trying to avoid real work.

    3. Re: "CryptoCurrency" Buzzword gets VC money by Monster_user · · Score: 1

      I just had a thought. A block chain is sticking a tracking device on a credential, and following it around the network. If that credential shows up going two different directions at once that would be a pretty big red flag. Might have some place in corporate or enterprise authentication methods.

  14. Publish your PUBLIC key. Keep passwords local. by Anonymous Coward · · Score: 0

    With regard to Single Sign-On, that's just a matter of disparate servers having access to a copy of your public key. That's how this stuff works, folks. You publish your public key in any way that you choose, and then authenticate yourself via an asymmetric key cryptographic challenge.

    How should this publishing be done? Well, that's up to the services in question; just upload a public key to them, I'd say. It's public information, by definition—it doesn't matter whether there's a centralized repository.

    The private key must be encrypted, and decrypted for some suitable definition of a "session"; that way, the password database is locally controlled (and possibly quite decentralized), rather than centralized for easy access by malicious entities. If you leave your "session" available to someone else, well, there's no way to protect against stupidity without implementing Tyranny.

    1. Re: Publish your PUBLIC key. Keep passwords local. by Monster_user · · Score: 1

      Using the private key for authentication, but not storing it on the machine, but rather in your head, is a very effective means of improving security even if the machine is not locked.

  15. Already exists by Anonymous Coward · · Score: 0

    We already have certificate logins, which do everything they say they want to do, and actually do it efficiently.

    1. Re:Already exists by Desler · · Score: 1

      But those don’t say blockchain in them so good luck getting a Silly-con Valley VC to take notice. This is all just a scam to score VC money from suckers.

  16. What demented nonsense is this? by gweihir · · Score: 1

    Of course, nothing like that is true or even desirable. This story is utter nonsense. Credentials (whether passwords, certificates or seeds for OTP mechnisms) are under company control so their servers can access them easily and so they can revoke them fast. The blockchain has absolutely no place here. Incidentally, when it comes to public identities, the blockchain is about as useful as the PGP server network, albeit more complicated and more expensive, i.e. useless. The one thing that makes these identities worth more is signatures of (at least somewhat) trusted third parties on the public key, but only if they actually verified the identity.

    Seriously, stop pumping Bitcoin with utterly stupid stories. Let it crash already and let those greedy and stupid enough to have bought late suffer.

    --
    Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  17. Blockchain: the new 2 mpg car by Anonymous Coward · · Score: 1

    These days a single bitcoin transaction uses as much electricity as a home does in a week. How long before it requires a similar amount of computation, and by extension, energy, just to open a goddamn file?

    1. Re: Blockchain: the new 2 mpg car by Anonymous Coward · · Score: 0

      These new CS grads keep making things take longer and longer by adding abstraction upon abstraction.

      Still beats the EE types who would try to use PLCs to authenticate users by encrypting by ROT13 an odd number of times.

    2. Re: Blockchain: the new 2 mpg car by Anonymous Coward · · Score: 0

      EE types? Technicians maybe... PLCs are made to be simple enough that gorillas can use them.

  18. Re: So 2 factor authentication? by Monster_user · · Score: 1

    So your describing it as part of a three factor authentication, where two factors are pairs of PGP keys, where both public keys are stored in publically accessible databases, and the third factor is a temporary secret?

    Where does the audit trail of blockchain come into play?

  19. A password does not a private key make. by Anonymous Coward · · Score: 0

    Buddy, you're not as clever as you think you are.

    Every private key is a password, but not every password is a private key.

    1. Re:A password does not a private key make. by sexconker · · Score: 2

      Every password is a private key.

      It may not be a private key in a public/private key pair. But it's a secret (key) that is known only to you (private).
      It may or may not be used in the same crytographic manner (hashing vs. encryption) as, a public/private key pair. But it's a secret (key) that is known only to you (private).

    2. Re:A password does not a private key make. by u801e · · Score: 1

      Every password is a private key.

      It may not be a private key in a public/private key pair. But it's a secret (key) that is known only to you (private).

      That's not correct. It's transmitted over the network and is also known by the server on the receiving end so that they can validate whether or not it's correct. A private key, on the other hand, is never known on the receiving server since it's never transmitted over the network, unlike a password.

  20. YOU sound idiotic. by Anonymous Coward · · Score: 0

    The password would be a local matter, handled by the phone/device in question, not stored in a central repository with a million other people's credential data where it can be accessed with a single attack.

