December 2003 - cypherpunks-legacy

Re: Hackers invade DOJ web site
by Nick West 17 Dec '03

17 Dec '03

At 06:46 PM 8/18/96 -0700, you wrote: >> CNN online (http://www.cnn.com) is reporting a "breakin" by "hackers" >> to the DOJ web site. They don't elaborate exactly how thy did it. > >Did anyone save a copy of the altered web page? I would like to see >what the crackers did. > >Thanks. > > They showed a the page on CNN Headline News yesterday. They might show it sometime again today, stay tuned. Nick West nick(a)multipro.com Member of the National Wild Turkey Federation and The Libertarian Party of Tennessee http://members.tripod.com/~NWest/index.html PGP Fingerprint= F9 F7 92 D9 D3 0B 56 3E FA 2A 78 59 27 32 7D 6F Public key available on request.

3 2

Re: Billy boy's satellites [Was - Floating DataHaven]
by tcmay＠got.net 17 Dec '03

17 Dec '03

At 12:16 AM 8/19/96, Vipul Ved Prakash wrote: >> >> Vincent Cate wrote: >> Doesn't the low orbit part mean that the satellites will have a low life >> expectancy, meaning a new launch every couple of days? (where's Gerald >> Bull when you need him - oh yes, now I remember ...) LEO (Low Earth Orbit) is only relative to the main alternative placement for broadcast satellites: geostationary orbit, out in the Clarke Belt. LEO satellites are well above significant drag of the troposphere and are expected to remain in orbit far longer than their lifetimes as useful transponders. >The Iridium Project was pretty neat in this respect. I don't know what >is happenning with it as of now. It had some 230+ satellites, and some >smart routing techniques, if a satellite goes down. Actually, far fewer. The original number of satellites (not counting spares) was equal to the atomic number of iridium, hence the name (though maybe they altered the number of satellites to match the atomic number?). They later reduced the needed number by a few, but kept the name. An Alta Vista search on "iridium AND satellite" produces 1000 hits, so there's plent of information out there. There are also two major competing systems, also not yet deployed. One is the Microsoft-McCaw Cellular project, another is being done by an aerospace company working with Qualcomm, or a subsidiary. Again, the Web should produce the information for anyone interested. Which of the three (and maybe more) systems will actually get deployed, and which will succeed in the market, is an interesting question. There's some obvious crypto/GAK/New World Order issues: many countries may not care for a communications system which allows citizen-units or enemy agents to make phone calls from the middle of the Kalahari desert or from within the jungles of Burma. (The Israelis, for example, are insisting that all handheld units have Explosives Escrow, for detonation with appropriate official orders.) --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(a)got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Licensed Ontologist | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

