ÿÿ00ÿ ÿ00ÿ01ÿ03ÿ01Privacy Issues in the Digital Worldÿ00 ÿ ÿÿ01ÿThat a society built upon democratic principles should defend privacy of the individual seems obvious. However, studying the new IT based information structures that are being formed in society, it becomes clear that privacy is something that cannot and should not be taken for granted.ÿ ÿ00 ÿÿ ÿ03ÿ02TOBIAS JANSSON ÿ02ÿ01ÿ01ÿ00ÿÿ By means of tracing an individual's electronic "footprints", a depressingly detailed image of the person in question can be gathered: health journals, phone bills, credit card payments and other economic transactions, police registers and, not least, the use of e-mail and World Wide Web all connect activities and sayings with the physical person. ÿ ÿÿ02ÿDeficient personal securityÿ ÿ00 ÿÿ01The most used netservice today is, supposedly, e-mail, electronic mail, the possibility of sending letters via Internet. The fastness, smoothness and flexibility of the system make it a day by day stronger growing competitor to ordinary mail. Though, few users of the service seem to know anything about its limits when it comes to security and integrity. To send e-mail is actually not much safer than to send a conventional postcard; the information is being exposed to basically anyone. System administrators have the same observation possibilities when forwarding e-mail as post personnel has when delivering postcards; the extent to which the privacy of the users is kept is in the hands of the administrators. In the problem also lies the fact that it is not only the administrators at the sender's and the receiver's computer systems that have these possibilities. An e-mail can be read by the persons responsible at ÿ02every ÿ01node it passes on its way to the receiver. The comparison with the postcard is striking: the post personnel at the different post offices a postcard passes have the same observation possibilities. Though, the difference turns obvious when one remembers that virtually nothing beyond vacation greetings and other in privacy matters less relevant messages are to be found on postcards, while delicate subjects and important issues very well may be discussed in e-mails. Another thing that ought not be forgot is the ease in going through electronic information. Conventional envelopes may be opened and the letter inside read, but the extra work and the economic loss a such procedure imply, results in its being used only in single, important cases. E-mails may easily be scanned for interesting key-words - on an automatic and routinely basis. To register e.g. political opinions is thus fairly simple. What is needed is some kind of "digital envelopes" that makes protection against observation possible. The supposedly only solution would be to use some kind of cryptography, something that however is connected with a lot of practical problems. Traditional methods for protecting information are based upon the use of a specific "key" that can encode and subsequently decode messages. The difficulty in this is the fact that the key must be transferred between the parties. If it is sent over the same not-secure channel as the encoded message there is nothing that stops someone from intercepting the transmission and pick up the key and use if to decode subsequent messages; this kind of cryptography scheme is only secure if the key can be transferred through a guaranteed secure channel - and to materialize a such is often practically impossible. Another solution is to have registers, digital stocks, where keys are stored. Though, the kernel of the problem remains: integrity is in the end dependent on a third party, here the ones responsible for the registers. In 1975 the 31-year-old computer wizard Whitfield Diffie, then at Stanford University, came up with an idea for a cryptography system which he named "public-key" cryptography. Diffie advocated decentralized computer systems where everyone literally has the key to his personal security. Furthermore, he predicted that there should come days when people not only would communicate electronically, but also make business by means of computers - something that would demand a digital counterpart to signatures for e.g. signing of contracts. In Diffie's system each user has his own "digital signature" that encrypts the messaged that is signed. The signatures make use of a ÿ02public ÿ01and a ÿ02private ÿ01key. The public key may and should be spread to the public, while the private one ought to be kept strictly secret. Code encrypted with either of the keys may always be decoded with the other. Suppose Alice wants to send an encrypted message to Bob (Alice and Bob are the favourite hypothetical characters in the cryptography world). She encrypts the message with Bob's public key and he decrypts it with his private key. Only Bob can decrypt the message as nobody else has access to his private key. If Alice only is in need of sending a message that the receiver has to know is authentic, she encodes - "signs" - the message with her private key. The receiver, say Bob, decodes with Alice public key, and is, if this action is successful, guaranteed that the message is authentic; if a text may be decrypted with Alice's public key there is no doubt that it has come from her and directly from her as nobody else has access to the private key that generates a code that can be decrypted with the corresponding public key. However, this procedure doesn't supply any safe security against observation as Alice's public key is available to just anyone - but such safety might not be wanted in all situations. These two procedures may be combined to obtain