<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Quick! Hit delete. &#8230;.</title>
	<atom:link href="http://www.p2pnet.net/story/9019/feed" rel="self" type="application/rss+xml" />
	<link>http://www.p2pnet.net/story/9019</link>
	<description>p2pnet.net - reader powered</description>
	<lastBuildDate>Tue, 24 Nov 2009 04:43:11 -0600</lastBuildDate>
	<generator>http://wordpress.org/?v=2.8.4</generator>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
		<item>
		<title>By: Reader's Write</title>
		<link>http://www.p2pnet.net/story/9019/comment-page-1#comment-46668</link>
		<dc:creator>Reader's Write</dc:creator>
		<pubDate>Mon, 12 Jun 2006 06:09:41 +0000</pubDate>
		<guid isPermaLink="false">#comment-46668</guid>
		<description>One way to deal with this situation is fairly simple, and can be fun as well:  Take advantage of the old computing adage:  GIGO.

Garbage In, Garbage Out.

The idea is to sufficiently contaminate the data repository with enough disinformation and false data that it can no longer be considered reliable.

The basic flaw in the architecture of these massive data mining schemes is that they use the vucuum cleaner approach to amassing data.  No attempt is made to verify that any of the data it is ingesting is verifiably accurate.  Just like a vacuum cleaner, it will suck up anything that comes in its path.  This would include dirt, lint, pet hair, potato chips crumbs, cigarette butts, etc.  It will also suck up things you don&#039;t want sucked up like coins, the contact lens you lost on the carpet last week, a small screw whose absence preventing reassembly of your wristwatch, the cat&#039;s tail, and it&#039;s own power cord.  The aim is to get the data mining system to do the latter, which is tantamount to having it commit suicide.

There have already been isolated cases of this happening with marketing and market research data bases based on so-called supermarket shopper &#039;Club&#039; cards, whereby you get a discount on certain items if you consent to having all of your purchases recorded, sifted, analyzed, digitized, catagorized, demographicized, etc.  Group opposed to such schemes have done things like organize &#039;card trading&#039; groups where every two weeks persons of considerably dissimilar demographics trade cards among themselves, so the elderly couple with four cats where Mrs has bladder control issues and Mr has false teeth trade cards with the 30-something Gay Male couple with two large dogs and a penchant for fresh pasta and premium deli, who trade cards with the young straight couple with 3 small children, one of whom is teething and has diaper rash, and the toddler has a gluten allergy, etc.  Within a few weeks, all of the data that&#039;s been collected on these people is totally worthless when any kind of market sector analysis algorithm is applied to their purchases. 

Catalina Marketing is the company who is the main purveyors of these club plans and provides sophisticated analyses to the makers of the various national and regional brand products found in the local supermarket.  There are certain specific markets they have just given up on because of such countermeasures being employed.  In other locales people have just simply boycotted the chain forcing those to pay a premium who chose not to participate in their snoopy scheme, in favor of a chain that trumpets that they do no such thing.  This also skews their data and makes it far less valuable because it eliminates an entire class of people for their data collection window and thus is no longer an accurate cross section of the market in that area.

There are something that you can&#039;t screw sround with too much, like credit data, but &#039;marketing&#039; information is fair game.  Take out a trial subscription to a magazine on a topic you totally loathe or know absolutely nothing about.  Take the 4 free issues and then write cancel on this bill, just like the ad says.  You&#039;re name is still duly recorded as being interested in Architecture of 13th Century Antarctica, or whatever the publication was about.  Take the blow-in cards that fall out of any magazine in an annoying manner, fill them out with an imaginary name, but a real address and drop it in the mailbox.  No postage required.  Oops, another small amount of &#039;dirt&#039; in the database.

Make hotel reservations in some distant location and cancel them a few days later.  Same with rental cars or a U-haul.  All of that stuff gets saved away in a database somewhere.

Sign up for Myspace and pretend you are a 24 year old woman, even if you are a 45 year old man or vice versa.  Create your own set of &#039;friends&#039;.  Create you own network of phantom &#039;friends&#039;.  Connect networks of ersatz networks.

If enough people do this to the data sources used for the data mining, the statistical probablility of accuracy will drop far enough where the whole system can no longer be considered &#039;reliable&#039;, certainly not reliable enough for &#039;probable cause&#039; or for any law enforcement action to be initiated (such as an investigation) based only on the results from a questionable system.

Don&#039;t worry though.  Filling out your coffee pot warranty card with false information is not a crime.  In fact, the Government is doing an end run around the requirement they get a subpoena to access certain information by simply purchasing it as the product of a data broker.

