Re: the calculus of encrypting non-textual data

[Peter Makholm <peter@makholm.net>]

Joel Rees <joel.rees@gmail.com> writes:

>>> Did you know that encrypting a picture sometimes results in a picture
>>> that looks like it has been through a random color-permuting filter?
>>
>> Can you proof it?
>
> Memory of coursework in encryption. The professor did some simple
> encryption on uncompressed images and showed how the results tended
> not to hide the things one would want hidden.

Encrypting uncompressed images (bitmaps) in ECB mode tends to reveal a
lot of the structure of the image. There is an example at wikipedia:

  http://en.wikipedia.org/wiki/ECB_mode#Electronic_codebook_.28ECB.29

This just illustrates that using ECB mode is really not any better than
any other form of simple substitution cipher even though the underlying
cipher is considered cryptographically strong.

I don't think 'simple encryptions of uncompressed data' is useful for
anything than an argument for doing it right.

> Then he pointed out that the parts of an image with the most
> information are the parts that are least likely to compress. And he
> pointed out that standard encryption methods tend to be byte-oriented,
> for speed.

I think block ciphers are the most common norm these days with 128 bits
being the nmost popular block size. But even more importantly is an
enhanced focus on how ciphers are applied to the plaintext in ways that
doesn't leak structural information like ECB mode does.

//Makholm