Tue, 10 Apr 2007 14:25:49 -0400
|
On 10 Apr 2007 at 13:11, Bill Brown wrote:
>I researched it a bit and looked at an email I had with a 6MB attachment. I
>searched for that same 4 letter word in the encoded attachment and gave up
>after 40 instances. I was only about halfway through the file.
That's a bit curious... I'm missing something. A B64 character set with
lowercase and uppercase letters means you have two chances out of 64 to
get an "f/F" or 1 out of 32. 4 letters would be 32*32*32*32 or 1 per MB,
so you should have seen 3 (in half of a 6MB file) not 40. I tried putting
a bunch of various mixed case (f-)words at different offsets within a file
and decoding it, but got back nothing interesting in ASCII. So, uh, I
guess there's something else going on... like a word processor uses some
binary sequence that causes The Word to be generated an order of magnitude
more often than I would have expected. Math Error?
-Kary
|
|
|