Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Resume of “Skew Line Encryptions” Programs.

22 views
Skip to first unread message

adacrypt

unread,
Feb 9, 2012, 6:04:37 AM2/9/12
to
The version of this program as it appears on my website is the
original development scheme as I wrote it piecemeal over the past
fifteen years. I have no intention of changing it in any way because
I believe the expansive simplicity of this first form is helpful in
the understanding of it all however banal it might seem today with the
benefit of hindsight.

The variable names may seem naïve and the procedures in need of
polishing up but that is the way I liked it first day and I am going
to leave like that. I could easily rewrite the whole lot and
introduce some smart short-form code that would create a slick program
but who wants that. I personally hate it.

I have put this cryptography in the public domain and invited people
to modify it to their heart’s content – please do that instead of
criticising what you see.

I take the view that my original mission was to amalgamate some
difficult mathematics, cryptography and programming into a successful
unit that works reliably. That has been done in a satisfactory way.

I appreciate it that if someone like say NSA or IBM decided to use my
cryptography they would immediately employ their massive resources to
change my programs into something quite different that would suit them
better, until that happens if ever, there is no point in me doing
anything to enhance my own work. I prefer it as it is.

The review of the Ada programming language in 2005 introduced some new
vector methods for people who asked for a mechanical means of doing
things in vector arithmetic that previously were done in user-crafted
longhand procedures albeit by the computer. It is not beneficial to
me to use these new methods because I have certain benefits still
emanating from the older procedures that I don’t wish to discard – I
do know about these new methods however.

Recapping.

The cryptographic strength of this cryptography is underpinned by
these basic tenets,

1) Each ciphertext item is the sum of two vectors. It is impossible
by any means known to mankind to deduce what these are, Alice and Bob
of course can do so. Decryption cannot proceed without that
knowledge => "ciphertext only" attack is foiled completely.

2) The ciphertext is a string of studiously crafted mathematical
discontinuities that trap all attempts at numerical inversion of the
ciphertext by such means as linear analysis, differential analysis,
plain ordinary numerical methods or indeed any probabilistic means
whatever => numerical attack is completely foiled.

3) I have just demonstrated that the ciphertext string can be made
impervious to all statistical attack by the ploy of making it a
scientifically random set => statistical attack on intercepted samples
of ciphertext data is foiled.

These are the three main precepts of the scheme as defence weapons
against the illegal cryptanalyst.

As Mr Kerchoff says in words to the same effect,

“As long as the entities keep their keys safe they can enjoy total
secrecy of communications.”

The entities operate from mutually synchronised databases that are
easy enough to keep safe after the one-off initial secure physical
delivery of the copy database has been made to Bob.

(please note – it is ludicrous to make hypothetical assumptions of
situations that seem to say that there is vulnerability through
cryptanalysis of the arrays of the databases that the entities use
*inside their secure databases in my crypto scheme. These databases
must be breached first and then there is no need for clever analysis.
This is like saying how badly stacked the gold is in Fort Knox and how
a thief may use this to advantage without explaining how you will
break into the place first of all. – how stupid can you get.)

Advice needed.

Does any reader have experience - knowledge - of type setting software
(methods they can recommend or other) such as Tex, Latex, in
particular can anybody review "Adobe Framework" software frome their
own experience. Your help would be greatly appreciated in this
matter.

- adacrypt

Mark Murray

unread,
Feb 9, 2012, 4:50:34 PM2/9/12
to
On 09/02/2012 11:04, adacrypt wrote:
> The version of this program as it appears on my website is the
> original development scheme as I wrote it piecemeal over the past
> fifteen years. I have no intention of changing it in any way because
> I believe the expansive simplicity of this first form is helpful in
> the understanding of it all however banal it might seem today with the
> benefit of hindsight.

"I reject all criticism of my code".

> The variable names may seem naïve and the procedures in need of
> polishing up but that is the way I liked it first day and I am going
> to leave like that. I could easily rewrite the whole lot and
> introduce some smart short-form code that would create a slick program
> but who wants that. I personally hate it.

"If I didn't do It, I'm not interested in it."

> I have put this cryptography in the public domain and invited people
> to modify it to their heart’s content – please do that instead of
> criticising what you see.

I did that. Above, you showed hostility and defensiveness.

BTW, learn to use terminology properly; you licensed the code with a
GNU license; that is NOT "public domain". Are you thinking of "open
source" perhaps?

> I take the view that my original mission was to amalgamate some
> difficult mathematics, cryptography and programming into a successful
> unit that works reliably. That has been done in a satisfactory way.

Except that its broken, and you are blindly ignoring that. Amusingly,
you are instead writing glowing reviews of your work, aparrently
imagining that this somehow makes a case.

