<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0"
xmlns:dc="http://purl.org/dc/elements/1.1/"
xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
xmlns:wfw="http://wellformedweb.org/CommentAPI/">
<channel>
<title>Techdirt. Stories filed under &quot;captcha&quot;</title>
<description>Easily digestible tech news...</description>
<link>http://www.techdirt.com/</link>
<language>en-us</language>
<image><title>Techdirt. Stories filed under &quot;captcha&quot;</title><url>http://www.techdirt.com/images/td-88x31.gif</url><link>http://www.techdirt.com/</link></image>
<item>
<pubDate>Mon, 8 Oct 2012 19:50:11 PDT</pubDate>
<title>Human Rights Group Deploys An 'Empathy Test' Captcha System To Help Sites Fend Off Trolls</title>
<dc:creator>Tim Cushing</dc:creator>
<link>http://www.techdirt.com/articles/20121006/12430920629/human-rights-group-deploys-empathy-test-captcha-system-to-help-sites-fend-off-trolls.shtml</link>
<guid>http://www.techdirt.com/articles/20121006/12430920629/human-rights-group-deploys-empathy-test-captcha-system-to-help-sites-fend-off-trolls.shtml</guid>
<description><![CDATA[ Fact: if you have a site with any amount of traffic and open comment threads, you&#39;re going to <a href="http://www.techdirt.com/articles/20110926/16014916101/trolls-dont-need-to-be-anonymous-not-all-anonymous-people-are-trolls.shtml" target="_blank">draw trolls</a>. There&#39;s no method that&#39;s been proven to completely rid your site of trolls, though not for a lack of trying. (<a href="http://drupal.org/project/misery" target="_blank">This one</a> is particularly mischevious.) Various sites have tried anything from aggressive moderation to requiring Facebook logins... all to no avail. (Although the latter method has proven that certain people are more than willing to troll without the protection of anonymity.)<br />
<br />
<a href="http://www.wired.com/threatlevel/2012/10/empathy-captcha" target="_blank">A new anti-troll tool from a rather unlikely source has just been unveiled</a>, one that hopes to combine the "fun" of solving captchas with something akin to a "blush response:"
<blockquote>
<i>A human rights group is introducing a new take on CAPTCHAs, those little boxes that make you type in a word to prove you are human before you can comment or register for a site. Their version doesn&rsquo;t just present a scrambled word to be deciphered, but instead forces a person to choose the right word to unscramble based on the proper emotional response to a human rights violation.</i><br />
<br />
<i>Civil Rights Defenders, the Swedish-based group that developed the tool, hopes the Civil Rights Captcha will help sites block spiders and bots, while letting humans in &mdash; and hopefully educating the humans at the same time...</i><br />
<br />
<i>But perhaps forcing a troll to repeatedly choose an empathetic response will, over time, soothe the ravages of comment sections around the net. Okay, that might also be asking too much, but at the very least spreading information about human rights abuses certainly can&rsquo;t hurt, even if the jerks of the internet (see, for example, YouTube comments) remain beyond help.</i></blockquote>
That&#39;s right. It&#39;s <a href="http://www.technovelgy.com/ct/content.asp?Bnum=126" target="_blank">Voight-Kampff</a> for comment threads. Instead of trying to parse a set of <a href="http://www.techdirt.com/articles/20070524/174116.shtml" target="_blank">badly scanned words</a> and <a href="http://tech.slashdot.org/story/12/03/29/2055228/google-using-recaptcha-to-decode-street-addresses" target="_blank">Street View cam house numbers</a>, <a href="http://captcha.civilrightsdefenders.org/#what_is" target="_blank">Civil Rights CAPTCHA</a> instead asks you how you <i>feel</i> about certain horrific situations, hoping that you&#39;ll make the "right" decision before spewing your vitriol and ignorance into the now-unlocked comment box.<br />
<br />
While its heart is certainly in the right place, the implementation still requires captchas, something most users would rather not encounter every time they make a comment. (Yes, I know. But sometimes, decent , non-trolling humans don&#39;t want to "create an account" or "enter an email address" in order to participate.)&nbsp;On top of that is the fact that each captcha has only one "right" answer, making the system more than a little heavy-handed in its moralizing. This assumes that your regular, non-troll commenters are going to be fine with being preached at while jumping through hoops. It also assumes that all dedicated trolls are morons incapable of deducing the (obviously) "right" reaction to each situation presented.<br />
<br />
