| Sysop: | Amessyroom |
|---|---|
| Location: | Fayetteville, NC |
| Users: | 23 |
| Nodes: | 6 (0 / 6) |
| Uptime: | 52:13:51 |
| Calls: | 583 |
| Files: | 1,139 |
| D/L today: |
82 files (1,646K bytes) |
| Messages: | 111,533 |
Path: archive.newsdeef.eu!mbox2nntp-comp.org.eff.talk.mbox.zip!nntp.gmd.de!stern.fokus.gmd.de
!ceres.fokus.gmd.de!zib-berlin.de!news.mathworks.com!hookup!swrinde!howland.reston.ans.net
!math.ohio-state.edu!caen!msunews!netnews.upenn.edu!netaxs.com!grendel
From: grendel@netaxs.com (Michael Handler)
Newsgroups: alt.religion.scientology,comp.org.eff.talk,misc.legal,alt.privacy.anon-server
Subject: Chaining Remailers (Was: Re: CoS and Dennis Ehrlich--the results so far)
Organization: Twenty minutes into the future
Lines: 101
Distribution: inet
NNTP-Posting-Host: unix1.netaxs.com
X-Newsreader: TIN [version 1.2 PL2]
Message-ID: <3ijlpn$bt2@netaxs.com>
In article <jktaberD4HD55.19D@netcom.com>,
John K. Taber (jktaber@netcom.com) wrote:
Uh, Tim, this is naive. Most likely some law enforcement agency will operate >> a popular remailer as a sting operation.
There is nothing to keep David Dirmeyer from setting up his own remailer,
as a hobby let us say, and secretly monitoring the traffic for law
enforcement purposes. He would be so shocked to discover porn.
The technique of "chaining" remailers, and utilizing PGP encryption on the >messages, is designed to prevent traffic analysis against attacks like
this. Consider:
We have Remailers 1 through 5, [ abbreviated R1 through R5 for brevity ]
all run by the Good Guys [tm].
Then, there is Remailer Theta [ abbreviated RT ], jointly run by David >Dirmeyer, Canter & Siegel, the NSA, the BATF, the CoS, and the Tri-Lateral >Commission. Definitely the Bad Guys [tm].
I want to send a message from <me@mysite.com> to a friend ><recipient@othersite.com>. I construct a remailer chain, like so:
<me> ==> R1 ==> R2 ==> R3 ==> R4 ==> R5 ==> <recipient>
What this means is: I create a "layered" PGP message, a somewhat complex >process. This "layered" PGP message has instructions for all the remailers
in the chain, hidden in layers of PGP encryption. The outermost layer is >encrypted for R1. I send the message to R1.
* R1 gets the message, undoes the encryption, sees that the message is >supposed to go to R2, anonymizes the message, and sends it to R2.
* R2 gets the message, undoes the encryption, sees that the message is >supposed to go to R3, anonymizes the message, and sends it to R3.
* R3 gets the message, undoes the encryption, sees that the message is >supposed to go to R4, anonymizes the message, and sends it to R4.
* R4 gets the message, undoes the encryption, sees that the message is >supposed to go to R5, anonymizes the message, and sends it to R5.
* R5 gets the message, undoes the encryption, sees that the message is >supposed to go to <recipient@othersite.com>, anonymizes the message, and >sends it to <recipient@othersite.com>
Since R1 through R5 are the Good Guys [tm], nothing bad can happen here.
But, what if, as some say, one remailer is a Bad Guy [tm] in disguise
trying to do traffic analysis from the inside? Well, let's see.
Here's a new remailer chain:
<me> ==> R1 ==> R2 ==> R3 ==> RT ==> R4 ==> R5 ==> <recipient>
In the middle of this chain, RT gets the message. What can RT see? It
knows the message came from R3, but has no idea where it came before
that -- all the previous remailer have anonymized it. It knows the
message is going to R4, but it can't undo the next encryption layer, so
it doesn't know the where the message is going after that. It can't do >*anything* with the message. The only possible `attack' is to drop the >message on the floor -- and any remailer that is known for "accidentally" >losing messages will be shunned like the plague.
Now, what if RT is at the *beginning* of the chain?
<me> ==> RT ==> R1 ==> R2 ==> R3 ==> R4 ==> R5 ==> <recipient>
They can see that I'm sending a message into the remailer network --
somewhat `suspicious', but not a crime (yet). They know the message is
going to R1, but after that...?
Finally, what if RT is the last remailer in the chain?
<me> ==> R1 ==> R2 ==> R3 ==> R4 ==> R5 ==> RT ==> <recipient>
Now, they can see the final message before <recipient> gets it. If I
encrypt the message with <recipient>'s public key, however, they can't
read that either. And they have *no* idea who it came from --- only R4.
Now, the beauty of the whole `remailer chaining' scheme is this: if >encryption is used on the intervening layers, it requires the cooperation
of *every* remailer operator in the chain to successfully determine where
a message came from. If even one of the remailer operators refuses to >participate, the rest of the chain cannot be determined.