both safety against observation and a guarantee that a message is authentic. If Alice wants to use this when she sends a message to Bob, she first signs it with her private key and then encrypts it with Bobs public one. Bob performs these steps in reversed order by fist decode the message with his private key and then check the signature with Alice's open one. 1977 a research group at Massachusetts Institute of Technology developed algorithms that implemented Diffie's scheme. These algorithms later got the name RSA after the scientists Rivest, Shamir and Adleman. RSA-cryptography made up a working public-key system that thereby not suffered from the weakness of all former cryptography systems, the need to safely exchange keys. The algorithms were patented by RSA Data Security that took the chance to sell cryptography programs to companies like Apple. Microsoft, WordPerfect, Novell and AT&T. ÿÿ02ÿPGP - an Alternativeÿ ÿ00 ÿÿ01"ÿ02You may be planning a political campaign, discussing your taxes, or having an illicit affair. Or you may be doing something that you feel shouldn't be illegal, but is. Whatever it is, you don't want your private electronic mail or confidential documents read by anyone else. There's nothing wrong with asserting your privacy. Privacy is as apple-pie as the Constitution.ÿ01" ÿ Phil Zimmermann ÿIn the first place, it is the privacy of the individual that get jammed in the digital world; big companies and governments can afford expensive cryptography systems while private persons and grassroot organizations are exposed to the deficiency of the communication systems. Phil Zimmermann, a politically engaged computer freak realized that tools like the RSA-algorithms should be available to everybody. He started to ponder over a way to give "the people" access to a security system that uses public-key cryptography. Zimmermann began to work on the project seriously in 1984, but as he wasn't a professional cryptographer he moved slowly. In 1986 he had implemented the RSA-algorithms and in 1991 the program was ready for release. He called his system PGP, for "Pretty Good Privacy". That there was an interest for cryptography systems among common users became clear when Zimmermann started to spread PGP. He placed the program on a handful of BBS:s and gave it to a friend who posted it on Internet. Within a couple of hours people all over the world were downloading it. "Like thousands of dandelion seeds blowing in the wind", he wrote, PGP spread through cyberspace. "It was overseas the day after the release. [...] I've gotten mail from just about every country on Earth." That NSA (National Security Agency), the authority in the US the task of which is to preserve the secret codes of the country and crack the ones of the enemy, should react when Zimmermann implemented RSA-cryptography in a program, which probably would be spread outside the country. For some reason, this never happened. This worried Zimmermann, since he thought this meant that PGP had some kind of weakness, a trapdoor which the government hadn't found. But after he had consulted a cryptographer, Zimmermann realized that PGP in many ways was ineffective, but it offered a protection at least as good as the American government-standard DES (Data Encryption Standard), and thus "pretty good privacy". ÿÿ02ÿGood and bad sides of cryptographyÿ ÿ00 ÿÿ01When someone encrypts his data, it seems like many people think that he has something to hide. To recompare e-mail with ordinary postcards, imagine a society where everyone followed the law, where people only use postcards when they send messages by post. If anyone starts using envelopes, other people would be suspicious. Maybe the government would open his mail to se what he wants to hide. But we live in a world where most people protects most of their mail with envelopes; and thus nobody becomes suspicious. Therefore daily usage of cryptography of all e-mail, e.g. PGP, would be a step in the right direction; see it as a kind of solidarity. Phil Zimmermann himself formulates the essence of the problem in the manual to PGP (1994): "If privacy is outlawed, only outlaws will have privacy." Another effect of cryptography is that it will be much harder to discover crime in a system where PGP is used. In that case, the government is not able to control the contents of e-mails. Conventional envelopes can always be opened, but to decrypt PGP-encrypted messages without the correct key, is not an easy task - if it at all is possible within a reasonable time. It would be desirable to maintain a balance between the individual integrity on one hand and the safety of the government - the country - on the other. In a speech in 1979, the NSA-chief Bobby Inman among other things said "nongovernmental cryptologic activity and publication ... poses clear risks to the national security". "Crypto Anarchy", i.e. a society where every citizen has access to powerful, easy to use encrypting tools, is something that scares those who prefer state control. "We have the capability of 100-percent privacy, but if we use this I don't think society can survive", computersafety expert Donn Parker says in an article by Steven Levy, under the entry "Crypto Rebels" in the magazine Wired (nr. 1.2, 1993). Dorothy Denning at Georgetown University claims that if we fail, by misstake or not, to carry out a "trapdoor" which the government can use, the computer networks will be a sanctuary for organized criminality; bootleggers, terrorists and other criminals can conspire and act free and unpunished. In the end we'll face an increasing number of serious crimes against the society, in