Cryptography-Digest Digest #812, Volume #9 Thu, 1 Jul 99 03:13:05 EDT
Contents:
Re: Can Anyone Help Me Crack A Simple Code? (mercury)
Re: Kryptos article ("Douglas A. Gwyn")
Re: A Quanitative Scale for Empirical Length-Strength (Jim Gillogly)
Re: Why mirrors invert left-to-right (was: Kryptos article) ("Douglas A. Gwyn")
Re: Can Anyone Help Me Crack A Simple Code? (mercury)
Re: Why Elliptic Curve Cryptosystem is stronger with shorter key length? (Greg
Ofiesh)
Re: new book ("Douglas A. Gwyn")
Re: Can Anyone Help Me Crack A Simple Code? (Jerry Coffin)
Re: Can Anyone Help Me Crack A Simple Code? (S.T.L.)
Re: Can Anyone Help Me Crack A Simple Code? (mercury)
----------------------------------------------------------------------------
From: mercury <[EMAIL PROTECTED]>
Subject: Re: Can Anyone Help Me Crack A Simple Code?
Date: Thu, 01 Jul 1999 01:12:51 -0400
Reply-To: [EMAIL PROTECTED]
> Douglas A. Gwyn Wrote:
>
> You can't, even knowing the data used, without
> making assumptions.
I am trying to discover an encryption algorithm
by observing how it behaves. If "you can't", then
why bother using any of the "strong" encryption
algorithms? After all, if it were impossible to
discover how an algorithm works without directly
seeing it, then any cocktail of XORs and MOD
functions would be unbreakable. (assuming the
algorithm itself is in an "indestructable" container
and can not be directly observed)
We all know this is not true. Breaking a code, by
definition, (correct me if I am wrong) MUST involve
making assumptions, building on them, then seeing
if the assumptions lead anywhere. If you have the
algorithm, then you are not breaking a code. You
are solving an algebra equation.
I should also point out that my problem, which is
finding out what a group of numbers has in common,
is ultimately a math problem.
f(X1) = Y
f(X2) = Y
f(X3) = Y
f(X4) = Y
f(X5) = Y
f(X6) = Y
...
f(Xn) = Y
I know six of n X values. Y is unknown, but all
functions are equal to Y. Find the function.
Does it look like some sort of a multi
dimentional math problem now maybe? Perhaps a
polynomial equation?
I may need all of the X values to solve this to
find an exact solution, but I would be happy
with a general solution. I'm sure someone in
this newsgroup is a better mathematician than I.
> [EMAIL PROTECTED] Wrote:
>
> I hope what you are trying to do is not illegal. ;-)
> John Savard Wrote:
> Look, we _know_ you're trying to bust the serial
number scheme for some software.
What I am doing is not illegal.
This code is not a "Software Crack". Software
cracks are created with soft-ice. You use tools
which reveal the algorithm. In my case, I can
not see the algorithm directly - I can only observe
its behavior.
Ed Yang Wrote:
> I called all of those telephone numbers and
> I got the same result for each one. I got a
> recording from the telephone company asking
> me to look up the number again and make sure
> it is correct.
Mental Note - Bash this guy with as many insults
as possible when I have more time.
William "Billy" Tanksley Wrote:
> Okay. Then the most helpful answer I can give is that
> none of us can help.
I'm sure there have been many discussions in
this newsgroup about how important it is to use
a well known, unbroken algorithm. Why? Because
Bruce Schneier says so?
PhD Mathematicians are quick to point out that
some algorithms are a joke and easy to break.
I would like to know how. Perhaps we can discuss
this further at a later time.
> BTW, the statement "it does have an answer" is not
> something which can be deduced from the rest of your
> question.
Agreed. If I have not seen the answer, how can I
know if it exists? (Please do not try to look
for "deep meanings" in this statement. I mean
exactly what it says, nothing more)
> It's quite possible that the numbers change randomly
> and constantly. Have you checked for that?
No. This is explained better below.
> High instance of eights. Got any failing
> numbers for us?
> One very useful chunk of information would
> be whether there are large blocks of bad numbers.
Almost all numbers fail. There are very few
"good" number codes. I might say there are
somewhere between 10 and 1000 "correct" codes,
but I am only guessing here.
> There are 10^10 numbers, too many for generalized
> brute force.
