December 2003 - cypherpunks-legacy

Re: Neural Nets
by geeman＠best.com 17 Dec '03

17 Dec '03

This comes up once in a while --- it appears an inappropriate approach, they say, since the solution space for the problem consists of exactly one spike, in the vast sea of all possible solutions ... there is no smooth contour over which to minimize the net's error function, and finding the one spike which is the correct result is no more efficient in such a case than any other exhaustive search. This argument would break down if there were detectable biases in the crypto algorithm that you could exploit. But then whether a nn would be the tool of choice in such a case may be uncertain. At 01:00 PM 12/12/96 GMT, you wrote: >Hi Folkd > >Just sat through a lecture on an itroduction to Neural Nets. > >Thought. Does anybody know of any tools that use Neural Nets to break >ciphers? > >Cheers > >Ste > >

2 1

Re: Java DES breaker?
by dlv＠bwalk.dm.com 17 Dec '03

17 Dec '03

More than one person writes: > Vulis wrote: > > "Mark M." <markm(a)voicenet.com> writes: > > > compiles Java to a stand-alone executable. Details at > > > http://webhackers.cygnus.com/webhackers/projects/java.html . > > It would be very foolish to touch any shit that comes out of Cygnus. > > Why? (specifically, I am about to try using a GCC port to WinNT, and > I would like to know _why_ you think their work is shit). First, because it's King John "Lackbrain" Gilmore's company. :-) Second, because they hire unqualified people (rather, people whose qualifications have nothing to do with the job) and they've already fucked up every project they've ever touched. Third, because they steal from the people who originally developed the free software by charging money for the non-existent "support". I'm quite happy with MS VC++ for NT. I doubt that Cygnus can do a good job porting gcc to NT. People much better than the assholes at Cygnus have not done a great job porting gcc to OS/2. If I want gcc, I just run it on the Linux box, or a Sun box. Fuck Cygnus. --- Dr.Dimitri Vulis KOTM Brighton Beach Boardwalk BBS, Forest Hills, N.Y.: +1-718-261-2013, 14.4Kbps

4 6

Puffer 2.1
by KDBriggs1＠aol.com 17 Dec '03

17 Dec '03

Puffer 2.1 is now available from http://execpc.com/~kbriggs Puffer is a shareware file and e-mail encryption program for Windows in 16-bit and 32-bit editions. Features 40-bit RC4 encryption in exportable shareware and international versions plus 160-bit Blowfish encryption in US/Canada registered version. A multi-pass data wipe feature is also available for files, file slack, and empty disk space. New for version 2.1: program settings saved in INI file, selectable decryption directory, improved ASCII PUF split option, fixed Windows 95 cache flush bug, other minor bug fixes, new uninstall utility. A patch program is also available from the web site for registered users to self-upgrade from version 2.0. Kent Briggs Briggs Softworks, http://execpc.com/~kbriggs

1 0

Re: Why PICS is the wrong approach
by Joseph M. Reagle Jr. 17 Dec '03

17 Dec '03

At 07:58 PM 12/11/96 -0500, you wrote: > PICS is the wrong approach becuase it oversimplifies the ratings of > content, because it places the ratings made by the author in the payload > itself, and because third-party ratings systems are cut out of the loop > (effectively). Perhaps I don't understand what you are saying. I just want to ensure that you understand that the PICS labels can be distributed in multiple ways. (document, server, label bureau.) I suspect you do, and what you are objecting to is that documnet-embedded labels will have a greater weight than those distributed by third parties: > agencies, and laws. But once set, the "binding" has been made. Later > reviews or reviews by other entities cannot affect the binding, at least > not for this distributed instance. And consequently authors have a greater responsibility/liability than you would like: > More importantly, the "payload" does not carry some particular set of > fairly-arbitrary PICS evluations. Binding by the censors instead of by the > originator, which is as it should be. In which case, I disagree. I think accurate, consistent, "objective" (I know this is an argument on the other thread, I think one can get relatively "objective ratings" see my RSAC case study for a break down on the qualities of rating systems on my ecommerce page (home page below)) well branded and reputable agents will have a greater weight, and will have a market motivation for accuracy exceeding regulatory pressure. (Plus, there is nothing preventing thresh-hold tolerances for use with multiple ratings.) _______________________ Regards, Restlessness and discontent are the first necessities of progress. -Thomas A. Edison Joseph Reagle http://rpcp.mit.edu/~reagle/home.html reagle(a)mit.edu E0 D5 B2 05 B6 12 DA 65 BE 4D E3 C1 6A 66 25 4E

