December 2003 - cypherpunks-legacy

The Virtues of Extremism
by Tim May 17 Dec '03

17 Dec '03

It seems to me we're seeing an obvious manfestation of the classic "inside the Beltway" mentality. The vocal SAFE suporters are so closely aligned with the legislative process that they can't see the forest for the trees. I won't even speculate too much about the dynamics of why this is so, but it's been seen time and time again. Maybe it's that their vision becomes tunnel vision, sort of like the way a pilot focusses on his altimeter, tapping it and not quite believing it, as his plane crashes into the ground. Maybe this is the only world they know, the world of lobbying. Maybe they do truly believe it's "the best deal we can get." Well, I am not inside the Beltway, and I don't think much of _anything_ Washington does. I refused to renew my membership in the National Rifle Association when it became clear they were "playing the game," going along with repressive, gun-grabbing legislation because "this is the best deal we can get." The NRA lost a lot of members with this stance, and saw gun ownership rights eroded at every turn. The new leadership, elected after this fiasco, pledged that no longer would they "work with" repressive legislation. We'll see what happens. Meanwhile, as with the Cypherpunks movement, vastly more radical steps are needed to sanction the gun grabbers like Swinestein and her leftist allies. (Hint: the more extremist organizations I now work with are probably one of the areas I'm apparently committing putative felonies in. We don't treat the gun laws as valid, and will defend our rights against any gun grabbers. That we use PGP means we could easily face the maximum CDT-approved sentence. As Marc Rotenberg reminded us, the CDT wrote: ""CDT strongly urges you to report H.R. 695, the SAFE Act, out of the Courts and Intellectual Property Subcommittee without amendment." Fuck Congress, and fuck CDT.) "Extremism in the defense of liberty is no vice." Now more than ever Cypherpunks need to take action. If SAFE becomes law, every single organization which supported it should be held accountable. That means PGP, Inc., if indeed PGP, Inc. supports this repressive legislation. Perhaps we should think about sending letters, summarizing the points we've been making here, to all of these organizations which signed the letter supporting SAFE. Perhaps many of the organizations spent little time thinking about the issues. Perhaps Roger Schlafly should speak to his mother (head of the Eagle Forum, as I recall) about the chilling effects of criminalizing crypto use when any of the tens of thousands of Big Brother-defined "prosecutable offenses" are allegedly committed. Perhaps we should ask Phil Zimmermann why PGP, Inc. supports this. Or perhaps he'll finally see the light when some of his anti-nuke friends are charged with using his own product, PGP, in connection with their "prosecutably offenses" at the Nevada Test Site! That would be ironic indeed. It's time we simply give up on Washington and its criminal actions. Washington is a cancer than can only be cured with a few megatons of nuclear disinfectant. --Tim May, still a felon, and soon to be even more of a felon if he uses crypto in support of his activities There's something wrong when I'm a felon under an increasing number of laws. ---------:---------:---------:---------:---------:---------:---------:---- Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay(a)got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Higher Power: 2^1398269 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

1 0

Re: TWO Letters on SAFE
by jimbell＠pacifier.com 17 Dec '03

17 Dec '03

At 12:07 5/02/97 -0400, Jonah Seiger wrote: >This is indeed true - CDT sent a separate letter and signed the IPC letter. > >As Marc Rotenberg knows very well, the legislative process is complicated. >Marc also knows that Goodlatte, Coble, Lofgren, and the other subcommittee >supporters of SAFE chose not to consider any amendments at the subcommittee >markup, but rather to take up our concerns at the full committee. You obviously meant, "ignore our concerns" rather than "take up our concerns," right?!? Because that is, more accurately, what they did and will do. >Goodlatte's staff and the other sponsors of SAFE had worked very hard to >prevent any amendments, hostile *and* friendly, at the subcommittee vote >for fear of hostile amendments from the Administration and opponents of >SAFE in the congress. A bill isn't worth much when it gets to the full >Committee if it gets gutted at the Subcommittee. Goodlatte wanted to get >some momentum behind it before going into the expected all out fight at the >full committee. I see a issue that's not being addressed. Presumably, in legislative negotiations the theory behind including such a provision (the criminal provision) is that there are people "out there" who would support it with that provision present, but who would oppose it with that provision removed. (Moreover, the theory would suggest that somebody believes that this bill would be MORE likely to pass with the provision present than absent.) Who, exactly, are these people? Aside from a few government thugs, I really don't see the groundswell. Numerically, I think I'm safe in concluding that there are in the citizenry, by far, more people who would support the SAFE bill only with the provision removed than those who'd support it only with the provision present. We should insist on a bill which reflects the desires of the public, NOT the government. One more thing. That idiot Clinton can easily veto this bill if he chooses to do so. The only way this bill is going to turn into law, I think, is if he _doesn't_ veto it, and that will mean that he prefers getting its "positives" (positive to him, at least), the criminal provision, and all those "legitimate needs of law enforcement", and tolerating its "negatives" (again, to him, at least.) He may publicly oppose it, but I think that position is a fraud. Jim Bell jimbell(a)pacifier.com

