Cryptography-Digest Digest #246, Volume #12 Wed, 19 Jul 00 09:13:01 EDT
Contents:
Re: PGP US Versions Broken,no good?? (jungle)
Re: RC4-- repetition length? (Bill Unruh)
Re: Good free stream cipher ? (Boris Kazak)
Re: PGP US Versions Broken,no good?? (Dave Ashley)
Stream Ciphers one way hash Question (Nemo psj)
Re: PGP US Versions Broken,no good?? (Stray Cat)
Re: Cipher Block Chaining (Mok-Kong Shen)
Re: RC4-- repetition length? ("Scott Fluhrer")
Re: Good free stream cipher ? ("Scott Fluhrer")
Re: Carnivore and Man-in-the-middle (Steve Rush)
Re: Good free stream cipher ? (Runu Knips)
TAGGED INFORMATION (+wuff)
how strong is my own encryption? ([EMAIL PROTECTED])
Re: Good free stream cipher ? (S�bastien SAUVAGE)
Re: how strong is my own encryption? (Eric Hambuch)
Re: Crypto source code library suggestions? (Bob Deblier)
Re: Crypto source code library suggestions? (Tom Anderson)
Re: Win2000 Encryption (Daniel James)
Project (Teo Li Xi)
Re: PGP US Versions Broken,no good?? (Richard Herring)
Re: Carnivore and Man-in-the-middle ("matt")
Re: how strong is my own encryption? (Runu Knips)
----------------------------------------------------------------------------
From: jungle <[EMAIL PROTECTED]>
Crossposted-To: alt.security.pgp
Subject: Re: PGP US Versions Broken,no good??
Date: Tue, 18 Jul 2000 23:34:17 -0400
you are showing by your writing,
that you don't have any idea, after reading my post, what I did say ...
complete lack of intelligence, at least ...
An Metet wrote:
>
> > viper wrote:
> > > I've heard from a few people(one who is a
> > > programer of encryption software) that the
> > > US versions of PGP(6.5.3 etc) are broken,
> > > no good and the US gov. can break them
> > > because these versions are made so they
> > > can be broken so the gov. can read anything
> > > encrypted by the US versions. Could just be
> > > an urban myth but I 've dumped my 6.5.3 for
> > > 6.5.1i(international)(supposedly safe)
>
> jungle wrote:
> > it's not what you did hear, but what you can confirm ...
> >
> > I did hear that v651i is broken ...
> > the reliable one is only v262 ...
>
> Jesus, jungle. I don't know why I keep using temporary kill
> filters on you. They expire after 14 days and I come back
> to yet more of your non sequiturs.
>
> You are a bona fide idiot. I'm making this filter permanent.
>
> *PLONK*
------------------------------
From: [EMAIL PROTECTED] (Bill Unruh)
Subject: Re: RC4-- repetition length?
Date: 19 Jul 2000 04:07:44 GMT
In <8l32co$q9t$[EMAIL PROTECTED]> "Scott Fluhrer" <[EMAIL PROTECTED]>
writes:
>- It is now known that RC4 is efficiently distinguishable from random data
>after 2Gb.
What is the distinguisability (Ie, how is it distinguishable).
------------------------------
From: Boris Kazak <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Good free stream cipher ?
Date: Wed, 19 Jul 2000 04:58:58 GMT
Runu Knips wrote:
>
> I'm looking for a good & free stream cipher algorithm.
> Does anybody have a suggestion ?
=========================
Recipe: Take BLOWFISH and run the key setup procedure.
You will have 5 arrays of subkeys.
P[72], S0[1024], S1[1024], S2[1024], S3[1024]
Now your stream cipher will look like following:
Ct[i] = Pt[i]^P[i%71]^S0[i%1019]^S1[i%1021]^S2[i%1023]^S3[i%1024]
(you can verify yourself that 71, 1019, 1021, 1023 and 1024 are
all mutually prime)
The period of this generator will be equal to the product of
all 5 numbers = 77380915780608 ~ 2^46.
Any intermediate byte from this series is available once you
know the index.
I hope that your message will be shorter than that, in any case
you always have the option of rekeying.
