<?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"
	>
<channel>
	<title>Comments on: Poor advice from SiteAdvisor</title>
	<atom:link href="http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/</link>
	<description>Security Research, Computer Laboratory, University of Cambridge</description>
	<pubDate>Sun, 06 Jul 2008 12:12:47 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5.1</generator>
		<item>
		<title>By: Susan W.</title>
		<link>http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23651</link>
		<dc:creator>Susan W.</dc:creator>
		<pubDate>Sat, 18 Aug 2007 21:21:25 +0000</pubDate>
		<guid isPermaLink="false">http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23651</guid>
		<description>It is not surprising that McAfee isn't capable of reviewing the intentions of a site maker, or the validity of its business claim (e.g., is this a real store?) -- they specialize in malware, and so that's what they know how to look for.  Besides, that analysis can be automated, whereas checking fraudulent activity is very human-intensive.  (Although they are welcome to ask aa419.org for permission to use our database!)

I found &lt;a href="http://www.siteadvisor.com/sites/aa419.org" rel="nofollow"&gt;SiteAdvisor's analysis of aa419.org&lt;/a&gt; to be amusing:
&lt;blockquote&gt;When we visited this site, we found that most of its links are to sites which are safe or have only minor safety/annoyance issues.&lt;/blockquote&gt;
Actually, we have thousands and thousands of links to sites that are dangerous and fraudulent ;-)  I can only assume they count links, and evaluate the ones which are most frequent (and/or are in SiteAdvisor's records already).</description>
		<content:encoded><![CDATA[<p>It is not surprising that McAfee isn&#8217;t capable of reviewing the intentions of a site maker, or the validity of its business claim (e.g., is this a real store?) &#8212; they specialize in malware, and so that&#8217;s what they know how to look for.  Besides, that analysis can be automated, whereas checking fraudulent activity is very human-intensive.  (Although they are welcome to ask aa419.org for permission to use our database!)</p>
<p>I found <a href="http://www.siteadvisor.com/sites/aa419.org" rel="nofollow">SiteAdvisor&#8217;s analysis of aa419.org</a> to be amusing:</p>
<blockquote><p>When we visited this site, we found that most of its links are to sites which are safe or have only minor safety/annoyance issues.</p></blockquote>
<p>Actually, we have thousands and thousands of links to sites that are dangerous and fraudulent <img src='http://www.lightbluetouchpaper.org/wp-includes/images/smilies/icon_wink.gif' alt=';-)' class='wp-smiley' />  I can only assume they count links, and evaluate the ones which are most frequent (and/or are in SiteAdvisor&#8217;s records already).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Feral Pundit</title>
		<link>http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23641</link>
		<dc:creator>Feral Pundit</dc:creator>
		<pubDate>Sat, 18 Aug 2007 10:48:23 +0000</pubDate>
		<guid isPermaLink="false">http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23641</guid>
		<description>I was tagged yellow for "Possibly advertising in junk emails" I have never advertised in any emails but according to the dings at SiteAdvisor..... When we visited this site, we found that it may advertise in junk e-mail.

&lt;a href="http://feralpundit.com/blog/2007/08/18/a-reply-from-a-siteadvisor-fan/" title="My Current Blog post about it" rel="nofollow"&gt;</description>
		<content:encoded><![CDATA[<p>I was tagged yellow for &#8220;Possibly advertising in junk emails&#8221; I have never advertised in any emails but according to the dings at SiteAdvisor&#8230;.. When we visited this site, we found that it may advertise in junk e-mail.</p>
<p><a href="http://feralpundit.com/blog/2007/08/18/a-reply-from-a-siteadvisor-fan/" title="My Current Blog post about it" rel="nofollow"></a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Walter</title>
		<link>http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23589</link>
		<dc:creator>Walter</dc:creator>
		<pubDate>Tue, 14 Aug 2007 13:41:32 +0000</pubDate>
		<guid isPermaLink="false">http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23589</guid>
		<description>When linking to sites like hotshopgood, you should consider using the rel="nofollow" tag so that they don't benefit from your PageRank.
http://en.wikipedia.org/wiki/Nofollow</description>
		<content:encoded><![CDATA[<p>When linking to sites like hotshopgood, you should consider using the rel=&#8221;nofollow&#8221; tag so that they don&#8217;t benefit from your PageRank.<br />
<a href="http://en.wikipedia.org/wiki/Nofollow" rel="nofollow">http://en.wikipedia.org/wiki/Nofollow</a></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Rob</title>
		<link>http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23585</link>
		<dc:creator>Rob</dc:creator>
		<pubDate>Tue, 14 Aug 2007 08:08:14 +0000</pubDate>
		<guid isPermaLink="false">http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23585</guid>
		<description>Richard said "However, it is a problem of their own making! McAfee advertise that they provide a particular service, and if it’s an impossible service to provide then the solution is for them to stop misrepresenting their abilities, not for others to find excuses for their failure."