1 0

[Meeting] Cypherpunks Shooting Club
by Lucky Green 17 Dec '03

17 Dec '03

Fellow firearms friends, The Cypherpunks Shooting Club will meet this Sunday, 12/14/96, the day after the monthly Bay Area Cypherpunks Meeting. From now on, we will meet once a month, on the day after the monthly meeting. Anyone interested in firearms is encouraged to attend. You don't need to own a gun and you don't need to know how to operate one. Of course, if you do, all the better. Various firearms (rifles, handguns, shotguns) will be available. We will gladly provide safety and operations instructions. When: 2:00 pm 12/14/96 Where: United Sportsmen Rifle Range, Ebora Road, Concord, CA, 510-676-1963 How to get there: o From SF, Oakland, the North Bay: [Cross the bridge to the East Bay if applicable]. Take I580 to Highway 24 to WALNUT CREEK. In Walnut Creek, the 24 merges with I680. Follow I680 NORTH. Past Concord, take Highway 4 EAST. Drive for a few miles. Note the nuclear weapons depot on the right. Just past the depot, exit on WILLOW PASS ROAD. Make a LEFT on Willow Pass Road. Cross underneath the highway. At the end of the road, make a RIGHT. Drive 3/4 of a mile. United Sportsmen is the second dirt road on the LEFT. If you reached some homes, you went too far. o From the Peninsula: Take San Mateo Bridge to the East Bay. Go NORTH for a few miles. Take the freeway EAST (number?) to LIVERMORE. At the PLEASANTON interchange, take I680 NORTH. Follow the instructions above. Why we are meeting in the East Bay: Several regular participants are already meeting Sunday morning in Berkeley. What if it rains: The outdoor range is an all weather range. We will meet unless it really pours. Cost: $5 range fee + ammunition. $15 total should suffice for a first timer. What to bring: o Eye protection [mandatory]. Prescription and sunglasses meet the requirements, though I would advise you to use something safer. Chemistry lab glasses work great. o Ear protection [mandatory]. If needed, you can buy some foam type ear protection at the range. It costs less than $5. o Any firearm you might want to bring. Please RSVP if you plan to attend. See you Sunday, -- Lucky Green <mailto:shamrock@netcom.com> PGP encrypted mail preferred Make your mark in the history of mathematics. Use the spare cycles of your PC/PPC/UNIX box to help find a new prime. http://ourworld.compuserve.com/homepages/justforfun/prime.htm

1 0

Re: Why PICS is the wrong approach
by E. Allen Smith 17 Dec '03

17 Dec '03

From: IN%"reagle(a)rpcp.mit.edu" "Joseph M. Reagle Jr." 12-DEC-1996 23:12:04.05 > > More importantly, the "payload" does not carry some particular set of > > fairly-arbitrary PICS evluations. Binding by the censors instead of by the > > originator, which is as it should be. >In which case, I disagree. I think accurate, consistent, "objective" (I >know this is an argument on the other thread, I think one can get >relatively "objective ratings" see my RSAC case study for a break down on >the qualities of rating systems on my ecommerce page (home page below)) >well branded and reputable agents will have a greater weight, and will have >a market motivation for accuracy exceeding regulatory pressure. (Plus, >there is nothing preventing thresh-hold tolerances for use with multiple >ratings.) Umm... I pointed out a while back the considerable problems with the RSAC attempt at objective ratings. See http://infinity.nus.sg/cypherpunks/dir.archive-96.05.09-96.05.15/0092.html for a review of my objections. The system in question is obviously much more subjective than, say, one that had: Does this page contain any female frontal nudity? Does this page contain any male frontal nudity? Does this page contain any female rear nudity? Does this page contain any male rear nudity? and so on. The parts on violence are particularly subjective. -Allen

1 0

Re: NEWS: Web Security Hole Revealed (opportunity?)
by msprague＠ridgecrest.ca.us 17 Dec '03