This is why Tim is so adamant about getting more remailers started up in >foreign countries. If someone wants to trace a message back, and some or
all of the remailers used in the chain are in foreign countries, it's
going to be a lot more difficult to work with, contact, or compel
remailer operators to do so. In fact, if you choose your countries
correctly, it may be impossible (or near impossible).
As it stands now, the remailer network is pretty weak. Most of the
remailers are on US soil -- a federal official might not have a difficult >time getting action taken against them all. WE NEED MORE FOREIGN REMAILERS! >I'm sorry if this explanation is a bit vague and non-linear. I've been >working on a better one, and if I ever get it done, I'll post it here. :-) >Michael
--
Michael Handler <grendel@netaxs.com> >Civil Liberty Through Complex Mathematics Philadelphia, PA >"Toi qui fais au proscrit ce regard calme et haut" -- Baudelaire * Skotoseme >PGP Key ID FC031321 Print: 9B DB 9A B0 1B 0D 56 DA 61 6A 57 AD B2 4C 7B AF [end quoted plain text]
...snip
Date: 24 Feb 1995 04:03:03 GMT
From: grendel@netaxs.com (Michael Handler)
Newsgroups: alt.religion.scientology,comp.org.eff.talk,misc.legal,alt.privacy.anon-server
Subject: Chaining Remailers (Was: Re: CoS and Dennis Ehrlich--the results so far)
Message-ID: <3ijlpn$bt2@netaxs.com>
In article <jktaberD4HD55.19D@netcom.com>,
John K. Taber (jktaber@netcom.com) wrote:
Uh, Tim, this is naive. Most likely some law enforcement agency will operate
a popular remailer as a sting operation.
There is nothing to keep David Dirmeyer from setting up his own remailer, >>> as a hobby let us say, and secretly monitoring the traffic for law
enforcement purposes. He would be so shocked to discover porn.
The technique of "chaining" remailers, and utilizing PGP encryption
Path: archive.newsdeef.eu!mbox2nntp-comp.org.eff.talk.mbox.zipsnip
!nntp.gmd.de!news.rwth-aachen.de!news.rhrz.uni-bonn.de!news.uni-stuttgart.de
!rz.uni-karlsruhe.de!xlink.net!howland.reston.ans.net!swrinde!ihnp4.ucsd.edu
!network.ucsd.edu!cthulhu.extern.ucsd.edu!user
From: lcottrell@popmail.ucsd.edu (Lance Cottrell)
Newsgroups: alt.religion.scientology,comp.org.eff.talk,misc.legal,alt.privacy.anon-server
Subject: Re: Chaining Remailers (Was: Re: CoS and Dennis Ehrlich--the results so far)
Date: Sat, 25 Feb 1995 23:34:29 -0800
Organization: Obscura
Lines: 64
Distribution: inet
References: <3ijlpn$bt2@netaxs.com> <gradyD4HoC6.DFs@netcom.com> <3ikj2d$s0p@netaxs.com>
NNTP-Posting-Host: cthulhu.extern.ucsd.edu
Message-ID: <lcottrell-2502952334290001@cthulhu.extern.ucsd.edu>
-----BEGIN PGP SIGNED MESSAGE-----
In article <3ikj2d$s0p@netaxs.com>, grendel@netaxs.com (Michael Handler) >wrote:
) In article <gradyD4HoC6.DFs@netcom.com>,
) Grady Ward (grady@netcom.com) wrote:
) > Except that if one were to sniff packets at a dozen or so backbones,
) > you could construct a pretty good case just from traffic analysis
) > about who sent what plaintext, no matter what number of PGP remailers.
)
) I deliberately didn't mention packet sniffing or other such attacks here;
) I couldn't come up with something on the fly to do the topic justice. I
) wanted to address these issues in the full document.
) As Grady suggests, there are other attacks that will work against even
) chained remailers. And, of course, there are techniques being developed
) to combat these attacks and more: cover traffic, generated by a script;
) the Mixmaster Type II remailer.
) --
) Michael Handler
<grendel@netaxs.com>
There is cover traffic already. I generate random length random content >messages, and chain them through random chains of remailers. These messages >are sent on average every 10 minutes. The messages use PGP for all[end quoted plain text]
remailers which support it. The traffic generator also pads the message for >each remailer which supports cutmarks. My remailer chaining script "Chain" >also has all of these features (the traffic generator is just a hack of >Chain).
Thanks for mentioning Mixmaster. A client for Sun SPARC stations (which >includes netcom) is available on my homepage along with a paper on the
design of mixmaster.
