GMiller.net » Captcha Game

Recaptcha used to be really cool. Not only was it the de facto standard for human verification on the web, but it also had a noble goal. Google, who was always thirsty for big data, was scanning the entire library of congress and attempting to digitize it. Computers equipped with OCR (Optical Character Recognition) software would read the scan images of the books and convert them to text. There was only one problem: sometimes the OCR software wasn’t good enough and it needed help from a human.

Enter recaptcha. Recaptcha attempts to verify a human user from a bot, by showing them two words which are difficult for a computer to read. The human would enter the words, and if correct, be allowed to proceed. There was just one catch that not many people knew about: although recaptcha showed two words, only one of the words was needed to pass the human-verification test. The second word, was actually passed on from the google-book scanning project. Let me reiterate: two words were given to each recaptcha user: one to verify that they’re a human, and one the OCR software couldn’t read!

The human would enter both words as they were told, and the recaptcha would use their interpretation of the world in their translated books! Pretty sneaky, but infinitely clever – Google figured out how to mine billions of people’s worth of human effort, while also providing a critical service to the internet at the same time.

Well, enough about why recaptcha used to be so cool. These days they’re pushing a new method which – is way easier than the old method, but doesn’t really help translate books.

I decided that I needed to learn how to use the Recaptcha library, and at the same time, I also wanted to learn the Facebook api. I came up with a funny idea – a game where you had to successfully pass a chain of recaptchas. Your score would be how many captchas you could successfully pass in a row. It sounds simple, but people often complained about having to enter captchas more that once, so it was simple but not necessarily easy!

After I learned how to get the Recaptcha API working, I dove into the Facebook API and it so users could login, and keep track of their high-score with Facebook. Further, in the high-scores area it would use their Facebook thumbnails.

I’ve always wanted to revisit this project, but I almost never have the time. I have some cool ideas for it, but I … maybe someday. For now it’s basically abandon-ware. I still own the domain name for a few more years and I don’t plan on letting it go. But it needs work. The Facebook API updated at one point, and now it’s kinda glitchy. I wasn’t as good at CSS or PHP or making websites in general when I made this, so it’s suppper rusty all-round. It still technically works, so give it a try!

Add Comment

Comments