Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
The Real & Final Enigma.
11 views
Skip to first unread message
adacrypt
Feb 10, 2012, 4:25:00 AM2/10/12
to
Okay, I have being labouring several points rather heavily,
excessively perhaps, in recent days but I cannot allow my very hard
work over many years to be trivialised by foul-mouthed ridiculous
claims.
This hopefully, will be the last piece of defensive posting necessary
to debunk the attempts at wrecking my work.
I set out my stall as follows.
My crypto scheme uses these keysets,
4 keysets of 1000 elements that are 95% random if not indeed 100% so.
1 keyset of 8000 elements that are also 95% random.
1 keyset that is 100% random
1 keyset of 36 equations that is also 100% random.
These keysets are stored as arrays in secured databases.
The keysets as arrays of data in my computer may be scrambled and
sliced afresh for every single message if needs be but in truth are
scrambled and sliced periodically but frequently instead (daily).
Every message is encrypted using a combination of all of these keys.
Furthermore, the ensuing ciphertext is carefully contrived to be 99.5%
random i.e. not having repeats of any single elemental ciphertext item
such that the collective ciphertext string itself is now also a
tertiary crafted keyset, that may be reckoned as further (recent and
novel) defensive entanglement.
Despite this vast maelstrom of possibility space of combinations and
careful usage, one reader claims to be able to forecast a general
solution that will enable the 99.5 % random ciphertext to be
accurately mapped to plaintext in a comparator library and thus
circumvent the proper decryption process which happens in passing, to
be a uniquely mathematical one-way function.
I would be fascinated if that possibility was even remotely true.
I have no doubt that readers will judge for themselves and see that
this is nothing more than destructive foul-mouthed propaganda – it
won’t catch on.
The defence rests here, this cryptography is complete, you won’t be
hearing any more promotional or defensive tracts on this matter from
me at least for some time. Thanks for your patience in listening.
- adacrypt
excessively perhaps, in recent days but I cannot allow my very hard
work over many years to be trivialised by foul-mouthed ridiculous
claims.
This hopefully, will be the last piece of defensive posting necessary
to debunk the attempts at wrecking my work.
I set out my stall as follows.
My crypto scheme uses these keysets,
4 keysets of 1000 elements that are 95% random if not indeed 100% so.
1 keyset of 8000 elements that are also 95% random.
1 keyset that is 100% random
1 keyset of 36 equations that is also 100% random.
These keysets are stored as arrays in secured databases.
The keysets as arrays of data in my computer may be scrambled and
sliced afresh for every single message if needs be but in truth are
scrambled and sliced periodically but frequently instead (daily).
Every message is encrypted using a combination of all of these keys.
Furthermore, the ensuing ciphertext is carefully contrived to be 99.5%
random i.e. not having repeats of any single elemental ciphertext item
such that the collective ciphertext string itself is now also a
tertiary crafted keyset, that may be reckoned as further (recent and
novel) defensive entanglement.
Despite this vast maelstrom of possibility space of combinations and
careful usage, one reader claims to be able to forecast a general
solution that will enable the 99.5 % random ciphertext to be
accurately mapped to plaintext in a comparator library and thus
circumvent the proper decryption process which happens in passing, to
be a uniquely mathematical one-way function.
I would be fascinated if that possibility was even remotely true.
I have no doubt that readers will judge for themselves and see that
this is nothing more than destructive foul-mouthed propaganda – it
won’t catch on.
The defence rests here, this cryptography is complete, you won’t be
hearing any more promotional or defensive tracts on this matter from
me at least for some time. Thanks for your patience in listening.
- adacrypt
adacrypt
Feb 10, 2012, 8:27:42 AM2/10/12
to
Supplement 1)
I can now confirm after taking the trouble to do so that the spread of
the keyset called “Change-of-Origin_J _Coefficients” in my cipher
database is,
The key set is comprised of 14250 keys as (i, j, k) vectors.
It is only necessary to demonstrate that one column of coefficients of
the vector set is random for the whole set to be random.
The column of ‘j’ coefficients was taken as the sample to do this
check.
It was found that,
14246 are non-repeated => 99.972 % of true randomness.