My first attempt at looking for similarities
in these numbers looked for groups of digits which
were all divisible by the same number in
all six number sets.
So, I took the first 2 digits in each code and
looked for a common (prime) divisor with a
modulus of zero. Then I took digits 2 and 3
in each group and did the same. Then I checked
in groups of 3,4,5,6,7,8,9, and 10 digits.
This took about 20 seconds. 10^10 is not all
that big. Unfortunately, I can not try all
possibilities on the actual "algorithm machine".
> My advice is the same: if you want help, give us the
> whole problem.
Okay. I'll do the best I possibly can to explain
everything I know. You will have to excuse the
metaphores, which I feel are necessary for political
reasons. I hope this does more clarifying than
confusing.
***********************************************
The Black Box Company
A company offers a line of Black Box products to
manufacturing companies. They offer one with a
Green Light, one with a Blue Light, one with a
Violet light, etc. All Black Boxes are also
equipped with a red light.
The Black Box company is a monopoly, so they use
tricks to get as much money as possible from the
manufacturing companies using them. After all,
Black Boxes are money makers and the Black Box
company wants to make more money as the
manufacturing companies make more money.
So the designers of the Black Boxes come up with
a system. They make the Black Boxes dependant
on keys. In order to make a Black Box work, you
must buy a key from the Black Box company. When
you buy a key, you specify what color light the
Black Box has.
The key is only good for one specific color light
The key is date coded, so it must be used within
a certain amount of time.
The key will only make the light come on a certain
number of times. After that, the Black Box
Logs that the key has been used and will not
accept the same key again.
So companies must buy a certain number of keys
based on how much they plan on using the black box
within the amount of time the keys are good for.
***********************************************
Design of the Black Box
The electronics are of sub-miniature type on multi-
layer circuit boards, which are very static-
sensitive. There is a ten digit key pad for entering
the codes. The key pad is integrated into one of
the main boards. The Black Box has no RS232 port.
So it is nearly impossible (with the equipment I
have) to enter the codes any other way besides by
hand with the key pad. It is not possible to try
a large number of codes.
The light bulbs have a built-in microchip. This
chip configures the Black Box for whichever light
bulb is being used. With the exception of this
chip, the electronics are identical in all Black
Boxes. (I MAY be able to retrieve the information
off of this chip. I may have answered the question
of where the algorithm is right here. I'll look
into this further.) The lightbulbs are only
installed or changed at the Black Box company.
***********************************************
Black Box Keys
The Black Box company has always used methods to
hide inner workings of thier equipment, but these
methods were designed to hide thier secrets from
the curious, not the determined. I have been able
to figure out just about everything on the earlier
models. I've even taught quite a bit to the Black
Box technitions, who always seem to be the last to
know. I highly doubt they are using anything more
than a simple routine for generating keys. Yes,
there is the possibility I am wrong, but I doubt it.
The Black Box company proudly claims that they are
100 percent Y2K compatible, and their equipment will
work until the year 2999. They are also known for
using as little memory as possible in their products.
Yes, there is the possibility that there is
a huge table somewhere which contains all random
codes for all color lights for all dates until
2999, but once again, I highly doubt it.
Phil Zimmerman does not work for the Black Box
company. It is logical to assume that the
Black Boxes take a ten digit code and run it
through an algorithm which converts it to a date
and color code, or returns an error if it is
not a "good" code.
********************************************
So I can get codes. The 6 ten digit keys I
posted are for the exact same color/date
combination. I can get other groups of
matching codes, but I will not be able to
get all of them for any color/date combination,
nor do I know how many possible keys there are
for any combo.
I will look into other possible methods in
getting the algorithm for this problem, but
this is the area of cryptography that interests
me the most. Cracking crackable codes. If
cryptographers insist that some codes are a
joke and easy to crack, why am I having such
a hard time finding someone who can crack
codes or information on how to do it?
Perhaps keeping the algorithm a secret is just
as safe as using a powerfull well known one.
------------------------------
From: "Douglas A. Gwyn" <[EMAIL PROTECTED]>
Subject: Re: Kryptos article
Date: Thu, 01 Jul 1999 05:17:24 GMT
John Myre wrote:
> Like, "I have found a truly wonderful break for Skipjack, but
> the proof is too large to fit in the margin. - D. Coppersmith".