That's what I was getting at with my earlier comment. For example, I buy a malware scanning service from ScanSafe. They market it as "Web malware scanning" not "protection from scams", which is how it should be.

Shane said "Or e-mail a criticism to complaints shift 2 siteadvisor.com" 

That's a nice way of munging an email address, but it assumes a US keyboard layout. The UK keyboard I'm typing on right now would require a shifted apostrophe instead ;-)</description>
		<content:encoded><![CDATA[<p>Richard said &#8220;However, it is a problem of their own making! McAfee advertise that they provide a particular service, and if it’s an impossible service to provide then the solution is for them to stop misrepresenting their abilities, not for others to find excuses for their failure.&#8221;</p>
<p>That&#8217;s what I was getting at with my earlier comment. For example, I buy a malware scanning service from ScanSafe. They market it as &#8220;Web malware scanning&#8221; not &#8220;protection from scams&#8221;, which is how it should be.</p>
<p>Shane said &#8220;Or e-mail a criticism to complaints shift 2 siteadvisor.com&#8221; </p>
<p>That&#8217;s a nice way of munging an email address, but it assumes a US keyboard layout. The UK keyboard I&#8217;m typing on right now would require a shifted apostrophe instead <img src='http://www.lightbluetouchpaper.org/wp-includes/images/smilies/icon_wink.gif' alt=';-)' class='wp-smiley' /></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Shane Keats</title>
		<link>http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23570</link>
		<dc:creator>Shane Keats</dc:creator>
		<pubDate>Mon, 13 Aug 2007 18:41:47 +0000</pubDate>
		<guid isPermaLink="false">http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23570</guid>
		<description>Hi, this is Shane Keats from McAfee SiteAdvisor. Thanks for bringing this site to our attention. You're right that our focus is on malware and spam. We systematically test for drive-by exploits, adware/spyware in downloads, phishing, spam, pop-ups and bad linking practices. 

We do some human review of misleading business practices (like offering free ring tones but bury in the fine print that the ring tone automatically subscribes the user to an expensive monthly subscription.) We are working to do more in this area. 

You're also correct that we don't test e-commerce sites by making purchases. One of the reasons for offering McAfee SiteAdvisor reviewers the chance to rate sites for their e-com experience is because we don't have an automated (or affordable!) way to test this.

We've got a bunch of failsafe's built into the reviewer system so that one bad shopping experience can't cause Amazon to go red, or even a small business to go red for that matter.

Also, to be clear we do not keep any personally identifiable information. Period. We don't purge it either because we don't keep it to begin with. We know what site you visit because we have to tell you our test results for it, but that's totally anonymous. Here's an analogy. You're a tourist in London. You walk up to a guide called McAfee SiteAdvisor and say "Is Hyde Park safe?" McAfee tells you its findings and then immediately forgets who asked it.

Feel free to send us feedback at http://www.siteadvisor.com/feedback.html

Or e-mail a criticism to complaints shift 2 siteadvisor.com

Thanks.</description>
		<content:encoded><![CDATA[<p>Hi, this is Shane Keats from McAfee SiteAdvisor. Thanks for bringing this site to our attention. You&#8217;re right that our focus is on malware and spam. We systematically test for drive-by exploits, adware/spyware in downloads, phishing, spam, pop-ups and bad linking practices. </p>
<p>We do some human review of misleading business practices (like offering free ring tones but bury in the fine print that the ring tone automatically subscribes the user to an expensive monthly subscription.) We are working to do more in this area. </p>
<p>You&#8217;re also correct that we don&#8217;t test e-commerce sites by making purchases. One of the reasons for offering McAfee SiteAdvisor reviewers the chance to rate sites for their e-com experience is because we don&#8217;t have an automated (or affordable!) way to test this.</p>
<p>We&#8217;ve got a bunch of failsafe&#8217;s built into the reviewer system so that one bad shopping experience can&#8217;t cause Amazon to go red, or even a small business to go red for that matter.</p>
<p>Also, to be clear we do not keep any personally identifiable information. Period. We don&#8217;t purge it either because we don&#8217;t keep it to begin with. We know what site you visit because we have to tell you our test results for it, but that&#8217;s totally anonymous. Here&#8217;s an analogy. You&#8217;re a tourist in London. You walk up to a guide called McAfee SiteAdvisor and say &#8220;Is Hyde Park safe?&#8221; McAfee tells you its findings and then immediately forgets who asked it.</p>
<p>Feel free to send us feedback at <a href="http://www.siteadvisor.com/feedback.html" rel="nofollow">http://www.siteadvisor.com/feedback.html</a></p>
<p>Or e-mail a criticism to complaints shift 2 siteadvisor.com</p>
<p>Thanks.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Richard Clayton</title>
		<link>http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23569</link>
		<dc:creator>Richard Clayton</dc:creator>
		<pubDate>Mon, 13 Aug 2007 17:04:56 +0000</pubDate>
		<guid isPermaLink="false">http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23569</guid>
		<description>Byron said: "Surely the only way McAfee could be sure this site was a scam was if they actually attempted to buy something from it themselves?"