2 keys are repeated once in the (j) coefficients column => 4 keys
absolute
(It is very unlikely that the same two are also repeated in the (i)
and (k) columns (not checked) of the same keyset items => it can be
taken safely that these two are not therefore de facto repeats per se
of the parent vectors => 100% random is a very safe assumption).
I feel I owe it to the readership to provide this feedback information
in the light of recent doubt-casting allegations that might yet again
arise. Left unanswered these allegations could be damaging to my
cryptography.
As it happens the cipher is secured by a mathematical one-way function
that is independent of this pure randomness anyway and the latter is
simply belt n’ braces over-kill, all good extra grist to the mill in
terms of crypto strength of course.
I will probably check the (i) and (k) columns another time but
meanwhile this is a very, very powerful cipher.
I will be pleased to send samples of plaintext and corresponding
ciphertext to any person who wants that.
- adacrypt
Globemaker
Feb 10, 2012, 10:21:47 AM2/10/12
to
Hello adacrypt,
You should provide a website page that defines the hard coded keysets
and how the user defined key is used to calculate the ciphertext and
to decrypt. Please prepare your website page to show the four 1000
word keys, the 8000 word key table and the other key tables you
mentioned. The website should have spell out the file name of your
software's current version.
You have so many versions of software that cryptanalysts do not know
which version you are writing about. Here is an example of a URL you
can post :
http://www.adacrypt.com/skewline/sl-v4.exe 32,765,987 bytes dated
January 22, 2012
Please put on this website page the cipher file name, file size in
bytes, and date when it is published. This helps us know if changes
were made. By providing these basic facts, cryptanalysts can be
assured that they are working on the software that you claim is valid.
People do not want to search your vast website for software that you
fail to provide links for. Do not expect people to read your source
code to find your 8000 words key table. Just show us what is in the
table. Post the 8000 words here on sci.crypt. I have no idea what file
to look at to see the 8000 words. Does the array have a name? Tell us
the name of the array and what file it is in.
4 keysets of 1000 elements that are 95% random if not indeed 100% so.
1 keyset of 8000 elements that are also 95% random.
1 keyset that is 100% random
1 keyset of 36 equations that is also 100% random.
Show us the elements here.
When you have prepared that basic information, please post the link to
that one page on sci.crypt. When I go to your website and click on the
skew line link, a download begins with 32 megabytes, but no web page
was seen for skew line documentation. Please prepare the requested
documentation on a web page, not in a download. List all file names in
the download. Then show us the link to your documentation.
Many people enjoy reading your teachings. Keep up the good work.
You should provide a website page that defines the hard coded keysets
and how the user defined key is used to calculate the ciphertext and
to decrypt. Please prepare your website page to show the four 1000
word keys, the 8000 word key table and the other key tables you
mentioned. The website should have spell out the file name of your
software's current version.
You have so many versions of software that cryptanalysts do not know
which version you are writing about. Here is an example of a URL you
can post :
http://www.adacrypt.com/skewline/sl-v4.exe 32,765,987 bytes dated
January 22, 2012
Please put on this website page the cipher file name, file size in
bytes, and date when it is published. This helps us know if changes
were made. By providing these basic facts, cryptanalysts can be
assured that they are working on the software that you claim is valid.
People do not want to search your vast website for software that you
fail to provide links for. Do not expect people to read your source
code to find your 8000 words key table. Just show us what is in the
table. Post the 8000 words here on sci.crypt. I have no idea what file
to look at to see the 8000 words. Does the array have a name? Tell us
the name of the array and what file it is in.
4 keysets of 1000 elements that are 95% random if not indeed 100% so.
1 keyset of 8000 elements that are also 95% random.
1 keyset that is 100% random
1 keyset of 36 equations that is also 100% random.
When you have prepared that basic information, please post the link to
that one page on sci.crypt. When I go to your website and click on the
skew line link, a download begins with 32 megabytes, but no web page
was seen for skew line documentation. Please prepare the requested
documentation on a web page, not in a download. List all file names in
the download. Then show us the link to your documentation.