    I'm not sure exactly why "Blockchain" is necessary, but public/private key is a great idea.

  21. Re:Um... that's exactly when Private Keys are best by sexconker · · Score: 2

    If you're always signing into something like that, then you should have already setup a public/private key solution for yourself, fool.

    Seriously. This is why our computing experience sucks; we've got fools like "fluffernutter" running things.

    Work smart, not hard.

    Many systems don't support such a setup. Most systems (servers, networking gear, UPS, building/infrastructure management, etc.) still require a simple password as the lowest level authentication mechanism.

    And how the FUCK would a key pair help? You still need to present the private key somehow. Carry it with you? Gee, better not carry it in plaintext, so you better encrypt it in some sort of reversible way. With a password.

    DERP.

    And finally, working smart doesn't mean you don't have to work hard. Why not work smart and hard?

  22. Re: Blockchain may not be all good, but it ain't a by Monster_user · · Score: 1

    As I understand it, blockchain is like a bunch of trees growing in synchonicity. Exact duplicates. If a tree doesn't grow exactly the same, it is considered defective and cut down.

    A blockchain might function as a web accessible smart card or key fob which functions for all accessible websites. This keyfob would need to be protected in some way.

  23. Re: So 2 factor authentication? by sexconker · · Score: 1

    No, it's 1 factor. It relies on 1 thing you know (your private key) presented over a single channel (the internet).

  24. Re: Blockchain may not be all good, but it ain't a by gweihir · · Score: 2, Insightful

    This keyfob would need to be protected in some way.

    While your picture is somewhat correct and somewhat wrong, this really is the key-point. Incidentally, this is already the key-point with a password, but there it is relatively easy to do. All those that got weakly protected customer passwords stolen in the last few years were just grossly incompetent in protecting them. It is well known (to experts) how to do that right: salt, hash, iterate and in newer times add a large-memory property. PBKDF2 was the standard since at least 2000 and and is still doing reasonably well with good parameters. Don't use it for new designs though. Argon2 is the new standard. Both are not hard to use, but you need to know about them and understand why they work and that relatively low level of expert knowledge was already not available in all these hacked companies.

    So the blockchain really has no place here as it does not solve the problem, and it also does not make it easier to solve.

    --
    Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  25. Re: So 2 factor authentication? by Orgasmatron · · Score: 1

    It is either 1 or 2 "factors" depending on user choice. That it isn't obvious says a lot about the uselessness of that system for classifying authentication schemes.

    I didn't really explain how the blockchain figured into it, but after that it is a textbook-standard public key authentication system with some dress-up for the web. We could do it today if we wanted to.

    The blockchain part is for key management, also known as "the hard part of PKI". Basically, you only need to communicate one identifier to everyone you do business with. You don't need to tell them when you revoke a key and you don't need to trust them to keep your key safe. You could use the same key for everyone, or different keys, and it makes no difference.

    Namecoin allows decentralized registration of identities and allows people to publicly attach metadata to them. You register a name and you own it as long as you keep renewing it, which is dirt cheap and getting cheaper over time (which was by design). Any time you want to, you can update the metadata attached to it, which is a JSON structure of arbitrary complexity. No third party can take it from you, or override your choice of what data to attach to it.

    --
    See that "Preview" button?
  26. hashchaining by Anonymous Coward · · Score: 0

    mod parent up, though recursive/iterative hashing - 'hashchaining' - tends to be commonly useful

  27. Re: Blockchain may not be all good, but it ain't by Monster_user · · Score: 1

    Makes sense, but still doesn't give me a complete picture.

    The block chain stores your/my public keys, and I guess any public keys of entities you/I do business with.

    Where do your/my private keys go? And how are they kept secured?

  28. Blockchain but what about SQRL by dougdbkv · · Score: 1

    There is currently a technology out there right now that addresses all the problems inherent in profile/passwords, i.e. they get shared, lost, site must hold a secret for user. SQRL (https://www.grc.com/sqrl/sqrl.htm) handles all that and more so why not just implement that tech? Admittedly there is going to be some cost but when/if it scales up it will the answer to all of those questions.

  29. If you are not worried about Blockchain by OYAHHH · · Score: 1

    You should be. The possibility of the ultimate loss of all privacy to the government should be scaring the bee-jeep-ers out of everyone!