Its brown, it smells like crap, it really is a turd.

> I appreciate it that if someone like say NSA or IBM decided to use my
> cryptography they would immediately employ their massive resources to
> change my programs into something quite different that would suit them
> better, until that happens if ever, there is no point in me doing
> anything to enhance my own work. I prefer it as it is.

"I have no interest in anyone else's contributions or improvements."

> The review of the Ada programming language in 2005 introduced some new
> vector methods for people who asked for a mechanical means of doing
> things in vector arithmetic that previously were done in user-crafted
> longhand procedures albeit by the computer. It is not beneficial to
> me to use these new methods because I have certain benefits still
> emanating from the older procedures that I don’t wish to discard – I
> do know about these new methods however.

Yet you choose not to even simplify your older-style code to remove
unused statements, useless code and obfuscatory style abuse. Your
"choice" smacks more of a fear of breaking that which you barely
understand in the first place.

> Recapping.
>
> The cryptographic strength of this cryptography is underpinned by
> these basic tenets,
>
> 1) Each ciphertext item is the sum of two vectors. It is impossible
> by any means known to mankind to deduce what these are, Alice and Bob
> of course can do so. Decryption cannot proceed without that
> knowledge => "ciphertext only" attack is foiled completely.

Not any more. I have dictionaries.

> 2) The ciphertext is a string of studiously crafted mathematical
> discontinuities that trap all attempts at numerical inversion of the
> ciphertext by such means as linear analysis, differential analysis,
> plain ordinary numerical methods or indeed any probabilistic means
> whatever => numerical attack is completely foiled.

Brute force dictionary construction, however, works. Your
"discontinuities" were a HUGE help, thank you very much!

> 3) I have just demonstrated that the ciphertext string can be made
> impervious to all statistical attack by the ploy of making it a
> scientifically random set => statistical attack on intercepted samples
> of ciphertext data is foiled.

No - you deluded yourself by computing an irrelevant statistic. I have
enciphered 700000+ repetitions of the letter "a" with no duplicates;
this is the basis of my dictionaries.

> These are the three main precepts of the scheme as defence weapons
> against the illegal cryptanalyst.

Sue me. "Illegal" HAH.

> As Mr Kerchoff says in words to the same effect,
>
> “As long as the entities keep their keys safe they can enjoy total
> secrecy of communications.”

Your keys aren't safe. In fact, it is a matter of some confusion on
your part what the keys actually ARE. Right now, your program is
a bad random number generator with a few tweakbles, and no documented
(or even stable) notion of a keyspace.

> The entities operate from mutually synchronised databases that are
> easy enough to keep safe after the one-off initial secure physical
> delivery of the copy database has been made to Bob.

Carry on repeating that to yourself, if it makes you happy.

To me, its funny to see how clue-hostile you can be.

> (please note – it is ludicrous to make hypothetical assumptions of
> situations that seem to say that there is vulnerability through
> cryptanalysis of the arrays of the databases that the entities use
> *inside their secure databases in my crypto scheme. These databases
> must be breached first and then there is no need for clever analysis.
> This is like saying how badly stacked the gold is in Fort Knox and how
> a thief may use this to advantage without explaining how you will
> break into the place first of all. – how stupid can you get.)

It is ludicrous to make specious claims about how safe your iron front
door is when the rest of the house is make of bricks that are simply
resting upon each other with no mortar. Sod the lock, I just pull it
out of its frame.

> Advice needed.
>
> Does any reader have experience - knowledge - of type setting software
> (methods they can recommend or other) such as Tex, Latex, in
> particular can anybody review "Adobe Framework" software frome their
> own experience. Your help would be greatly appreciated in this
> matter.

I know and use LaTeX, and can hightly recommend it. There are plenty of
books on it, and its usually free. A What-You-See-Is-What-You-Mean
editor called LyX works very well with LaTeX, and is also free. In the
Windows environment, LyX+MikTeX is the usual combination.

Warning: LaTeX has a learning curve that is steeper than the usual
WYSIWYG products, but with a MUCH higher gain at the end. If you are
battling with Ada, I suggest you give it a miss.

Save yourself the money and mental trauma and use MS-WordPad. Its at
about the right level.

M
--
Mark "No Nickname" Murray
Notable nebbish, extreme generalist.

Bruce Stephens

unread,
Feb 9, 2012, 5:53:30 PM2/9/12
to
Mark Murray <w.h....@example.com> writes:

> On 09/02/2012 11:04, adacrypt wrote:

[...]