This particular captcha service might prove useful in limited situations, like being pre-loaded with questions related to a particular cause or event being discussed/promoted at the website deploying it. It also might prove popular with the sort of people who are willing to annoy a certain percentage of their community in order to "raise awareness." It will become a form of penance for those involved, much like forwarding "concerned" emails and switching Facebook statuses to show support. You know, the sort of thing that will morph into "I solved Captchas for world peace. What will YOU do?" t-shirts popping up on Cafepress.<br />
<br />
I can&#39;t see this solving the troll problem, but I can see it annoying most of a user base, leaving the site deploying it with a smaller audience consisting of people who like being moralized at frequently. Like any other captcha, the spambots and trolls will find a way around it, with the only ones affected being decent human beings, which would seem to be the sort of "demographic" you&#39;d want to annoy <i>less</i>. Pushing them through a "think our way or hit the road" filtering system doesn&#39;t make trolls any less prevalent or make non-decent human beings any more "decent."<br /><br /><a href="http://www.techdirt.com/articles/20121006/12430920629/human-rights-group-deploys-empathy-test-captcha-system-to-help-sites-fend-off-trolls.shtml">Permalink</a> | <a href="http://www.techdirt.com/articles/20121006/12430920629/human-rights-group-deploys-empathy-test-captcha-system-to-help-sites-fend-off-trolls.shtml#comments">Comments</a> | <a href="http://www.techdirt.com/articles/20121006/12430920629/human-rights-group-deploys-empathy-test-captcha-system-to-help-sites-fend-off-trolls.shtml?op=sharethis">Email This Story</a><br />
 ]]></description>
<slash:department>because-i-love-being-talked-down-to-by-a-dialog-box</slash:department>
<wfw:commentRss>http://www.techdirt.com/comment_rss.php?sid=20121006/12430920629</wfw:commentRss>
</item>
<item>
<pubDate>Wed, 2 Nov 2011 17:00:00 PDT</pubDate>
<title>DailyDirt: We've Got The Droids You're Looking For...</title>
<dc:creator>Michael Ho</dc:creator>
<link>http://www.techdirt.com/articles/20101215/03505112285/dailydirt-weve-got-droids-youre-looking.shtml</link>
<guid>http://www.techdirt.com/articles/20101215/03505112285/dailydirt-weve-got-droids-youre-looking.shtml</guid>
<description><![CDATA[ Software is getting better and better at mimicking human behavior on the internet. People are fooled every day by automated messages that seem like they come from actual humans, and sometimes real human messages are mistakenly thought to be composed by computers. Here are just a few more examples of internet bots getting smarter.
<ul>
<li> <a title="http://www.networkworld.com/news/2011/110111-researchers-defeat-captcha-on-popular-252620.html" href="http://bit.ly/sUHPDS">Software called Decaptcha is getting better at solving CAPTCHA challenges.</a> Decaptcha can't foil CAPTCHAs from Google and reCAPTCHA, so it's not going to help digitize books (but maybe someday it will). [<a href="http://www.networkworld.com/news/2011/110111-researchers-defeat-captcha-on-popular-252620.html">url</a>]</li>
<li> <a title="http://searchengineland.com/google-can-now-execute-ajax-javascript-for-indexing-99518" href="http://selnd.com/sORKbw">Google's indexing spider, GoogleBot, can now execute AJAX or JavaScript to get into dynamic commenting systems from Facebook and Disqus.</a> When GoogleBot achieves self-awareness, hopefully it won't judge the human race from what it's read in blog comments... [<a href="http://searchengineland.com/google-can-now-execute-ajax-javascript-for-indexing-99518">url</a>]</li>
<li> <a title="http://news.cnet.com/8301-1009_3-20128808-83/socialbots-steal-250gb-of-user-data-in-facebook-invasion/" href="http://cnet.co/tBDz6P">Socialbots are infiltrating social networks and collecting as much personal information as they can, using fake accounts with attractive profile pictures to befriend unsuspecting people.</a> If a man you've never met before suddenly gives you flowers... (or pokes you on Facebook) Guess what? It's probably spam or an ad for something. [<a href="http://news.cnet.com/8301-1009_3-20128808-83/socialbots-steal-250gb-of-user-data-in-facebook-invasion/">url</a>]</li>
<li><b>To discover more interesting robot-related content, <a title="http://www.stumbleupon.com/to/stumble/topic:29" href="http://bit.ly/h0iGmR">check out what's currently floating around the StumbleUpon universe.</a></b> [<a href="http://www.stumbleupon.com/to/stumble/topic:29">url</a>]  <a title="what's this?" href="#" class="whatsthis help_ddstumble">&nbsp;</a>
</li>
 

By the way, StumbleUpon can recommend some good <a title="http://www.stumbleupon.com/to/stumble/stumblethru:www.techdirt.com" href="http://bit.ly/fagV8c">Techdirt</a> articles, too.
