A couple of times in the last few days I’ve had to deal with Captchas. Those are the distorted images of letters and numbers, supposedly readable by real people, but not by the software spammers use to pollute websites with their trash.
Twice now I’ve had real problems making out what I was supposed to type and have chosen to listen to the audio for some help.
I was truly shocked by what I heard. This audio came with the image attached to this post: Captcha audio (200Kb, MP4).
While the second word in the image above is ‘Louise’, the first word is totally unrecognisable. And as for the audio file: with all that mess of spoken words going on how on earth is anyone meant to identify which ones are the text for the Captcha?
I kept pressing the refresh button on the Captcha until I found something I could confidently type, but even when doing that on some recent occasions I still failed and was asked to repeat the process once or twice more.
I’ve come across a few places that use other methods to identify users as humans — such techniques as asking for the answer to a simple addition, such as 2+4, or a binary-answer questions, such as ‘Is ice hot or cold?’.
These still have their problems, of course, requiring users to be numerate and literate to a certain degree, and relying on cultural knowledge. But I’m inclined to think that those techniques are far preferable to the distorted image and sound methods.
What do you think? Have you had unpleasant or annoying experiences with Captchas? What about good experiences? Have you come across any great techniques for sifting out spammers? Please let us know in the Comments.