the same time as we get fewer chances to stop them. On way to implement such a "trapdoor", is to force everyone who wants to use public-key cryptography, to deposit their private keys in a register which is controlled by a third party, which in case of e.g. a trial, has authority to hand out a key. This forces the users to trust a third party with self interests. So, the same paradox occurs here, as the one earlier discussed in relation to conventional cryptography methods. Whitfield Diffie's idea is based upon private keys that must not be spread in any circumstances to anyone. In the same way, USA has come up with a proposal that all communication products, telephones, fax machines, wired computers and so on, must have a build-in cryptography chip - "Clipper" - witch makes it possible to send encrypted information without any special software. The catch is that every chip must have a unique key which the government has a copy of. To think about this kind of technology in the hands of a totalitarian regime is terrifying. In combination with the next step - to illegalize all non-controlled encryption, there's a risk that the society will develop into a probation-dystopia, similar to those described by writers like George Orwell and Karin Boye; every false step the individual takes, will be registered. ÿÿ02ÿElectronic Moneyÿ ÿ00 ÿÿ01The bill has through centuries been the dominating way of paying as it under the prevailing conditions has been easy to handle, but its days are to be considered counted. To handle conventional money is not economically justified. "Cash is a nightmare", says Donald Gleason, manager for a department of Electronic Payment Services Inc, USA, in another article by Steven Levy (E-Money (That's What I Want)) in Wired (nr. 2.12, 1994). "It costs money handlers in the US alone approximately US$60 billion a year to move the stuff." The introducing of high quality photo copy machines threatens the safety of paper money. Usage of credit cards becomes more and more popular, but these systems today lack adequate security devices and integrity protection; they make observation of the economic situation of the individual possible and make it easy to trace transactions. In the same rate as Internet is expanding, the demand for online payment in a convenient and safe way is increasing. The founder of the American company DigiCash, David Chaum, has a lot of ideas on how electronic money, "e-money", could - and should - be implemented. He says in an article in Scientific American (August, 1992) that the best solution would be to create "digital bills" that should work analogy with ordinary bills. Digital signatures is used to verify the authenticity of the "e-bills". A unique such digital signature, or serial number, is connected to each bill - the number itself is the bill. When the number - bill - is used for payment or other transactions it is verified by way of comparing the number to a central register, that the e-bill is authentic and that it hasn't been used before. Conventional cash guarantees anonymity; one may spend printed bills knowing that nobody can trace them. May e-money be made anonymous? And, if so, should they be? Chaums company supplies a solution that is based upon the use of "blind signatures", a process with the help of which e.g. a bank may verify the authenticity of a serial number, so that it can be used as an e-bill - without the bank knowing who owns the bill and therefore cannot trace it. When receiving a number that may be used as a bill, there is a possible of changing the number and at the same time keep up the authenticity of the bill as a payment unit. The system guarantees a wholly anonymous way of paying. The drawbacks are, however, many. "It would be dangerous and unsound public policy to allow fully untraceable, unlimited value digital currency", says Kawika Daguio at American Bankers Association in Levy's article. "It opens up opportunities for abuse that aren't available to criminals now. [...] In the physical world, it is possible to follow people, so a kidnapper can potentially be caught if the currency is marked, if the money was being observed on location, or if the serial numbers were recorded." A kidnap situation would, moreover, become even more difficult if the money exchange took place online, in cyberspace on a place that doesn't exist in the physical world. To trace where anonymous electronic money is spent is - even if they are spent offline - virtually as hard as to find out who spent them; marking e-money is hardly realizable if a special possibility for this is not built into the system - and a such possibility would contradict the purpose to keep absolute anonymity. If one, contrary to this, imagines a world where all money is electronic and traceable, one realizes that it this money contains the most effective crime fighting weapon in history. To protect the individual when transferring and storing information in the digital world, the only solution seems to be to use some form of cryptography. In connection to this we stand before an important crossing, primarily between different technological systems, secondarily between different future societies. If we do not succeed in maintaining balance between the freedom of the individual and governmental control of electronic communication systems, we risk on one hand to get a society with an increase in criminality, on the other hand a "Big Brother is watching you"-system a la that described in Orwell's dismal future vision 1984. ÿ01ÿ02ÿ03THANKS TO MIKAEL ADOLFSSON FOR VALUABLE HELP WITH TRANSLATION DURING SOME HECTIC HOURS AT THE PARTY.