2 1

Re:Why BlackNet *IS* a Data Haven
by tcmay＠got.net 17 Dec '03

17 Dec '03

At 8:28 AM 8/19/96, Jim McCoy wrote: >Tim May writes: >> Without splitting too many semantic hairs about the precise definition of >> "data haven," let me examine some ways in which BlackNet behaves >> identically to a conventional data haven. > >I would be willing to concede the point if you would take a few seconds >to examine the issue of complete lack of persistence in BlackNet. There are various amounts of persistence, depending on which kind of "pool" is used: * Usenet newsgroups have a persistence of days to weeks, depending on one's particular site. This is enough for parties to find each other _if_ they are watching (or having agents watching). And of course Usenet archives are expected--Alta Vista and Deja News has Usenet articles going back several months in many cases. * Mailing list message pools (a la Cuperman) have an effectively infinite persistence, if parties archive the messages. * Web-based message pools, which to my knowledge have not been used yet, would have a persistence as long as the messages are archived...which could be very long. Rather than _persistence_, I'd say _access time_ or _latency_ is a more serious criticism of my approach. In contrast to a large library or bookstore, with access times of ~minutes, the access times for material on BlackNet may be ~days to ~weeks. Or, of course, "never." >Your descriptions of BlackNet as a data haven seem to be completely >based upon the presumption that an anonymous contact service and contract >exchange is the functional equivalent to a data haven. Here are a few >reasons why I would disagree: > >As a publisher of "naughty bits" I do not have the ability to just toss >data up and assume that it will be there when someone wants it. I am forced >to continuously monitor the appropriate newsgroups to find messages from >people asking me to post the blueprints to the orbital mind control lasers >or kiddie porn. I cannot put my data onto "the Net" with an expectation >that any arbitrary user will be able to get the bits one month later. To >maintain persistence I need to constantly repost my data, making it easier >for authorities to trace me through simple taffic analysis if nothing else. I grant that my version is not like a "Mega-Barnes-and-Noble" bookstore, with vast amounts of stuff available for browsing. It is more like a "stacks"-based library. (Many universities have lesser-used books, sometimes _all_ books, in "stacks," accessible only by qualified librarians, or by permission.) >> The classical data haven is closely identified with "place." To many >> people, they naturally assume "data haven" = a haven for data, a "harbor" >> (same IE root as haven) = a physical place. >> >> But is "place" important? > >No one has really claimed place is important, in fact the ideal data haven >has no physical existence whatsoever. This is a given. As a practical >matter it is a lot easier if you start off in a "place" because there are >fewer complications but this has never been a necessity. Some have focussed on "place," by focussing on ideas like offshore buoys, orbiting broadcasters, and, of course, on finding accomodating jurisdictions that will tolerate data havens in their midst. You and I may agree that data havens need to be virtual. I'm just proposing an architecture--and certainly not the final version!--that actually works. Others can suggest their own alternatives, or build on versions that are out there. >> A person in the U.S. seeking the Necronomicon posts a message to BlackNet >> (or any similar forum, using the same methods) asking for a copy of it, or >> offering to pay for it. (Whether the information is free or for a fee is >> not central to the idea.) This request is, of course, untraceable. >> >> Anyone, anywhere in the world, with a copy of this banned material on his >> or her private machines may see this request and respond, either giving the >> material away, or negotiating a fee. (As I said before, the absence of a >> robust digital cash system, bidirectionally untraceable, is a known >> limitation of all such systems.) > >Now you reveal the objection I had to BlackNet being a data haven. What if >only one person has a copy of this banned material? It may not be in this >publishers interest to have the data available to anyone for posting in >response to the query ("Information does not want to be free, it wants to >be expensive and liberated...") and some data is not widespread enough or of >interest to enough people to assume that multiple copies exist to those who >read BlackNet postings. Therefore the only way for a publisher to maintain >availability of their data is to constantly monitor the appropriate newsgroups >and republish for each request, persistence is maintained only through >eternal vigilence (much like liberty, only requiring a lot more effort :) What if only one person has a copy? Nothing in *my* concept of data haven says anything about information being free, or cheap, or widely available. Books aren't usually free, of course, so why should information be free The issue of what happens if someone buys an item and then republishes or redistributes it is, of course, orthogonal to the discussion here. It's a tough problem, and the basis of the "information wants to be free" aphorism. But not a key issue for differentiating forms of data havens, as I see things. >> It's a data haven. > >No, it is an anonymous contact service. To claim this is a data haven is >like claiming that the classified ads in a newspaper are the equivalent to >a mall; you could probably find the same goods if you looked long enough, but >there is a reason that manufacturers sell goods through stores rather than >just posting classified ads across the country. When one does not have the >time to check the classified ads, wants to goods from a reputable source, and >wants the goods in a timely fashion they will go to a shopping center. I agree with the "classified ads vs. shopping mall" distinction. In fact, it neatly summarizes the latency/availability issues. So be it. Until the "shopping mall" exists, I'll settle for the "classified ads." If BlackNet and similar systems function adequately as an anonymous classified ad system, this will be an improvement over what exists now, which is, of course, "nothing." --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(a)got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Licensed Ontologist | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

1 0

Indonesia detains democracy activist after post to mailing list (fwd)
by gimonca＠skypoint.com 17 Dec '03

17 Dec '03

Forwarded message: > Date: Mon, 19 Aug 1996 07:05:33 -0700 (PDT) > From: Declan McCullagh <declan(a)eff.org> > To: cypherpunks(a)toad.com > Subject: Indonesia detains democracy activist after post to mailing list > HotWired, The Netizen > 19 August 1996 > > Trouble in Paradise > by Declan McCullagh (declan(a)well.com) > Washington, DC, 18 August > > Indonesian democracy activists have taken their fight for freedom > to the Net, and the government doesn't approve. > > After distributing email messages about riots in Jakarta last month to > an international Indonesian-politics mailing list, Prihadi Beny > Waluyo, a lecturer at Duta Wacana Christian University, was arrested > and interrogated by the military. Since then, the mailing list has > been banned from the country and Waluyo has returned to his house, > where he remains under surveillance. > [dan lain-lain...] Exactly which mailing list was "banned from the country"? All the Indonesian mailing lists I'm on, including apakabar(a)clark.net, are functioning normally, with no unusual complaints or interruptions.