1 0

Building a Tempest Interceptor (fwd)
by Ray Arachelian 17 Dec '03

17 Dec '03

This was on the Surveillance List... by Frank Jones aka SpyKing(a)thecodex.com http://www.thecodex.com Copyright 1997 The is very little public knowledge about TEMPEST. Most infomation, including how to safeguard against monitoring, has been classified by the government. You can view an article I wrote on TEMPEST at: http://www.thecodex.com/c_tempest.html and another at: http://www.thecodex.com/rise.html In order to protect computers and CRTs from electromagnetic emission monitoring (interception, recontruction and viewing) it is necessary to use shielding. This shielding is available in several forms but that is another article... Once the shielding is installed, how do you test it for leakage? It is necessary to see if you can monitor the device yourself. Since TEMPEST monitoring equipment is normally very expensive and hard to come by, here is a simple way to construct a simple monitoring device that will help you detemine if your computer can be snooped on by outside sources... this device was designed for defensive purposes and should NOT be used offensively... it does not have the distance capability of my DataScan device but it WILL work as a laboratory benchtop demo device... Like Van Eck and Moller, I have deliberately flawed the article for the technically challenged... After all, we dont want everyone to have one of these, do we? TEMPEST Interceptor - Follow my instructions... You need to learn something about electronics to make this happen. The first lesson will be to construct a sync generator to drive the computer monitor without any signal or video. To start this you must take your spectrum analyzer, turn off every appliance in the area. Put the analyzer near the computer monitor, turn on the monitor but NOT the computer. Start searching around 15Khz. You should get a strong signal somewhere between 14Khz & 17Khz. Log that frequency. Next try looking around 30Hz to 60Hz. Which is stronger 30 or 60? Log that. Next were gonna build a computer sync generator (video gear will not work). Most technicians who have tried to build a TEMPEST interceptor have been trying to use off-the-shelf products. Don't believe what you read or hear, they won't work. Youve got to build a computer sync generator. It isn't really so hard to do. Get yourself a copy of the "CMOS Cookbook". It's around. In there it shows how to build a 555 timer. It is a small and simple circut. Maybe 5 components. The 555 has in it all the clock and electronics to make a timer from 1Hz to 10Mhz. Follow the instructions and make a variable timer (it is really simple). Make the first one run at the exact frequency of the monitor, say 16 Khz. Next make a seperate one at the second frequency say 30 Hz. Then tie them together with a .o1 Mfd cap. The output should drive your monitor (generator is done). Two minor things I forgot to mention. The output of the 555 is about 5 to 12 volts. You will need to attenuate it to about .3 volts. Several resistors will do. The other thing is that you need a negative pulse out of the 555. The normal output of a 555 is a square wave. On and off . You must vary the duty cycle to create a square wave which is most of the time on and the sync duration off. You now need a seperate dedicated computer monitor. Open the input plug and find the shielded cable inside. Attach the output of your circut to the center-conductor and the battery minus from your circut to the shield. You have a solid black screen on the monitor. Turn up the brightness. There should be no rolling and weaving. If so... adjust the pot in YOUR CIRCUIT not the monitor, to stop it. Next step. Collect the signal you hear and see on your anaylzer/receiver. Take the output of your reciever tuned to those frequencies. Connect the output through a .01 mfd cap to the output of the sync generator and viola you should start to see the monitors radiated signal on your monitor. Shielding your viewing monitor from the recieving monitor and small adjustments in your generator frequency should do it. In fact no adjustments should still do it. The antenna that you use is critical. Make sure you use the right one... ;-) ************************************************************************** Send Postings to: spyking(a)con2.com with "Post" in subject field with subject ************************************************************************** Subscribe to: spyking(a)con2.com In the subject field type: subscribe-surveillance list e-mail address **************************************************************************

