December 2003 - cypherpunks-legacy

In Defense of Anecdotal Evidence
by nobody＠huge.cajones.com 17 Dec '03

17 Dec '03

At 12:09 AM 12/12/1996, E. Allen Smith wrote: >From: IN%"mjmiski(a)execpc.com" "Matthew J. Miszewski" 11-DEC-1996 19:43:51.88 >>If I wanted to I could repeatedly issue heart-wrenching stories of >>poverty in America (similar, of course, to politicians using "real >>world examples" in speeches). You seem to assume that this would be >>"wrong". > > As did Ronald Reagan in talking about "welfare queens"... as >could I in discussing how my grandparents got out of poverty and have >two children with MDs and one with a PhD. Statistics are preferable >to anecdotal evidence for just this reason; I've seen that over and >over again in science. Anecdotes are for lawyers talking to juries >and demagogic politicians talking to the masses. Statistics are a useful tool, but they have their problems. Their accuracy is often in doubt. Most scientific data comes with an error analysis so you can tell what the figure means. For some reason statisticians never do this so we cannot tell whether their numbers are accurate to within 0.1%, 1.0%, 10%, or even worse. There are many other problems. For instance, users of statistics assume they have a random sample, even in cases where that is far from clear. Social statistics are a black art. There was a study awhile back which claimed gun ownership reduced violent crime. That is a surprising result. It was apparently obtained by subtracting out all the other factors that could explain the differences in the areas studied. This process must involve some real stretches statistically. I can't imagine how cultural differences are determined and subtracted - I assume it is a subjective process. Another problem with statistics is that they are difficult to verify. We may wish to verify the information in cases of deception - sometimes well meaning - but also for cases of statistical incompetence. It is also hard to explore the details of the study if the authors are unavailable. The advantage of first hand experience is that it is primary evidence. You know it's true because you were there and saw it. The advantage of anecdotal evidence (in the sense we have been using it) is that the person who is telling you the anecdote was there and saw it. You can cross-examine them and get a full understanding of the evidence provided. Specific examples, that is anecdotal evidence, also provide a nice framework for discussing our abstract beliefs about what is morally right or wrong, or what various parties should be expected to do or say in particular situations. Specific examples also make it possible for the participants in the conversation to deepen their own understanding of their experiences. You might have seen something and interpreted it in a particular way. Somebody else might be able to show you how you misinterpreted what you saw. Red Rackham

3 2

RE: !! Point 'n Crypt -- Win95 Privacy for Everyone !!
by Walt Armour 17 Dec '03

17 Dec '03

The focus for the current PnC offering is more privacy than security. To avoid any issues of ITAR and export we chose to make the first version weak. There will be future offerings that use stronger, non-exportable crypto. As for the posting, I have seen the occasional commercial post float through. This will be the only time that the post will show up on Cypherpunks. Now if we can keep any long threads from getting kicked off (an all too common occurrence). Perhaps replies could be taken to e-mail... walt ---------- From: Gabe Kostolny[SMTP:gabe@ixlabs.com] Sent: Thursday, December 12, 1996 2:45 AM To: Walt Armour Cc: coderpunks(a)toad.com Subject: Re: !! Point 'n Crypt -- Win95 Privacy for Everyone !! > > Point 'n Crypt uses 40-bit DES-CBC (exportable), salted SHA passphrases, > and conforms to PKCS #5 and PKCS #7. > It seems to me that anything that's exportable is pretty damn useless, if you really want to keep something secret. Also... I kinda got the impression that this list wasn't for silly commercial traffic? -gabe

2 1

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

New E-commerce Paper:
by John Young 17 Dec '03

17 Dec '03

A new electronic commerce paper: "Internet Markets: Emerging Business Models" By Dan Yurman December 11, 1996 Abstract What are business models for profiting on the net? Perhaps the best metaphor to describe Internet business is that it is like the California gold rush of 1849. A wave of entrepreneurial start-ups have entered the field, but all are struggling to find the gold. The miners are going broke while the saloon and general store owners, and the brothels, are prospering. Anyone who wants to make real money in mass markets is going to have to break the mold of "business as usual" and offer real value which is not available offline. Meanwhile, business-to-business commerce is growing using the time-based value of data exchange as a pricing mechanism. This brief paper considers the business models and related success factors which favor profitability for doing business on the Internet. It addresses four areas: + Mass Markets + Business-to-Business Markets + Online Banking + Advertising ----- http://jya.com/emarkets.htm (44 kb)

2 1

Re: WEB: Yahoo/Firefly Website recommendation service
by Alexander Chislenko 17 Dec '03

17 Dec '03

At 06:52 PM 12/11/96 -0800, Dale Thorn wrote: >Alexander Chislenko wrote: >> Firefly Network Inc. has just launched a public beta of our website >> recommendation service on My Yahoo! This service is the result of a >> partnership between Yahoo! Inc. and Firefly Network, Inc. in application >> of Automated Collaborative Filtering (ACF) technology to the Web. >> It allows users to find interesting websites interest and like-minded >> people, and otherwise help the user navigate the vast domain of sites >> and people in an intelligent and personalized way. > >I tried Firefly. What a waste. Unless your tastes in most things >entertainment-wise are pretty mundane (music=Pearl Jam, Prince, >other mainstream drek), they won't be able to find a match at all, >no matter how much information you give them! > You are talking about our first site, with music and movie recommendations. I personally rated not-at-all-mainstream movies by Kurosawa and Tarkovsky and got a similar advice. Though, I agree, the recommendations are not perfect. Partly, because the most of the audience are mainstream. Partly, because the site <www.ffly.com> runs the first version of our ACF software server. The website recommendation uses the feature-guided ACF server that I hoped would solve most of the algorithmic problems we encountered in the music and movie domains. I'd be interested to know what you think of it. Also: I created a group profile, name: cypherpunk, password: group that people on this list might use together as a common interest view, or a joint bookmark list. --------------------------------------------------------------------------- Alexander Chislenko <sasha1(a)netcom.com> www.lucifer.com/~sasha/home.html Firefly Network, Inc.: <alexc(a)firefly.net> www.ffly.com 617-234-5452 ---------------------------------------------------------------------------

2 1

Re: Redlining
by E. Allen Smith 17 Dec '03

17 Dec '03

From: IN%"dthorn(a)gte.net" "Dale Thorn" 12-DEC-1996 14:56:52.69 >Actually, there is not only good evidence for the environmental argument, >but you can reason it out yourself if you give attention to some things >that don't make it into most discussions on this topic. Yes, there are strong arguments for the environment being the determining factor... there are also strong arguments (such as interracial adoption still leaving blacks below the average IQ of adopted siblings) for it being genetics. We won't be able to find out which is which until we know what the genetic determinants of intelligence are, which will take some time. (Using current techniques, several hundred years at the minimum... but I'm not prepared to predict how good techniques will get). As I previously stated, I don't believe it is any part genetic. -Allen

2 1

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