3 2

Re: Why BlackNet *IS* a Data Haven
by tcmay＠got.net 17 Dec '03

17 Dec '03

At 2:03 PM 8/19/96, Hal wrote: >mccoy(a)communities.com (Jim McCoy) writes: >>Your descriptions of BlackNet as a data haven seem to be completely >>based upon the presumption that an anonymous contact service and contract >>exchange is the functional equivalent to a data haven. Here are a few >>reasons why I would disagree: > >I think part of the confusion here is the name "BlackNet". As I envision >the concept, BlackNet is not really an anonymous contact service, or in >fact a network of any sort. Rather, it is a vendor. It buys and sells >information. The name, while provocative, is a bit misleading in this >regard. (This is just my model, and may not actually correspond with >Tim's or anyone else's idea. But I think it more closely matches the >data haven concept, and in fact is more consistent with the original >announcement.) Hal is right that BlackNet was presented as a _specific vendor_. But the announcement also described how _anybody_ could set themselves up in the same way, by sending out a similar announcement with their own public key. My purpose was to: -- demonstrate the coming future -- use a concrete example, "BlackNet," to drive home the point (people often pooh-pooh an abstract idea until a concrete, reified version is produced) -- stimulate debate about anonymous information markets I'll comment on a few of Hal's points, but, to save space, will not comment on most of them. >BlackNet has a public key, and a known virtual location in the form of >certain newsgroups that it monitors. Anyone can initiate a >communication interchange with BlackNet by posting a message to those >groups, encrypted with BlackNet's key. Presumably in that message will >be included return address information in the form of a key and a set of >locations that will be monitored for replies. In this way ongoing >conversations can be maintained between BlackNet and customers who are >either buying or selling to it. Exactly. And this is how it was used for several two-way communications, back in Sept-Nov '93. An article on the cover of "Information Week" last summer triggered new interest, and a couple of new messages addressed to "BlackNet"--the ones I tried to read apparently used a spoofed key, or the one Dettweiler created and placed ahead of mine on the MIT keyserver (the shorter key that the MIT group was able to eventually break). >BlackNet would not be used (as I see it) for direct communication >between buyers and sellers of information. How would the BlackNet >public key fit into this model? The existence of a specific BlackNet >public key is part of what drives me to picture it as a vendor. >Rather, BlackNet will buy information (plus unrestricted rights to >disseminate that information), add it to its catalog, and then >advertise its availability and price. It could be used for direct communications, via pools, a la the classified ad analogy that Jim McCoy just used (and that, in fairness, I used to describe it in a talk at Hackers several years ago). There is "BlackNet the company" and "BlackNet the abstraction." Inasmuch as the recipe is easy to duplicate, "BlackNet the company" would face heavy competition. ....[stuff elided]... >This is a little different from my picture of BlackNet, as I wrote above. >I would see BlackNet as being a particular seller of information, who >will respond to this message. It could have competitors like SafeHaven, >StrongHold, InfoBase, etc., each of which will offer data for a price, >and each of which will have its own reputation for reliability. Exactly. Hal quoting Jim: >>Now you reveal the objection I had to BlackNet being a data haven. What if >>only one person has a copy of this banned material? It may not be in this >>publishers interest to have the data available to anyone for posting in >>response to the query ("Information does not want to be free, it wants to ... >Here is where BlackNet as an information middleman makes the most sense. >Its business model includes the costs of this sort of vigilance, which >after all can be automated. All sorts of automated vigilance can be done: scripts that scan newsgroups and message pools, even Alta Vista-type spider searches, agents, etc. Depending on the type of message pool, whether Usenet newsgroup, mailing list, Web site, etc., various kinds of automation are possible. Which will prove popular of course depends on a lot of factors. ...[more elided]... >Actually we now have "virtual malls" online. These are in their infancy >but eventually they could become as easy to use and reliable as regular >malls (for appropriate kinds of goods). All that BlackNet (as I picture >it) lacks is a WWW interface, and even that could be provided if the >gateway server could be made immune to legal pressure and if various >technicalities about anonymous WWW connections could be dealt with. Agreed. >As for reputations, if BlackNet is one of several vendors of >information, like its competitors, they can all develop reputations of >their own for reliability, honesty, availability, etc. There may be >problems if the testimonials of customers are all anonymous, but in >some cases such methods as signed transcripts of information exchanges >can be used by one side or the other to justify claims that the other >side has cheated. Again, agreed. Evolutionary learning will take place, reputations will be strengthened and weakened, as always. Certainly some fraud will occur, as in all markets. The point being that this information market will be anarchic, in that no government or official hierarchy will rule on legality of data. Various access mechanisms will be tried. The Usenet and mailing list pools are somewhat slow, but have already worked for things like "I'd like a copy of the Church of Scientology secret documents." (In fact, in addition to "alt.religion.scientology," for discussions and requests, there is the newsgroup "alt.binaries.scientology," for posting anonymized copies of restricted documents.) That the latency is not as low as some other markets seems to be more a function of nascency (is this a word? it should be) than intrinsic limitations. ---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(a)got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Licensed Ontologist | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