1 0

chained stego & interfaces to remailers
by Adam Back 17 Dec '03

17 Dec '03

Introduction (Skip down to "Chained Stego" if this intro is obvious for you) Some users are in countries where crypto is illegal, and would still like the ability to communicate with others on topics which are considered illegal in their countries (an example being documenting civil rights abuses). For them using encryption is not an option unless they combine with steganography. Also in countries where using crypto is legal, people still would like to have the ability to discuss topics which are considered illegal in their country (an example exporting cryptographic code, perhaps) For these users they do have the option of using encryption. However for some serious users of crypto, they may wish to hide the fact that they are sending encrypted communications to anyone at all, as well as hiding _who_ they are communicating with via mixmaster remailers, and hiding the contents of their communications via the use of encryption. Using remailers and using encryption shows up under traffic analysis even in freer countries, and if you were using steganography, you would be above suspicion, apparently never having sent an encrypted message at all, or at least not having sent one during the time in question. A useful capability to address the needs of these classes of users would be to be able to send messages to a remailer by steganographically encoding them in publically published material, Examples being: messages posted to USENET, messages posted to mailing lists, material made available on web pages (text, images, audio files), CUSeeMe video streams, etc. Such a system would allow a person in a country where crypto is outlawed to participate anonymously in discussions in the remaining free parts of the world without appearing to be using cryptography. The simple approach is to set up stego to remailer gateways which scan newsgroups and mailing lists. (It might be a useful to integrate this function into the Eternity server, as it is already doing quite a bit of scanning of newsgroups, and scanning newsgroups is relatively expensive, so we may as well scan for multiple message types while we are scanning). So, to send an anonymous message, a person PGP encrypts their message, and manipulates the encrypted message to remove the statistical properties identifying it as a PGP message (by removing "-----BEGIN..." headers, the fixed PGP formatting inside that, and the bias caused by the use of RSA), PGP stealth version 2.x would be suitable for this purpose. Next a suitable steganographic encoding is required. A simple method would be perhaps an in-your-face approach, to include a randomly distributed, randomly changing .sig: == Use a random number, go to jail, so lock me up already. QGRjcy5leC5hYy51az4gKEhpZ2ggU2VjdXJpdHkpiQCVAgUQMli+7B98EdWB2LS9 AQFF0gQAjiAOPPCs7s0VCHoFI2IWMEcAeQInmnl2p+6rpsvIxjX1v3wBqqstgBu5 aCLY9Uns+iKjzcnt5DTj6NPhJ8EOlefwgHUssiBLTsw7tOvT9fQwcIXOE5ikGP7j RObTq3a2Vtz4/O/YgN0KQnWcqTDuadeP17cJ2bbaWJpZiGDyWGSJARUDBRAySOsk The stego remailer gateway has to be able to put back together the multiple parts in the correct sequence. The Date: field from the post would be enough information to allow this. Chained Stego However there is another problem: the remailer gateway knows the identity of the stego users. The stego users therefore have to trust the stego gateway not to tell the authorities in their country. They also have to trust that the gateway is not actually be run by their own countries censorship board. The price for misplacing trust and revealing your use of encryption may be death in some of these countries. It would be nice if we could chain the stego function, so that the amount of trust required of each element could be reduced, in a similar way to chaining through remailers acheives this. So (here we get to the actual content of this post, the above having been introduction), we could arrange that the stego user encrypts with a selection of stego gateways public keys. The steps to send to gateways A and then B are: 1. encrypt with B's public key, normalise output (with pgp stealth) 2. encrypt with A's public key, normalise output 3. encode normalised output from 2 into an innocuous looking USENET post The process by which the stego message is recovered is then more complex, and has higher overhead. Each stego gateway has to scan newsgroups for stego messages, and forward it's potential partially decrypted stego messages to the other stego gateways. Then each of those remailers decrypts, and if the result is more stego noise, again forward back to the other stego gateways. The problem here is when do we stop! Remember if we pick a message, and apply our stego decoding method to it we get something which looks indistinguishable to a stego message whether there exists a stego message or not. Also how much of USENET do we scan? The stego gateways will be sending a lot of partially decrypted potential stego messages between themselves. One potential solution to the infinite loop problem is to include with the message a note to the gateway stating how many times this has been decrypted. After a few hops it can be discarded. If the max hops is 2, and there are 3 gateways A, B and C. The messages created by the single encrypted message are: round 1: A, B, and C decrypt and forward to each other round 2: B, and C decrypt C discards the output; B forwards the message to the requested remailer (We presume that multiply encoding to the same remailer is not useful, if A can decode by denormalising and decrypting twice, you may just as well have addressed it to A normalised and encrypted once). B does not know which person's USENET post this came from. He does not know where it is going (to another USENET group via a chain of remailers, or to an indivdual) However B does know that remailer A knows who the message is from. Presume there are n remailers. For hops which go through h remailers, collusion is required by all h remailers to obtain the identity of the poster. If the attacker owns k of those remailers, that is k of the remailers are colluding to discover the identity of the stego user, then the if k < h the attacker will never discover the identity of any posters using chains of length h. If k >= h the attacker will discover the identity of a stego user only in cases where the chain includes all of the hops. (C(n,k) is choose k of n function computed with n!/((n-k)!k!) k!(n-h)! Prob = C(k,h)/C(n,h) = -------- n!