The Future Of... Data Encryption

The Future Of... Data Encryption

Summary: Digital memories are long. Emails, images, and documents sent today can resurface years from now, but new software could help ensure that what happens online, doesn't have to live there eternally. ZDNet correspondent Sumi Das explains how 'Vanish," the work of researchers at the University of Washington, Seattle, uses peer-to-peer networks to create unique encryption keys.

SHARE:

Topics: Hardware, Browser, Collaboration

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

15 comments
Log in or register to join the discussion
  • Deletion like a phone call???

    The government monitors and sometimes records phone calls.How does this compare? The government could require web-based email to be stored.Better to "wipe" the info off of your computer.Some expert software maker should invent software to store this info,that when you want to "wipe" it off,they can "wipe" that part of the hard drive that it is stored on.
    razzamatazzer@...
    • An email service needs to developed w' P2P

      and off the grid. that would be cool.
      pcguy777
  • RE: The future of... data encryption

    I agree with razzamatazzer; depending on whom you are
    hiding from, this isn't a completely reliable solution.
    I'll stick with invisible ink on the wrong side of post
    cards
    J044NY8
  • Again?

    Again?

    Didn't I refute this video not long ago?

    The data only needs to exist in plain text ONCE and be copied ONCE and you lose all benefits of this "vanishing encryption." If somebody really wants to copy the message and store it permanently - they can.
    CobraA1
    • So all SSL encrypted credit card transactions, bank transactions, etc

      is all useless, because it was in plain text "ONCE?"

      ...
      T1Oracle
      • In transit vs at destination

        Information in transit is an entirely different matter. It's very well protected by encryption.

        BUT - that's not what this is trying to solve.

        What this is trying to solve is the idea that information can be saved permanently at the information's destination.

        But that's where it fails. Because at the destination, it can be decrypted and copied before it has degraded.
        CobraA1
        • I thought they were talking about points on the way to the destination

          where it would still be encrypted, but may stick around for longer than you would like.

          An example would be the server storing and backing up (perhaps multiple times) your encrypted e-mail.
          T1Oracle
          • Generally not an issue

            If it stays encrypted, then it's not really not much an issue.

            Current encryption algorithms are considered infeasible to crack.

            Currently even if computation power doubles every year, it'll likely take well over 100 years to crack some of our current algorithms. Unless a serious weakness is found in the algorithms, the information is secure.

            And that's a really, really low estimate - every time a single bit is added to the key length, the amount of computation needed to crack it via brute force doubles. Key lengths can easily grow faster than Moore's law with very little impact on performance.

            "An example would be the server storing and backing up (perhaps multiple times) your encrypted e-mail."

            In which case only you would have the key anyways, which you can destroy at any time you wish. No reason for them to be storing your key.
            CobraA1
        • control alt print screen

          yikes !

          but its still a good idea, because it could not be proven a forgery either way -- (non-repudiation). like if someone created a fake account or two, and said you sent this etc ( a faked doc/text etc). You wouldn't be the creator. They would (the forger). So a screen shot wouldn't prove nonrepudiation. So still a clever concept.
          pcguy777
          • thoughts

            "but its still a good idea, because it could not be proven a forgery either way -- (non-repudiation). like if someone created a fake account or two, and said you sent this etc ( a faked doc/text etc). You wouldn't be the creator. They would (the forger). So a screen shot wouldn't prove nonrepudiation. So still a clever concept. "

            Generally, digital signatures are much better for that.
            CobraA1
  • Ahead of the curve!

    We have been pioneering this for years!




    http://fakesteveballmer.blogspot.com
    Windowsseven
  • My Idea

    Every home user would have a vpn gateway built into their home router. Every connection to every resource would generate a random synchronous one time (pad) key for that session via the gateway (hardware). So isp's would only know where packets are being routed to by IP or DNS info and thats it. Im saying this would be cooler than SSL certs because this would be built in to all SOHO tech, and server gateways by default. As Server memory gets bigger and bigger, maybe something like this could be a reality someday. an entirely encrypted cloud !
    pcguy777
    • one time pads and keys

      The problem with the one time pad is the length of the key - the key literally has to be as long as the data, and it has to be kept a secret. Now you have the issue of exchanging keys in a secure fashion.
      CobraA1
  • Video not working

    The video stops in the middle, tried again and again but it simply stops playing.
    malcarada
  • RE: The future of... data encryption

    i dont see what the big deal is, "if you dont want to resurface, dont put it in electronic form" has been a secure way of thinking since the days of "Sneaker Networking" (the bad ole days of 6Khz processors and dual 5-1/4" drives.
    glockmi