-----BEGIN PGP SIGNATURE-----
Version: 2.6
iQCVAwUBL1AupFVkk3dax7hlAQEHdAP9G/xLl2O+3YV9A/RwrSRLdzM58nMpI+c2 >CXESBGxgSbIrp61B3nWvGAlh/BRj05V/4fAXYaraRoH0B1gQke8NXWIG0YG9xYzG >rgoHHjZ/Ny9WuZl+/RVa+GbOiOHN1j6Myg+rF5n9aqrvyxzB25O02QFdRnFD6SyM
zYheXQA5Ts0=
=ms/b
-----END PGP SIGNATURE-----
--
Lance Cottrell loki@obscura.com
PGP 2.6 key available by WWW or server. Encrypted mail welcome.
Home page http://nately.ucsd.edu/~loki/
Check out the next generation remailer Mixmaster on the WWW page.
For remailer info: remailer@nately.ucsd.edu Subject: remailer-help
"Love is a snowmobile racing across the tundra. Suddenly
it flips over, pinning you underneath. At night the ice
weasels come."
--Nietzsche
David Tytler Astrophysics Group Home Page[end quoted plain text]
Our group is located on the campus of the University of California, San Diego at
the Center for Astrophysics and Space Sciences (CASS). We are studying many >aspects of Quasar Absorption Lines with data we obtain with the Hubble Space >Telescope, the W. M. Keck 10-m Telescope on top Mauna Kea on the Island of >Hawaii, and the Shane 3-m Telescope at Lick Observatory near San Jose, CA.
To see what we've done recently, check out some of our papers here:
List of available papers.
The Members of the Group are
David Tytler Our Merciless Task Master
Scott Burles Galley Slave
Lance Cottrell Galley Slave
http://nately.ucsd.edu/~loki/Welcome.html
David Kirkman Galley Slave
Slave, Student, it's such a fine line.
Some Useful Astronomy Links:
Center for Astrophysics and Space Sciences at UCSD
Hubble Space Telescope
American Astronomical Society
National Optical Astronomy Observatories
Harvard ADS Abstract Service
National Aeronautics and Space Administration
Astrophysics Preprint Server
Astronomical Society of the Pacific
University of California Observatories
European Southern Observatory
Send Comments Here , Galley Slaves like mail too. (619) 534-8575
Obscura[end quoted plain text]
http://obscura.com/~loki/Welcome.html
Free Speech Online: Blue Ribbon Campaign[end quoted plain text]
This is my Home Page
There are many others like it
But this page is mine.
No left turns un-stoned!
Alerts!
Penet remailer shuts down! Between unfounded acusations of spreading child >pornograph and the likelyhood of being forced to hand over names to the >Scientologists, Julf reluctently shut down his anonymous remailer at >anon.penent.fi and issued a press release explaining the situation.
Mixmaster for DOS and Windows is now available! To download it go through the >export restricted page, and look in the DOS directory. You should also download
the latest, Mixmaster enabled, version of Private Idaho. Mixmaster is difficult
to use without it.
Netscape: Why? The threat of GAK (Government Access to crypto Keys) (was >"Netscape May Support GAK!"). Update: IBM releases "Lotus Notes" with GAK! >Stuff of interest
Mixmaster & Remailer Attacks. An essay on attacks against anonymous remailers,
and a discussion of how Mixmaster defeats some of those attacks. Now updated for
Mixmaster 2.0.X (and about time too!).
Mixmaster Remailer FAQ
Andre Bacard's Anonymous Remailer FAQ
Keep your list of remailer and their keys up to date. You can download these >master lists and use them to replace your current list. Many thanks to John Perry
for maintaining these files.
type2.list
pubring.mix
Export restricted
The long awaited Mixmaster The worlds best anonymous remailer. You must
download Mixmaster to use Mixmaster remailers. Version 2.0.3 released Nov 27,
1995
Winsock Remailer: by Joey Grasty. Finally an anonymous remailer you can run
from your windows machine. In the DOS directory in the hidden directory. Last
updated 20 August 1996.
Secshar: A a secret sharing program. Last update 10/15/94. This program is in
alpha testing.
Sites that I use all the time.
Raphael L. Levien's "reliable remailer list" This is what I use to make
remailer.list for the chain script.
BAL's Mosaic based PGP key server at MIT
Bellcore's Trusted Software Integrity System.
Links to other Crypto/Remailer/Cypherpunk sites that I use frequently
Some other interesting links
The Tytler Astrophysics Group. This is where I am when not doing the
Cypherpunk thing.
Idea Futures. This is a stock market for ideas. You buy and sell shares based
on your oppinion of the idea. The price represents the consensus likelyhood of
the conjecture through supply and demand.
PGP Stats. A really nice analysis of the PGP web of trust. Take a look and see
how well connected you really are.
Emerging Japanese Encryption Policy by Stewart A. Baker. A detailed analysis
of the development of encryption policy in Japan.
Obscura information security.
The Center for Democracy and Technology (CDT)
The Electronic Privacy Information Center (EPIC)
Directory of the U.S. Congress
The NCSA Mosaic Home Document
Suggested Starting Points for Internet Exploration.
Lance Cottrell loki@obscura.com
PGP key
Hits since 2/12/96
This site is hosted by the hardest working ISP on the Internet: Infonex