Hacker News Reader

Comment by josho | Hacker News Reader

Interestingly I activated a new gmail account today and during the signup process I experienced the obligatory captcha. It was in two parts, the first looked strikingly like a street view picture of a house number, while the second looked like a traditional captcha.

I suspect that google has been using techniques like this to validate their computer vision conclusions. Which makes their 99% assertion even more interesting, because it's likely 99% confirmed by a very large crowd sourced data set, not simply a staff member going through several hundred samples to come up with the success rate.

View on HN

That's been going on for two years: http://techcrunch.com/2012/03/29/google-now-using-recaptcha-...

Isn't this like free mechanichal turk services? Or forced labour or something. Cynically brilliant outsourcing, tho.

I highly recommend watching the TED talk by Luis von Ahn, creater of CAPTCHA and reCAPTCHA. He calls this "human computation". He is also the founder of Duolingo.

https://www.youtube.com/watch?v=-Ht4qiDRZE8

He's also one of the few famous entrepreneurs from Guatemala.

That's fantastic. I was trying to guess from the name. Latin American? German? Vietnamese?

Most likely of German descent. I seem to remember when travelling through Guatemala about 10 years ago that one of the locals that I got to know explained that the wealth in Guatemala, especially around Antigua, is concentrated in a number of immigrant families of European (primarily German) descent. There are about 13 large families and they tend to marry within that group.

Humans recognize images, machine learning makes use of that in ways no human fully understands, and that makes it "google's computer vision"? AI has really come a long way :P

http://en.wikipedia.org/wiki/The_Turk

It has been like this for a while now. AFAIK you have either

  - the actual captcha and a word from a book (Google Books)
  - the actual captcha and a house number from Street View (Google Maps)

Only the actual captcha has to be typed in correctly.

Do people ever intentionally type in nonsense for the non-captcha part, out of spite, for instance? If you could persuade enough people to provide non-answers for the non-captcha part, would Google stop using it (the non-captcha part)? My guess is it's like spam. As long as it works a small fraction of the time, it will still exist.

I did when Recaptcha came out and it was consistently obvious which was the unknown part. It's no longer so obvious, in my experience.

I believe ReCaptcha presents the same unknown value to several users, and rejects outliers.

In my experience, it's still extremely obvious. It's always the word that's less readable (and doesn't look like it came from a book). And if you have numbers, it's even easier.

Recently they added special captchas for untrusted users. If they think you might be a bot, you'll get much harder captchas.

http://googleonlinesecurity.blogspot.com/2013/10/recaptcha-j...

Actually often it can be the opposite of what you would think. I don't think Google will distort the "unknown" word as much as the "known" word, making it where the "known" word can be harder to read for a human because it's passed through more filters.

I always refresh until they give me a ``hard'' captcha because the numbers, although easier to read, are harder to type.

At one time, there was this effort on 4chan to spam recaptcha with "penis" for the second word. When this got to some online news site, they asked the recaptcha guys who confirmed that the traffic was irrelevant.

I suspect the nonsense won't matter. You're drowned out.

I'm sure people do intentionally type nonsense sometimes - I just did for example. Google can get around that by looking for the same answer a few times. Like if the responses are 0000, 4513 and 4513 it's fairly evident which is correct.

I don't think I've typed a whole captcha in over a year. It is always obvious which part it's necessary, and now that most of the time the unnecessary part is to improve Google's bottom line by checking addresses instead of helping digitize books I will continue typing dhgd instead. typing an intentionally wrong word is just as much a waste of time as typing the right word

All this speculation on whether or not you can type anything for the other word when you can just try it.

http://www.google.com/recaptcha/learnmore

Only worked once for me, and only when I typed a word that was of similar length to the original, but I was only able to get it to work that one time. Leaving it blank or with a period or only four letters would be an extremely obvious thing to check and deny access to.

The 99.8% assertion comes from synthetically generated CAPTCHAs. We don't need crowd-sourced ground-truth -- we know what the true answer is because we generated it.

Reply on news.ycombinator.com