(k-h)! Some sample values, n = 8 values of h = hops, and k = no of rogue gateways h= \k| hops\| 1 2 3 4 5 6 7 --------------------------------- 1 | .13 .25 .38 .50 .63 .75 .88 2 | .00 .04 .11 .21 .36 .54 .75 3 | .00 .00 .02 .07 .18 .36 .63 4 | .00 .00 .00 .01 .07 .21 .50 5 | .00 .00 .00 .00 .02 .11 .38 6 | .00 .00 .00 .00 .00 .04 .25 7 | .00 .00 .00 .00 .00 .00 .13 Now interestingly it seems to me that this figure is per stego message, so if you chose different gateway chains on each message your cumulative chance of being caught would be worse than if you just chose a chain and stuck to it. You're trying to minimise your cumulative chance of being identified. If you use a chain, and the jack boots don't kick down your door for a few weeks, you hope it is a good chain. Of course there is a chance that gateways could be taken over and your chain would then become a compromised chain. An additional danger if a gateway is compromised and it's key is revealed, the attacker can then go back through it's saved messages and look for messages which came from the newly acquired gateway, or went to it, to see how many more chains this compromises. It would seem like a good idea for the gateways to ensure forward secrecy by re-keying reasonably frequently. To do better than this, we could make the stego gateways to act like middlemen remailers. That is a gateway will always communicate with other gateways via a remailer. Now gateway B does not know which of gateways A or C decoded the message from the newsgroup. So he has to attack them all. Better still, the gateways can be completely hidden. They can use a message pool to communicate. This has higher overhead, but now the fact that gateway B knows either A or C decoded the message does not help much because gateway B can't physically locate A and C. An alternative to the suggestion of including a count of the number of decryptions that have taken place so far to avoid infinite loops, is to do the job probablistically. Say that each remailer says that it will only accept mesages where the low order 3 bits after decryption are "101". This rule doesn't apply to the first hop, which knows that the message hasn't circulated anyway, as it has just pulled it from the newsgroup. So the first gateway decodes, forwards to the other gateways. Each gateway decodes, and if the desired bit pattern comes out, sends the decrypted message to the other remailers. This way we place no restriction on the number of hops, avoid revealing who the first hop is explicitly, but ensure that the messages which are not stego at all die naturally, and don't propogate for ever between the gateways. Also the "101" is likely enough to come up quite by chance, and the number of actual stego messages is small compared to just ordinary USENET posts that seeing a "101" doesn't prove anything conclusive after 1 hop. However after several hops (if an attacker owns several of the gateways), the probability of getting multiple "101"s in a row points increasingly to a stego message, whereas if we relied on a header stating how many hops, no information would be leaked by the attacker handling multiple hops in the middle of the chain. So it's a trade off between lower overall overhead whilst allowing multiple hops, whilst increasing the susceptability of the system to leaked information to an attacker who owns multiple rogue gateways. However these are all very high overhead because they have DC net like overheads, you send a copy to everything to everyone always, so you have a perfect mix. You can trade off this security to obtain lower bandwidth. Here's one suggestion of how to do this. Obviously we can't include with each partial decryption instructions on who to remail it to (because if you found some instructions, that would reveal immediately that it was a stego message). However, we can interpret the random bit patterns as remailer instructions. Say there are n gateways, we can number the gateways consecutively and take the decrypted message mod n - 1 to produce the index number of a gateway to send to next. We can include a count of decryptions so far to reduce overhead. Another way to avoid infinite loops with this variation is to take the decrypted message mod ((n - 1) + j), anything over n-1 gets discarded. Messages naturally die out statistically at a speed determined by the ratio of j / n - 1, after h hops there is probability: (j / n - 1) ^ h that the message will get discarded. Provided the stego users account for a small amount of traffic, their use would likley be lost in the large pool of noise even if they stuck to a particular chain. Or they could themselves use normally distributed selections of random chains. However if they did this, their cumulative chance of sending through only rogue gateways on one post would increase. Interestingly this setup has different properties to mixmaster mixes, with this arrangement traffic analysis gives you nothing, as the inputs are hidden, and the traffic is perfectly normally distributed. (Presuming that the stego users create random chains also). Cover traffic between gateways adds nothing. You probably need cover traffic on output however, so that the gateways send constant, or at least evenly distributed amounts of traffic off into the remailer net, so that one persons posting can not be correlated with exits from the gateway network. Some delay in the gateway network would help also. Random delay is different to mixing in this instance because there is already lots of mixing going on, the point is to ensure that an exit from the gateway network is not observable at fixed time periods after posts from particular suspected stego users. If there are many users this should average out. Note also, that when a message does decode to a message, the gateway should generate a random replacement message and forward that to a random remailer with the same probability as for other non-decoding messages, otherwise receiving a message would skew the statistics. Given that the stego user has an interest in sticking to a chain, this is less secure than sending copies of partially decrypted potential messages to all gateways. Either he doesn't stick to his chain, thus increasing his odds of being detected, or he uses his fixed chain and runs the risk that this statistic shows through. If there are many stego users each with different fixed randonly chosen chains, the fact that chains are fixed would be less likely to reveal information as the random chains would tend to cancel out, and themselves have the same normal distribution as the non-stego traffic. Comments? More efficient ways to get around the weakness of a single remailer gateway hop? Adam -- Have *you* exported RSA today? --> http://www.dcs.ex.ac.uk/~aba/rsa/ print pack"C*",split/\D+/,`echo "16iII*o\U@{$/=$z;[(pop,pop,unpack"H*",<> )]}\EsMsKsN0[lN*1lK[d2%Sa2/d0<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<J]dsJxp"|dc`

