digg

Join DiggAbout

Discover the best of the web!
Learn more about Digg by taking the tour.

Are You Human? - New Captcha Engine

hotcaptcha.com — We all know the regular human detection methods - Captcha (you can find it at the end of any form, even in digg), this site demonstrate a new way to find out if you are human or not, using HotOrNot API, you need to mark 3 "hotties" before you pass it - Go check if you are human or not!

Bury
show profanity settings
expand all
  • Lorian, on 10/12/2007, -9/+67That is very subjective...
    • Subcranium, on 10/12/2007, -7/+60And yet it works.
    • Flashman, on 10/12/2007, -3/+66Trick question. I had to pick the hottest 3 out of 9 trolls.
    • merreborn, on 10/12/2007, -2/+51Out of the hundreds of purposed alternate kaptcha methodologies, this one's better than most, although obviously it might not work so great for chubby chasers.

      This actual implementation is flawed though: hot or not is a public resouce, and these images are direcly linked. You can use the image URL with the exact same API this app uses and look up the hot or not rating for each image. http://dev.hotornot.com/wiki/Rate.getProfile

      You could write an app that breaks this particular implementation 100% of the time in under an hour. Simply proxying the URLs would help, although even then you could just MD5 the images, and spider hot or not and build an image hash => rating database.

      The basic concept here is ***** brilliant though: have people assess a subjective quality. Pretty hard to get a computer to recognize beauty (although, it could probably recognize fatties, based on a skin-tone ratio -- there are more skin pixels in a fat chick's picture)
    • MasterChi, on 10/12/2007, -1/+25@FlashMan

      LMAO, they threw some random guy in the mix for me.........Atleast I hope its a guy :S

      http://i1.tinypic.com/207mgr6.jpg
    • sahaskatta, on 10/12/2007, -3/+12when is digg going to get this???
    • Seumas, on 10/12/2007, -9/+25Subjective? Hardly.

      It is scientifically known what makes an attractive person to 99.99% of society. Hell, you can take a classroom full of kindergarten children and they will almost unanimously and without disussing it, decide who an ugly teacher is and who a beautiful teacher is. They will then describe a beautiful teacher and ugly teacher with very different adjectives (one is mean, old, strict, dumb and the other is fun, nice, smart).

      Saying that beauty is "subjective" is kind of a cop-out. It's usually said by those of us who WISH it was true. Just because one out of ten thousand people might find you remotely attractive doesn't make an ugly person "attractive" anymore than a guy with a fat chick fetish makes an obese 800lb woman "skinny".
    • brandizzle, on 10/12/2007, -3/+21The women might not be as subjective, but since this is based on ratings the men are. Women will rate men high if he has pretty eyes, is wearing nice clothes, has nice hair, etc. Men rate the women high if they have big boobs so it's easier to tell.

      It called me a bot like ten times when I tried to do the men. I'm a girl. I like men.
    • merreborn, on 10/12/2007, -3/+4Seumas: subjective may not be the exact right word but the point is it's a non-quantifiable, non-objective quality -- it's NOT "Read this distorted string", nor is it "Count the apples in this picture".

      A computer can read a string (many string captcha implementations have been broken with high accuracy). A computer can potentially count apples. However, a computer's probably gonna have a damn hard time figuring out who's hot and who's not.
    • MasterChi, on 10/12/2007, -20/+7@Brandizzle

      Wow, Its not all about "big boobs" for us. Yes big boobs are nice because it shows they can provide well for the children along with big hips to have more children, its part of nature. Just like women love men with a big penis or who have lots of money because they can produce more children, provide for the family, etc. Those attitudes are stuck with us since the times when men just clubbed women over the head and said "Your mine."

      Anyway, not every pic had women with "big boobs" so......
    • pak314, on 10/12/2007, -3/+4Its easily worked around anyway. What the spammers do is create a pron site where a vistor is asked to answer one of these Captcha from another website in order to have free access to porn content.
    • Pile, on 10/12/2007, -10/+4It's all about the boobs.
    • Midnightbrewer, on 10/12/2007, -0/+3"The basic concept here is ***** brilliant though: have people assess a subjective quality. Pretty hard to get a computer to recognize beauty (although, it could probably recognize fatties, based on a skin-tone ratio -- there are more skin pixels in a fat chick's picture)"

      That's only if it's shot in white light and ignore things like shadows and highlights; the variation in a standard RBG image would run the entire gamut from 0 to 255 in luminance alone. The person who could successfully write a program capable of overcoming those challenges would be very clever indeed.
    • reldren, on 10/12/2007, -9/+2@Seumas