>> Does any reader have experience - knowledge - of type setting software
>> (methods they can recommend or other) such as Tex, Latex, in
>> particular can anybody review "Adobe Framework" software frome their
>> own experience. Your help would be greatly appreciated in this
>> matter.
>
> I know and use LaTeX, and can hightly recommend it. There are plenty of
> books on it, and its usually free. A What-You-See-Is-What-You-Mean
> editor called LyX works very well with LaTeX, and is also free. In the
> Windows environment, LyX+MikTeX is the usual combination.

Agreed that LaTeX (perhaps using LyX) is a decent thing to try. LaTeX is
relatively widely known, at least to the extent needed to understand
simple equations and expressions. Because it's relatively well known,
it's offered as a syntax in other systems like
<http://www.mathjax.org/>.

> Warning: LaTeX has a learning curve that is steeper than the usual
> WYSIWYG products, but with a MUCH higher gain at the end.

Presuming he meant Adobe FrameMaker rather than Adobe Framework
(whatever that is), FrameMaker's non-trivial to learn, too. On the whole
I'd suggest LaTeX through LyX is just as easy, and free.

And anyway, I suspect the basic problem is the content rather than the
presentation, specifically that there's much too much fluff obscuring
whatever content might be present.

[...]

David Eather

unread,
Feb 10, 2012, 10:16:09 PM2/10/12
to
On Fri, 10 Feb 2012 07:50:34 +1000, Mark Murray <w.h....@example.com>
wrote:
If he really wants to output PDF he should use Open Office - about the
same difficulty as MS Office but it will output PDF. He really has no
excuse for all his crap and delay.
--
We have failed to address the fundamental truth that endless growth is
impossible in a finite world.

Mark Murray

unread,
Feb 11, 2012, 3:15:34 AM2/11/12
to
On 11/02/2012 03:16, David Eather wrote:
> If he really wants to output PDF he should use Open Office - about the
> same difficulty as MS Office but it will output PDF. He really has no
> excuse for all his crap and delay.

He may not have one, but he needs one so very badly.

David Eather

unread,
Feb 11, 2012, 3:19:58 AM2/11/12
to
On Sat, 11 Feb 2012 18:15:34 +1000, Mark Murray <w.h....@example.com>
wrote:

> On 11/02/2012 03:16, David Eather wrote:
>> If he really wants to output PDF he should use Open Office - about the
>> same difficulty as MS Office but it will output PDF. He really has no
>> excuse for all his crap and delay.
>
> He may not have one, but he needs one so very badly.
>
> M

I fear the most extreme case of Dunning-Kruger ever to become public.

Mark Murray

unread,
Feb 11, 2012, 4:24:26 AM2/11/12
to
On 11/02/2012 08:19, David Eather wrote:
> I fear the most extreme case of Dunning-Kruger ever to become public.

Yup. Up there with Harris.

adacrypt

unread,
Feb 11, 2012, 8:34:13 AM2/11/12
to
Thanks for your advice re. LyX and MikTex – I have installed these now
and will start familiarising myself with it straight away.

Re “Skew Line Encryptions” – I could change your mind in the space of
just one hour of one-to-one if that was possible

*This is extremely difficult to discuss from a distance.

For instance, the first step in disseminating this crypto is for me to
sell the new mathematics to professional mathematics over here. That
has only just started. Trying to convince somebody such as your self
from a distance is just too much.

The application of this to cryptography is very convincing once you
understand it.

Randomness is secondary really.

You are looking at the wrong ciphers at your end.

See the package “Free Software - Gnu General Public License – Skew
Line Encryptions. etc -

See in particular the Ada packages there regarding keysets such as,
encryption numbers, change-of-origins etc for your analysis but note
well that this cipher does not depend on randomness alone although it
is important in a secondary role to obviate statistical mapping of the
ciphertext as the last ditch attack by a cryptanalyst.

I cannot keep up any tit-for-tat correspondence on this stuff because
it inevitably becomes acrimonious after a few exchanges because of the
misunderstandings that must always inevitably arise.

This requires a lot of chalk and talk in a classroom - there is no
alternative which is why I am try to learn desk top publishing hoping
that this might solve the problem.

Regards - adacrypt

rossum

unread,
Feb 11, 2012, 12:48:04 PM2/11/12
to
On Sat, 11 Feb 2012 18:19:58 +1000, "David Eather" <eat...@tpg.com.au>
wrote:

>We have failed to address the fundamental truth that endless growth is
>impossible in a finite world.
Year 1, grow 100%
Year 2, grow 50%
Year 3, grow 25%
Year 4, grow 12.5%
...

Zeno was an economist. :)

rossum

Mark Murray

unread,
Feb 11, 2012, 1:19:51 PM2/11/12
to
On 11/02/2012 13:34, adacrypt wrote:
> Thanks for your advice re. LyX and MikTex – I have installed these now
> and will start familiarising myself with it straight away.