1 0

Re: A new system for anonymity on the web
by Hal Finney 17 Dec '03

17 Dec '03

Adam Back, <aba(a)dcs.ex.ac.uk>, writes: > Hal Finney <hal(a)rain.org> writes: > > Theoretically, at least for long downloads, passing through multiple > > slow links shouldn't slow down the transmission, > > I'm not sure that this is so. > [...] > When Alice and Bob are online, consider that they are actually using > their bandwidth most of the time. I know I do; if I'm not using it I > hang up, with pay per second phone lines, you're likely to. Actually here in the U.S. most people have access to unmetered Internet access these days. Local and national ISP's are almost always that way, and AOL offers that option now as well. I find that I tend to browse in "flurries", paging around a bit, then settling down to read for a while. But you're right, I did not consider the interference among multiple paths running through a jondo. That issue applies to the higher speed links as well. If average path length is n, then on average there will be about n paths going through each jondo (assuming all "home" jondos have set up paths). So the question will be whether the average person uses more than 1/n of the bandwidth available during the time he is connected. This will no doubt depend on the pricing model for Internet service, as you suggest. > > just increase the latency. I don't have a very clear picture about > > how long it would take to snake in and out of a bunch of people's > > modem lines en route to the web server. > > I'm sure it's going to increase the latency too. My ping times are > 200ms from the PPP link alone (ie pinging the PPP server machine > itself). Add to this that the members of the jondo / crowds pool may > not be on the same ISP, and you've got the additional overheads of > whatever latency is added by the cumulative latency between each of > the hops in the chain. Yes, latency would be cumulative, and I just tested mine and found it was 160-220 ms, about the same as what you saw. So running through say 5 jondos at the end of modem paths would add about a second of latency. I think this would be fine if it only happened once per web page, but almost intolerable if it was once per tiny picture. > I wonder if you could improve the security of this by trading off > against some additional bandwidth consumption for the ISP. Say have > split the jondo in half, with two cooperating half-Jondos acting as a > single virtual jondo in such a way that someone with root access on > one machine but not the other, can't extract any useful information by > spying on that half of the jondo. This is an intriguing idea. Secure multiparty calculation protocols allow calculations to be made such that neither machine would have access to all the data. It is an interesting question how this could be applied to the anonymous communication problem. In this specific case, if the Crowds system were enhanced so that end-to-end encryption was used (which seems very practical and useful), you could run a jondo on your PC whose only function was to link to the jondo on the ISP and then set up the additional path through the jondo net. The path between your local PC jondo and the end one in the path would be encrypted, so even root on the local ISP could not see the contents of what you send down the jondo path. He might still be able to see when you were sending and when not, but he couldn't tell where it was going. Hal