Best wishes BNK
------------------------------
From: Dave Ashley <[EMAIL PROTECTED]>
Crossposted-To: alt.security.pgp
Subject: Re: PGP US Versions Broken,no good??
Date: Wed, 19 Jul 2000 04:53:03 GMT
If I understand PGP correctly, there is an easy way to check.
Just encrypt the same file or text using the same key on both products.
If the output is the same, the later version has not been "broken".
Hope this makes sense.
Dave.
In article <[EMAIL PROTECTED]>,
jungle <[EMAIL PROTECTED]> wrote:
> you are showing by your writing,
> that you don't have any idea, after reading my post, what I did
say ...
>
> complete lack of intelligence, at least ...
>
> An Metet wrote:
> >
> > > viper wrote:
> > > > I've heard from a few people(one who is a
> > > > programer of encryption software) that the
> > > > US versions of PGP(6.5.3 etc) are broken,
> > > > no good and the US gov. can break them
> > > > because these versions are made so they
> > > > can be broken so the gov. can read anything
> > > > encrypted by the US versions. Could just be
> > > > an urban myth but I 've dumped my 6.5.3 for
> > > > 6.5.1i(international)(supposedly safe)
> >
> > jungle wrote:
> > > it's not what you did hear, but what you can confirm ...
> > >
> > > I did hear that v651i is broken ...
> > > the reliable one is only v262 ...
> >
> > Jesus, jungle. I don't know why I keep using temporary kill
> > filters on you. They expire after 14 days and I come back
> > to yet more of your non sequiturs.
> >
> > You are a bona fide idiot. I'm making this filter permanent.
> >
> > *PLONK*
>
>
--
====================================
Dave Ashley, [EMAIL PROTECTED]
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED] (Nemo psj)
Subject: Stream Ciphers one way hash Question
Date: 19 Jul 2000 05:41:17 GMT
I have read on this board several times that some stream ciphers wich have
been mixed with a form of a one way hash function are very secure. Im curious
how secure are they and is there something better?
------------------------------
From: Stray Cat <[EMAIL PROTECTED]>
Crossposted-To: alt.security.pgp
Subject: Re: PGP US Versions Broken,no good??
Date: Wed, 19 Jul 2000 05:41:41 +0000
On Wed, 19 Jul 2000 04:53:03 GMT, Dave Ashley <[EMAIL PROTECTED]>
wrote:
>If I understand PGP correctly, there is an easy way to check.
>
>Just encrypt the same file or text using the same key on both products.
>
>If the output is the same, the later version has not been "broken".
>
>Hope this makes sense.
>
>Dave.
The output will not be the same because the random session key used
will not be the same. Even if you encrypt the same file twice using
the *same* version, the output will not be the same. Try it.
--
You can contact me by posting to alt.anonymous.messages, ATTN: Stray Cat
New PGP Key. All others are obsolete.
PGP Key: 0x0CC6E051 finger: [EMAIL PROTECTED] for a copy.
Nym address is disabled. Don't send mail there.
DSS/Diffie-Hellman PGP Keys Will NOT Be Accepted.
------------------------------
From: Mok-Kong Shen <[EMAIL PROTECTED]>
Subject: Re: Cipher Block Chaining
Date: Wed, 19 Jul 2000 07:59:12 +0200
Vinchenzo wrote:
> I would like to know how to decrypt a message encrypted with a block algo
> (IDEA) mixed with the CBC (cipher block chaining) method?
>
> I understood that for the encryption you must initialize a vector 'v' and
> then xor it with my first block of plain data before encrypting the result
> with the block algo. After I display the result of the encryption and I xor
> it with the next block...etc...
>
> Now if I have all the encrypted block and the initial vector how should I
> decrypt the message?
Why don't you consult standard texts like Schneier's AC or even better
do an example to figure out the solution to the issue you have at hand
(for it gives you better insight)?
M. K. Shen
------------------------------
From: "Scott Fluhrer" <[EMAIL PROTECTED]>
Subject: Re: RC4-- repetition length?
Date: Tue, 18 Jul 2000 22:55:39 -0700
Bill Unruh <[EMAIL PROTECTED]> wrote in message
news:8l39ig$sbu$[EMAIL PROTECTED]...
