🚀 go-pugleaf

markh@csd4.csd.uwm.edu (Mark) writes:
>   Supposing I have a complete description of DES in text form, and I split it
>into 8 parts, with file N containing every 8th character modulo N.
>
>   If 100 people each decide to post one of these files (they make a random
>choice without knowing about the other 99 people even being there), and posts
>it to a place that can be publically accessed outside the US, then have any
>restrictions on DES been violated?  If so, who violated them?  How?

Don't be silly.  If transmitting a complete description of DES is against some
law (which it is not, according to recent discussions; flames to /dev/null),
then participating in a transmission violates that same law even if your part
is not sufficient to complete the act by itself.  The only use of one of the
eight portions is to reconstruct the original file, so the poster is obviously
intended that the original file be reconstructed.

Are armed robbery laws unsound because one person can point the gun and another
can shout "give me all your money"?  How about the fact that ten people could
agree that they will each randomly decide which of these roles to play without
discussing it with each other, but they'll show up at the same place at the
same time (post via the same medium)?  (Or to make the analogy complete, you
advertise in a newspaper for people to do this, so that no one can be
ABSOLUTELY sure that the others will show up, but it's clear for some reason in
your town that the community mind is such that it would be extremely unlikely
for no one to be interested.)

Don't try to tell me that it would just be coincidence that all 100 people
decided to post those files and that there was no exhortation on your part for
them to be posted.  It's not credible that you split up the file in that way
for no reason.

If you're interested in getting a description of DES out, simply post one.  Or
post a journal article reference.

>   Supposing I have a complete description of DES in text form, and I split it

It is my understanding that a description of DES in text form is not
restricted, as long as it is not a software implementation of DES.  Byte 
Magazine already published one quite a few years back.  (I forget whether 
there was a full software implementation included or not.  There probably 
was, in something like 6800 assembly language.)  And it's my understanding 
that the Feds would think nothing of responding to a request for the 
appropriate FIPS or whatever standard, with the return address of KGB 
headquarters in Moscow, by sending the document.  This was back when
the Soviet Union existed and was considered an "Evil Empire".

>into 8 parts, with file N containing every 8th character modulo N.

>   If 100 people each decide to post one of these files (they make a random
>choice without knowing about the other 99 people even being there), and posts
>it to a place that can be publically accessed outside the US, then have any
>restrictions on DES been violated?  If so, who violated them?  How?

I believe this is called a conspiracy.  It's probably illegal REGARDLESS OF
WHAT THE TEXT CONTAINED, even if it's a complete but PG-rated description 
of Clinton's first post-marital sex, the Constitution of the United States, 
or the text of one of Bush's speeches.  I think you're legally much safer
posting the plaintext.

						Gordon L. Burditt
						sneaky.lonestar.org!gordon

From article <1992Sep5.164646.21443@uwm.edu>, by markh@csd4.csd.uwm.edu (Mark):
>    Supposing I have a complete description of DES in text form, and I split it
> into 8 parts, with file N containing every 8th character modulo N.
> 
>    If 100 people each decide to post one of these files (they make a random
> choice without knowing about the other 99 people even being there), and posts
> it to a place that can be publically accessed outside the US, then have any
> restrictions on DES been violated?  If so, who violated them?  How?


As I understand things...

   The description of DES is public knowledge.  Do what you want with it.

   However, software _implementations_ of DES have been deemed "munitions",
   and are thus restricted by the US Gummint ("We're your friends! Trust
   us!").

-- 
John G Dobnick                          ATTnet: (414) 229-5727
Computing Services Division             INTERNET: jgd@uwm.edu
University of Wisconsin - Milwaukee     UUCP: uunet!uwm!jgd

"Knowing how things work is the basis for appreciation,
and is thus a source of civilized delight."  -- William Safire

In article <1992Sep5.164646.21443@uwm.edu> markh@csd4.csd.uwm.edu (Mark) writes:
>
>   Supposing I have a complete description of DES in text form, and I split it
>into 8 parts, with file N containing every 8th character modulo N.
>
>   If 100 people each decide to post one of these files (they make a random
>choice without knowing about the other 99 people even being there), and posts
>it to a place that can be publically accessed outside the US, then have any
>restrictions on DES been violated?  If so, who violated them?  How?

You can do far better than that:

1. Split the article using the original and seven one-time pads (which must
   have good statistics), such that
   DES = A Xor B Xor C Xor D Xor E Xor F Xor G Xor H

2. Hide each of A..H in a form such that they all have some use other than
   re-constructing the original DES code.  (For example, embed them as
   noise in the low-order bits of a picture).

This has the advantage that in isolation, none of the messages transmit
any information at all, and even with all eight it is impossible to prove
which was the original and which were merely keys.