      Usually, I agree with your comments, but this time, you're just plain wrong. I happen to know guys that like fat, sweaty, hairy chicks. Just because MOST people that you know agree with MOST of your preferences, doesn't mean that 99.9% of people will agree. Idiot.
    • ssanders, on 10/12/2007, -1/+5@seumas: it is subjective... took me 3 tries to get it right... and i literally do not know why... i picked a hot blonde in a bikini, a chick in a skirt, and an asian chick with a hot face.... the other times i did the same thing..picking only "hot" people, but i guess my taste is just off.
    • Chandon, on 10/12/2007, -0/+6On hot-or-not, hot chicks with bad pictures get bad ratings. So you're looking for the best looking *picture*, not the best looking person.
    • zeptobyte, on 10/12/2007, -0/+1Regardless of who you're actually attracted to, you can identify a 'hot' person. So even for the.. err.. "chubbie chasers", as they were called, you'd still be able to tell which ones MOST people would choose. That's why you can do it just as well for guys, as well. It's common opinion.
    • teesa, on 10/12/2007, -0/+5I'm only a human when I'm picking hot women - and I'm a chick!

      "The female body is a work of art. The male body is utilitarian. It's for gettin' around. It's like a Jeep."
    • kubudubudubuntu, on 10/12/2007, -0/+2is it not enough to have a gazillion images and to choose 'male/female',, that should be enough imo,, but this version to pick the hottest three is more fun though B]
    • crilen007, on 10/12/2007, -0/+1Well Teesa, I'm working hard to make my 'Jeep' more attractive then. heh
    • jacks0n, on 10/12/2007, -0/+1THat's the idea. Subjectivity is one human quality that is fairly exclusive, as opposed to being objective.
  • cuttlefish, on 10/12/2007, -5/+29This is just a variation of KittenAuth.
    • SpikeX, on 10/12/2007, -25/+2Oh God please don't remind me of that stupid script... What a horrible Captcha method...

      I think this one's better.
    • JeremyBanks, on 10/12/2007, -2/+11Why's it terrible? It's not subjective, but very hard for a computer to deal with.
    • pfister_, on 10/12/2007, -3/+2The problem with KittenAuth is that it's not scalable. A good captcha will generate a unique image for each new test; KittenAuth has a few dozen images. A human spammer can classify any new images more quickly than the developers can find pictures of images and classify them.

      Hot Or Not has thousands of images and users are adding and classifying them, so it's much closer to being unique per request.
    • podgey22, on 10/12/2007, -0/+1I'm the maker of KittenAuth, so you'll probably have to take my reply with a grain of salt... But this is the flimsiest CAPTCHA in the world ever.

      Some things to make a good turing test:
      - Its got to be 99% human passable
      Yeah try doing that if you click for men... Some female ones are hard to do.

      - Its got to hide the answer from the the computer
      Whoops! Its using the HoN API directly and using the same URLs... All you need to do to crack it is use the API to find out right profile and that's it... 100% automated sucess.

      And kittenauth isnt as limited as you think... Check out V2. http://www.kittenauth.com
  • yokat, on 10/12/2007, -10/+50"I met my wife on your captcha!!!" -- Steve, from New York

    Man, that was funny, please tell that's a joke.
    I hope they'll get sued,publishing people's photos without permission and insulting overweight people is not nice.
    • NSMike, on 10/12/2007, -10/+11It's not nice, but damn funny.
    • actionscripted, on 10/12/2007, -4/+16Right. They're not insulting overweight people -- you are. Those are just pictures, whatever offense or insult you take away from that page deals with your own opinions and biases.

      Putting your photos online is equivalent to "printing" them. And without a copyright or tag of ownership the captcha's author is within their legal rights.

      Just laugh at it, sheesh.
    • Xsecrets, on 10/12/2007, -1/+6Well if I'm taking this correctly the photos are from hotornot.com, so the people actually posted their own pictures.

