Page 24 - index

Anyone who creates encryption algorithms will already know that 2^128 is 10^38 and 2^256
is 10^77. The highest estimate for the number of subatomic particles in the universe is 10^87
(10^90 possibly).

When I selected my infinite algorithm the key was efficiency. I could have reduced the
number of steps to allow quicker calculation but it would have been far less secure. I aimed
for an efficient algorithm to reduce the number of calculations needed rather than increase
the number of calculations to increase security.

There must be a sufficient safety margin built in. On the other hand you must not go too far
above it because it would take too long and only the very top cryptographers will have the
correct understanding of where this is. The algorithm must be super-efficient, not messy in
any way, because we are dealing with huge, huge, huge numbers of permutations.

The quickest algorithm is not necessarily the most efficient one. It would be an incomplete
algorithm. I could not get away with a deficient algorithm and my progress would have
halted. The five which reached stage 2 of AES work in different ways and will not have all
been equally efficient.

When you program a computer you can get away with a not-so-efficient system because it is
a computer that has to do all of the work. If you expect an average football-supporting
human to do all the calculations then you cannot get away with a not-so-efficient system.
That is why I had to devise an efficient infinite algorithm which could be used by ordinary
people. The average person can only memorize a certain number of calculations so the
algorithm's efficiency is going to have to do most of the work.

Once you know exactly how efficient an algorithm is supposed to be then and only then you
can correctly decide on the number of steps required in each step of the process.

But, that's just my opinion. Yours may differ.

With data encryption algorithms many different ones can be created (which will work
successfully). With the password encryption algorithm, however, there is only one (because
of a variety of factors). So you have to find one out of infinitely many rather than one out of
millions out of infinitely many.

Cryptographer Bruce Schneier (whose team of eight men submitted Twofish) has expressed
his view that each person can create an encryption algorithm that he/she cannot crack.
Everyone is entitled to their own belief and I respect his. I would like to express my own view
here, which is that I think it is harder to create an encryption algorithm than it is to crack one.
There are plenty of math geniuses around but there are infinitely many wrong encryption
algorithms. How many algorithms failed? How many of them didn't work efficiently enough?

Did any of the teams who submitted algorithms for AES create a password encryption
algorithm? Did any of the people who attempted the two Eternitys? Did you?

I like the last question. Ask the person who asked the question in the first place. I should
answer the question too. I did create a password encryption algorithm and all of the world's
best mathematicians did not. It's just not the sort o' problem they deal with.

! " $
! # ! "