To confuse matters further, add an extra step:

3. HARMLESS1 = A Xor B Xor C Xor D Xor I
   HARMLESS2 = E Xor F Xor G Xor H Xor J

(where HARMLESS1 and HARMLESS2 are meaningful and useful, but of no
significance re. export laws).  Then publish I and J.

Note, however, that as has been pointed out by another poster, US export
laws are concerned with computer programs and hardware devices, not mere
descriptions of DES.

I would also note that whereas this (and several other intellectual
property laws) suffer from a confusion between data flow and information
content, I would not trust the US authorities to worry too much about
legal or indeed logical nicities if they really wanted to get you.

Graham Shaw
Department of Physics, Durham University, England

markh@csd4.csd.uwm.edu (Mark) writes:
       
   "Supposing I have a complete description of DES in text 
form, and I split it into 8 parts, with file N containing every 8th
character modulo N. I"

[a scheme to evade the U.S. export prohibitions on crypto 
technology]

--------------------


The group would undoubtedly be charged with criminal conspiracy 
in addition to violating the Munitions and Foreign Armaments
act, in addition to whatever other trumped-up charges could be 
applied.  I would guess, looking at the typical punishment 
meted out in the U.S. Code you would be threatened with 20-30 
years in prison and a half-a-million dollar fine, each.  
Consider this the U.S. government's contribution to your 
academic freedom and creative state-of-mind.  :-)

However, take heart, the Internet is already the most immediate 
and powerful voice of free technical international exchange on 
the planet.  Given that the people in all countries linked by 
it are at least as smart on average as U.S. and Canadian crypto 
researchers, and that crypto research in general requires no 
vast infrastructure (unlike the superconducting supercollider,
for example), just sharing our thoughts on sci.crypt and 
international conferences such as EuroCrypt, will soon 
guarantee that the crypto sophistication outside North America 
will surpass that on the inside.

The NSA will slowly become just a poor technical arm of the 
FBI, since only citizens in _this_ country will be stuck with 
crypto/telephony systems that can be broken...

(Stuck on citizens and industry that don't have the technical 
expertise to read and understand sci.crypt, that is. ;-))

For all you crackers out there -- this is a dream manifesto: 
the "official NSA/FBI approved" industrial crypto and digital 
telephony systems will be delayed and use less effective crypto 
systems than available in the E.C. or the Pacific Rim world.  
That undoubtedly will mean there will be holes that can be 
exploited by the crackers, while the cracker _themselves_ can 
use wholly effective cryptographic means picked up at any 
international souk to conspire and pool their computational and 
analytic skills.

What can you do to resolve these issues?

I've already internationally posted DES and RSA source to this 
group; I'm writing a new short article describing a disciplined 
approach to combining conventional crypto systems in a way that 
each "hide" each others canonical weaknesses: a method that I'm 
calling "laminated" cryptography.

By all of us amateurs and private industry professionals 
working as hard as we can to create and disseminate crypto 
information, I hope that we can stimulate the immune systems of 
our FBI/NSA to become innovative on their own and figure out 
how to fulfill their statutory missions without stifling 
cryptographic/information-theoretic research.  But I'm not 
optimistic.

Demanding innovation from any bureaucracy with its built in 
"smoothing function" is always difficult, but with the NSA/FBI 
it is going to be exponentially hard because of the lack of 
effective oversight by the citizens they serve.

I'd have posted this to sci.crypt.politics, but, alas, the 
technical and political issues seem utterly conflated.  :-(


(* Grady Ward Public Key, Mathematica format
338 decimal digit modulus]*)
e=13278340892377147290021581995910566265630745404772050295976\
531169652074968462728692947088224740474086880014187741018\
827679282936360488374332586621372969992125015540674307926\
963403599898215432920593460759568145460543791116449305640\
269575860706334126070662560984930042322363643567821283666\
166102594784564767603179111052467026453547876550953;

n=20601487053579422679033128126836668550296101243048136963408\
859700715907596530150758174233318688216293109628630361192\
968088290065732167528721908401881409490304922316321979613\
722426616580603195157882720031084346896241421285868732184\
267763239722864343244932750154734013576761572454860859591\
356619745333866669144310472787794900107034134707229;
to send me a private message, e-mail me PowerMod[message,e,n]

In article <Bu5597.3x4@sneaky.lonestar.org> gordon@sneaky.lonestar.org (Gordon Burditt) writes:
>I believe this is called a conspiracy.  It's probably illegal REGARDLESS OF
>WHAT THE TEXT CONTAINED, even if it's a complete but PG-rated description 

Not in the U.S.  Conspiracy is when you meet with someone else and
agree to commit a crime.  In some cases, you also have to take steps
towards actually carrying out what you conspired to do.  But I can get
together with a friend and conspire to meet him at McDonald's for
lunch, because meeting at McDonald's for lunch isn't a crime.