Many people enjoy reading your teachings. Keep up the good work.
adacrypt
Feb 10, 2012, 11:03:17 AM2/10/12
to
something definitive later when all current statistical checks are
complete - I will then point to something conclusive on my website -
may even consider a fresh website altogether - thanks for your
feedback - adacrypt
adacrypt
Feb 10, 2012, 12:09:56 PM2/10/12
to
On Feb 10, 3:21 pm, Globemaker <alanfolms...@cabanova.com> wrote:
This cryptography is complete as far as I am concerned – I am totally
satisfied with the results but I agree that it needs proper
documentation now that will make it readable by even the most casually
interested person in the future.
Unfortunately I am a bit absent-minded when it comes to writing
technical reports.
What I would suggest is that I prepare a draft for publication by you
on “Popular Cryptography Magazine” that you could criticise and have
me adjust and fine tune until it is perfected. Your magazine could
then be the home so to speak that readers in sci crypt could zoom in
on.
I have a connection here with a maths society also that may take some
duplication of the finished work also if you don’t mind that.
Your magazine would host everything including a downloadable program
in one place that readers could go to.
This would be quite large – maybe as much as 50 megabytes. Could take
me several months.
I never seem to get it right no matter how I try and this would save a
lot of confusion if two minds could work on it.
I envisage some kind of low-key discreet exchanges of material.
I have asked about typesetting software with a view to posting in the
establishment but no replies yet from sci crypt readers.
Please consider publishing in “Popular Cryptography Magazine” as the
better alternative.
Please treat this as a problem for me that I want to resolve in the
best possible way for others to benefit by a highly transparent
solution.
- adacrypt
Mark Murray
Feb 10, 2012, 3:18:48 PM2/10/12
to
On 10/02/2012 09:25, adacrypt wrote:
> Okay, I have being labouring several points rather heavily,
> excessively perhaps, in recent days but I cannot allow my very hard
> work over many years to be trivialised by foul-mouthed ridiculous
> claims.
Nor will you even consider fixing its very many flaws.
> Okay, I have being labouring several points rather heavily,
> excessively perhaps, in recent days but I cannot allow my very hard
> work over many years to be trivialised by foul-mouthed ridiculous
> claims.
> This hopefully, will be the last piece of defensive posting necessary
> to debunk the attempts at wrecking my work.
> I set out my stall as follows.
>
> My crypto scheme uses these keysets,
>
> 4 keysets of 1000 elements that are 95% random if not indeed 100% so.
of them.
> 1 keyset of 8000 elements that are also 95% random.
files has poor entropy. You amusingly store this in 3 separate
files, each ontaining 14500 values (of which you only use 8000,
which is good, because you typoed one of them.)
> 1 keyset that is 100% random
> 1 keyset of 36 equations that is also 100% random.
to constants?
> These keysets are stored as arrays in secured databases.
You have lots of constants, very many of which spew out of functions,
but which work just as well (but faster) in normal arrays. Its all
source code, so nothing "secure" about it.
> The keysets as arrays of data in my computer may be scrambled and
> sliced afresh for every single message if needs be but in truth are
> scrambled and sliced periodically but frequently instead (daily).
source. For daily use, surely you'd be reading this from a "daily
scrambe parameters" file?? The requirements for these scrambling
constants are INCREDIBLY security sensitive; wrong choices here and
your adversary's job gets REALLY easy (as opposed to just "easy").
> Every message is encrypted using a combination of all of these keys.
> Furthermore, the ensuing ciphertext is carefully contrived to be 99.5%
> random i.e. not having repeats of any single elemental ciphertext item
> such that the collective ciphertext string itself is now also a
> tertiary crafted keyset, that may be reckoned as further (recent and
> novel) defensive entanglement.
achievement. What you have failed to achieve is "no usable structure".
> Despite this vast maelstrom of possibility space of combinations and
> careful usage, one reader claims to be able to forecast a general
> solution that will enable the 99.5 % random ciphertext to be
> accurately mapped to plaintext in a comparator library and thus
> circumvent the proper decryption process which happens in passing, to
> be a uniquely mathematical one-way function.
all the image set outputs into separate input sets, with no duplicates.
> I would be fascinated if that possibility was even remotely true.
Show me where I can upload the dictionaries. Email is no good, they
are too large for that.
> I have no doubt that readers will judge for themselves and see that
> this is nothing more than destructive foul-mouthed propaganda – it
> won’t catch on.
