Re: Re: Reformated: How secure....

Firstly, I would like to thank Cypherpunks members for helping me 
evaluate and critique my document on PrivaSoft's security. I would like 
to address several of the issues raised on list list. 

    1. Alot of question would be answered if you had a copy of 
PrivaSoft to look at or at least a scrambled page to examine. Both of 
these are available at our WEB site 
(http://www.privasoft.com/privasoft) this is not an advertising plug, 
it really helps to see the thing in action to better understand its 
capabilities. 
    2. Re: the 9 digit key.
        Partially why you misunderstand the strength of the 9 digit key 
in PrivaSoft is b/c Graphical Encryption is very different form data 
encryption. Graphic encryption takes an image of your document and 
scrambles the tiles accross the page in a pseudo-random order based 
both on the encryption algorithm and the key used, Each key produces a 
completely different scramble. Further, descrambling entails moving  
tiles and recognition of readable text. An edge detection machine would 
not be helpful with PrivaSoft, as far as I know, OCR or human 
interface, or the PrivaSoft descrambler( with the correct key) are the 
only methods that can be employed. And all three of these take 
considerable time. Time is a key factor here (no pun intended). the 
strength of an encryption relies heavily on the time it takes to 
bruteforce check different keys. as such, even the addition of 10 
miliseconds to the process is a significant increase in security when 
taken on the large scale of how many keys need to be checked. 

    3. The key extension is useful when documents are scrambled using 
the same default key. This feature adds 2 digits to your key, thus, 
even the same exact document scrambled with the same key but a 
different key extension would produce a radically diofferent scramble. 

    4. With regard to the increased font size, this is not necessary 
but is helpful in that the tiles themselves would contain less 
recognisable features of letters. 

    5. As a fax security product, their is nothing comparable to 
PrivaSoft in ease of use, strength, verstility. I'm sure you have all 
recieved faxes that have come in skewed, with heavy line noise, a black 
line running accros the page, or any other combanaiton of fax problems. 
These do not effect PrivaSoft's capability to descramble the fax, 
PrivaSoft has several patented features that produce readable text even 
after moderate damage to the page. Also, if you haven't figured this 
out yet, PrivaSoft is the only encryption software that provides 
Hardcopy encryption and decryption, ie. you can print out a hardcopy of 
the scrambled image, and fax it, store it, even snail mail it and then 
on the recieving end, fax or scan it back into the pc and decrypt. 

    6. re: custimization:
        This refers to our capability, for corperate clients and the 
like, to produce batch operation or need specific features. Also we can 
increase the digits of the key if the client requires such, as in gov't 
agencies etc, however this is not necesary for day to day use as 
PrivaSoft is strong enough for most transmissions over the Net and via 
fax regardless of the prevalent comments by many members of this group. 
I do not take offence at the many slurs on our product. I fully believe 
that if any of those who believe that PrivaSoft is a fly-by-night or 
bogus product have not looked at or tried the software and therfore 
speak from less than a knowledgable position. Please do not take this 
the wrong way, I do not mean to insult anyones inteligence or ability, 
I just mean to say that many here are quick to judge, and they have a 
right to be suspicious of everything and everyone, that is the nature 
of security, but I do challenge the naysayers to truly check out the 
software before completely condeming its use and capabilities. I have 
personnally demonstrated are software to several key individuals in the 
gov't, military, and corperate arenas both executive and technical 
persons. All were impressed have taken the software back for further 
testing. I have recieved positive responces from them and many have 
recomended / are presently using PrivaSoft.

    Once again, thank you for your help in editing and restructurin g 
the How secure document and i will post an updated version when it is 
ready. Also, if anyone would like a scrambled page E-mailed to them to 
look at / try to crack please E-mail me as such, I do not wish to post 
one only b/c many of you might not want to recieve it. If you have 
further questions or comments, I do appreciate and encourage them, 
please feel free to E-mail me at privsoft@ix.netcom.com or post them 
here.

Steve O.

*************************************************
PrivaSoft TM                                    *
1877 Springfield Ave PO BOX 600                 *
Maplewood NJ 07040-0600                        *
Tel. 201-378-8865  Fax. 201-762-3742            *
Http://www.privasoft.com/privasoft              *
E-mail: privsoft@ix.netcom.com                  *
*************************************************

Steve writes:
>     2. Re: the 9 digit key.
>         Partially why you misunderstand the strength of the 9 digit key 
> in PrivaSoft is b/c Graphical Encryption is very different form data 
> encryption.