2 1

lambda 2.10
by jt＠freenix.fr 17 Dec '03

17 Dec '03

netizen's --> Lambda Bulletin 2.10 <-- contents flash bulletin + French Constitution censors Internet control + Singapore and the Censorship Proxy Server + G7 and the EC take strong steps for Key Escrow Encryption * * * * * French Constitution censors Internet control As we speculated in our last bulletin (2.09), the French Conseil Constitutionnel, watchdog of the 1958 Constitution, censored 2 articles in a new telecom act which were intended to establish a kind of administrative control over Internet speech and online services. The nine "Sages" said that the creation of the Conseil Superieur de la Telematique, which would have make guidelines on Internet content, breaches article 34 of the Constitution which states that the Parliament can act alone to dictate rules concerning "civic rights and fundamentals garanties given to citizens for a fair exercice of public liberties". The CST could have undermine these principles, because the law didn't specify clearly how it would have taken its decisions. So the Conseil recognized the particular state of the Internet, which is not a basic medium. Only one section remains in the censored law : it obliges ISPs to give their clients "technical means" to forbid or select access to online services, software that allows a so-called "parental control". * * * * * Singapore and the Censorship Proxy Server A communique from SingNet, Singapore's main Internet Service Provider, states that "ALL SingNet customers will be required to connect to SingNet's proxy server by the deadline September 14th 1996, failing which you would not be able to access the web without the proxy. This applies to all dial-up, ISDN, and leased line customers (STIX customers are exempted). " The proxy server will ease the ISP to "Deny access to blacklisted sites supplied by SBA". SingNet said that "Access to sites banned by the SBA will prompt the message : The site you requested is not accessible For more information on Singapore's Internet regulation, please check http://www.gov.sg/sba/netreg/regrel.htm". On July 11 the Singapore government passed the Singapore Broadcasting Authority Act (Chapter 297) in which it announced a "Class Licence Scheme" aimed "to encourage responsible use of the Internet while facilitating its healthy development in Singapore. It encourages minimum standards in cyberspace and seeks to protect Net users, particularly the young, against the broadcast of unlawful or objectionable materials. ... SBA will focus on content which may undermine public morals, political stability and religious harmony of Singapore. However, SBA recognises that it is impossible to regulate the Internet fully. ... Singaporeans can help SBA in the identification of objectionable sites in order to keep cyberspace clean. SBA welcomes public feedback on objectionable content found on the Internet. Members of the public can write to SBA, call its toll-free hotline ... or post their views on the SBA homepage at http://www.gov.sg/sba. " China, which has created its own Internet regulations aimed at controlic data traffic and urged netizens to declare themselves to the authorities, approved the Singapore Act and an official was quoted as saying, "China has a lot to learn from Singapore's experience" (source : Fight-censorship mailing list). * * * * * G7 and the EC take strong steps for Key Escrow Encryption The European Commission's DG-13 division on information security (Infosec) opened on July 30th a "call for tenders" for "preparatory works" towards regulating encryption procedures. The plan is aimed to test the introduction of the Europe-wide network of Trusted Third Party Services (ETS). Observers saw in this move the so-called "guidelines" the EC was to propose last year, when press reports (Nature, Sept. 28, 1995) argued the Commission and the Council of Europe in Strasbourg were willing to regulate encryption use through the creation of TTPs. The Infosec call for tenders, which will end by September 30, is to "identify, define and verify ... operationnal, technical, regulatory and legal aspects ... to assess the effectiveness, economics and acceptability of Trusted Third Party Services." Other voices in European talks, however, said these "preparatory works" are to push for EU countries to adopt TTPs and the principles of key-escrow encryption. Nordic countries such as Finland, Denmark and Sweden, are said to be opposed to change encryption legislation, as France and Britain took steps in June and July to enforce the creation of TTPs in their own country. On July 30 G7 countries agreed policies that would "accelerate consultations on encryption that allows, when necessary, lawful government access to data and communications in order to prevent or investigate acts of terrorism, while protecting the privacy of legitimate communications". The EPIC, in Washington, DC, said "stronger measures sought by the US to restrict information on the Internet and limit the availabilioty of encryption were apparently not adopted by the G7 countries". Among other industrialised nations, Japan and Australia are said, like Nordic countries in Europe, to oppose key escrow as a mean to regulate the free flow of information. Remember the OECD talks in June, were the US tried to impose key-escrow legislation to the 27-countries' club of the industrial world (see lambda 2.09) ***** Soon archived on www.freenix.fr/netizen ----- Jerome Thorel =-= Journaliste/Free-lance Reporter =-= Paris, France =+= the lambda bulletin --> http://www.freenix.fr/netizen =+=