OK - be careful not to assume that has much in commom with standard
WYSIWYG word processors. It is NOT WYSIWYG; layout and font decisions
happen at print time, not edit time. If you get used to the idea that
LaTeX has more in common with programming than editing, where LyX
disguises this just a bit, then you'll be OK.

Amazon sell a few LaTeX books; there are a bunch of them that are often
sold together; Lamport, Goossens and Kopka are three books that are on
many LaTeX-users' shelves, and the combined price is not bad.

> Re “Skew Line Encryptions” – I could change your mind in the space of
> just one hour of one-to-one if that was possible

I would be happy to listen, if this was to be a dialogue, not a lecture;
IE I also get to have my input taken seriously.

> *This is extremely difficult to discuss from a distance.

Not my experience; if one listens as much as one speaks, long-distance
discussion does not have to be especially hard.

> For instance, the first step in disseminating this crypto is for me to
> sell the new mathematics to professional mathematics over here. That
> has only just started. Trying to convince somebody such as your self
> from a distance is just too much.

Have professionals not already explained what they need? Have you taken
any of their comments into account?

What do the "buyers" actually want? have you researched your
marketplace?

> The application of this to cryptography is very convincing once you
> understand it.

Careful of circular logic:

Premiss 1: If you understand this cryptography, you will
understand how good it is.
Premiss 2: You think this cryptography is bad.
Conclusion: You don't understand this cryptography.

The premisses and the conclusion rely on a fallacy and a highly
questionable hidden premiss:

Premiss 0: This cryptography is undeniably good

First demonstrate premiss 0. So far this is an assertion, and you are
ignoring counterevidence.

Then tackle Premiss 1; it is a "poisoning the well" fallacy; you are
setting yourself up to discredit anyone who disagrees with you as
simply not understanding you, and dismissing the notion that they
may have valid objections.

It it not my responsibility to understand it, it is yours to prove it,
and dismissing inconvenient counter-evidence is not the way to do this.

> Randomness is secondary really.

Your code and argument heavily use a strange understanding
of randomness; if, as you say above, this is seconday, why not
remove it altogether and just expose the primary debate?

> You are looking at the wrong ciphers at your end.
>
> See the package “Free Software - Gnu General Public License – Skew
> Line Encryptions. etc -

Er, thats what I am looking at.

> See in particular the Ada packages there regarding keysets such as,
> encryption numbers, change-of-origins etc for your analysis but note
> well that this cipher does not depend on randomness alone although it
> is important in a secondary role to obviate statistical mapping of the
> ciphertext as the last ditch attack by a cryptanalyst.

You are trying to show "vector cryptography" right? For me to look at
that, I have to download some 10's of MB of stuff, the HUGE majority of
which I delete straight away.

I don't need an Ada compiler, I have one! Put that up for download by
all means, but please don't pad every darn zipfile with it; let folks
download it _once_if_they_need_it_.

Remove all the generated files (*.exe *.ali *.o *.c), and please remove
duplicate files. All the *_mark*.adb files with only comment or other
minor differences are a waste of my time to read.

The reason for the "tutorial" directory is unclear; it looks like more
duplicates and padding; more near identical code to plough through,
looking for (sometimes non-existant) differences.

PLEASE PLEASE PLEASE simplify the code, DRASTICALLY; if you want me to
read it and appreciate your work, please don't make it hard for me:

o Write proper documentation.

o Separate out functions into separate files, AND reuse these
instead of copying them the whole time.

o Pick a code style and stick to it; haphazard code is hard to read.
Write idiomatic code; your "key function" class things are simply
bizarre; rewrite those as static arrays - that way they are faster
and simpler.

o Remove all code that isn't demonstrating your core concepts. In
fact review the lot and remove all "do-nothing" code, its only
acting as a barrier to acceptance.

o Abstract properly; writing a simple vector class is no harder than
what you've already done; I did it in 20 lines or so.

o Use variables that correspond to the documentation, and that
otherwise make sense. Use them in a consistent way.

o Drop "baggage"; your ASCII-only fixation, for example.

> I cannot keep up any tit-for-tat correspondence on this stuff because
> it inevitably becomes acrimonious after a few exchanges because of the
> misunderstandings that must always inevitably arise.

Do you want help from people? They (we!) are doing this for free in
valuable spare time. Please show some respect by not making it hard,
entering into dialogue and listening as much as you talk.

> This requires a lot of chalk and talk in a classroom - there is no
> alternative which is why I am try to learn desk top publishing hoping
> that this might solve the problem.