> In <8l32co$q9t$[EMAIL PROTECTED]> "Scott Fluhrer"
<[EMAIL PROTECTED]> writes:
>
> >- It is now known that RC4 is efficiently distinguishable from random
data
> >after 2Gb.
>
> What is the distinguisability (Ie, how is it distinguishable).
In short, digraph statistics. For the long answer, see:
S. Fluhrer, D. McGrew, Statistical Analysis of the Alleged RC4 Keystream
Generator, Fast Software Encryption Workshop 2000
--
poncho
------------------------------
From: "Scott Fluhrer" <[EMAIL PROTECTED]>
Subject: Re: Good free stream cipher ?
Date: Tue, 18 Jul 2000 23:01:48 -0700
Boris Kazak <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Runu Knips wrote:
> >
> > I'm looking for a good & free stream cipher algorithm.
> > Does anybody have a suggestion ?
> -------------------------
> Recipe: Take BLOWFISH and run the key setup procedure.
> You will have 5 arrays of subkeys.
> P[72], S0[1024], S1[1024], S2[1024], S3[1024]
>
> Now your stream cipher will look like following:
>
> Ct[i] = Pt[i]^P[i%71]^S0[i%1019]^S1[i%1021]^S2[i%1023]^S3[i%1024]
> (you can verify yourself that 71, 1019, 1021, 1023 and 1024 are
> all mutually prime)
> The period of this generator will be equal to the product of
> all 5 numbers = 77380915780608 ~ 2^46.
And this can be broken (that is, you can reconstruct the array contents that
the stream cipher uses) with 71+1019+1021+1023+1024 = 4158 known stream
cipher outputs, by Gaussian elimination.
--
poncho
------------------------------
From: [EMAIL PROTECTED] (Steve Rush)
Date: 19 Jul 2000 07:41:08 GMT
Subject: Re: Carnivore and Man-in-the-middle
[EMAIL PROTECTED] (wtshaw) misunderstood the last paragraph of my reply:
>[EMAIL PROTECTED] (Steve Rush) wrote:
>
>> Of course, these devices would be activated on a particular line only with
>a
>> court order.
>
>It matter greatly sho does these thing and if there is oversight. Secret
>and automatic processes without assurance of the spirit of
>constitutionality are reason for getting alarmed.
Just after that sentence, I mentioned the plutonium experiments on unsuspecting
citizens, believing that that would be enough to indicate the sarcasm.
The zeroth priority of every government, no matter what its founders intended
or its constitution says, is protecting and expanding its power. I wonder if
the various {whatever}gate scandals are intended to distract public attention
from matters that should be getting a lot more attention. The last time I read
the Fourth Amendment, it didn't include any language to the effect of "Yelling
'Drugs!', 'National Security!' or 'Child Pornography!' loud enough suspends
this amendment."
==========================================================================
==============
If it's spam, it's a scam. Don't do business with Net abusers.
------------------------------
Date: Wed, 19 Jul 2000 10:27:53 +0200
From: Runu Knips <[EMAIL PROTECTED]>
Subject: Re: Good free stream cipher ?
wtshaw wrote:
> In article <[EMAIL PROTECTED]>, Runu Knips <[EMAIL PROTECTED]> wrote:
> > I'm looking for a good & free stream cipher algorithm.
> > Does anybody have a suggestion ?
> Text or binary data?
What are you talking about ???
A good cipher == a modern cipher which resists
linear and differential cryptanalysis -> ergo
Vigenere doesn't work.
------------------------------
Date: Wed, 19 Jul 2000 10:50:07 +0200
From: +wuff <[EMAIL PROTECTED]>
Subject: TAGGED INFORMATION
I use "tagged information" to track connections at times, and it never
fails to work. I weave additional info into something I like to see if
it makes its way around... and this additional info can not be removed
if the recipient does not know about that.
Does anyone have systematic information about that ? Was anything
written up in a book about it ?
+swiss+wuff+ http://www.swisswuff.ch
------------------------------
From: [EMAIL PROTECTED]
Subject: how strong is my own encryption?
Date: Wed, 19 Jul 2000 08:53:15 GMT
Hi everybody!