The key is to have a sufficient number of angry people with a subversive vein participate in the scheme to make it effective.</description>
		<content:encoded><![CDATA[<p>One way to deal with this situation is fairly simple, and can be fun as well:  Take advantage of the old computing adage:  GIGO.</p>
<p>Garbage In, Garbage Out.</p>
<p>The idea is to sufficiently contaminate the data repository with enough disinformation and false data that it can no longer be considered reliable.</p>
<p>The basic flaw in the architecture of these massive data mining schemes is that they use the vucuum cleaner approach to amassing data.  No attempt is made to verify that any of the data it is ingesting is verifiably accurate.  Just like a vacuum cleaner, it will suck up anything that comes in its path.  This would include dirt, lint, pet hair, potato chips crumbs, cigarette butts, etc.  It will also suck up things you don&#8217;t want sucked up like coins, the contact lens you lost on the carpet last week, a small screw whose absence preventing reassembly of your wristwatch, the cat&#8217;s tail, and it&#8217;s own power cord.  The aim is to get the data mining system to do the latter, which is tantamount to having it commit suicide.</p>
<p>There have already been isolated cases of this happening with marketing and market research data bases based on so-called supermarket shopper &#8216;Club&#8217; cards, whereby you get a discount on certain items if you consent to having all of your purchases recorded, sifted, analyzed, digitized, catagorized, demographicized, etc.  Group opposed to such schemes have done things like organize &#8216;card trading&#8217; groups where every two weeks persons of considerably dissimilar demographics trade cards among themselves, so the elderly couple with four cats where Mrs has bladder control issues and Mr has false teeth trade cards with the 30-something Gay Male couple with two large dogs and a penchant for fresh pasta and premium deli, who trade cards with the young straight couple with 3 small children, one of whom is teething and has diaper rash, and the toddler has a gluten allergy, etc.  Within a few weeks, all of the data that&#8217;s been collected on these people is totally worthless when any kind of market sector analysis algorithm is applied to their purchases. </p>
<p>Catalina Marketing is the company who is the main purveyors of these club plans and provides sophisticated analyses to the makers of the various national and regional brand products found in the local supermarket.  There are certain specific markets they have just given up on because of such countermeasures being employed.  In other locales people have just simply boycotted the chain forcing those to pay a premium who chose not to participate in their snoopy scheme, in favor of a chain that trumpets that they do no such thing.  This also skews their data and makes it far less valuable because it eliminates an entire class of people for their data collection window and thus is no longer an accurate cross section of the market in that area.</p>
<p>There are something that you can&#8217;t screw sround with too much, like credit data, but &#8216;marketing&#8217; information is fair game.  Take out a trial subscription to a magazine on a topic you totally loathe or know absolutely nothing about.  Take the 4 free issues and then write cancel on this bill, just like the ad says.  You&#8217;re name is still duly recorded as being interested in Architecture of 13th Century Antarctica, or whatever the publication was about.  Take the blow-in cards that fall out of any magazine in an annoying manner, fill them out with an imaginary name, but a real address and drop it in the mailbox.  No postage required.  Oops, another small amount of &#8216;dirt&#8217; in the database.</p>
<p>Make hotel reservations in some distant location and cancel them a few days later.  Same with rental cars or a U-haul.  All of that stuff gets saved away in a database somewhere.</p>
<p>Sign up for Myspace and pretend you are a 24 year old woman, even if you are a 45 year old man or vice versa.  Create your own set of &#8216;friends&#8217;.  Create you own network of phantom &#8216;friends&#8217;.  Connect networks of ersatz networks.</p>
<p>If enough people do this to the data sources used for the data mining, the statistical probablility of accuracy will drop far enough where the whole system can no longer be considered &#8216;reliable&#8217;, certainly not reliable enough for &#8216;probable cause&#8217; or for any law enforcement action to be initiated (such as an investigation) based only on the results from a questionable system.</p>
<p>Don&#8217;t worry though.  Filling out your coffee pot warranty card with false information is not a crime.  In fact, the Government is doing an end run around the requirement they get a subpoena to access certain information by simply purchasing it as the product of a data broker.</p>
<p>The key is to have a sufficient number of angry people with a subversive vein participate in the scheme to make it effective.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Reader's Write</title>
		<link>http://www.p2pnet.net/story/9019/comment-page-1#comment-45966</link>
		<dc:creator>Reader's Write</dc:creator>
		<pubDate>Fri, 09 Jun 2006 22:37:47 +0000</pubDate>
		<guid isPermaLink="false">#comment-45966</guid>
		<description>so now the government is building a database of 13 year old girls and their friends.......still don&#039;t see how this stops terrorism since somebody planning an attack probably wouldn&#039;t be a myspace member.</description>
		<content:encoded><![CDATA[<p>so now the government is building a database of 13 year old girls and their friends&#8230;&#8230;.still don&#8217;t see how this stops terrorism since somebody planning an attack probably wouldn&#8217;t be a myspace member.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