A problem is that your program and your existant explanation do not
correspond; your documentation talks about vectors; your program spends
99% of its code on shuffling arrays, and obfuscates the vector
code so severely that it takes a VERY bored person[*] to be interested
enough to figure it out. The fact that you are happy with the code is
one thing; you now want other people to accept it; have the good grace
to present them with polished work, not the current overwhelming mess.

I've rewritten the skew-line encryptor using most of the above code
improvements. It produces the same output as yours (your decryptor can
decrypt its output), and it clarifies the vector work into about 5 lines
of easy-to-find code.

Would you like to see this?

adacrypt

unread,
Feb 11, 2012, 1:53:13 PM2/11/12
to
Yes please. That's the whole idea of making it available. It sounds
ok if the decryption program can decrypt it ok. Very good for you .
Do you use Ada often for other work?.

I am not going for gold -this is a development program that
demonstrates the feasibiliy of displacement cryptography only - there
are lots of people out there who can write beautiful programs but I
take your point.

Digressing, do you know if its possible to insert
'mathtype' (dedicated mathematical software) equations from other
documents into LyX docs.?

Bruce Stephens

unread,
Feb 11, 2012, 2:07:29 PM2/11/12
to
adacrypt <austin...@hotmail.com> writes:

[...]

> Digressing, do you know if its possible to insert
> 'mathtype' (dedicated mathematical software) equations from other
> documents into LyX docs.?

According to <http://en.wikipedia.org/wiki/MathType>, it's probably
possible:

MathType also supports the math markup languages TeX, LaTeX and
MathML. LaTeX can be entered directly into MathType,[nb 1] and
MathType equations in Microsoft Word can be converted to and from
LaTeX.[2] MathType supports copying to and pasting from any of these
markup languages.

adacrypt

unread,
Feb 11, 2012, 2:17:31 PM2/11/12
to
Forgot to say also I have quite consciously layered on huge extra
entanglemnt options right left and centre during he development stages
with the view that this can always be discarded if not required. You
obviously are of that frame of mind - that's great.

The diagnostic program was never required be me in thousands of cases
- never once did this program fail in practice. I enclosed it as a
tutorial hoping it would benefit some users.

I think you are better placed to explain this crypto to sci crypt
readers now maybe than me ?

Please do this if you can.

The new mathematics is "Vectoring Factoring" - you will need to get
your head around this and notice how it is applied to cryptography in
th Ada program if you are ever going to teach it academically ? - a
lucrative prospect in the future - could well be!

Good Luck if you do.

-adacrypt

Mark Murray

unread,
Feb 11, 2012, 2:32:45 PM2/11/12
to
On 11/02/2012 18:53, adacrypt wrote:
> Yes please. That's the whole idea of making it available. It sounds
> ok if the decryption program can decrypt it ok. Very good for you .

Progress! I'll post a link to it when its as close to windows-friendly
as I think I can get it. (I use *ix, not Windows)

> Do you use Ada often for other work?.

Not yet. At least not now, but I have plans.

> I am not going for gold -this is a development program that
> demonstrates the feasibiliy of displacement cryptography only - there
> are lots of people out there who can write beautiful programs but I
> take your point.

The MASSIVE majority of this program is NOT "displacement cryptography",
its messing around with arrays of hard-coded constants that have little
to do with the vector work, and essentially form a second-tier of
encryption (and a very poor one at that).

The "displacement cryptography" bit can be done in 2-3 pages maximum.

> Digressing, do you know if its possible to insert
> 'mathtype' (dedicated mathematical software) equations from other
> documents into LyX docs.?

If MathType can do MathML, then yes, but it has an equation-entry mode
of its own which (once learnt) is very efficient. You are in for a steep
initial learning curve; I suggest you read the tutorial; it gets you
going pretty well. Hang in there; early hard work will pay off later.

Mark Murray

unread,
Feb 11, 2012, 2:42:08 PM2/11/12
to
On 11/02/2012 19:17, adacrypt wrote:
> Forgot to say also I have quite consciously layered on huge extra
> entanglemnt options right left and centre during he development stages
> with the view that this can always be discarded if not required. You
> obviously are of that frame of mind - that's great.

Well fine, but as you are trying to make a point, its best for YOU
to remove all the arbitrary extra stuff to clarify that point.

> The diagnostic program was never required be me in thousands of cases
> - never once did this program fail in practice. I enclosed it as a
> tutorial hoping it would benefit some users.

Enclosed with no documentation and no explanation it was no use to me.

> I think you are better placed to explain this crypto to sci crypt
> readers now maybe than me ?

No. My position is that its insecure. You proposed it, you defend it.

> Please do this if you can.

I'll post my version of the source code. As for defending the algorithm,
thats your job.