That didn't seem to work very well for Fermat's "theorem".
------------------------------
From: Jim Gillogly <[EMAIL PROTECTED]>
Subject: Re: A Quanitative Scale for Empirical Length-Strength
Date: Wed, 30 Jun 1999 22:59:34 -0700
Douglas A. Gwyn wrote:
>
> Jim Gillogly wrote:
> > In fact, I believe Courville showed that double transposition is
> > equivalent to single transposition with a much longer key.
>
> Sure, but that's a trivial observation that doesn't help crack the
> system.
Agreed -- just pointing out that the opposite conjecture was false.
--
Jim Gillogly
8 Afterlithe S.R. 1999, 05:58
12.19.6.5.16, 7 Cib 4 Tzec, Eighth Lord of Night
------------------------------
From: "Douglas A. Gwyn" <[EMAIL PROTECTED]>
Subject: Re: Why mirrors invert left-to-right (was: Kryptos article)
Date: Thu, 01 Jul 1999 05:16:02 GMT
> Then, P violation comes into the picture.
What P violation? The famous experiment that resulted in a Nobel
Prize didn't exhibit P violation, it exhibited a remarkable
failure of the analysis to properly treat magnetic field as what
is usually called a pseudovector, not a true vector.
(Pseudovectors are inverted upon reflection of the coordinate
system.)
------------------------------
From: mercury <[EMAIL PROTECTED]>
Subject: Re: Can Anyone Help Me Crack A Simple Code?
Date: Thu, 01 Jul 1999 02:11:48 -0400
Reply-To: [EMAIL PROTECTED]
Douglas A. Gwyn wrote:
> mercury wrote:
> > I am trying to discover an encryption algorithm
> > by observing how it behaves.
>
> It's not truly encryption; it is a hash into a single bit.
> Too much information is lost.
>
"Hash" may be a more appropriate term for what is
going on here. The input is ten digits. Since more
than one input reveals the same output, it is reasonable
to assume it is somehow getting reduced to an output
smaller than ten digits.
However, the output is a code which represents a
date, as well as other information, so the output must
be larger than one bit.
This would mean the input codes all contain the
same information which is not lost in the "hash",
then there is some added insignificant information
which is lost.
So we have 6 identical numbers with some extra
garbage to obscure their meaning.
------------------------------
From: Greg Ofiesh <[EMAIL PROTECTED]>
Subject: Re: Why Elliptic Curve Cryptosystem is stronger with shorter key length?
Date: Thu, 01 Jul 1999 06:10:06 GMT
> > Who is NIST?
>
> The National Institute of Science and Technology.... In matters
related to
> encryption, the NSA has advisory and, ultimately, veto power over the
decisions they
> make.
Then how can anyone take their recommendations seriously? I thought
this would be the answer and I would never touch what they recommend.
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: "Douglas A. Gwyn" <[EMAIL PROTECTED]>
Subject: Re: new book
Date: Thu, 01 Jul 1999 04:52:09 GMT
John Savard wrote:
> ... it is not elitist to hope that an assessment of "too technical"
> may be an indication of real content.
Yup. If you saw the recent Today Show segment on Kryptos, you know
that their on-site reporter tried, fairly successfully, to present a
*very simple* explanation of substitution and transposition
encipherment, with clear, tiny examples, and the studio commented
that they thought that went over the heads of their audience.
(I suspect they were right about that, more's the pity.)
There are far too many books these days that try to "popularize"
technical topics, including cosmology, superstring theory, etc.
by "dumbing down" the subject, resorting to unexplained floating
abstractions and misleading analogies, but heaven forfend there
should be any equations. It seems almost impossible to find books
written for what we used to call "the educated layman". This may
have something to do with the general failure of our public
educational system to educate the public.
So when a review says that a book is good but too technical for
the general public, I take that as a hopeful sign for the book.
------------------------------
From: [EMAIL PROTECTED] (Jerry Coffin)
Subject: Re: Can Anyone Help Me Crack A Simple Code?
Date: Thu, 1 Jul 1999 00:35:27 -0600
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
[ ... ]
> I am trying to discover an encryption algorithm
> by observing how it behaves.
By definition, an encryption algorithm is reversible -- I.e. starting
with the encrypted text, there is SOME what to retrieve the original
plaintext. With what you've posted so far, what you've got is more or
less a hash, though it may be open to some argument whether a one-bit
output qualifies as a hash or not.