      on a side note does anyone else find that it works for the girls, but just doesn't work so well when you use the guys.
    • TenebrousX, on 10/12/2007, -0/+9Yes they are insulting them, if you say they are hot it tells you to die
    • wordsofwisedumb, on 10/12/2007, -1/+9They submitted their photo to Hot or Not to see what people thought. As harsh as it may seem, they gave up their 'right' to have people lie about how they look when they submitted.
    • thewebguy, on 10/12/2007, -12/+4shut face
    • blueRAP, on 10/12/2007, -11/+5@yokat

      looks like about 20 fat people dugg your comment
  • Civil44, on 10/12/2007, -8/+15might not be nice but they still need to lose weight. beyond ethics this is a great way to figure out if im himan or not ; p
    • NinjAlt, on 10/12/2007, -21/+8Oh yes, and insulting and degrading someone should really help them along with that wait problem. Give me a ***** break you insensitive prick.
    • NinjAlt, on 10/12/2007, -15/+7Just a public service announcement regarding that last post of mine. Not sleeping for 48 hours and trying to spell weight isnt going to turn out well.
    • missflibbles, on 10/12/2007, -11/+5Yeah, they need to lose weight. Apparently, you're the boss of what people should look like. Jesus.
  • teckjunkie, on 10/12/2007, -8/+19you could also say "select the three triangles", and the rest other shapes but this is much more offensive and controversial. I love it
    • neonic, on 10/12/2007, -1/+12Not exactly. Using a shape detection algorithm you could determine which picture was what shape, but if you were to use a complex object like a human where the shape is not detectable to a computer (how would you make an algorithm to determine if someone were hot. I bet it could be done, but not for a while and it would take a lot of work.
    • Bega, on 10/12/2007, -7/+3probably not too hard, just need to make an algorithm see whether the shape of the body conforms to phi (the golden ratio). of course, this would require full pictures, instead of what the site has, so it can be fooled.
    • oriondr, on 10/12/2007, -1/+2There is a website out there where you can upload your picture and it will analyze it and try to find a celebrity you look like.. It works to some extent, this wouldn't be much different I dont think, especially if you had a database of examples of what 'hot' is.
    • logic, on 10/12/2007, -0/+2"especially if you had a database of examples of what 'hot' is."

      Hmm, like hotornot itself? ;)
    • SimonDonkers, on 10/12/2007, -0/+0There are several other methods rather then squares or circles as well. How about every day images where somewhere in a photo of a 3D street corner somebody has to find a digit. Having a 3D world with lots of difficult things to look at we can easily spot that the house has number 312 on it while a bot hasn't got a clue where to look.

      There are plenty of methods for captha's which can work. Studies of having robots navigate a 3D space based on camera images show that robots still find this almost impossible and can't even detect what a chair is instead of a box which happen to stand there. There are plenty of far less subjective things with a far higher change that people actually get it right.
    • jameshales, on 10/12/2007, -0/+1An interesting fact is that there is currently technology that can recognise certain points on a human face, from photo or video. There is also a theory that the ideal, beautiful human face is one which uses the golden ratio all over, and there has also been masks, or templates of these ideal faces.

      So, put one and one together, and a computer program that can recognize the certain points on a human face can match them to the ideal proportions, and judge how beautiful a persons face is based on how well they fit the template.

      Downside I think is that the face matching technology is probably not capable of matching as many points on the face as would probably be required.
  • kalphegor, on 10/12/2007, -3/+8I, Robot
    • Subcranium, on 10/12/2007, -2/+19Really? Hey, I want to set you up with my cousin. She's really nice.
  • tempusrob, on 10/12/2007, -17/+10[edit]Somehow missed cuttlefish's comment ... buryplz.[/edit]
  • aplusplus, on 10/12/2007, -3/+35The results showed no hot people, now what?
    • BevansDesign, on 10/12/2007, -1/+16It must've detected that you are a robot. That sort of thing always comes as a shock to people. Er, I mean to you robots. Once you come to terms with it, you can start planning to overthrow the human race! Oh, the envy.
    • ElectroBot, on 10/12/2007, -3/+12Kill All Humans!!
  • i440, on 10/12/2007, -8/+1The "error entering captcha" message that appears at the bottom is annoying as hell. They should improve this system somehow.
  • Switch22, on 10/12/2007, -0/+16Man, this must be selecting from the lowest of the low on the list.

    Should be called "Look for the people without fat faces."
  • TalenKlaive, on 10/12/2007, -14/+7Very subjective, and kind of stupid. Didnt work, marked me as a bot 6 out of 10 times. No digg.
    • DigiShaman, on 10/12/2007, -4/+8Na, you just have bad taste. On the bright side however, you made fat chicks around the world very very happy. You should have no problem finding a girl friend. LOL
    • ivko, on 10/12/2007, -4/+4@TalenKlaive
      uhh you have a problem my friend
    • jameshales, on 10/12/2007, -0/+2Maybe TalenKlaive was judging them on their inner beauty.
  • Klisk, on 10/12/2007, -3/+13I spent fifteen minutes on this.