> The new mathematics is "Vectoring Factoring" - you will need to get
> your head around this and notice how it is applied to cryptography in
> th Ada program if you are ever going to teach it academically ? - a
> lucrative prospect in the future - could well be!
>
> Good Luck if you do.

Careful. You are poisoning the well again here. My disagreement does
not automatically invalidate my position due to musunderstanding. You
need to make an unambiguous, clear case.

adacrypt

unread,
Feb 11, 2012, 3:43:20 PM2/11/12
to
This is becoming too accusative already - I have placed this crypto on
the table for all and sundry on a take it or leave it basis - for me
the job satisfaction was in realising the intellectual core into a
working program - its job done to my personal satisfaction at least -
I don't have to stand over anything - there 's no money back
guarantee.

I seek nothing in fame or fortune except that I expect my name to go
on to this invention and I would like it known that it was registered
at the Library of Congress in Washington some ten years ago as well as
at Stationers' Hall London just prior to that - that's all.

Go ahead and optimise this cipher as much as you can - I suggest that
you give the Ada program a rest and instead have a good look at the
"factoring of Vectors " theory before you go much further. This is
essential to understanding the cipher.

I understand that intensive coding in a cipher program can seem to
shift the goal posts in the apparent modus operandi a lot but at the
end of th e day it must stand up to the independent mathematical paper
attacks.

Good Luck with everything you do in any case.

- adacrypt

Bruce Stephens

unread,
Feb 11, 2012, 3:50:54 PM2/11/12
to
adacrypt <austin...@hotmail.com> writes:

[...]

> This is becoming too accusative already - I have placed this crypto on
> the table for all and sundry on a take it or leave it basis -

And, as on previous occasions, nobody seems inclined to take it.

You don't seem inclined to give any motivation to do so (for example by
fixing a particular cipher and (briefly and concisely) describing
it). Or even more simply, by indicating how you think it's better than
OTP. (Because at first glance it looks much worse: more complex, less
efficient, less general.)

[...]

adacrypt

unread,
Feb 11, 2012, 4:17:46 PM2/11/12
to
On Feb 11, 8:50 pm, Bruce Stephens <bruce+use...@cenderis.demon.co.uk>
wrote:
Bye - adacrypt

Mark Murray

unread,
Feb 11, 2012, 4:44:03 PM2/11/12
to
On 11/02/2012 20:43, adacrypt wrote:
> This is becoming too accusative already - I have placed this crypto on
> the table for all and sundry on a take it or leave it basis - for me
> the job satisfaction was in realising the intellectual core into a
> working program - its job done to my personal satisfaction at least -
> I don't have to stand over anything - there 's no money back
> guarantee.

What??! "Accusative"? We are debating both sides of a point; there
is no need to be hair-trigger-sensitive about that.

Do you want folks to help you? If so help them to do so!

> I seek nothing in fame or fortune except that I expect my name to go
> on to this invention and I would like it known that it was registered
> at the Library of Congress in Washington some ten years ago as well as
> at Stationers' Hall London just prior to that - that's all.

Your copyright is on your work the moment you publish it; EU and US law
makes it so, and there's no need to work at that. The law and history
are well-and-truly on your side here.

> Go ahead and optimise this cipher as much as you can - I suggest that
> you give the Ada program a rest and instead have a good look at the
> "factoring of Vectors " theory before you go much further. This is
> essential to understanding the cipher.

Please don't get repetitive over this "understanding is essential".
I've read it, and I've given you feedback (ignored near as I can tell).

Declaring anyone who disagrees with you as having "misunderstood" has
essentially wrecked your argument; you portray yourself in a very poor
light by doing so.

Now, are you prepared to continue debating in a mature manner or is
it throw-wobbly time?

> I understand that intensive coding in a cipher program can seem to
> shift the goal posts in the apparent modus operandi a lot but at the
> end of th e day it must stand up to the independent mathematical paper
> attacks.

Correct. Pity you won't play nice.

> Good Luck with everything you do in any case.

... because you are too busy kicking and screaming to have a mature
discussion?

Mark Murray

unread,
Feb 11, 2012, 4:44:46 PM2/11/12
to
On 11/02/2012 21:17, adacrypt wrote:
> Bye - adacrypt

How very mature of you.

David Eather

unread,
Feb 11, 2012, 5:47:50 PM2/11/12
to
On Sun, 12 Feb 2012 06:43:20 +1000, adacrypt <austin...@hotmail.com>
wrote:
What is wrong with you Adacrypt? Mark has been the most patient,
hospitable and accepting a person I or anyone else can imagine. If he
doesn't agree with you, it does not mean he is attacking you. In this
case, it just means that YOU have not made your case. HE is *STILL*
working with you in good faith.