</ul><br /><br /><a href="http://www.techdirt.com/articles/20101215/03505112285/dailydirt-weve-got-droids-youre-looking.shtml">Permalink</a> | <a href="http://www.techdirt.com/articles/20101215/03505112285/dailydirt-weve-got-droids-youre-looking.shtml#comments">Comments</a> | <a href="http://www.techdirt.com/articles/20101215/03505112285/dailydirt-weve-got-droids-youre-looking.shtml?op=sharethis">Email This Story</a><br />
 ]]></description>
<slash:department>urls-we-dig-up</slash:department>
<wfw:commentRss>http://www.techdirt.com/comment_rss.php?sid=20101215/03505112285</wfw:commentRss>
</item>
<item>
<pubDate>Fri, 25 Feb 2011 17:00:00 PST</pubDate>
<title>DailyDirt: Games For People... Not Computers</title>
<dc:creator>Michael Ho</dc:creator>
<link>http://www.techdirt.com/articles/20110121/10460612769/dailydirt-games-people-not-computers.shtml</link>
<guid>http://www.techdirt.com/articles/20110121/10460612769/dailydirt-games-people-not-computers.shtml</guid>
<description><![CDATA[ Computers can be programmed to play all sorts of games, but these machines don't <i>enjoy</i> playing -- or even winning.  It'll be quite the feat to create artificial intelligence that actually understands which games are fun to play... and what games are boring.  Game designers aren't guaranteed to create fun games, so it's not exactly an easy task for humans to figure out.  But when a game is fun, people seem to naturally know it.  That's not to say that every popular game is fun for everyone, but there seems to be some quality of good games that can't just be replicated easily.  Here are a few quick links on games designed just for us humans.
<ul>
<li> <a title="http://venturebeat.com/2011/01/20/game-guru-jane-mcgonigal-says-gamification-should-be-hard-not-easy/" href="http://bit.ly/h2FaqZ">Gamification should make tasks challenging, so that humans actually feel a sense of accomplishment when they finish them.</a>  On the other hand, a bot doesn't feel accomplishment. <i>It can't be reasoned with. It doesn't feel pity, or remorse, or fear. And it absolutely will not stop, ever...</i> [<a href="http://venturebeat.com/2011/01/20/game-guru-jane-mcgonigal-says-gamification-should-be-hard-not-easy/">url</a>]</li>
<li> <a title="http://www2.parc.com/istl/projects/captcha/history.htm" href="http://bit.ly/h4df4e">Humans can try to come up with simple tasks that computers can't do -- and the history of the CAPTCHA apparently starts in 1997.</a>  If only CAPTCHAs were more fun to do... [<a href="http://www2.parc.com/istl/projects/captcha/history.htm">url</a>]</li>
<li> <a title="http://www.jenovachen.com/flowingames/introduction.htm" href="http://bit.ly/eUjFD4">Game designers are always trying to get into the "flow zone" to make their games more addictive.</a>  Hmm. "<i>The famous GRE test is a good example of design based on the concept of the Flow Zone.</i>" [<a href="http://www.jenovachen.com/flowingames/introduction.htm">url</a>]</li>
<li><b>To find some cool online games, <a title="http://www.stumbleupon.com/to/stumble/topic:117" href="http://bit.ly/ifsJE4">check out what StumbleUpon has found to play.</a></b> [<a href="http://www.stumbleupon.com/to/stumble/topic:117">url</a>]  <a title="what's this?" href="#" class="whatsthis help_ddstumble">&nbsp;</a>
</li>
</ul> 

By the way, StumbleUpon can also recommend some good <a title="http://www.stumbleupon.com/to/stumble/stumblethru:www.techdirt.com" href="http://bit.ly/fagV8c">Techdirt</a> articles, too.<br /><br /><a href="http://www.techdirt.com/articles/20110121/10460612769/dailydirt-games-people-not-computers.shtml">Permalink</a> | <a href="http://www.techdirt.com/articles/20110121/10460612769/dailydirt-games-people-not-computers.shtml#comments">Comments</a> | <a href="http://www.techdirt.com/articles/20110121/10460612769/dailydirt-games-people-not-computers.shtml?op=sharethis">Email This Story</a><br />
 ]]></description>
<slash:department>urls-we-dig-up</slash:department>