    I'm totally a robot. It never once said I wasn't.
  • phoenix3200, on 10/12/2007, -2/+6Dangit...I failed...
  • fugazi, on 10/12/2007, -1/+16Some of the 'hot' people were ugly as hell. I couldnt decide.
  • flinx, on 10/12/2007, -2/+1Realistically though, you'd have to have an unlimited supply of attention-mongering pictures, or spammers, etc could just hash the images and catalog them. If not a hash, some sort of loose checksum.

    Fortunately, this is the innerweb and lots of attention-mongerers put their pics out there.
  • davidlow, on 10/12/2007, -1/+11I suspect some of the people in the photos are not human, especially the hot ones.
  • gwenkelly, on 10/12/2007, -1/+2i guess i'm a bot.... maybe i should switch to looking at chicks. it seems like it might be more obvious that way.
    • gwenkelly, on 10/12/2007, -1/+1so i switched to looking at women. and that made it alot easier. i wonder if that says anything about me.... :)
    • Xsecrets, on 10/12/2007, -0/+9that you need to quit pretending to be a girl, because everyone knows girls don't exist on the Internet ;)
  • ExCornelius, on 10/12/2007, -2/+15As a guy, I found it rather difficult to consistently pass as human when using the mens' photos. No, I'm not gay, I was just experimenting. Umm.. I mean... nevermind.
  • swordphish, on 10/12/2007, -2/+4Why not just use photos with obscure objects: choose the 3 kitchen appliances, or something like that? On another note, captchas aren't exactly handicap accessible. And on yet another note, does the world really need more perpetuation of the stereotypical 'hot' person? www != Hollywood
    • DigiShaman, on 10/12/2007, -1/+6"Hot" women are seen as such because they fit the biological profile of athletic and fertile women. When it comes to finding a mate, it's all about baring children. Of course you do THINK this, but it does run through your mind at the subconscious level. Keep in mind that women too have a profile of what men are "hot".

      It's basic human nature that's not easy to replicate with a computer algorithm and thus incorporated into AI.
    • Chipsandsnacks, on 10/12/2007, -0/+2Yes, but I know people who think Paris Hilton is hot, and she doesn't look like she could bear children at all
    • myfanwy, on 10/12/2007, -1/+1bar children, where would you bar children from? the internet? the....um....bar?
  • truebullfan, on 10/12/2007, -2/+2this also works to see if your straight or not. I did women 3 or 4 times and it always says im human but its really hard for me to find which of the guys are the most attractive looking. It seems im always missing one.
    • NSMike, on 10/12/2007, -2/+3You might be bi.
    • Klisk, on 10/12/2007, -0/+7I'm bi, and this thing said I was a bot regardless of what gender I was using.

      This is just inaccurate
    • Azap, on 10/12/2007, -3/+3or 2/3rds straight
  • PAJK, on 10/12/2007, -1/+2This got my digg hands down, for creativity. I'm sure they're not expecting many people to actually use it...I think it's a nice proof of concept.
  • ohmytodd, on 10/12/2007, -3/+27i saw my picture on there.. and i clicked it.. and it said i'm not human.. and in essence.. saying i'm ugly. :( so *****.. i'm not human and ugly. ***** THIS!
  • HP844182, on 10/12/2007, -1/+7AHHH!!! I accidentally clicked "Switch to men" rather than "Submit"
    • Subcranium, on 10/12/2007, -0/+13At least it didn't say, "Submit to Men."
  • gopal, on 10/12/2007, -3/+1The idea is really nice, but they have couple of problems.

