Because... circled numbers can't be read by OCR software?

Without some visual obfuscation, I fail to see how this is an improvement against bots. Simple OCR + location of match = 100% success rate (just click the middle). Improvement for users, perhaps, especially due to its simplicity... but I don't know if that's worth it.

Sometimes it ask you to choose the biggest square, not only biggest/smallest questions. The challenge varies all the time.

For example, I can ask you to choose the biggest triangle out of some boxes, circles and stars, and the question sentence can be displayed using a line of traditional CAPTCHA

I can see where that'd help (significantly), but I don't see how a captcha-reader can't also parse the key words, and interpret the image appropriately. "size-est" and "shape" / "color" is easy to rip out, and a basic OCR to detect shapes and sizes and a number is pretty basic.

And remember that the site lists Captchas as being OCR-able, and that the question sentence must be readable - quickly - by humans, so it can only be weakly obfuscated, unlike some of the shorter Captchas. That weakens an already-claimed-as-weak system, so it's easy to assume the instructions can be read. Detecting a couple key words via a smallish dictionary seems simple at worst.

A few mistakes can be ignored, because the server must allow a few - people will make them too. After all of the above, what's the success rate on a dumb interpreter? Say, one that can't understand order of words, merely their existence.

I'm not seeking to totally shoot this down - I think it's an interesting idea, and with a large number of objects and some visual obfuscation could possibly supersede Captcha. But it's achieved by adding logical complexity; to retain ease of use, visual complexity / some other form has to be sacrificed, which makes it easier to read by machines. Or, potentially, making the image quite a bit bigger, so the images / objects used can be made significantly more complex ("click the dog" when given several images).

Infinite! Just this of the possibilities with that kind of system. You could keep adding different questions (just minor variations) and it won't be difficult to keep ahead of spammers. Plus there is always a trade off between security and ease of use. I think the ease of use with clickcha is well worth the slight loss of security.

I'll give them two guesses who will win most of the time.

It doesn't have to be circled numbers (or any other text), that is just one of the possibilities.

On the minus side, it is then totally useless aside from hassling actual users.

Ordinary captchas exclude blind and other visually-impaired people. Audio captchas exclude deaf and hard-of-hearing people. Clickcha excludes people with muscle-control disabilities.

Of course, there would be alternatives for people with disabilities.

It suffers from the same issues of forcing commentators to stop and think about something utterly unrelated to the comment they are trying to make, and of being easily defeated by third world workers paid a pittance or first world porn addicts.

on one scenario you 'll select a 24x24 square out of a 200x100 pixel picture, the success rate is 2.8%

I am the developer of clickcha. Appreciate the input, but you seem to be forgetting that bruteforcing can be avoided by temporarily banning the IP for some time after certain number of incorrect attempts.

50x traffic is too obvious to ban :)

Edit: what's wrong with this?

Because if not you'll have a hard time understanding what I'm saying. (And if it is you have no excuse for the bad grammar.)

Basically you come across as (you sound like) a "cheerleader". You want him to hire you, so you are praising his software. But your praise is false, and doesn't add anything to the discussion.

Telling the whole world you want to do a ruby on rails port doesn't add anything to the discussion either.

Thanks for the clarification, it indirectly helps me to understand both the culture and idioms of the english language.