random files from Layer 1, we chose 25 random files from the other

higher layers. What's in these files? We'll tell you briefly:"

Without knowing what is going on here, I can see an advantage of doing leaks a particular way.  Let's say a person has information to leak, he could separate it into a large number of sections, let's say "n" in number.  (I'm thinking perhaps 1000 parts).   Do a cryptographic hash of each portion, and reveal each of the hashes.  Select a provably random number of those sections (using input from the public, perhaps as many people who want to contribute), and reveal them.  The Public will be able to see the information, and it will also see that the information has the hash values previously claimed.  

The leaker won't be able to select which sections to leak, and therefore the material exposed will be representative of the total.  Due to sampling theory, any substantial amount of 'fake' sections would be exposed, if they exist, in the first round of releases.

As subsequent tranches are released, it will become more certain that all of the initial hashes are genuine.

                  Jim Bell