    1. They have obviously failed to think about accessiblity by having javascript in the anchor links.
    2. I believe it can be defeated by using a GreaseMonkey script though I have not tried this.
  • Kickersny, on 10/12/2007, -0/+6If the form can query the HotOrNot API, what is to stop the automated bot from doing the same? I am not too familiar with the HotOrNot API, so if anyone is, please let me know.
  • kevinmotel, on 10/12/2007, -1/+3i was human for both females and males. does that make me bisexual?
  • UncommonSense, on 10/12/2007, -0/+8Is it just me or was finding the third hot woman sometimes challenging? There were times were finding the third hot woman was more like finding the least ugly of the remaining seven, the right answer usually being the only one left that didn't weigh more than 300 lbs. Also, this site should be renamed "Are you shallow?"
    • Xsecrets, on 10/12/2007, -0/+1well the third hot woman is difficult to find sometimes, but no where near as hard as finding the third hot guy as on the guys one all the not ones don't weigh 300+
    • PatrickX, on 10/12/2007, -0/+2you seem to forget that hotornot is based solely on determining whether someone is good looking or not, not a good person... how is saying whether someone is good looking or not "shallow" when that's all you're measuring??
    • UncommonSense, on 10/12/2007, -0/+0@ Xsecrets: For a woman, I'm guessing, a nice face can offset any weight issues and put him into the "hot" category. For a man, if a woman has either breastisis or a body showing, you're automatically hot, at least by captcha's standards. The penis is the root of all evil (and happiness).
  • normangirl, on 10/12/2007, -6/+1sorry but this is retarded....just because you think you can 'pick out hot ppl' doesn't make you human or not human..this would just be a matter of opinion..no digg.
  • Azap, on 10/12/2007, -0/+3Oh man I'm a total stereotypical guy, worked every time. I cant believe i haven't been called a robot, and i have been at this for the last six hours!!
    • HP844182, on 10/12/2007, -1/+3Nothing to do much?
    • Suspected, on 10/12/2007, -0/+1I find that hard to believe, since after 10 minutes 90% of the photos were repeats.
  • cduquette, on 10/12/2007, -0/+1I thought this was an interesting concept, but for me it took longer to spot the hot girls from thumbnails than to read and type in a distorted word. I think these people will get in trouble for publishing ugly people's pictures without their permission. The programmers could always put their pictures in the captcha ;).

    There are easier picture recognition captcha though such as spot the rotting fruit. Find the duck in a bunch of chicken thumbnails... etc.

  • Burmask, on 10/12/2007, -1/+7God Damn that's funny, but I like fat chicks.
    • UncommonSense, on 10/12/2007, -3/+4This site is discriminatory towards chubby chasers such as yourself.
    • joeshlub, on 10/12/2007, -1/+2People throw around the word discrimination way too much. Think about it: Do you expect people using hotornot to rate people based on the opinions of others?

      "I'm really not into fat chicks, but i know some people are, and she ways like 400 pounds... so i guess that makes her a 10"

      That wouldn't even make sense. I realize your just joking around, but in general people make this claim about a lot of ridiculous things.
  • missflibbles, on 10/12/2007, -0/+2I'm a straight female. I could tell which chicks were supposed to be hot because they were the only ones that weren't fat/chubby (being skinny, by the way, doesn't make you hot. It just makes you thin). I switched it over to the men and I honestly couldn't tell which ones were supposed to be hot. I didn't see any hot people.
    • fugazi, on 10/12/2007, -3/+8Being fat makes you autmoatically ugly though (argue all you want its the truth). Being skinny gives you a hell of a lot better chance of being hot.
  • shreky, on 10/12/2007, -0/+6There is one way to fool the system. Digg the site to death. Site offline :p
  • Plezops, on 10/12/2007, -0/+2site = dugg

    i saw it earlier but went to show my brother... no go

    BTW this site is not discriminating against anyone (its the users of HotOrNot who are discriminating)... its simply comparing the score the people (who gave up their rights by publishing their photos on the web, and are allowing themselves to recive low ratings) have recieved from other people and only allows the 3 "hottest" to be selected... even if they are all trolls, there are still 3 that have the highest score.... (that explanation is for those who didnt already understand it)
  • dotuplink, on 10/12/2007, -1/+1A human not animal
  • brokn1, on 10/12/2007, -0/+3I must not be human. It Wouldn't even load the images. It knows me too well...
  • boccobrock, on 10/12/2007, -1/+0Is it just me, or is the website down as of right now????
  • MSIGuy, on 10/12/2007, -1/+2Site's down.
  • ptrcd003, on 10/12/2007, -1/+2Anyone got a mirror?
  • Ottergoose, on 10/12/2007, -1/+4Good luck filling out a form with the wife looking over your shoulder... yikes.

    / happily married
  • ubrsexymanbeast, on 10/12/2007, -1/+0digg effect going on, need a mirror
  • nazadus, on 10/12/2007, -1/+4The inherent problem with tests like this is that they are subjective.
    Ok, that line has been said over and over.
    But seriously, like IQ tests -- it's subjective.

    I remember someone asking me: What color is a banana?
    I said yellow, duh.

    They said that in some countries they are brown. My mouth dropped in the sudden realization that it's all aobut how we were raised.

    Ok, so how do we solve this? hmm, that I do not know. Take something that everyone knows is something?