1 0

Agents, Spiders, Linda, and BlackNet
by tcmay＠got.net 17 Dec '03

17 Dec '03

Key Phrases: agents, spiders, linda, blacknet, indexing, hierarchy, alternatives to hierarchy, emergent indexing, aptical foddering, tuple spaces, virtual communities, shared environments A swirl of terms, related in some interesting ways. This mini-essay is inspired by the debate whether BlackNet is or is not a meaningful instantiation of a data haven. How can information be retrieved from the Net? Is organization needed? Who does the organizing? As the ARPANet evolved, under various names, and as UUCP and other machine-to-machine protocols evolved, the Usenet came into being. The first "message pool." A basic classification existed, mostly of fairly reputable top-level topics (sci, soc, comp, etc.). Then came "alt," largely created by our own John Gilmore. There are of course now more than 20,000 newsgroups. Searches and greps of the newsgroup list are a way to find potentially relevant newsgroups for posting a message or finding messages of interest. (As is well-known, the Usenet Cabal gets its orders from the Bilderbergers as to which newsgroups fit in with New World Order sanctioned epistemologies.) Some are saying there is an alternate method. With the advent of search engines which can index messages on the Usenet (and in Webspace, but the idea is the same), why not this alternative: put your message in a bottle and just throw it into the "sea" of possible messages. Let search engines find the messages of interest (modulo a day or two of latency, as the spiders reach the space where the message was placed). No newsgroups needed. The "keywords" list at the beginning of this message would help the search process, though of course the body of the message should have sufficient keywords; a formal keyword list or field serves mainly to remind the author to add some keywords (Schelling points) that he might not have included in his message per se. Conversations and threads would take place in a virtual meeting place, even more so than today. This is of course largely happening already, and even more clearly with mailing lists which get cc:ed to other mailing lists, e.g, the way the e-spam list forwards some of our stuff to their list and then replies pull in the orginal author. (The connection with "Linda" is the connection with David Gelernter's "Linda" system, based on "tuple spaces" into which messages are placed. A kind of sea of messages in this tuple space. The connection with the Unabomber is left as an exercise for the reader.) Sometimes hierarchy is useful. Library call numbers and indices make finding books easier than searching at random; however, sufficiently fast "library crawlers" could find even randomly-placed books. (And friends of mine are working on small RF "localizers" which, if small enough, could be placed on books. One could type in "Find "Robinson Crusoe," and a book anywhere in the library could chip "Here I am." Obviously the problem is more easily solvable for data.) With the rise of more powerful search engines, of distributed geodesic networks, and with the decentralization of naming power, I see Linda-type seas of objects as more and more attractive. This helps BlackNet-type information markets and virtual data havens. Just some ideas. Nothing new, to me at least. But I thought some of the newer list members might not have seen some of these ideas, part of the assumed culture to we crypto anarchists. --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(a)got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Licensed Ontologist | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