-- 
Marc Unangst                | Real men don't use Windows.  Real men use X. 
mju@mudos.ann-arbor.mi.us   | Only a real man would use a GUI where the
                            | shift keys after "Alt" are "Super" and "Hyper."

In article <1992Sep6.071901.21881@uwm.edu> jgd@csd4.csd.uwm.edu writes:
>   The description of DES is public knowledge.  Do what you want with it.
>
>   However, software _implementations_ of DES have been deemed "munitions",
>   and are thus restricted by the US Gummint ("We're your friends! Trust
>   us!").

Obviously only Americans are capable of taking an algorithm
description and turning it into working code. The next thing you know,
the US government will be revoking the passports of anyone suspected
of being able to program...

Phil

In article <1992Sep7.051916.16197@qualcomm.com>, karn@servo.qualcomm.com (Phil Karn) writes:
|> 
|> Obviously only Americans are capable of taking an algorithm
|> description and turning it into working code. The next thing you know,
|> the US government will be revoking the passports of anyone suspected
|> of being able to program...

This of course not an old idea. Look up any decent history book and read all
about the industrial revolution and English emigration law. :-)  If anyone
believe this worked they should also believe that the Merrimack Manufacturing
Company was an illusion.

On the other hand, the people who laugh at the export restrictions probably
haven't done any real integration testing of a really large software construct.
At least there's a fair chance that the foe fouls up...

---
Haakon Styri			***   Std. disclaimer applies	***
Norwegian Telecom Research

Maybe DES itself is against some law? :-> *flameflame*

	There is a nice version of DES in pascal included in "Computer Networks"
by Andrew S. Tanenbaum. Anyone have a version of DES with a 128 bit key size? Or
alternative choices for S-boxes? Or just maybe the NSA doesn't want you to crypt
something that the gov't can't uncrypt.

who@gandalf.UMCS.Maine.EDU (William H. Owen) writes:


>	There is a nice version of DES in pascal included in "Computer Networks"
>by Andrew S. Tanenbaum. Anyone have a version of DES with a 128 bit key size? Or
>alternative choices for S-boxes? Or just maybe the NSA doesn't want you to crypt
>something that the gov't can't uncrypt.
Uh, read your first line. You have the source- change the Sboxes to your
heart's content. Double, triple, qudruple encrypt with different keys,
and different permutations. If NSA had really wanted to make sure you
couldn't use anything but their weak(???) DES, they would never have
agreed to release the details of DES for you to change them.

In article <1992Sep6.094022.28262@uwm.edu> markh@csd4.csd.uwm.edu (Mark) writes:
>(4) If it's described in English, is it regarded as software?  English is
>    executable under a suitably designed specialized natural language
>    interpreter.

When we really have compilers capable of executing free-form English, the
government may adjust their notion of the distinction between software and
human-oriented description.  Since the description of DES *is* in a public
domain document (a US FIPS), it's clearly not considered to be a software
implementation that is subject to export restrictions.

In answer to the other posting about US citizens being the only people able
to turn descriptions into implementations, that's not the issue.  The
government doesn't want us to facilitate the creation of munitions by our
enemies (encryption devices are considered to be weapons -- remember, the
development of computers was catalyzed by the use of encryption during
WWII), and selling working implementations of DES is considered
facilitation.  There are also published articles that detail how to build
bombs, yet the government also controls their export as well.

Of course, these days encryption is more useful for commercial uses than
military, and the government is shooting the economy in the foot by
restricting its sales.
-- 
Barry Margolin
System Manager, Thinking Machines Corp.

barmar@think.com          {uunet,harvard}!think!barmar

In <Bu5597.3x4@sneaky.lonestar.org> gordon@sneaky.lonestar.org (Gordon Burditt) writes:

>It is my understanding that a description of DES in text form is not
>restricted, as long as it is not a software implementation of DES.  Byte 
>Magazine already published one quite a few years back.  (I forget whether 
>there was a full software implementation included or not.  There probably 
>was, in something like 6800 assembly language.)

There was 6502 assembly language included.  Interestingly enough, no library
in New Zealand (as far as I could find out) has a copy of the two issues it
was published in, although issues on either side of the two were available
freely...... I'll leave you to draw your own conclusions :-)

>enemies (encryption devices are considered to be weapons -- remember, the
>development of computers was catalyzed by the use of encryption during
>WWII), and selling working implementations of DES is considered
>facilitation.  There are also published articles that detail how to build
>bombs, yet the government also controls their export as well.

Hey, that means that there won't be laws against encryption in USA,
since the second amendment guarantees to right to keep and bear arms.
Or perhaps it was a misprint and the right was only to keep bear arms
after all, since other kinds of arms have many restrictions even
today.  Bummer - but perhaps someone can design an encryption
algorithm which is to be used against bears, and so you guys in USA
would be guaranteed to use that freely?

//Jyrki