h a l f b a k e r yYou want a piece of this?
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
This method of cryptography would not be suited for large scale or public use, as it helps if it is not known how the information is encrypted. It could be used instead for a small, secrative group or highly classified govornment communication.
Some sort of pad would be used that can record handwriting
digitally, like the things people write signatures on to be digitally stored or printed (though much larger), to write out a message. On the pad, a randomly generated shape would be displayed that is constructed of vertical and horizontal lines, but is much more complex than a rectangle. All of the writing must be kept within this figure. The writing would be recorded as a bitmap, but only the bits within the shape would be recorded. Because of this, you would have to know the dimensions of the shape that was used in order to properly display the bitmap. The dimensions for the shape would be stored in a way that is readable by the program used to decode the message and would be cut into smaller pieces and stored in certain places within the bitmap that are known by the decoding program. Certain bits in the file (a pretty large portion of them; determined with a simple, predefined formula) would be inverted (this is to remove long sequences of same bits) and the whole thing would be further encrypted with traditional cryptography (it wouldn't have to be very good).
The reciever would use a program that would decode the normal cryptography, uninvert the inverted digits, read and remove the dimensions of the shape, reconstruct the shape, and display the bitmap in it.
This method would be easily usable with simple drawings and symbols, which would often make the message more clear and understandable.
Zebra Randomiser
Zebra_20Randomiser [phoenix, Jun 26 2008]
Striped Fish Randomizer
Striped_20Fish_20Randomizer [phoenix, Jun 26 2008]
True Random Number Generator
True_20Random_20Number_20Generator [phoenix, Jun 26 2008]
[link]
|
|
This seems unsafe, complicated, and cumbersome. (Compared to, say, using PGP.) What problem are you trying to solve? |
|
|
So cut up an image and encrypt it. Combining security through obscurity with existing encryption methods (which are already secure) isn't going to do you any better than plain encryption, in fact it may just weaken it. |
|
|
...and you've got all the nasty problems of key exchange to deal with (i.e. securely communicating to your counterparty how to rearrange the bits of the picture). Security types will hate it because it's really hard to gauge how secure this is - it might come down to something like how big or legible your handwriting is. |
|
|
With handwriting like mine, who needs encryption? |
|
|
The truth is that your process would allow people to gain access to the stronger parts of your encryption protocol, as implied by [Bukka...]. |
|
|
This smells like a Gordian knot. Probably looks like one too. |
|
| |