1 0

The "Best" as the Enemy of the "Pretty Good"
by tcmay＠got.net 17 Dec '03

17 Dec '03

Pretty Good Privacy, Pretty Good Remailers, Pretty Good Digital Cash, Pretty Good Data Haven, ..... One of the main lessons of evolutioary learning theory (aka evolutionary game theory, market learning, Darwinian selection, and related variants) is that getting something out early is often more important than getting it "right." Rigor is important, but, interestingly, rigor is often best-established in an evolutionary learning environment. (We build machines and buildings not based on first doing exhaustive analyses for centuries, but on building actual instances and learning from mistakes...bridges that fail, buildings that collapse, planes that crash.) A recent example of this is the Xanadu project, whose members worked for many years (and spent something like $7 million) to get all the long-range, rigorous details of hypertext "right"...and were then "scooped" by Tim Berners-Lee with his simple and straightforward HTML/URL approach. (I am not basing this analysis on the hatchet jobs done in the press on Ted Nelson and the other Xanafolks, but on personal contacts with many of them, including an identical analysis from Mark Miller at an Extropaganza this past Saturday.) Another example is that of remailers. There is no denying that "DC-Nets" are a more elegant approach than "mixes," but mixes (remailers) can be easily implemented in Perl and deployed rapidly, while I know of not a single, actual, operational DC-Net. And of course we cannot forget Phil Zimmermann's "Pretty Good Privacy." Had a "pretty good" version not come out, where would we be today? (And "pretty good" does not mean PGP is weak or has been broken.) Just a reminder that often the best is the enemy of the good. --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(a)got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Licensed Ontologist | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

1 0

Naked woman decapitates man on Internet!!!
by Rich Graves 17 Dec '03

17 Dec '03

You just gotta love the headline. Anyone have the URL? Seen on Newspage headlines; I assume the full text is on ClariNet. NAKED WOMAN DECAPITATES MAN ON INTERNET - Pictures of a naked young woman decapitating a man with a saw are available on the Internet, to the consternation of police in Munich who say they are unable to do anything about it. [AGENCE FRANCE PRESSE, 175 words] -rich who thought the associated press was bad

8 8

search engine improvement
by bryce＠digicash.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Keywords: distributed ratings systems, search engines, spiders, spiderspace, idea futures, The Shockwave Rider, John Brunner You know there is a trick that might greatly improve the effectiveness of a search engine at almost no cost to the end user. It is the well-known heuristic of "If Person A likes X and Y, and Person B likes X, then Person B probably likes Y.", combined with passive polling (which is getting information about people's opinions just by watching their actions, instead of by asking them). A first simple implementation would keep a table of the pages that people choose, keyed from the query that they originally submitted. Those pages that people choose most frequently from the list of matching pages (and/or those pages that people "stop" on-- that they do _not_ follow by further searching), would get bumped up a little in the list. This would be massively expensive in networking, storage, and computation, giving those hi-tech Alpha clusters at AltaVista something to do... :^) There are plenty of extras and refinements that could be added (for example, put some keywords identifying your "affiliation" in a separate field. It will only consider the results from other people who entered the same affiliation keywords when weighting your search results.). And there are some good topics for further discussion, such as is it worthwhile to distinguish between "relevancy" and "value"? I don't have a comprehensive list of people who are already working on this area (distributed ratings) (if I did, I might have Cc:'ed them), but I know that many people are. I hope that they and the search engine people get together and make cool stuff soon. There is the interesting issue of whether this will cause self-reinforcing "degeneration", where people (or an "affiliation"-keyed group of people) accidentally overlook a worthy page early in the game, and then, using each other's behavior to influence their own, reinforce that mistake. As a final attribution note: John Brunner thought of this idea idea in his prophetic novel _The Shockwave Rider_ in the 75. There is a wonderful line which I can't find right now, about how it turned out to be a flywheel instead of an oracle, merely aggregating human mistakes and successes. Regards, Bryce P.S. ObCryptoRelevance: Um... you could get paid for your ratings using Chaumian ecash, and even have your ratings popped into the right "affiliation" using Chaumian credentials... P.P.S. CryptoRelevance isn't very Ob anymore, is it? Just as well, IMESHO. -----BEGIN PGP SIGNATURE----- Version: 2.6.2i Comment: Auto-signed under Unix with 'BAP' Easy-PGP v1.1b2 iQB1AwUBMhjA+EjbHy8sKZitAQGpIQMAyDcdHUgK9/KhNskvUG8AAbourl1Hg6J5 ZIzo7aTnDq3ZGN9RnqKRkBRRmk4hjN1rFFWvQUYtA3XQQl85scE2XVGG/oURBoTW EU4WwB2oMSsAVGkYHn02B4gFn8gO6hmA =tZn3 -----END PGP SIGNATURE-----

4 3