Re: Compression and crypto

daw@xxxxxxxxxxxxxxxxxxxxxxxx (David Wagner) wrote in

My complaint, again for clarity, is that while decompressions will lead
to strings with the correct alphabet [e.g. ASCII, binary, symbols of
your choosing] they won't follow a grammar [or even make sense].

For instance, you could map all random strings to decodable JPEGs but
will you also ensure the content of the JPEG makes sense and isn't just
white noise? Replace JPEG with a binary image [e.g. program] or text
document or HTML or PDF or ...I'll be able to tell the key was wrong by
checking if the decoded output makes any sense.

For what it's worth, I believe tomstdenis is absolutely correct here.
(He has made this point many times before, but somehow it doesn't seem
to have sunk in yet.)

For what its worth you picked a perfect time to reply in this thread
to support one of your fans. At least years ago he was. Tommy is trying
his best to CHANGE from the example I give. It would be very hard to
if not impossible to check all keys and see how many make a nice jpeg.
If you notice I do what most people do I show easy example where bijective
means and a small file would help. You could argue that Unicity distance
and other considerations are minor thats your choice. But the fact you
comment here and not on the simple example makes me and I hope others
wonder what your motives are. The simple fact is Tommy has failed many
times to make a point there But somehow you lack the objectivity to see it.
You seem to imply that he is right about my example. He is not he
changed the subject but I feel your implying he was right about my
example. Can you be straight forward and say I am wrong or not about
the example I gave.

Let me attempt to clarify it. I original stated with by bijective
programs and BICOM I could take a one byte output file and get more
than one possible valid english text message that compress and encrypt
to it. Dave is that wrong is this is one of Tommy's many times makeing his

I then went on to state why that even if I don't use my routines
but only Matt's BICOM that the one byte would map into many millions
of possible files depending on the key and that it was trival to
produce more than 256 such files. Tommy implied that was wrong
is that one of his points he made. Be honest and say if it was.
Now of course it would be much harder in those millions of possible
outputs to find resonable english since without my routines or
other bijective compressors tuned to english text there would be
many mamy files without ascii in them.

Or are you going to say BICOM which is not my code is something you
never looked at. So again I feel your condemming at least what I say
with out every checking if BICOM can do what I say it does. It was
written in way to be portable Matt did a good job. Again BICOM
is not my code it was written as a stand alone PPM compressor
with CBC AES encryptor that maps from all files to all files.
It could be uses as a component in an encryption system that
inculdes add random data or authemtacation. These last two are not
part of it. But it would be childs play to add them in.

Below is a part of message back in thread back:

x As I stated before using even somthinga as simle as by bijective
x compressors with BICOM. I could porduce an output file as small as
x one byte with the valve of zero this all 8 bits are zero. When you
x decrypt and decompressing using the correct password you get the ascii
x english message I encrypted. If you do the wrong key must likely you
x will get a text that is close to like english as you test millions
x of key you get millions of message from this one byte file. The main

So much for the poor wording but I think its decodeable to the average
human brain with a slight understaing of english.

David A. Scott
My Crypto code old version
My Compression code
**TO EMAIL ME drop the roman "five" **
Disclaimer:I am in no way responsible for any of the statements
made in the above text. For all I know I might be drugged.
As a famous person once said "any cryptograhic
system is only as strong as its weakest link"


Relevant Pages

  • Re: Bijective - an explanation please?
    ... >> English text to an entropy density that is asymptotically ... > about 2 bits per character. ... Are you talking about only general-purpose compressors, ...
  • Re: Bijective - an explanation please?
    ... >>This gives the computationally-unbounded attacker an advantage over ... I couldn't find the best one listed (the 1.2957 bpc value from ... for English is not exactly fair. ... representative of good compressors on English prose. ...