<wfw:commentRss>http://www.techdirt.com/comment_rss.php?sid=20110121/10460612769</wfw:commentRss>
</item>
<item>
<pubDate>Tue, 23 Mar 2010 04:54:19 PDT</pubDate>
<title>Is A Captcha DRM? Craigslist Wins Default Judgment Claiming Yes</title>
<dc:creator>Mike Masnick</dc:creator>
<link>http://www.techdirt.com/articles/20100322/0316398655.shtml</link>
<guid>http://www.techdirt.com/articles/20100322/0316398655.shtml</guid>
<description><![CDATA[ Last year, we wrote about a troubling set of lawsuits <a href="http://www.techdirt.com/articles/20091008/2324416469.shtml">filed by Craigslist</a> that seemed very dangerous, as it was pushing the boundaries on a series of legal concepts, all of which could come back to haunt Craigslist (and others) at a later date.  For example, we noted that there was a "weak" DMCA claim that said that the captchas used by Craigslist to get people to prove they were human were actually "technological protection measures," and circumventing them violated the anti-circumvention provisions of the DMCA.  While it's not the same lawsuit (apparently Craigslist had filed even more such lawsuits), Ray Dowd has the <a href="http://copyrightlitigation.blogspot.com/2010/03/beware-of-default-judgments-captcha.html" target="_blank">details of Craigslist winning a default judgment in a similar lawsuit</a> after the company sued didn't bother to defend itself.  This is why the concept of default judgments always concerns me.  Now we have a ruling on the books that finds captchas are like DRM, and getting around them <i>even if for perfectly legal purposes</i> (can't read 'em?) may count as violating the DMCA.<br /><br /><a href="http://www.techdirt.com/articles/20100322/0316398655.shtml">Permalink</a> | <a href="http://www.techdirt.com/articles/20100322/0316398655.shtml#comments">Comments</a> | <a href="http://www.techdirt.com/articles/20100322/0316398655.shtml?op=sharethis">Email This Story</a><br />
 ]]></description>
<slash:department>uh-oh...</slash:department>
<wfw:commentRss>http://www.techdirt.com/comment_rss.php?sid=20100322/0316398655</wfw:commentRss>
</item>
<item>
<pubDate>Wed, 22 Apr 2009 00:33:25 PDT</pubDate>
<title>Spammers Solving Difficult AI Problems With An  Underground X Prize</title>
<dc:creator>Mike Masnick</dc:creator>
<link>http://www.techdirt.com/articles/20090420/0228044558.shtml</link>
<guid>http://www.techdirt.com/articles/20090420/0228044558.shtml</guid>
<description><![CDATA[ <a href="http://tech.slashdot.org/article.pl?sid=09/04/19/024213&#038;from=rss" target="_new">Slashdot</a> points us to an interview with Luis von Ahn (who we're a <a href="http://www.techdirt.com/articles/20060901/164218.shtml">big fan</a> of), where he talks about how spammers who are frustrated by various types of CAPTCHA tests <a href="http://www.newscientist.com/article/dn16973-innovation-harnessing-spammers-to-advance-ai.html" target="_new">have set up their own sort of "innovation prize,"</a> offering up somewhere in the range of $500,000 for software that can automatically pass CAPTCHA and reCAPTCHA reading tests (the things where you have to fill in a series of letters to sign up for a service or post a comment).  As von Ahn points out: "If [the spammers] are really able to write a programme to read distorted text, great -- they have solved an AI problem."  It is, effectively, an "X Prize" for optical character recognition.  Not that we like to encourage spammers, but it is rather fascinating how the underground business seems to mirror the above ground innovation world as well.<br /><br /><a href="http://www.techdirt.com/articles/20090420/0228044558.shtml">Permalink</a> | <a href="http://www.techdirt.com/articles/20090420/0228044558.shtml#comments">Comments</a> | <a href="http://www.techdirt.com/articles/20090420/0228044558.shtml?op=sharethis">Email This Story</a><br />
 ]]></description>
<slash:department>fascinating</slash:department>
<wfw:commentRss>http://www.techdirt.com/comment_rss.php?sid=20090420/0228044558</wfw:commentRss>
</item>
</channel>
</rss>