    --
    Caution: Contents under pressure
  30. Yet another end to passwords? by QuietLagoon · · Score: 1

    How many times have we heard about some new technology that is going to obviate the need for the lowly password?

    1. Re:Yet another end to passwords? by Desler · · Score: 1

      Yeah but this scheme is a hot way to score VC money.

  31. Password can be SHARED secret by Anonymous Coward · · Score: 0

    Well, a password can be a shared secret, so that's not a private key, even by your own sorry definition.

    So, the Anonymous Coward is the one who's correct.

    1. Re: Password can be SHARED secret by Monster_user · · Score: 1

      As long as the password is not stored using reversible encryption, then it is for most practical purposes a private key, and not a shared secret.

      However, more important is the claim I was making that a password is an effective solution for certain security vulnerabilities created by other solutions.

  32. Re: So 2 factor authentication? by Monster_user · · Score: 1

    I don't see an advantage to blockchain in your comment. Blockchain is used to make PKI management eaiser? I can't visualize the entire concept, start to finish. In order to understand what advantage blockchain has over other options, or to understand where vulnerabilities might occur. I wouldn't know how to implement blockchain to manage PKI for my own use, not in a manner I trusted to be secure.

  33. So... by XSportSeeker · · Score: 1

    Another idiot who doesn't understand the tech proclaiming that it'll replace a tried and true standard when he doesn't really understand the scenarios where his product works or not.
    Join the masses of idiots who said biometrics are going to replace passwords, among others.

    Should've ignored it. Here's the red flag: "the geeks are working on it right now"

  34. Re: Um... that's exactly when Private Keys are bes by Lobachevsky · · Score: 1

    Proof of identity isn't the same as SSO. Whenever you access "https" the server is proving its identity to you, since you access its public key (certificate) and trace it up to the root certificate that you already have installed. The server does not "sign on" to your desktop to prove its identity or use some kind of password or login authentication.

    The blockchain can eliminate the need for getting blessed by a root certificate like Verisign (Verisign is very expensive, at $400/yr). That can open the door to consumers self-signing their data (no sane consumer would pay $400/yr to Verisign), and eliminates the need for "logging in". Any server can verify your identity through your own digital signature the same way your browser verifies a server's identity through its digital signature.

  35. Re:Um... that's exactly when Private Keys are best by Lobachevsky · · Score: 1

    If you've ever used kerberos or any other key management system, you'll realize that the password is only asked once when it has to read your password-protected private key from cold storage (disk) and thereafter it uses ephemeral keys stored in volatile memory and never bothers with asking for your password again until you reboot or shut down.

  36. Private key? by RobinH · · Score: 1

    Yes, your private key is protected by a ... password. And nobody's ever had their bitcoin wallet hacked. I'm not saying blockchain technology brings nothing to the table, but it's certainly not a panacea.

    --
    "I have never let my schooling interfere with my education." - Mark Twain
  37. Here's a clue for y'all by Baron_Yam · · Score: 1

    Whenever someone says 'blockchain', they're almost certainly leading into selling you an inefficient solution that doesn't apply to the problem they think it does.

    Just say 'no' to blockchains.

    1. Re:Here's a clue for y'all by Anonymous Coward · · Score: 0

      Yes, this.

      "Using our exciting new hi-tech blockchain system, each time you log in to a website it will use enough electricity to power an average US home for 1 week!"

  38. Passwords aren't the problem; centralization is. by Anonymous Coward · · Score: 0

    The problem is that passwords are now centralized.

    Millions of people's authentication credentials (or information that is sensitively related such credentials) are stored in a single place that is very easy to steal through a single security breach.

    In contrast, managing a private key, even with a password, decentralizes that information, and makes it the sole responsibility of the user who cares about it.

  39. Re: Um... that's exactly when Private Keys are be by Monster_user · · Score: 1

    Not really sold on the verification by way of digital signature.

    Blockchain seems like a good idea for maintaining the integrity of public information.

    However I can't fathom a solution for a consumer for which blockchain provides sufficient security. Who or what is the ultimate authority? Is it turtles all the way down?

  40. Centralization of credential info is the problem. by Anonymous Coward · · Score: 0

    When people say "private key", they mean the kind of data that can be used in public–private key cryptography; that information is stored by the user who cares to keep it secret, not by some third-party.