> If "you can't", then why bother using any of the "strong"
> encryption algorithms? After all, if it were impossible to
> discover how an algorithm works without directly seeing it,
> then any cocktail of XORs and MOD functions would be
> unbreakable. (assuming the algorithm itself is in an
> "indestructable" container and can not be directly observed)
First of all, you're aided considerably in decryption by the fact that
the original plaintext can be obtained from the output. It's not
possible to take a simple yes or no, and recover the numbers you put
into the algorithm you're working with.
Second, when you're working at breaking strong encryption, you
generally start with knowledge of the algorithm, NOT just the output
by itself. Third, when you devise a way of breaking an algorithm, you
may find that you have to collect a substantial amount of data before
the break can be completed. For example, you might start with
hundreds or thousands (or considerably more than that) known
plaintexts along with their enciphered counterparts. Out of these,
you might need to find plaintexts with specific characteristics to be
of any use at all.
In your case, you're providing a half dozen inputs that produce a
specific output. To be able to get very far with NO knowledge of the
algorithm, we'd probably need to be able to collect information on a
LARGE amount of input. For example if I were working on it, I'd
probably start with an input of 0. I'd then feed it inputs with one
bit set, rippling the set bit across the width of the input number.
I'd then do some playing with two bits set, three bits, and so on.
Eventually, I'd start to see things like the minimum number of bits
that have to be set before I got a positive output. When I started to
see specific patterns, I'd try to isolate some boundary conditions by
testing whether what I thought I saw was really a pattern, or simply a
coincidence in the testing thus far.
With the six specific inputs you've provided, it's impossible to do
intelligent testing along that line. One person made one post about a
(rather minimal) degree of similarity between a couple of the inputs
you provided. It might be on the right track, or it might be merely a
coincidence. Six inputs is simply too few to say much for sure -- you
could start to do some testing with other numbers that fit the same
general criteria and see what came out of it.
> We all know this is not true. Breaking a code, by
> definition, (correct me if I am wrong) MUST involve
> making assumptions, building on them, then seeing
> if the assumptions lead anywhere. If you have the
> algorithm, then you are not breaking a code. You
> are solving an algebra equation.
Not true. A well-designed cipher allows the attacker to know the
algorithm without being able to break the encryption easily.
> I should also point out that my problem, which is
> finding out what a group of numbers has in common,
> is ultimately a math problem.
>
> f(X1) = Y
> f(X2) = Y
> f(X3) = Y
> f(X4) = Y
> f(X5) = Y
> f(X6) = Y
> ...
> f(Xn) = Y
>
> I know six of n X values. Y is unknown, but all
> functions are equal to Y. Find the function.
> Does it look like some sort of a multi
> dimentional math problem now maybe? Perhaps a
> polynomial equation?
I've already pointed out that given only the inputs you've shown so
far, there are about as many f(N)'s as you'd care to devise. At the
_very_ least you need to give some other values that give the other
output. This will at least eliminate SOME possibilities.
> I may need all of the X values to solve this to
> find an exact solution, but I would be happy
> with a general solution. I'm sure someone in
> this newsgroup is a better mathematician than I.
You probably don't need ALL the values of X, but you ABSOLUTELY DO
need more than 6 values of X that all produce the same output.
> I will look into other possible methods in
> getting the algorithm for this problem, but
> this is the area of cryptography that interests
> me the most. Cracking crackable codes. If
> cryptographers insist that some codes are a
> joke and easy to crack, why am I having such
> a hard time finding someone who can crack
> codes or information on how to do it?
Because you're not providing information sufficient to accomplish
anything. People have asked for more information, and instead of
providing it, you've simply drafted a long complaint about how nobody
has done any magic on your behalf yet.
> Perhaps keeping the algorithm a secret is just
> as safe as using a powerfull well known one.