There is indeed a problem in being sure about assessments, and there may well be sites that are far less clear than this one -- where the prices, the payment method, the community opinion, the history of related sites all point to a single conclusion.

However, it is a problem of their own making! McAfee advertise that they provide a particular service, and if it's an impossible service to provide then the solution is for them to stop misrepresenting their abilities, not for others to find excuses for their failure.</description>
		<content:encoded><![CDATA[<p>Byron said: &#8220;Surely the only way McAfee could be sure this site was a scam was if they actually attempted to buy something from it themselves?&#8221;</p>
<p>There is indeed a problem in being sure about assessments, and there may well be sites that are far less clear than this one &#8212; where the prices, the payment method, the community opinion, the history of related sites all point to a single conclusion.</p>
<p>However, it is a problem of their own making! McAfee advertise that they provide a particular service, and if it&#8217;s an impossible service to provide then the solution is for them to stop misrepresenting their abilities, not for others to find excuses for their failure.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Sami</title>
		<link>http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23566</link>
		<dc:creator>Sami</dc:creator>
		<pubDate>Mon, 13 Aug 2007 13:59:57 +0000</pubDate>
		<guid isPermaLink="false">http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23566</guid>
		<description>Speaking of reputation systems and the web, you might be interested in &lt;a href="http://www.mywot.com/" rel="nofollow"&gt;WOT&lt;/a&gt;. Still work in progress though.</description>
		<content:encoded><![CDATA[<p>Speaking of reputation systems and the web, you might be interested in <a href="http://www.mywot.com/" rel="nofollow">WOT</a>. Still work in progress though.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Rob</title>
		<link>http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23565</link>
		<dc:creator>Rob</dc:creator>
		<pubDate>Mon, 13 Aug 2007 13:36:07 +0000</pubDate>
		<guid isPermaLink="false">http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23565</guid>
		<description>I think Byron has hit the nail on the head with this last. Plenty of companies offer, e.g. malware filtering and market it as such. In this case, the marketing seems to have written cheques that the product can't cover.</description>
		<content:encoded><![CDATA[<p>I think Byron has hit the nail on the head with this last. Plenty of companies offer, e.g. malware filtering and market it as such. In this case, the marketing seems to have written cheques that the product can&#8217;t cover.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Byron Thomas</title>
		<link>http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23564</link>
		<dc:creator>Byron Thomas</dc:creator>
		<pubDate>Mon, 13 Aug 2007 12:49:08 +0000</pubDate>
		<guid isPermaLink="false">http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23564</guid>
		<description>Sorry, last comment should have said "their intentions".

However, another thing occurs to me: it will be hard for McAfee to make decisions on who is a scammer and who isn't. For example, if they react to a single negative comment to change their rating, I could potentially extort money from, e.g. Amazon, by posting a negative comment about with SiteAdvisor. Where do you draw the line between protecting customers and protecting honest retailers. Surely the only way McAfee could be sure this site was a scam was if they actually attempted to buy something from it themselves? Indeed, this might be what the average user takes from the message "we tested this site", since how can you test a shopping service without actually shopping? McAfee could of course use behaviour indicators to judge, e.g. uses Western Union, likely to be a scam, but I think they need to change the public perception of what this service does (by changing messages displayed) or change the service to match up to the public perception (likely to be difficult to achieve).</description>
		<content:encoded><![CDATA[<p>Sorry, last comment should have said &#8220;their intentions&#8221;.</p>
<p>However, another thing occurs to me: it will be hard for McAfee to make decisions on who is a scammer and who isn&#8217;t. For example, if they react to a single negative comment to change their rating, I could potentially extort money from, e.g. Amazon, by posting a negative comment about with SiteAdvisor. Where do you draw the line between protecting customers and protecting honest retailers. Surely the only way McAfee could be sure this site was a scam was if they actually attempted to buy something from it themselves? Indeed, this might be what the average user takes from the message &#8220;we tested this site&#8221;, since how can you test a shopping service without actually shopping? McAfee could of course use behaviour indicators to judge, e.g. uses Western Union, likely to be a scam, but I think they need to change the public perception of what this service does (by changing messages displayed) or change the service to match up to the public perception (likely to be difficult to achieve).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Byron Thomas</title>
		<link>http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23563</link>
		<dc:creator>Byron Thomas</dc:creator>
		<pubDate>Mon, 13 Aug 2007 12:43:33 +0000</pubDate>
		<guid isPermaLink="false">http://www.lightbluetouchpaper.org/2007/08/12/poor-advice-from-siteadvisor/#comment-23563</guid>
		<description>Fair enough, I clearly misunderstood your intentions. You're right that if the public think this service "keeps you safe from online scams", then sites such as the one you found must be flagged.</description>
		<content:encoded><![CDATA[<p>Fair enough, I clearly misunderstood your intentions. You&#8217;re right that if the public think this service &#8220;keeps you safe from online scams&#8221;, then sites such as the one you found must be flagged.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