    Furthermore, sure, that "private key" is secured through a regular old password, but guess what? That password info is managed locally on just the computing systems of that one user; the password information isn't being managed in a centralized repository that holds the same sort of info for millions of other people. This decentralization of the password info makes it much harder for an adversary to gather and process it.

    Get it, yet?

  41. Re: Blockchain may not be all good, but it ain't by gweihir · · Score: 1

    I have no idea.

    --
    Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  42. Re: Blockchain may not be all good, but it ain't by Monster_user · · Score: 1

    I'm not sure I do either, getting totally confused by this blockchain concept. I think my comment was intended for another poster who said the blockchain/wallet would act as some kind of PKI management database. Facilitating the use of public and private key pairs for the general public at a lower cost than the current going rates of SSL certificates. So if this thing works, SSL cert vendors might go out of business, as "root" authorities will no longer be required. The blockchain will be the root CA, in some fashion.

    SSL public key signatures in blockchains are used as signatures in the history. Like a list of names and signatures on a title deed, it traces the ownership of the currency. There is little to be gained in breaching older segments of the block chain. The history of the chain is used for integrity verification and remidiation of theft attempts.

    And that is how I view blockchain, a remidiation tool, not a preventative measure.

    So in taking a blockchain which tracks the transfer of nothing, it would appear to reduce or eliminate the remidiation effect. Which leaves me befuddled as to what advantage this blockchain has over a database? Is this simply a peer 2 peer network of public key databases? Or is this a public profile of a person based on the companies they typically do business with?

  43. Re: Centralization of credential info is the probl by Monster_user · · Score: 1

    Nope. I generally don't store my passwords in a centralized database. Merely a salted hash of the password using irreversible encryption. You know, a "public key".

  44. Re: Um... that's exactly when Private Keys are be by vtcodger · · Score: 1

    "Not really sold on the verification by way of digital signature."

    Me either. But surely it's good enough for sites like slashdot where I really don't much care if someone logs in as me.

    --
    You can't see ANYTHING from a car, You've got to get out of the goddamned contraption and walk...Edward Abbey
  45. Re: Centralization of credential info is the prob by Anonymous Coward · · Score: 0

    Generally...

    I'm just passing here to say the previous AC is saying is right.

    Even hashed and salted, a password scheme is still a shared secret.

    A keypair of public and private key is a whole other beast.

    (given the pain it is typing on a phone, I'll let to the interested reader the acquisition of various texts to educate himself)

  46. Re: Blockchain may not be all good, but it ain't by Anonymous Coward · · Score: 0

    People are going on and on about how blockchain is the answer to just about everything, even problems it has little to do with.

    While it's definitely a useful engineering technique, people are starting to look at it as a hammer and every potential problem suddenly looking like a nail. There are a lot of problems it does not overcome, such as the fact that you need people willing to make up a network of nodes to process the blockchain in the first place. In some ways it's similar to "the cloud" in that way - you can build an in-house solution, but if not, ultimately you're using someone else's computer. Cryptocurrencies manage to justify this because of the potential payoff of running a node on their network, but a lot of potential applications for blockchain don't give that incentive. And quantum computing might render much of it completely useless, anyway. That isn't an unrealistic threat considering IBM has made a lot of progress with quantum computers lately.

  47. Friend, you're not as clever as you think. by Anonymous Coward · · Score: 0

    See subject.

  48. But password management is local to user. by Anonymous Coward · · Score: 0

    The whole point is to decentralization the management of passwords; rather than storing sensitive account-related information for millions of people in one centralized, easily hacked database, the responsibility for managing sensitive information is pushed out onto the people who really care to keep it secret, namely each user.

    It's not about getting rid of passwords from the perspective of the user; it's about getting rid of passwords from the perspective of servers.

  49. https does not prove identity. Only cert presence. by Anonymous Coward · · Score: 0

    Proof of identity isn't the same as SSO. Whenever you access "https" the server is proving its identity to you, since you access its public key (certificate) and trace it up to the root certificate that you already have installed.

    https... unless the certificate has been stolen. In which case you may get encryption, but you certainly don't get identity. Well, you may think you do, but of course it is false. Such theft happens often enough that it makes the whole identity thing moot; and the reason that happens is because idiots keep saying that https equals proof of identity, so others tend to rely on that, and eventually they get screwed when the cert's been compromised.

    The entire idea that these things "prove" identity is false. They provide encryption, and they prove the cert is present - and that's all they prove. Well, that, and that tech people were gullible enough to let Network Solutions and its ilk rip them off for a huge amount of money for a very long time.