2 1

SafePassage Web Proxy follow-up
by SafePassage Sales 17 Dec '03

17 Dec '03

Dear SafePassage Evaluator: It's been a while since you downloaded the evaluation version of SafePassage, and we just wanted to check in to see how things were going. If you have problems installing SafePassage, using SafePassage with your current browser, or any other questions about our SafePassage product line, please email: safepassage-sales(a)c2.net or safepassage(a)ukweb.com If applicable, please include some recent samples of your log files, and detailed information on your platform and the version of SafePassage you're using. If you have sales or ordering questions, please feel free to give us a call at 510 986 8770. Our hours are 10 am to 6 pm, Pacific Daylight Time. You can also contact UKWeb, our European partner and master distributor, at +44 113 222 0046. -- Join our Early Adopter Program: For the following applications, we are offering a flat-fee license of $45,000 per year, provided payment is received by May 15, 1997: (a) Banks, for distribution to their customers. (b) Internet Service Providers, for distribution to their customers. (c) Companies, for internal use.

1 0

Artikel about XS4ALL in New York Times (fwd)
by cpunks＠algebra.com 17 Dec '03

17 Dec '03

----- Forwarded message from tank ----- >From cpunks(a)manifold.algebra.com Fri May 2 05:04:57 1997 From: tank <tank(a)xs4all.nl> Message-Id: <199705020912.LAA11297(a)xs1.xs4all.nl> Subject: Artikel about XS4ALL in New York Times To: tank(a)xs4all.nl Date: Fri, 2 May 1997 11:12:37 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-cypherpunks(a)toad.com Precedence: bulk THE NEW YORK TIMES/CYBERTIMES www.nytimes.com April 29, 1997 By BRUNO GIUSSANI For an ISP in the Netherlands, Controversy Is an Old Friend AMSTERDAM -- April has meant business as usual at XS4ALL (access-for-all), the third-largest Internet service provider in the Netherlands. The month started off with the XS4ALL Web site being out of reach for a substantial portion of the European online population, and it is ending with the provider's adding a notch to its track record as a champion of freedom of speech. On April 11 the Dutch Web site was blocked by the German academic network, Deutsche Forschungsnetz (DFN), which serves about 400 universities and research organizations and provides Internet access to a half million people. DFN acted under pressure from the Federal Criminal Investigation Bureau pointing out the illegal -- in Germany, though not in the Netherlands -- content of Radikal Magazine, which is housed on the XS4ALL server. Radikal, a left-wing underground magazine, advocates "militant and armed interventions" to overthrow the government, and has published a "Short guide to hindering railway transports of all kinds" -- a handbook describing how to attack and damage tracks. Since selectively barring single home pages is technically impossible, the DNF action cut off all 6,000 pages on the XS4ALL servers, including those of Serbian opposition radio station B-92 and several scientific databases. Thus, while blocking illegal material, the German network was also hampering scientific work -- which DFN has been established to nurture. Not to mention that skilled Internet users could route around the obstacle by using a remailing system or a proxy server located abroad. What's more, Radikal Magazine can be found on several dozen "mirror sites" around the world. Ten days later -- after being flooded by protests -- the German network lifted the ban on XS4ALL. "An effective blockage of illegal information has not been within the realms of possibility," Klaus-Eckart Maass, a DFN spokesman, conceded in an interview with The Associated Press. Now, this happened the very same week as the indictment of a German manager of CompuServe, a leading international online service, for the transmission of allegedly illegal materials over the Internet and as the German Parliament opened discussions on a new multimedia law. The new bill would place the responsibility for content on the supplier of the data, thus Internet service providers would not be held liable for illegal information that could pass over their wires unless they have been alerted and "have the technical ability" to delete or block it -- the very same scheme the German academic network found impossible to enforce. It was not the first time XS4ALL had been at the forefront of an Internet skirmish. Last September, most German ISPs blocked XS4ALL for a month after complaints by a regional prosecutor about Radikal. (In January Angela Marquardt, a Bavarian socialist politician, was indicted for linking to the banned magazine from her personal home page). "After a couple of weeks, the censored information was mirrored on some 50 Web sites around the world and voluntarily removed by our user from the XS4ALL server," the company's founder, Felipe Rodriquez, 28, explained. "After the block had ended, our user put the documents back on his page." Along with German prosecutors, the Amsterdam-based provider has lined up a fair list of other adversaries: the McDonald's fast-food chain, the Serbian government and the Church of Scientology, just to name a few. XS4ALL's roots reach into the hacker movement. The venture