Announcement

Collapse
No announcement yet.

Developer Claims: "A New, Fast & Unbreakable Encryption Algorithm"

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Developer Claims: "A New, Fast & Unbreakable Encryption Algorithm"

    Phoronix: Developer Claims: "A New, Fast & Unbreakable Encryption Algorithm"

    An unheard of independent developer has proclaimed designing a new, fast, and unbreakable encryption algorithm. While he admits to not being a mathematician or cryptoanalyst, he's wanting to get this encryption algorithm in the mainline Linux kernel and distributions...

    http://www.phoronix.com/scan.php?pag...akable-Encrypt

  • #2
    Ehm, how about no?

    Comment


    • #3
      Embrace yourself for the impact. Negative comments coming!

      Comment


      • #4
        Well, if he's no mathematician, he actually doesn't know whether this thing is unbreakable or not.

        Comment


        • #5
          he never heard of Schneier's law : https://www.schneier.com/blog/archiv...eiers_law.html

          Comment


          • #6
            *grabs popcorn*
            ## VGA ##
            AMD: X1950XTX, HD3870, HD5870
            Intel: GMA45, HD3000 (Core i5 2500K)

            Comment


            • #7
              If it is not written by Bruce Schneider or Daniel J. Bernstein, then I do not want it!

              Comment


              • #8
                As I have very little clue about cryptography I of course cannot judge the algorithm, but it looks a lot like a case of Dunning-Kruger effect.

                Anyhow, has it been published in a peer-reviewed cryptography journal yet?

                Comment


                • #9
                  Originally posted by soulsource View Post
                  As I have very little clue about cryptography I of course cannot judge the algorithm, but it looks a lot like a case of Dunning-Kruger effect.
                  I'm not a psychologist but yes, it seems.
                  ## VGA ##
                  AMD: X1950XTX, HD3870, HD5870
                  Intel: GMA45, HD3000 (Core i5 2500K)

                  Comment


                  • #10
                    Originally posted by thorgal54 View Post
                    I doubt he's heard of Bruce Schneier.

                    This is security through obscurity, "Bet you can't figure out what I did!" except the code was posted along with it, so it's more of a logic exercise than a meaningful new form of encryption.

                    I mean really, this code is just pasted clear-text onto this blog. There's no file, no header, no API, not even a description of how to use it! And this is expected to be merged into the Kernel?

                    For anyone who's curious, I found this interesting bit in the code sample:
                    Code:
                    [COLOR=#222222][FONT=Arial]/* UNOPTIMIZED VERSION for BETTER UNDERSTANDING OF THE FUNCTIONING OF THE ALGORITHM. IT IS NOT USED IN REAL LIFE. USE OPTIMIZED VERSIONS![/FONT][/COLOR]
                    [COLOR=#222222][FONT=Arial] * Encrypts or decrypts InOutBuf[/FONT][/COLOR]
                    [COLOR=#222222][FONT=Arial] * KeyBuf is the raw key buffer[/FONT][/COLOR]
                    [COLOR=#222222][FONT=Arial] * KeyCheckSum is 8 bit CRC checksum: Used to prevent "Related key attacks". If some bits of the key changes, entire cyphertext changes[/FONT][/COLOR]
                    [COLOR=#222222][FONT=Arial] * InOutDataLen is the length of the data to be encrypted or decrypted[/FONT][/COLOR]
                    [COLOR=#222222][FONT=Arial] * InOutBuf is the pointer to the data to be encrypted or decrypted[/FONT][/COLOR]
                    [COLOR=#222222][FONT=Arial] * Salt(or nonce) is a 4 bytes random number array.[/FONT][/COLOR]
                    [COLOR=#222222][FONT=Arial] * This logic ensures us this: An original key is created with an original salt value, for example for an online communication[/FONT][/COLOR]
                    [COLOR=#222222][FONT=Arial] * for each distinct packet, in the packet header, we can transmit a specific salt value for that packet and we can encrypt it with original key and salt[/FONT][/COLOR]
                    [COLOR=#222222][FONT=Arial] * when the receiver receives the packet, decrypts the new salt value with the original salt value of the key and passes that salt value to function,[/FONT][/COLOR]
                    [COLOR=#222222][FONT=Arial] * and decrypts packet body with that salt value. This method prevents "known plaintext" attacks amongst othersZ.[/FONT][/COLOR]
                    [COLOR=#222222][FONT=Arial] */[/FONT][/COLOR]

                    Comment

                    Working...
                    X