    Any server can verify your identity through your own digital signature the same way your browser verifies a server's identity through its digital signature.

    That doesn't verify your identity. It verifies that a machine with that data is connecting. It might be you - it might be your teenager, SO, or a visitor to your home or office.

    The benefit of a password can be - if used properly - that it's *not* on the machine, and the only time you get to log in on a target system, hopefully via an encrypted channel, and hopefully one that's not been compromised by a stolen cert, is when you drag it out of your memory. Your memory, so far, is difficult to secretly compromise if you apply some discipline to your password choices and utilization. It protects you from the people around you using your "authorized" machine from logging into your stuff.

    Blockchains will never replace password functionality for this very reason. Which is not to say that blockchains won't replace passwords - after all, the tech community embraced the fiction that certs provided identity, when of course, they don't. You can't fix stupid.

  50. Use your phone as One True authenticator device... by Anonymous Coward · · Score: 0

    ...lose everything WHEN - not "if" - it gets hacked or stolen.
    The Stupid is strong in this one.

  51. No. by Anonymous Coward · · Score: 0

    I secure my wallet with a password.

    No. Just.... no.

  52. Re:https does not prove identity. Only cert presen by Lobachevsky · · Score: 1

    Encryption without identity is vulnerable to a man-in-the-middle attack. The whole point of root certificates is verifying your identity and checking documentation of ownership of a website before they give you a certificate. If it were just encryption, sans identity, then https wouldn't accomplish anything, and any one in the middle can just encrypt with their own self-signed certificate and you haven't achieved a secure end-to-end session. A stolen certificate is rare, and when it does happen, there's a revocation process. This is similar to someone else finding out your password and you have to change the password.

    Your illustrated problem of a teenager or someone else gaining access is not unique to digital certificates, the same is true with passwords, since passwords are generally used only once per session and the then the rest of the session assumes continuity of identity. If you go to the bathroom or leave your desk, those websites will assume whoever sits down next has the same identity. This is why most offices require employees to lock their terminal before going to the bathroom or leaving for a meeting.

    In fact, passwords are generally considered less secure than crypto signatures due to the prevalence of keyloggers. Most secure websites generally recommend 2FA. You can essentially flip the precedence, so the primary login can be the crypto signature instead of the password, and the secondary authentication can be a pin, or mother maiden's name or any arbitrary security question & answer, only when you're doing some thing that requires 2FA like transferring money out of your bank account or changing the security settings.

  53. Re:Um... that's exactly when Private Keys are best by fluffernutter · · Score: 1

    Keys aren't allowed in my environment, too easy for private keys to get out never to be changed.

    --
    Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
  54. About blockchain, I feel... by shic · · Score: 1

    I've been intrigued by blockchain for months... but feel frustrated by the (lack of) technical material I can find on the subject.

    I definitely want to use (something with what I understand to be the properties of) blockchain for a few different purposes:

    • Not-easily-falsifiable audit. I would like my (internet facing) application to record every relevant event - and provide confidence that the events could not (reasonably) have been tampered with ex-post. Essentially, I'd like to be able to demonstrate that an event didn't happen. An example of such an event might be to grant access to a service.
    • Distributed sequencing of events. I really like the concept of IPFSt... but I can't find much information about how to publish a sequence of documents using the technology... Each document only gets its contents once it exists in its final digital form. For IPFS to be useful, a publisher needs to be able to associate new documents with the existing one - and to do so in such a way as to retain control about association of future documents. As a consumer of published data, I need to acquire access to the latest documents associated with a document of interest.

    For both of these use cases, it seems that BitCoin has already tackled the questions. While, I accept, I could review Bitcoin source code - that would be time-consuming and provide an insight into only one implementation of BlockChain technology.

    I'd like to know: are there any good technical resources that tell me how BlockChain technology is implemented - in order that I can establish the effort required to adopt such a technology in the context of a specific application?

  55. You don't say by OneHundredAndTen · · Score: 1

    This from a guy with a deep, deep interest for that assertion to be true.

  56. Re: Um... that's exactly when Private Keys are be by Anonymous Coward · · Score: 0

    So you wouldn't care if Creimer cracked your PW, logged in as you, and posted his shitty affiliate links everywhere? Got it.

    Go ahead Creimer. We got the go ahead from the man himself.