--

Mark Murray

unread,
Feb 11, 2012, 6:10:29 PM2/11/12
to
On 11/02/2012 22:47, David Eather wrote:
> What is wrong with you Adacrypt? Mark has been the most patient,
> hospitable and accepting a person I or anyone else can imagine. If he
> doesn't agree with you, it does not mean he is attacking you. In this
> case, it just means that YOU have not made your case. HE is *STILL*
> working with you in good faith.

*Blush* Thanks! :-)

adacrypt

unread,
Feb 11, 2012, 8:15:13 PM2/11/12
to
I should have known better from the past that it would all go awry but
not so soon - Good Luck again with your efforts but count me out here
and now - adacrypt

Mark Murray

unread,
Feb 12, 2012, 2:35:43 AM2/12/12
to
On 12/02/2012 01:15, adacrypt wrote:
> I should have known better from the past that it would all go awry but
> not so soon - Good Luck again with your efforts but count me out here
> and now - adacrypt

You only have yourself to blame; when you are ready to talk again,
minus the chip on your shoulder, I'll be here.

When you do so, DON'T expect your audience to make your case for
you; that's YOUR job. For as long as you have MASSIVE barriers to
folks' comprehending your work, it will remain opaque.

adacrypt

unread,
Feb 12, 2012, 4:25:21 AM2/12/12
to
Youv'e got it all wrong - I don't need anyone - I have already
delivered - I'm being altruistic to all others in this matter - Your'e
pushing against an open door here - the maths are sreaming at you.

It will be interesting to see how you managed this massive
compression of my stuff into 20 lines of source code.

I think you are probably doing a static arithmetic simplification of
the results of this one-off particular encryption session by me and
not taking into account that it must stand up dynamically in all the
billions of future repetitive sessions by users worldwide if it is
ever to survive as a working cipher.

Again , Good Luck to you if you can succeed, I can't see it happening
but I am keeping an open mind nonetheless.

That's it from me.

- adacrypt

Mark Murray

unread,
Feb 12, 2012, 1:34:56 PM2/12/12
to
On 12/02/2012 09:25, adacrypt wrote:
> Youv'e got it all wrong - I don't need anyone - I have already
> delivered - I'm being altruistic to all others in this matter - Your'e
> pushing against an open door here - the maths are sreaming at you.

A fundamentally important point in scientific and engineering
research is an open mind; constantly cherishing the notion that
you avn very easily be wrong.

When the position does not allow for this possibility, it is no
longer a scientific argument, it is a dogma.

That is what your position has descended to.

> It will be interesting to see how you managed this massive
> compression of my stuff into 20 lines of source code.

Learn to listen properly.

What I have is an encryption program that is compatible with yours
in that your decryptor will recover the original input, exactly as
if your encryptor (not mine) had been used.

The "twenty lines" is a guess at how long the program would be IF
all the obfuscating array-mangling was removed, and the program
reduced to the vector work ONLY. NOTE THIS POINT!!!

Of course, this would require compatible changes to the decryptor,
and I haven't done that either. NOTE THIS POINT TOO!

> I think you are probably doing a static arithmetic simplification of
> the results of this one-off particular encryption session by me and
> not taking into account that it must stand up dynamically in all the
> billions of future repetitive sessions by users worldwide if it is
> ever to survive as a working cipher.

That makes only partial sense.

I rewrote the program using proper programming techniques. Why not
see what I did instead of writing semi-comprehensible speculations?

> Again , Good Luck to you if you can succeed, I can't see it happening
> but I am keeping an open mind nonetheless.

You? Open mind? Joke?

I'll be posting my version of your code in an hour or two. I would have
done it earlier, but I was out all day.

> That's it from me.

Nope. You'll be back, original argument intact, chip firmly on shoulder.

Mark Murray

unread,
Feb 12, 2012, 3:14:47 PM2/12/12
to
On 12/02/2012 09:25, adacrypt wrote:
> I think you are probably doing a static arithmetic simplification of
> the results of this one-off particular encryption session by me and
> not taking into account that it must stand up dynamically in all the
> billions of future repetitive sessions by users worldwide if it is
> ever to survive as a working cipher.
>
> Again , Good Luck to you if you can succeed, I can't see it happening
> but I am keeping an open mind nonetheless.

I'm an Ada neophyte; I've no doubt committed many /faux pas/'s
of Ada Idiom. I'm working on this; clubat gratefully accepted!

My version of the code is at:

http://people.freebsd.org/~markm/skew_line_encryption.adb

Its all in one file (except the obvious Ada library stuff),
and I've preserved as much as I'm able (given the dearth of
information) of what I think constitutes the key material.