17 Dec '03

>At 11:14 PM 12/11/96 -0500, you wrote: >>X-Sender: okeefe(a)olympus.net >>Mime-Version: 1.0 >>Date: Wed, 11 Dec 1996 19:32:32 -0800 >>To: N E W S R E L E A S E <IPS(a)olympus.net> >>From: "Steve O'Keefe" <IPS(a)olympus.net> >>Subject: NEWS: Web Security Hole Revealed >> >>BREAKING NEWS >>For Release Thursday, December 12, 1996 >> >>MAJOR WEB SECURITY FLAW REVEALED >> >>(New York) -- Edward Felten, head of Princeton University's >>Safe Internet Programming Team (SIP), today revealed a >>major security flaw in the Internet's World Wide Web. >>Called "web spoofing," the breach allows any Internet >>server to place itself between a user and the rest of the >>web. In that middle position, the server may observe, steal >>and alter any information passing between the unfortunate >>browser and the web. (deletia) Wait a minuite; This sounds like an opportunity! I see a possibility for a machine confugured to tx/rx PGP encrypted packets to re-broadcast them with the machines IP. This would become an anonymous ISP

1 0

Re: Why PICS is the wrong approach
by Joseph M. Reagle Jr. 17 Dec '03

17 Dec '03

At 11:51 PM 12/12/96 EDT, E. Allen Smith wrote: > Umm... I pointed out a while back the considerable problems with >the RSAC attempt at objective ratings. See I agree that it is not a purely descriptive system, however it is much moreso than others. I thought the following was a useful breakdown for my own purposes: Caveat on Vocabulary Before proceeding with the analysis we must first discuss some of the terms used in the analysis. The usage of terms "objective" and "judgmental" can be rather contentious. To address this, we disassociate any of these terms with any pejorative meanings ( opinionated gut feelings about Web content can be very useful ) and posit that there are three variables with which content labeling systems can be considered: descriptive/judgmental - does the label describe the content, or provide an opinion about the "appropriateness" of the content. deterministic/non-deterministic - is the previous process a deterministic process, or is it "gut" based, and voluntary, mandatory, or third party - does the author label his works voluntarily, is he required to label his works by some other agency, or can other services label his content. No rating system we discuss is purely descriptive or deterministic. Rather, each system varies with respect to where it falls between extremes. _______________________ Regards, Restlessness and discontent are the first necessities of progress. -Thomas A. Edison Joseph Reagle http://rpcp.mit.edu/~reagle/home.html reagle(a)mit.edu E0 D5 B2 05 B6 12 DA 65 BE 4D E3 C1 6A 66 25 4E

1 0

Re: Why PICS is the wrong approach
by E. Allen Smith 17 Dec '03

17 Dec '03

From: IN%"reagle(a)rpcp.mit.edu" "Joseph M. Reagle Jr." 13-DEC-1996 00:19:47.19 To: IN%"EALLENSMITH(a)mbcl.rutgers.edu" "E. Allen Smith" CC: IN%"cypherpunks(a)toad.com" Subj: RE: Why PICS is the wrong approach Received: from RPCP.MIT.EDU by mbcl.rutgers.edu (PMDF #12194) id <01ICXWWYIT0GAH4L7K(a)mbcl.rutgers.edu>; Fri, 13 Dec 1996 00:19 EDT Received: from dialup-273.lcs.mit.edu by rpcp.mit.edu with SMTP id AA05830; Fri, 13 Dec 1996 00:20:21 -0500 Date: Fri, 13 Dec 1996 00:17:54 -0500 From: "Joseph M. Reagle Jr." <reagle(a)rpcp.mit.edu> Subject: RE: Why PICS is the wrong approach To: "E. Allen Smith" <EALLENSMITH(a)mbcl.rutgers.edu> Cc: cypherpunks(a)toad.com Message-id: <3.0.32.19961213001747.00937d30(a)rpcp.mit.edu> X-Envelope-to: EALLENSMITH Content-type: text/plain; charset="us-ascii" X-Sender: reagle(a)rpcp.mit.edu X-Mailer: Windows Eudora Pro Version 3.0 (32) Mime-Version: 1.0 At 11:51 PM 12/12/96 EDT, E. Allen Smith wrote: > Umm... I pointed out a while back the considerable problems with >the RSAC attempt at objective ratings. See > I agree that it is not a purely descriptive system, however it is much >moreso than others. I thought the following was a useful breakdown for my >own purposes: The below (from your essay) is a reasonable way to look at it. I would tend to compare a system to obviously possible systems as well as to simply what else is out there, however. >descriptive/judgmental - does the label describe the content, or provide an >opinion about the "appropriateness" of the content. In this regard, I would call the system in question about midway between a truly descriptive system and such obviously judgemental systems as SafeSurf. This partially judgemental nature is probably unavoidable in systems in which one uses a non-binary rating scheme; if the presence of something inevietably means that the system gives a "high" rating, then the system is judging that the something is more important than other factors. In this case, the judgement is pretty obviously that they deem the something in question (such as "hate speech" that calls for "harm" to some class - although I doubt they'd include pro-Affirmative-Action speech in that...) to be worse than the "lower" rated actions. >deterministic/non-deterministic - is the previous process a deterministic >process, or is it "gut" based, and I would agree that RSAC's system is pretty deterministic; the choices of what is labelled are rather arbitrary, but that falls under the description vs judgement category above. >voluntary, mandatory, or third party - does the author label his works >voluntarily, is he required to label his works by some other agency, or can >other services label his content. As currently set up, RSAC is either voluntary or mandatory - a government could require that it be used as a mandatory system, directly or indirectly (e.g., under threat of lawsuits for "corrupting minors"). >No rating system we discuss is purely descriptive or deterministic. Rather, >each system varies with respect to where it falls between extremes. Agreed. -Allen

