Here's a conclusion I reach at the end of this piece: "There is a reasonable chance the Supreme Court would see the overall absurdity of a situation where the knowledge is freely available to 200 million adult Americans, with no restrictions whatsover on publication, discussion, etc., and yet uttering this knowledge in front of a foreigner is a crime." At 1:47 AM 4/3/96, Hal wrote:
The next step is printed source code. There are fonts (or other tricks, such as per-line checksums) which can be used to make scanning this in relatively reliable. I don't have enough experience to know how good it can get. But let's suppose it were practically error-free.
Nearly error-free entry of credit card slips has been done for many years in various offshore data entry locations (Barbados was once a major center of this, with jets carrying in raw credit card slips to be processed by the Barbadans, who were highly literate and well-trained in English, and yet who worked cheap). Interestingly, according to a friend of mine (Mike Ward, of San Jose), it is much cheaper (or was when he did the analysis a couple of years ago) to have a chunk of text entered two or more times by humans and then XORed for errors than to OCR the text. This form of error correction--redundant entry--would presumably work well for Schneier's code, for example. What Phil Karn and others are doing is a useful and interesting experiment, and I join others in thanking Phil and the others. However, I think it pushes the envelope without really touching the real issues. To wit, here is my bold submission: Despite the ease with which printed text can be entered by cheap labor in nearly any country, I'll bet a lot that not a single person or foreign government got their crypto for actual use from a book! It just never was a credible claim that books like "Applied Cryptography" could represent a "leakage" of valuable crypto knowledge. (In terms of actual code--the book itself, like other books, is itself an incredibly "damaging" book for evil enemies to use to educate their cryptographers, but such is the case with nearly all technical books.) The MIT Press book of PGP's code is closer to being an important example, as approval of it for export could lead to offshore versions which are "identical" to PGP and thus eport of PGP would have been de facto approved. I submit that the issue has nothing to do with the OCR font used, modulo the points made above about a determined effort to input the text. In terms of realpolitik, a handwritten edition of the PGP code might be ignored, where an OCR version is deemed "close enough" to being actual runnable code that its export will not be approved. A distinction without a difference in my book. What we are really about to run into is the "export of knowledge" issue, which I think the Constitution will have some pretty important things to say about. (It has long been the case that certain armaments knowledge, and atomic secrets as per the Atomic Energy Act, were restricted. But this knowledge was _also_ not in the public domain. There have not been many cases that I know of where knowledge was freely discussable within, say, the United States, by any citizens (and maybe others, such as permanent residents, foreign students at colleges, etc.) and yet this knowledge could not be "exported" outside the U.S. Except with our crypto laws, and some related Munitions Act laws. There is a reasonable chance the Supreme Court would see the overall absurdity of a situation where the knowledge is freely available to 200 million adult Americans, with no restrictions whatsover on publication, discussion, etc., and yet uttering this knowledge in front of a foreigner is a crime. I don't think this would pass Constitutional muster, as the lawyers like to say. (The British at least have an Official Secrets Act. Much as I dislike that Act, at least they are more consistent in the sense of classifying things as being secret. How can the U.S. argue that knowledge available in any large library or bookstore to anyone who wants it, citizen or not, may not be "disclosed" to foreigners? If it's common knowledge, it's common knowledge!) --Tim May Boycott "Big Brother Inside" software! We got computers, we're tapping phone lines, we know that that ain't allowed. ---------:---------:---------:---------:---------:---------:---------:---- Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay@got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Higher Power: 2^756839 - 1 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."