That indicates to me that you don't understand anything at all about
cryptography. I'm not trying to be cruel. People around here have seen
me cruel. Its just that bits are bits and it doesn't matter what you
are encrypting. You seem to think that because images "need humans" to
determine if they make sense that you have somehow won -- in fact,
the statistics of real vs random images are so different that I can
hardly see how one could have an easier time of it.

Amateurs pretending that they are professionals going out and selling
snake-oil crypto are one of the biggest threats in our business. I
doubt that you are going to relent -- folks in your position have an
amazingly stubborn ability to ignore reality -- but I think anyone
buying your product is crazy.

> I fully believe 
> that if any of those who believe that PrivaSoft is a fly-by-night or 
> bogus product have not looked at or tried the software and therfore 
> speak from less than a knowledgable position.

Reading your messages is sufficient, actually. If I read something
from someone pretending to be a doctor and he notes that his new blood
thinning machine is useful for curing baldness, I don't really have to
try the product, now, do I.

I suspect it would take a real cryptographer very little effort to
break your system, but that no one will bother doing so because it
isn't really worth anyone's time.

> I have 
> personnally demonstrated are software to several key individuals in the 
> gov't, military, and corperate arenas both executive and technical 
> persons. All were impressed have taken the software back for further 
> testing.

Who cares? I know places where you can demonstrate psychic surgery and
get dozens of people to vouch for it.

Perry

"Perry E. Metzger" writes:
>Amateurs pretending that they are professionals going out and selling
>snake-oil crypto are one of the biggest threats in our business.

and then...

>I suspect it would take a real cryptographer very little effort to
>break your system, but that no one will bother doing so because it
>isn't really worth anyone's time.

Well, Perry, if it's really such a threat, isn't it worth someone's
time to combat it?

Wasn't that point driven home by the Netscape PRNG problems?

Scott Brickner writes:
> "Perry E. Metzger" writes:
> >Amateurs pretending that they are professionals going out and selling
> >snake-oil crypto are one of the biggest threats in our business.
> 
> and then...
> 
> >I suspect it would take a real cryptographer very little effort to
> >break your system, but that no one will bother doing so because it
> >isn't really worth anyone's time.
> 
> Well, Perry, if it's really such a threat, isn't it worth someone's
> time to combat it?
> 
> Wasn't that point driven home by the Netscape PRNG problems?

Netscape is in wide use. If a substantial number of people bothered to
use the Privsoft, then it might be worth breaking it. As it stands, I
don't think its worth my while. Maybe someone out there wants to
bother to do it as an exercise. It unfortunately has a bit of added
complexity because you have to learn a bit about image statistics in
order to do a good job of segregating the images, but it also looks
like you might be able to use multiple anagramming to get a nice
handle on the thing so your statistics don't have to be particularly
good.

Perry

> Further, descrambling entails moving tiles and recognition of
> readable text.

Why do you make this claim?  If PrivaSoft's transposition cipher is
even superficially ok, a wrong-key decryption will look like a random
permutation of the input pixels, i.e. an image with the same
black/white statistics as the original (a slight weakness, IMHO) but
with none of its spatial coherence.  Look at the distribution of run
lengths, or of the size of connected components.  

I just went and looked at your "PrivaSoft in action" example, and I'd
have to say that the cipher is not "superficially ok".  The ciphertext
is visibly structured: there are visible fragments of letters (an "e",
an "n", a "k", the top of an "S"); there are evenly-spaced vertical
lines of dashes and crosses; I can see the bold text of the original
(what's more, it's only diffused over a small extent, not the whole
ciphertext); and, um, was the letterhead text supposed to be
unreadable, or just dirtied up a little?

Since the algorithm doesn't break up small-scale structure very well,
a more robust way of testing for correct decryption would be to
count the number of black pixels on each scan line, and examine this
for periodicity.  Even with some noise and scan skew, there will be
obvious periodicity for unencrypted text, and little for an incorrect
decipherment.

I don't mean to be unnecessarily hard on your software.  It's probably
fine against casually nosy people and for protecting mildly
embarassing information, and it's conveniently exportable.  But if you
represent it as suitable for high-value secrets, you're misleading
your customers.

--
   Eli Brandt
   eli+@cs.cmu.edu