1 0

Re: Redlining
by Dale Thorn 17 Dec '03

17 Dec '03

E. Allen Smith wrote: > From: IN%"dthorn(a)gte.net" "Dale Thorn" 13-DEC-1996 00:55:54.44 > >It is possible to understand intelligence as "pattern matching" skills, > >without having to have attendant math and statistics to define it more > >precisely or clinically. This understanding (if you have it) is the key > >to knowing that Black persons do *not* have less IQ than White persons, > >regardless of the standardized tests. Something I know about, as I've > >scored in the top 1/1000 of one percent, etc. > Umm... so have I, and I've also done some extensive reading on > the subject. The IQ tests in question do a good job (better than any > other tests - to study & compare you've got to have some sort of test, > and an IQ test doesn't see what color you are) at correlating with > things such as job performance, educational success (e.g., GPA), > likelihood even within given racial groups of being in poverty, etcetera. I have not associated closely with any Black persons since 1981-1983, but at that time, when I had worked with several such persons in Los Angeles, in close quarters (me being the only White person), I was profoundly impressed with the diversity and complexity of the issue (I was curious about the issue due to my circumstances, although I certainly did not mention it) in dealing with real people. My experience tells me that persons who administer intelligence tests and other related tests, or persons who subscribe to study of such things, should get up close to (for example) Black persons for an extended period of time, and I think they (the White persons) would then seriously question any test results which showed a consistent superiority for Whites. > >Nice that you have good intuition on this point, however, the big question > >in your paragraph is not wherefore the genetic determinant, rather, it is > >the understanding of intelligence in its overall context, which I dare say > >most researchers in that field are probably ill-equipped to grasp. > Actually, my conclusions re: genetic differences are an educated > guess... just one that isn't currently testable. My guess comes from such > factors as that some groups that, to all appearances, have high IQs are > actually pretty closely related to some groups that have low IQs; take > hispanics (mixed caucasian and (via American Indian) asian ancestry). > The debates over what is intelligence, etcetera are quite > thoroughly ongoing in the field in question; multiple books have been > written on the subject. I have recently (Oct. 1996) more-or-less inherited three half-Navajo children, upon the death of their father (mother died 1993). The kids are 9, 11, and 13. I have described to various people (schoolteachers, parents, computer programmers, and so on) that raising kids of this age, at this point in my life, is no more difficult than just living my own life without them. Getting past the emotional weaknesses that can hobble you as a parent or guardian is paramount, and once you do that, the rest is a cakewalk. But what I really wanted to say was that my experience with these children now (whom I didn't know until mid-1996) has convinced me more than ever that intelligence is not only not primarily genetic, but may also be recoverable to a large extent up to the beginning of the teen years, if not beyond that. I would guess that children from most any background, young enough and lacking actual physical brain damage or extremely severe psychological trauma, can probably be brought up to the point where they can perform on an equal level with the best our society has to offer. The main and overwhelming factor in developing children who can accomplish at a superior level is the personal attention of a caregiver, and what that attention consists of. There are some things you can read about in a book and guess what they mean, and there are some things you can know when you're staring them in the face...

1 0