I may sound like an idiot, looking at all the algorithms here and
there, I'm sure my algorithm isn't something, anyway here's the
question:
How long would it take to crack down this type of encryption
(suppose you have the source code of the encryption software):
There is some kind of password (5 to 8 chars), I want to encrypt
a simple text file (in English). And here's what I do:
I somehow process all the text adding to each letter (ASCII code)
a password char (ASCII code) (every time it's the next char, and
I do the loop).
For example:
if the original text is:
"A little dog is in here!"
(ASCII codes: A=65, space=32, l=108, i=105 and so on...)
and the password is:
"dog"
(ASCII codes: d=100, o=111, g=103)
algorithm will do this thing:
"65+100 32+111 108+103 105+100..." and so on....
(A+d , " "+o , l+g , i+d ....)
the encrypted file will look like:
"165 143 211 205..." and so on....
(Note: there is no way the cracker will know the original password,
unless he/she figures it out)
and here's another question, what if I increase the password from
5 to 8 chars to let's say 20 chars, will it help or not?
Thanks,
Yuri Margolin
http://flybum.hypermart.net
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
Subject: Re: Good free stream cipher ?
From: [EMAIL PROTECTED] (S�bastien SAUVAGE)
Date: Wed, 19 Jul 2000 09:00:41 GMT
[EMAIL PROTECTED] (Runu Knips) wrote in <[EMAIL PROTECTED]>:
>I'm looking for a good & free stream cipher algorithm.
>Does anybody have a suggestion ?
I think ISAAC would fit.
Free, long period, large seed and large internal state,
uniformly distributed and unbiased, reasonably fast.
Java, C, C++, Delphi, Fortran and Modula-2 implementations available.
http://burtleburtle.net/bob/rand/isaacafa.html
--
S�bastien SAUVAGE - [EMAIL PROTECTED]
http://www.bigfoot.com/~sebsauvage
------------------------------
From: Eric Hambuch <[EMAIL PROTECTED]>
Subject: Re: how strong is my own encryption?
Date: Wed, 19 Jul 2000 11:03:40 +0200
[EMAIL PROTECTED] wrote:
>
> Hi everybody!
>
> I may sound like an idiot, looking at all the algorithms here and
> there, I'm sure my algorithm isn't something, anyway here's the
> question:
>
> How long would it take to crack down this type of encryption
> (suppose you have the source code of the encryption software):
With a computer and some KB of ciphertext: some seconds. This can be
done by a simple statistical analysis of the ciphertext.
> and here's another question, what if I increase the password from
> 5 to 8 chars to let's say 20 chars, will it help or not?
No, not really.
Eric
------------------------------
From: Bob Deblier <[EMAIL PROTECTED]>
Subject: Re: Crypto source code library suggestions?
Date: Wed, 19 Jul 2000 11:11:07 +0200
Kirk Ellett wrote:
> Hello,
>
> I am in need of a non-platform dependent encryption library in ANSI C
> source code. I only need conventional strong encryption/decryption, not
> public key, key management, or compression. The most important
> requirements for my project are speed and a generic code base, one not
> dependent on any libraries other than standard C. I also need it in a C
> library API format, not a standalone executable. Can anyone point me to
> any publicly available code legal for use in the US?
>
> Kirk
Dear Kirk,
As the author I can definitely recommend BeeCrypt; it's small and fast,
available under the LGPL license, and downloadable from the Netherlands, so
you can legally use it outside the US. See the homepage at
http://beecrypt.virtualunlimited.com/
Sincerely
Bob Deblier
Virtual Unlimited
------------------------------
From: Tom Anderson <[EMAIL PROTECTED]>
Subject: Re: Crypto source code library suggestions?
Date: Wed, 19 Jul 2000 10:19:18 +0100
On Tue, 18 Jul 2000, Kirk Ellett wrote:
> I am in need of a non-platform dependent encryption library in ANSI C
> source code.
have a look at:
http://dmoz.org/Science/Math/Applications/Communication_Theory/Cryptography/Programming_Libraries/
particularly:
http://www.cs.auckland.ac.nz/~pgut001/cryptlib/
the cryptlib web pages don't say it comes with source, but they do say:
<quote>
If you make any changes to the code, you should send a copy of the changes
to the author or authors to allow them to integrate them into the code.