And watch out for the hypocrasy; I seem to recall you trying to call
me mentally ill for using my name in a public forum.
> The defence rests here, this cryptography is complete, you won’t be
> hearing any more promotional or defensive tracts on this matter from
> me at least for some time. Thanks for your patience in listening.
begun to make a case!
As for not hearing any more, you have at least three follow-ups to this
thread that I have seen, all full of the same hubris and denialism.
M
--
Mark "No Nickname" Murray
Notable nebbish, extreme generalist.
Mark Murray
Feb 10, 2012, 3:34:13 PM2/10/12
to
On 10/02/2012 13:27, adacrypt wrote:
> (It is very unlikely that the same two are also repeated in the (i)
> and (k) columns (not checked) of the same keyset items => it can be
> taken safely that these two are not therefore de facto repeats per se
> of the parent vectors => 100% random is a very safe assumption).
How may years have you been using these numbers, and you don't know
> (It is very unlikely that the same two are also repeated in the (i)
> and (k) columns (not checked) of the same keyset items => it can be
> taken safely that these two are not therefore de facto repeats per se
> of the parent vectors => 100% random is a very safe assumption).
basic stuff like this about them??!
> I feel I owe it to the readership to provide this feedback information
> in the light of recent doubt-casting allegations that might yet again
> arise. Left unanswered these allegations could be damaging to my
> cryptography.
Why are there 14500 numbers in each file when you only use 8000 of them?
> As it happens the cipher is secured by a mathematical one-way function
> that is independent of this pure randomness anyway and the latter is
> simply belt n’ braces over-kill, all good extra grist to the mill in
> terms of crypto strength of course.
problems?! Riiiiight! This is military-grade BS.
So if the randomness is independant, separable and overkill, why not
remove it to demonstrate the central concept? You have HUGE amounts of
obfuscation, where the central theory can be concisely written in about
a screenful of code. Please accuse me of lying, so I can mail it to you!
> I will probably check the (i) and (k) columns another time but
> meanwhile this is a very, very powerful cipher.
> I will be pleased to send samples of plaintext and corresponding
> ciphertext to any person who wants that.
You cannot be trusted to play fair; with Paulo you rigged the test
that was set up with him.
Mark Murray
Feb 10, 2012, 3:42:46 PM2/10/12
to
On 10/02/2012 17:09, adacrypt wrote:
> This would be quite large – maybe as much as 50 megabytes. Could take
> me several months.
50MB??! Only if you pad it with the usual intermidiate file crap that
> This would be quite large – maybe as much as 50 megabytes. Could take
> me several months.
just waste space!
Take several months? Yest you can write screeds of self-praise in
an hour or two! This rather looks like you are chickening out again.
> I never seem to get it right no matter how I try and this would save a
> lot of confusion if two minds could work on it.
</sarcasm>
> I envisage some kind of low-key discreet exchanges of material.
be shy about.
> I have asked about typesetting software with a view to posting in the
> establishment but no replies yet from sci crypt readers.
for the answer, or, if you are not prepared to read what people write,
DON'T ASK THE BLOODY QUESTION!!
adacrypt
Feb 10, 2012, 5:52:00 PM2/10/12
to
On Feb 10, 3:21 pm, Globemaker <alanfolms...@cabanova.com> wrote:
This keyset is the wrong one I referred to and should in fact be
another keyset of 14250 elemnets which I can proved is 100 % random.
It is quite safe to use the 8000 however one inconjunction with other
keysets but it is now redundant since the larger keyset was developed
and is the one that will always be used in the future.
There was nothing wrong really with it in the first place but I think
it was hastily constructed and may have had some repeats .
A characteristic of vector cryptography is the enormous flexibility of
it to adjust to any situation internally or externally - such as this
one - there's an easy solution to hand for every situation that
arises.
Have no fears of anything regarding the crypto security strength.
- adacrypt
Mark Murray
Feb 10, 2012, 6:44:43 PM2/10/12
to
On 10/02/2012 22:52, adacrypt wrote:
> Have no fears of anything regarding the crypto security strength.
... for as long as the message is shorter than the key entropy, and
> Have no fears of anything regarding the crypto security strength.
the key is never re-used. I bet you don't know the key entropy.
I bet you don't know how to work out the key entropy.
0 new messages