But how can they verify that what was typed in, is the correct answer? For a CAPTCHA to work, the issuer has to know what the image has on it.
Two words are shown as opposed to one. One is the verifier and is known by the computer; the other is equally as hard-to-read and unknown by the computer.
The logic is that if you get one right, you're going to get the other. Successful guesses are then added to the database to decipher that word.
People will see the same CAPTCHAs every now and again. The more the words are correctly repeated, the more accurate the literature will be.
For example, the CAPTCHA would show 'dog cat' in it's scrambled form. The user has to enter that correctly. The computer will understand the word 'dog', and use that for verification - but you are informing it that the second word is 'cat'.
Myself, I think it's quite genius and works really well. Visiting the reCAPTCHA homepage will show a live example so you play with it all you like.
You're in Easy Mode. If you prefer, you can use XHTML Mode instead. |