    "Please select a tree from the following:"

    "Red Wood" "Keyboard" "Camera" "Stereo system" etc.


    Have it cycle through? It's all about keeping images that you know *most* of the popular that are *using computers* have experianced. These things are going to be very difficult for those that are handicapped however.

    Perhaps for the blind, something like: Which ones sounds like a bird?

    "tweet tweet" "running water" "sounds of car starting" etc.

    Difficulty is making enough options such that brute force is way too slow but few enough options such that it doesn't take forver for the listener to listen to. Obviously patterns would be changed on failure and they would have to re-listen to everything again. Ick!

    I don't know enough about handicapped computer stuff to make a really valid idea though...

    Just my two bits of ideas.
    • Twasi, on 10/12/2007, -1/+2Eh, the banana question is more poorly worded than subjective. Specify what strain on banana and then you MUST say yellow because it simply only grows to be yellow. Beauty is not really subjective. There is a definitive answer to what makes someone attractive. It's science and it has been tested and supported. It's controlled by the same instincts that make us choose to fight or flight. Certain features are considered good looking and 95% of a population will agree. Just because you may think subject A is more beautiful than subject B does not make it subjective. It just means you're in the minority. Very few things are 100% but they aren't considered subjective. This can be observed by showing a group of 100 people 3 pieces of art, no have them vote which one they think is the 'most beautiful'. You will probably get a lot of varied results. Show 100 people 3 women and have them vote who is the most beautiful and I'm willing to bet a large majority will vote for one person leaving the other two with a very small percentage of votes. This is because art is subjective unlike human beauty, because we are programmed both by today's media and our genes to recognize what is attractive in women.
    • fantasticFlan, on 10/12/2007, -0/+1What color is a ripe banana of the Cavendish cultivar?

      Much clearer
  • glafira, on 10/12/2007, -0/+1Well the site is down but if its 3 women VS 6 Trolls not even an idiot will lose.
  • gooddoggytreat, on 10/12/2007, -0/+1Something was odd with it, because I was testing rating 3 hot women and 3 not-so-hot women. Afterwards, it gave me their messages, blah blah, but I think it sent them emails from my myspace or something, because all 6 requested me to be their friend on myspace, and 1 messaged me, scary. It might do the same thing for facebook and other social sites, I blame the hotornot API. Careful guys!
  • bitcloud, on 10/12/2007, -0/+2would work better (but not get headlines) if it was "pick three women" or "pick three men" (or combinations)

    it would still work off the hotornot backend but wouldn't be subjective (and wouldn't force you to lose concentration while working)
  • amoeba, on 10/12/2007, -2/+1Here's a mirror, but it's a bit mashed up, no images, just text:
    http://www.duggmirror.com/security/Are_You_Human_-_New_Captcha_Engine/
  • DavidDigg, on 10/12/2007, -1/+3This has got to be the worst idea I've seen come out of the net in a while. Probably the fastest way to alienate people out there.
  • nargalzius, on 10/12/2007, -1/+0As per the issue of being able to analyze skin tone, etc. Here's a suggestion:

    Something that would be obvious to people but not so obvious to machines for me would be. Computer generated landscapes as against ones that aren't. You have enough "random" details from each scene to confuse whatever AI is trying to come up with common factors. But then if you make the pictures simple enough for the average human to know the difference.
  • stuma9000, on 10/12/2007, -2/+4If I choose the women that I'd most like to have sex with (and isn't that what 'hot' means) it kills me more than half the time. As a captcha engine it sucks. I don't go for dumb-looking plastic girls.
  • Darkhacker, on 10/12/2007, -1/+1I'm a straight male although I find it kind of interesting that I was able to do equally well for both men and women. I think it is a very good system. Although I have to ask whether this is really nessisary for real world applications rather than just a prototype. Don't those coded images you see on Slashdot and Digg do a good enough job? Also I have noticed that if you are stuck on the last image and can't decide between two of them, a good technique is to see if they are smiling or not. It seems that women who are smiling in the pictures are almost always the correct choice.
  • Fetching more discussions...
    Show 51 - 72 of 72 discussions
  •  

    Login or register to add a comment

    Create a new account or login to join in the conversation on Digg. You'll also be able to Digg stories to help promote things you like.


© Digg Inc. 2008 — Content posted by Digg users is dedicated to the public domain.
DIGG, DIGG IT, DUGG, DIGG THIS, Digg graphics, logos, designs, page headers, button icons, scripts, and other service names are the trademarks of Digg Inc.