Not even close. Turn the machine (or whatever it is) over to somebody
competent so they can provide it with some meaningful inputs and
actually do some useful testing, and they'll probably break it fairly
quickly. If, OTOH, you simply continue to whine, you're not going to
get anywhere at all. Above, I've tried to outline a very simple
approach to starting to learn at least a little about what you're
working with. It's not the only one, but at least it stands some
chance of producing some meaningful results. Simply plugging in
random numbers, and seeing what comes out may work WITH ENOUGH random
numbers. For example, feed it a few million random numbers and graph
the outputs for ALL of them -- NOT just the ones that give a positive
output, but the ones that give a negative output as well. You may
find some patterns in the graph that'll give some insight into the
operation. You are, however, dealing with 10-digit numbers. That
means you've provided approximately .00000006 percent of the output in
the range you're dealing with. Worse yet, ALL of the numbers you've
presented produced the same output. It's simply impossible to
extrapolate anything meaningful from that amount of data.
This is bit like my telling you how many children my brothers and
sisters have, and asking you to use it to tell me the current
population of the USA. It's somewhat related, but so minimal as to be
absolutely meaningless. If you think I'm exaggerating the situation,
consider that my two nephews and one niece constitute about 100 TIMES
as large of a sample as you've given, taken on a percentage basis.
------------------------------
From: [EMAIL PROTECTED] (S.T.L.)
Subject: Re: Can Anyone Help Me Crack A Simple Code?
Date: 01 Jul 1999 06:33:05 GMT
<<Sorry It took so long to reply, but I've spent the last twenty minutes
downloading
your signature file. What is your server? Oh, AOL ... I see.>>
Hay! Not _everyone_ who uses AOL is a newbie moron. I'm not a newbie moron.
Sadly, many others are. Be glad that my .sig is not Kibo's. Actually, I think
that my signature is quite funny. You are free to have your own opinion.
<<The algebra for RSA is quite easy to solve, once you understand how to
treat the MOD function algebraicly. The difficulty in breaking RSA is
not algebra, but in factoring. >>
Oh, I see. When you say, "solve by simple algebra", I thought you meant crack
the algorithm using simple algebra only. I.E. not factoring, not anything.
There's no need to be a goofball. I know how to work RSA quite well. I even
implemented it on a calculator.
Oh, yeah, my .sig. Have fun.
-*---*-------
S.T.L. ===> [EMAIL PROTECTED] <=== BLOCK RELEASED! 2^3021377 - 1 is PRIME!
Quotations: http://quote.cjb.net Main website: http://137.tsx.org MOO!
"Xihribz! Peymwsiz xihribz! Qssetv cse bqy qiftrz!" e^(i*Pi)+1=0 F00FC7C8
E-mail block is gone. It will return if I'm bombed again. I don't care, it's
an easy fix. Address is correct as is. The courtesy of giving correct E-mail
addresses makes up for having to delete junk which gets through anyway. Join
the Great Internet Mersenne Prime Search at http://entropia.com/ips/ Now my
.sig is shorter and contains 3379 bits of entropy up to the next line's end:
-*---*-------
Card-holding member of the Dark Legion of Cantorians, the Holy Order of the
Catenary, the Great SRian Conspiracy, the Triple-Sigma Club, the Union of
Quantum Mechanics, the Polycarbonate Syndicate, and People for the Ethical
Treatment of Digital Tierran Organisms
Avid watcher of "World's Most Terrifying Causality Violations", "When Kaons
Decay: World's Most Amazing CP Symmetry Breaking Caught On [Magnetic] Tape",
"World's Scariest Warp Accidents", "World's Most Energetic Cosmic Rays", and
"When Tidal Forces Attack: Caught on Tape"
Patiently awaiting the launch of Gravity Probe B and the discovery of M39
Physics Commandment #6: Thou Shalt Always Obey CPT Symmetry.
------------------------------
From: mercury <[EMAIL PROTECTED]>
Subject: Re: Can Anyone Help Me Crack A Simple Code?
Date: Thu, 01 Jul 1999 02:56:17 -0400
Reply-To: [EMAIL PROTECTED]
S.T.L. wrote:
>
>
> Oh, I see. When you say, "solve by simple algebra", I thought you meant crack
> the algorithm using simple algebra only. I.E. not factoring, not anything.
>
> There's no need to be a goofball. I know how to work RSA quite well. I even
> implemented it on a calculator.
>
> Oh, yeah, my .sig. Have fun.
How did you solve for an RSA decryption key on your calculator? And how
did you test to see if your numbers were prime? I have never seen this done.
How large are the keys you are creating? I am curious.
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and sci.crypt) via:
Internet: [EMAIL PROTECTED]
End of Cryptography-Digest Digest
******************************
