Sometimes reCAPTCHA asks you to identify a slightly crooked number and a Google Street View scanned street number. Other times, it's mixed-case, barely legible text.

Is there a reason for this?


There is a reason for that. One of the words is already known by the machine and the other one isn't. The app is using human task force to decipher words that OCR can't. Therefore, the words are completely randomized and don't really take into account how difficult they are.

Find more details here: https://stackoverflow.com/a/1435707

| improve this answer | |
  • But how does this explain the occasional crazy captcha? – Simon Kuang May 19 '14 at 3:12
  • That crazy word was supposely present in a document that ReCAPTCHA is trying to convert from image to text. – Renan May 19 '14 at 3:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.