This is to allow a central consistent version to be maintained.
</quote>
which seems to imply that it does.
HTH.
tom
------------------------------
From: Daniel James <[EMAIL PROTECTED]>
Subject: Re: Win2000 Encryption
Date: Wed, 19 Jul 2000 11:30:16 +0100
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>, Daniel James wrote:
> ... encrypts the symmetric key under the public key of each user
> and each group that has access to the file (presumably in the file's ACL)
Hmm. I've checked this, and it doesn't seem to do what I thought I understood
it would. It appears that the file key is only ecrypted under the public key
of the current user (the owning user?) and possibly also some system user(s)
for escrow/recovery. Simply marking a file as encrypted in Windows Explorer
makes that file unavailable to other users who previously had access (even
"Full Control" access) to the file.
Now, at the API level there are functions such as AddUsersToEncryptedFile
which (according to an article from Windows NT Magazine reproduced onthe MSDN
CD) will give additional users access to an encrypted file by storing a
cryptogram of the file's encryption key for that user - so what I described
is possible, but apparently isn't done automatically by the system.
This is a pity, as what it does do isn't particularly useful.
Cheers,
Daniel.
------------------------------
From: Teo Li Xi <[EMAIL PROTECTED]>
Subject: Project
Date: Wed, 19 Jul 2000 19:02:29 +0800
Dear all:
I am currently working on a project to come up with a GUI using
Visual C++. This GUI is supposed to be able to give a small tutorial to
the user on the latest encryption technologies, and must also be able to
let the user try out some of the existing available encryption
algorithms.
Has anyone done this before?
Is there a place where I can actually go and download as well as to
learn how to run the encryption algorithms? Thanx !
LX.
------------------------------
From: [EMAIL PROTECTED] (Richard Herring)
Crossposted-To: alt.security.pgp
Subject: Re: PGP US Versions Broken,no good??
Date: 19 Jul 2000 11:29:11 GMT
Reply-To: [EMAIL PROTECTED]
In article <8l3c7a$v75$[EMAIL PROTECTED]>, Dave Ashley ([EMAIL PROTECTED]) wrote:
> If I understand PGP correctly, there is an easy way to check.
> Just encrypt the same file or text using the same key on both products.
> If the output is the same, the later version has not been "broken".
This doesn't work. In its public-key modes, The random session keys
will be different each time, and so will the output ciphertext,
for the same input, even when nothing is broken.
--
Richard Herring | <[EMAIL PROTECTED]>
------------------------------
From: "matt" <[EMAIL PROTECTED]>
Subject: Re: Carnivore and Man-in-the-middle
Date: Wed, 19 Jul 2000 20:14:57 +0800
<[EMAIL PROTECTED]> wrote in message
news:F98d5.35841$[EMAIL PROTECTED]...
> Ahh, I see said the blind man. From what I had read in the news,
there
> were two big concerns. First, that it was administered by the FBI,
and
> not the ISP (in contrast to phone taps, which are placed by the
telco
> at the presentation of a court order). Then, that it read the
sender,
> recepient, and subject of every message.
>
> I then assumed it was working above the packet level, since that's
the
> way I would have built a machine to solve the problem. ;) It doesn't
> seem unreasonable, however. This design would be a hell of alot
> simpler than reassembling fragments into packets into complete
> messages.
I think I've read the same news, I guess we get cut and paste articles
here in Australia.
Wouldn't the simplest implementation be to simply place it in line
with the mailserver (POP3 host) or whatever, and then use that? This
would also solve problems with speed, as if there was a glut of
messages, it could presumably then check back with the server for any
missed messages when the traffic slows a bit...
Matt.
------------------------------
Date: Wed, 19 Jul 2000 14:45:39 +0200
From: Runu Knips <[EMAIL PROTECTED]>
Subject: Re: how strong is my own encryption?
[EMAIL PROTECTED] wrote:
> [...]
That cipher is the third weakest possible cipher besides sending
the plaintext (the null cipher) and using a fixed offset (the
caesar cipher). It is absolutely worthless. Don't use it.
------------------------------
** 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
******************************