Ultimately, this key material needs to be properly checked
and generated by software; it looks like it was originally
done by hand, and its rather badly flawed. Scrambling loops,
for example should likely be LFSRs or the like; instead they
display naive structure. This is a distraction anyway; it
has nothing to do with the purpose of the program; to show
vector crypto in action, and secure in its own right.
The little security of the vector methods is destroyed by
very poor choice of constants; in fact it is overwhelmingly
likely that poor understanding of the "mixing" constants
on the part of a user will wreck any security thats left.

Where the "skew line" bit comes in is entirely unclear.

David Eather

unread,
Feb 13, 2012, 2:28:16 AM2/13/12
to
On Sat, 11 Feb 2012 19:24:26 +1000, Mark Murray <w.h....@example.com>
wrote:

> On 11/02/2012 08:19, David Eather wrote:
>> I fear the most extreme case of Dunning-Kruger ever to become public.
>
> Yup. Up there with Harris.
>
> M

I disagree. This is much, worse. After persistent failure for a few months
Harris would admit his failure, apologize for his rudeness and then go
away.

It is true he would come back in a few months with the same stuff, but he
did have something of a reason for his bad memory.

Mark Murray

unread,
Feb 13, 2012, 3:19:19 AM2/13/12
to
On 13/02/2012 07:28, David Eather wrote:
>> Yup. Up there with Harris.
>>
>> M
>
> I disagree. This is much, worse. After persistent failure for a few
> months Harris would admit his failure, apologize for his rudeness and
> then go away.

Hmm. I only ever saw harris apologise twice; once was over his rubbish
travelling salesman algorithm, and the other was over the inevitable
failure of "surrogate factoring".

In both cases "apologiese" was a bit of a stretch; he did admit that
his work was specious, but his regrests over his behaviour were quickly
dispatched and not dwelled over.

On twitter, a year or three later, he ruminated on his past errors,
with the biggest mention going to the much-mentioned "I WILL GET MY
FUCKING MONEY!!!!" posting. An apology? Perhaps. A heartfelt one?
I didn't get that.

> It is true he would come back in a few months with the same stuff, but
> he did have something of a reason for his bad memory.

This is true :-)

David Eather

unread,
Feb 13, 2012, 9:56:49 AM2/13/12
to
On Mon, 13 Feb 2012 18:19:19 +1000, Mark Murray <w.h....@example.com>
wrote:
He sounds more angry than I remember him - perhaps he has started chewing
his tin-foil hat?

(I think he is too old for all composite fillings)

Mark Murray

unread,
Feb 13, 2012, 2:15:03 PM2/13/12
to
On 13/02/2012 14:56, David Eather wrote:
> He sounds more angry than I remember him - perhaps he has started
> chewing his tin-foil hat?

Very much so!

For a while he was running the planet from his Twitter account.
I guess he got better meds, because he stopped that. Now he's
infrequently writing very boring whines about how hard-done-by
he is on G+. Not worth a visit - promise!

> (I think he is too old for all composite fillings)

Dunno; Mercury diffusion could explain him rather well.

David Eather

unread,
Feb 13, 2012, 5:46:07 PM2/13/12
to
On Tue, 14 Feb 2012 05:15:03 +1000, Mark Murray <w.h....@example.com>
wrote:

> On 13/02/2012 14:56, David Eather wrote:
>> He sounds more angry than I remember him - perhaps he has started
>> chewing his tin-foil hat?
>
> Very much so!
>
> For a while he was running the planet from his Twitter account.
> I guess he got better meds, because he stopped that. Now he's
> infrequently writing very boring whines about how hard-done-by
> he is on G+. Not worth a visit - promise!
>
>> (I think he is too old for all composite fillings)
>
> Dunno; Mercury diffusion could explain him rather well.
>
> M

:D

On a different note, he was lucky I live a long way away. One day he
proudly boasted that he had received a place in a gifted student program.
I nearly exploded. The pig had that great opportunity, that some kids
would give their left arm for, and the best he could do was become an
Internet pest. I'm a bit more sanguine these days. You never know what
pressures there was to make someone pop a valve.

Mark Murray

unread,
Feb 14, 2012, 2:35:25 AM2/14/12
to
On 13/02/2012 22:46, David Eather wrote:
> On a different note, he was lucky I live a long way away. One day he
> proudly boasted that he had received a place in a gifted student
> program. I nearly exploded. The pig had that great opportunity, that
> some kids would give their left arm for, and the best he could do was
> become an Internet pest. I'm a bit more sanguine these days. You never
> know what pressures there was to make someone pop a valve.

Here's one to raise your schadenfreude glass to; He's unemployed (has
been for over a year), and his previous job was something like Data
Entry Clerk.

The world's greatest thinker has found his place.
0 new messages