started out in 1993 "to give anyone the possibility to access the Internet." At that time there were no commercial access providers in Holland. It has grown into a respectable and very successful business in less than four years, yet the principles on which it was created have not changed: "Internet for the masses" is still its motto. XS4ALL currently has 55 employees, boasts 21,000 subscribers and hosts some 6,000 home pages. "A few years ago, we would have been portrayed as a band of dangerous anarchists, bent on disrupting society," Rodriquez told the Dutch daily Trouw. "But now they have come to see that we are nice and quiet people really." The company was instrumental in the creation of the Amsterdam Digital City project, a community networking initiative backed by taxpayer's money, and Rodriquez himself played a key role in setting up the Dutch anti-child-pornography hotline, the first of its kind, last year. "Before we started the hotline, Holland had a reputation of being a kid-porn freehaven," he said in an interview last week. "We designed it as a non-censoring form of self-regulation." The hotline is run by Internet users and providers. Unlike Britain's Internet Watch Foundation, the Dutch hotline doesn't censor any information nor does it ask the provider to do so. Hotline operators contact the author of the information and ask him to remove the offending content. "If the author does not comply, we report him to the police, and he'll be prosecuted," said Rodriquez, who also is chairman of the Dutch Providers Association. "The Internet Watch Foundation forces the provider to remove the illegal content," he added. "This is a fundamentally different approach to responsibilities on the Internet. We think the author of the information is responsible for his own actions, not the provider." That's why XS4ALL didn't take any steps against the customer who posted Radikal Magazine on its server. "Our policy is that as a provider we are not in the position to judge whether this magazine is illegal in the Netherlands, therefore we do not interfere with our users' freedom of speech," Rodriquez stated. "If there is any doubt about the legality of the publication in Holland, a Dutch court of law would be the proper place to remove these doubts," he added. This was the case when, in September 1995, the Religious Technology Center -- better known as the Church of Scientology -- filed for the seizure of all the XS4ALL computer equipment "because one of our users had put on his home page some information to which Scientology said it owned the copyright." The document -- the now famous "Fishman affidavit" -- is the actual transcript of a testimony given by Steven Fishman in a Los Angeles Court in which he accused the church of having forced him to act illegally. "We denied any responsibility for the content on our users' pages; they decide for themselves what they will publish," Rodriquez recalled. "We won the litigation." The user preferred to take down the controversial document. Yet when violation of the law is flagrant, XS4ALL doesn't hesitate to comply, as it did a few weeks ago when it shut down a customer's home page called Neuroroom, which sold marijuana and other soft drugs in Holland and abroad. The company's commitment to support free expression and democratization of the Internet doesn't stop here. Last fall when the Serbian government censored radio station B-92, XS4ALL helped design an Internet campaign and started to carry news broadcasts (in RealAudio format) that kept the rare opposition voice alive and the international public informed through independent accounts of the events occurring during the mass demonstrations in Belgrade. "They gave us disk space, donated network traffic and helped in training people," said Frank Tiggelaar, a Dutch activist for democratization in former Yugoslavia. "Basically any project we like gets free resources from XS4ALL," Rodriquez commented. The campaign of Helen Steel and Dave Morris is among the projects XS4ALL's old hackers do like. The two British environmentalists are the main characters of a civil case that started in 1990 when the McDonald's restaurant chain sued them for distributing flyers pointing at what they called the company's economic and ecological "ravages." The trial is not over yet -- but it has spawned a large Internet-based support network and fed a huge anti-McDonald's Web site called McSpotlight hosted, not surprisingly, by XS4ALL. EUROBYTES is published weekly, on Tuesdays. Click here for a list of links to other columns in the series. Related Sites Following are links to the external Web sites mentioned in this article. These sites are not part of The New York Times on the Web, and The Times has no control over their content or availability. When you have finished visiting any of these sites, you will be able to return to this page by clicking on your Web browser's "Back" button or icon until this page reappears. Bruno Giussani at eurobytes(a)nytimes.com welcomes your comments and suggestions. Copyright 1997 The New York Times Company -- XS4ALL Internet BV - Felipe Rodriquez-Svensson - finger felipe(a)xs4all.nl for Managing Director - - pub pgp-key 1024/A07C02F9 ----- End of forwarded message from tank -----

