Anti-snooping operating system close to launch (developed to defeat Brit snooping laws)

New Scientist ^ | May 28, 2002

[proxy_user]

There are many ways to do this, depending on what you want to accomplish. Some are illegal, such as using chains of overseas proxies without permission of their owners. Others are legal, or at least not illegal. Things like public anonymizers, mail-to-news gateways, the Cyberpunk and Mixmaster systems, nymservers, Freenet, Crowds.

You need some technical sophistication to use these things succesfully.

[Wurlitzer]

"You need some technical sophistication to use these things successfully."

There is a very low-tech yet unbreakable method. You and someone agree on a book in both of your homes. You send a series of numbers that reference a page, paragraph, sentence, and letter or punctuation position. To break the code you must know the book title, which number is the page, which is the paragraph (if you use it), which number is the sentence, and which number is the actual character etc.. In using this method, repeating the same letter will generate a completely different sequence of numbers each time.

Unless you or the other person tells the government what book you are using they cannot break the code.

For example if I sent you 2,1,34,6,10,7,1,26,4,22,4,20,3,9,1,37,1,1 I doubt if any of our alphabet government agencies could tell you what I sent.

Sending a unique value say 78 would tell the person to change the order of page,paragraph,sentence,character. If you want to protect it further you could apply some encryption to it but that usually means you have to have a copy of the code somewhere and the whole purpose of this method is to prevent anyone from finding the encryption key.

The only caveat with this method is to choose a book or books that would be common to many homes not something unique like the collected wisdom of Hillary Clinton, which only has 1 blank page.

[Redcloak]

Someone has already suggested a similar approach, though a bit more high tech. What you're describing is a series of "one time pads"; keys that are used only once and then tossed. These are great unless someone compromises the book of code sheets. A researcher recently suggested using a continuous stream of data as the source for the one time pads. Users would simply pick a particular moment in time to start collecting data from the stream to generate the keys. When they're done, they discard the data. The next time the want to encrypt data, they go back to the stream and collect more data for a fresh set of keys. So long at the agreed upon time to start collecting data is kept secure, the given message is secure; even though the data stream is being broadcast to the general public. (Except, of course, from brute force attacks. But all codes are eventually suceptible to brute force attacks.)

[Wurlitzer]

"But all codes are eventually suceptible to brute force attacks.) "

You may be right Redcloak but to my knowledge, the method I described has not been broken without one of the parties giving up the book.

[Redcloak]

It's very secure. However, if one has the resources of an unlisted government agency and a basement full of Cray computers at their disposal, even one time pads can be beaten. A defense would be to encrypt garbage along with the "good stuff". This wastes "Agent Smith's" valuable computer time decoding things like grocery shopping lists and your 5-year old's poetry.

[Doctor Stochastic]

Kahn's "The Codebreakers" shows how this can be broken. It's not very secure.

[Doctor Stochastic]

No, onetime pads cannot be broken. One+epsilon time pads can be broken. See the venona pages on the internet.

[Britton J Wingfield]

I believe that the NSA uses white noise from a radio telescope as a one-time key. A 650-meg CD of random electromagnetic crap from the sun makes for a hell of a secure key.

[Free the USA]

Bump to Index

[Wurlitzer]

Kahn's "The Codebreakers" shows how this can be broken. It's not very secure. "

Strongly disagree! If used as described it has not been broken. Only if letters in the same sentence are reused (if your lazy) or you limit yourself to a single page (lazy again). The permutations are huge.

2,7,16,73,45,67,11,42,5,3,51,48,2,6,35,88 is "AAAA" encrypted. All you have to know is the book I used, which number is the page, which number is the sentence, and which is the character. (I didn't use paragraph I'm lazy it also is usually a smaller number and could be spotted.)I also changed the sequence somewhere in the string.

[general_re]

It's a slightly more complex version of the classic book cipher. Given enough ciphertext, it's crackable. You use that method long enough, and whoever's watching you will assemble enough of your messages to begin deciphering them.

[Doctor Stochastic]

Don't bet your life on it. Codes of this type have been broken without even knowing the book.

There are several pitfalls. A part of the book cannot be reused in other messages else the overlap allows unfolding of the code. It's not easy to select a book for two people to share and that is not obvious to the adversary; Bible, Quotations of Mao, Shakespeare, Book of Common Prayer, Gone With the Wind, Dictionary, etc., are all rather obvious. It's easy to check the code against many known books. Some books just do not have enough words (even The Dictionary) and using the book to select letters rather than words leaks even more information. The adversaries may also be suspicious that a suspected spy has a copy of a particular book.

[John Jorsett]

There was someone using digital images of working lavalamps to generate random sequences for people on the net to use. I don't think this is still available, but it was an interesting idea, and probably one that the average person could implement.

[general_re]

'Zackly. And along with the shortcomings of the method itself, there's still the problems inherent in any secure communications - how do you know your key exchange isn't compromised? How do you know the black hats aren't watching you and observing that you always pull out a copy of "Little House on the Prairie" when you get e-mail from "John"?

One-time pads are uncrackable, given two conditions. One, they must be random. Two, they must be uncompromised. The Venona stuff failed the first test - the Soviets got lazy and started re-using pads when they should have tossed them. John Walker was possibly the most dangerous spy in the history of the United States. Why? He was was giving the Soviets the pads and the "word of the day" to tell them what pad to use - there was a ten-year period during which the Soviet Union read every single piece of flash traffic that the US Navy generated.

[Wurlitzer]

"Don't bet your life on it. Codes of this type have been broken without even knowing the book."

You say yes, I say no so we will have to disagree! Noodle this around. If raghead #1 wants to tell raghead #2 that tomorrow is the day he does not send out a 100 page document. He encrypts "tomorrow" using the selected book. Using only the bare minimum of code, out of a book with 200 pages, 40 lines per page, 80 characters per line. Any of the 5 characters can come from any page, any line with a source of 640,000 possible characters. To decrypt you need some pattern or some type of mathmatical repetition. Raghead 1 can use any "t", any "o", any "m", any R, and any "w" in an entire book.

Calling all number crunching type people what are the odds of finding the sequence of just 5 characters in 640,000 in the correct order?

The page,line,character sequence must be know. P,L,C or C,L,P or C,P,L etc then the book title must be known. You have to also know that this is the type of code being used.