  57. WHAT are you talking about? by Anonymous Coward · · Score: 0

    The server in question need not accept your authentication by a compromised certificate.

    Truly, man, you're not thinking.

  58. You really have no idea what you't talking about. by Anonymous Coward · · Score: 0

    EACH person should have a certificate that authenticates THAT PERSON.

    An employee leaves? Revoke his certificate; tell the server not to accept him.

    Man, you need to stop talking out of your ass; if you don't understand something, then keep quiet.

  59. DIY Cryptocurrency Mining by Anonymous Coward · · Score: 0

    If you want mine your own crypto currency, you need a motherboard with 19 PCIe 1X slots to plug in 19 GPUs and a couple of 1200W PSUs.

  60. Re: Blockchain may not be all good, but it ain't by Anonymous Coward · · Score: 0

    Neither. This is the literal idea of "throw X at problems until it sticks."

    This is utterly useless for replacing passwords. Why?

    Authentication is a centralized concept. There is an "authority" that vouches for the client, and those that respect that "authority" will honor it's response. By it's very nature, giving control of such trust to some random third party is ripe for abuse. No matter how hard you try to make abuse difficult.

    In this case, you'd have a bunch of servers everywhere validating everyone's logins, each one verifying them and coming to a consensus, then giving the result to the system that asked for it. But even those these servers would be everywhere you have multiple problems:

    1. Most current authentication methods assume local(ish) auth. Requiring internet access to verify a logon will break a lot of things (go ask any education IT staff about chromebooks and their internet logins.), further it adds complexity which gives more opportunity for hackers. Never mind the issues when internet access isn't available for whatever reason.

    2. Remote auth is vulnerable to DNS poising, and server hijacking. If this were to ever take off, you'd have every alphabet soup agency in the world trying to get their fingers in it. (What better way to frame people or keep tabs on them than by just logging in as them?) Don't assume the server operators will be immune, they'll be the first targeted.

    3. Blockchain isn't effective unless it only tracks the key changes. If it were to track every login attempt in the world, the resulting chain would quickly become too much for a local machine or individual server to verify. Leading to some other authority to be needed to verify everything back up to the current point in which the local machine / server can verify on it's own. Rendering the entire concept of using blockchain for verification pointless. Tracking only the key changes though does not prevent someone from obtaining and reusing them, nor does it permit for remediation should such a breach occur. Once again, this renders the concept pointless.

    4. Should such a breach occur, you have no info about it. The most economical method for this doesn't tell you about login attempts it only verifies them. (You wouldn't have the processing power to check them all in a timely fashion anyway.) So the end result would be a breach could occur and you'd never have any info about it until well after the point that you could do anything about it. (Just look at all of the breaches involving user creds recently, how many of them were within 24 hours? Or hell even within a week? Turns out most external authentication providers love buying as much time as possible to cover their asses before telling you your info was stolen.)

    5. External auth is vulnerable to human nature. There is nothing like the world turning on you because of a comment on social media. How many of these server operators could actually be neutral parties? Hell taking the US as an example, you'd outright require a majority of verification servers outside the US to avoid being locked out of everything due to social media outrage. We already have people who loose their jobs over crap they did on their days off, like speaking their minds, now imagine being locked out of your bank account if that happened. Could you imagine defaulting on loan payments because you couldn't access your own money to pay them?

    In the end the best thing for authentication is rolling your own SSO, and keeping the authority in-house. Using someone else to verify it, has already been done miserably, we call it HTTPS.

  61. Re: Blockchain may not be all good, but it ain't by gweihir · · Score: 1

    Indeed. I can actually not see any added value compared to a simple distributed database, like the PGP key servers. To me, it seems that the efforts to justify the value of the blockchain for _something_ are getting more and more desperate. A typical hype-cycle.

    --
    Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
  62. DIY Cryptocurrency Mining... by Anonymous Coward · · Score: 0

    If you want mine your own crypto currency, you need a motherboard with 19 PCIe 1X slots to plug in 19 GPUs and a couple of 1200W PSUs.

  63. Re:Passwords aren't the problem; centralization is by hoggoth · · Score: 1

    > the sole responsibility of the user

    My mom just called me because she lost "her yahoo" again. Tell me again about people being responsible for their private key infrastructure again?

    --
    - For the complete works of Shakespeare: cat /dev/random (may take some time)
  64. Oct 29, 1969 to Dec 14, 2017. :( by mark-t · · Score: 1

    [nt]