1 0

Re: FCPUNX:Layoffs at PGP
by jamesd＠echeque.com 17 Dec '03

17 Dec '03

> [arrow] PGP Pretty Good on Privacy of Layoff Details > by Kristi Coale > > [...] > > > "Internet Fast-Forward is not as closely related to the tools we > develop for individuals for privacy," he said. And the fact that > Fast-Forward filters ads which are an "inimical part of some of > the emerging Web commerce models ... we don't want to hinder this > development." I think this tells us everything we need to know about the management of PGP inc. --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.jim.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the state. | jamesd(a)echeque.com

1 0

Cyberstrike
by harka＠nycmetro.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- May be of interest... In> From: cleacy(a)execulink.com (Chris Leacy) In> Newsgroups: comp.infosystems.www.announce In> Subject: SOFTWARE: CyberStrike 1.13a Ad filtering/Cookie cutting Proxy In> Date: 30 Apr 1997 23:28:43 -0400 In> CyberStrike: Version 1.13a First Public Release In> The first public alpha of CyberStrike was released today. In> It can be found at <URL:http://www.execulink.com/~cleacy> In> CyberStrike is a banner ad filtering, cookie cutting, HTTP In> header overriding proxy of a program. In> Whether for privacy concerns, or just to speed up your In> browsing, CyberStrike performs admirably well at it's task. In> Chris Ciao Harka /*************************************************************/ /* This user supports FREE SPEECH ONLINE ...more info at */ /* and PRIVATE ONLINE COMMUNICATIONS! -> http://www.epic.org */ /* E-mail: harka(at)nycmetro.com (PGP-encrypted mail pref'd) */ /* PGP public key available upon request. [KeyID: 04174301] */ /* F-print: FD E4 F8 6D C1 6A 44 F5 28 9C 40 6E B8 94 78 E8 */ /*<<<<<<<<<<<<<<<<<<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>*/ /* May there be peace in this world, may all anger dissolve */ /* and may all living beings find the way to happiness... */ /*************************************************************/ ___ Blue Wave/386 v2.30 [NR] -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQEVAgUBM2mQeTltEBIEF0MBAQEbUAf+NQOMPrml38ijLwv8MIPSkTUbtbNQebJy R4LaeYhXI6Z/OzJdX6AQC2GisxMyPiz3QWWBoAqxGkunWDHbQzDfZVWW6nZa4zLY do2r11CEWFPba1igX+8JtCm9gd11S8Zlff5Pw5eTYOMcsVSHf6Z/+iAT5aqnIdcb YU26JswL1MofhWlNhbNXYc/9wQ3uUn1hU3lin+Lujfa5+cXXlTyfz6eFcJlwD7ZY KdK7UfPMuCeZnUX/VJqBdJ1idXfC5YOlwVMBNIyQK7PgCy59FvP5CiiXhpui7uSj WijKBSMUsz2HuQIuQJf6zvW/bi5DGWr6PerTXnuVaUkjNHr4Fp13Mg== =5TtH -----END PGP SIGNATURE----- If encryption is outlawed, only outlaws will have encryption...

1 0

Re: Bombs Away, from The Netly News Network
by azur＠netcom.com 17 Dec '03

17 Dec '03

At 3:50 PM 5/1/97, Jeff Barber wrote: >Declan McCullagh writes: >> "A >> member of the DoJ committee accessed a single web site >> on the World Wide Web and obtained the titles of 110 >> different bomb-making texts." > >I wonder if this was a bookstore. I did a search for "explosives" >on amazon.com and came up with at least that many titles. Only a >handful were of the Anarchist's Cookbook ilk. Most were academic or >industrial studies; a few were from US Govt or Govt-affiliated >organizations. Probably Paladin Press. Publisher of such notables, as Hit Man: A technical manual for independent contractors, a semi-fictional work with lots of details (written by an anonymous woman, BTW). See http://www.paladin-press.com PGP mail preferred Fingerprint: FE 90 1A 95 9D EA 8D 61 81 2E CC A9 A4 4A FB A9 Key available on BAL server, http://www-swiss.ai.mit.edu/~bal/pks-toplev.html --------------------------------------------------------------------- Steve Schear | tel: (702) 658-2654 CEO | fax: (702) 658-2673 First ECache Corporation | 7075 West Gowan Road | Suite 2148 | Las Vegas, NV 89129 | Internet: azur(a)netcom.com --------------------------------------------------------------------- "I know not what instruments others may use, but as for me, give me Ecache or give me debt."

1 0