Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Captchas Are Becoming Ridiculous

blog.andrewmunsell.com

Post reply on HN
Print view View original

101–110 of 238 posts

Re: Captchas Are Becoming Ridiculous

#101

Post by mgurlitz »

dchichkov wrote: post #94

I see a possible solution that may work in a short term: Instead of displaying a static captcha, display a dynamic one, with letters going through elastic transformations. Humans are pretty good with video sequences. Computers are not. As a side effect this may help pushing computer vision algorithms to working with videos, rather than static images :)

Captchas are hard because there's only so much an algorithm can extract from spatial information. Computers are excellent with temporal data, given essentially unlimited memory for past video frames. Computers benefit more from increased information than humans do.

Re: Captchas Are Becoming Ridiculous

#102

Post by dchichkov »

rorrr wrote: post #51

I see two possible long-term solutions to the captcha problem 1) Ask user to do a relatively expensive computation. This can be done in the background while the user is typing his post. 2) Request a small amount of money (10c) per comment. Good websites will return the money to non-spam commenters, will keep spammer's money. This however requires working microtransactions.

1) Asking a computer to do a computation is not such a great idea. Low powered devices (think iPads) running javascript would be at a great disadvantage to highly efficient botnet clusters that spammers own.

2) Requesting a small amount of money may work. Alternatively requesting a user to do some useful task (like Amazon Turk HIT) to get some funds.

Re: Captchas Are Becoming Ridiculous

#103

Post by Shenglong »

I have no expertise in this area, but how about patterns? Since we're already parsing so much text, why not take strings of legible text (with a minor visual distortion), and strip a connecting word such as a preposition? The user is then prompted to enter the connecting word. Synonyms could be matched.

For example: "I can't stand all ____ your lies" (of)

or: "This is all too ____ for me to handle!" (much)

Sure, some strings would be hard to solve, but I feel like I'd be improved over time. Plus, captchas already take a few tries to solve anyway. I suppose it may impose on foreigners though.

Re: Captchas Are Becoming Ridiculous

#104

Post by dchichkov »

mgurlitz wrote: post #101
dchichkov wrote: post #94

I see a possible solution that may work in a short term: Instead of displaying a static captcha, display a dynamic one, with letters going through elastic transformations. Humans are pretty good with video sequences. Computers are not. As a side effect this may help pushing computer vision algorithms to working with videos, rather than static images :)

Captchas are hard because there's only so much an algorithm can extract from spatial information. Computers are excellent with temporal data, given essentially unlimited memory for past video frames. Computers benefit more from increased information than humans do.

This is computer vision, an I'm somewhat an expert in the area. I can tell that video sequence recognition is _much_ harder problem than image recognition.

For example, if you would show an letter made out of random noise moving through random noise, current computer vision algorithms would not be able to recognize anything. And you would pick out that letter immediately. Human visual subsystem is really amazing in that sense.

Re: Captchas Are Becoming Ridiculous

#105

Post by dchichkov »

adrianN wrote: post #98
dchichkov wrote: post #94

I see a possible solution that may work in a short term: Instead of displaying a static captcha, display a dynamic one, with letters going through elastic transformations. Humans are pretty good with video sequences. Computers are not. As a side effect this may help pushing computer vision algorithms to working with videos, rather than static images :)

Do OCR on every frame, perform majority voting on the result. You just made the spammer's task easier.

Here is an example of a problem that would be hard for computer to solve:

http://www.youtube.com/watch?v=4G4y79ZbaBs

Re: Captchas Are Becoming Ridiculous

#106

Post by hk__2 »

tjpaxton wrote: post #50
streptomycin wrote: post #28

It's easy to criticize captchas. That's why there are so many articles like this, which we all knowingly nod along to as we read. It's much harder to provide productive criticism that leads to an actual improvement over the status quo. (Although it is easy to suggest bad alternatives to captchas, which is why they appear in every comment thread about captchas on Hacker News, including this one.)

Take a look at our product at areyouahuman.com We launched in January and are using games to make them easier for people. Some of our early testing showed captchas can decrease signups by up to 25% and we're able to recover almost all of that. We also monitor how you play the game (like mouse movement) so we can ramp up our security without having to make the task more difficult for people. Read more here http://arey…

This is so stupid. Your CAPTCHA is unusable by blind people, and your audio CAPTCHA is inaccessible (you have to see the element to know where to click to access to the audio CAPTCHA).

Do you consider that blind people are bots?

Re: Captchas Are Becoming Ridiculous

#107

Post by wvs »

Shenglong wrote: post #103

I have no expertise in this area, but how about patterns? Since we're already parsing so much text, why not take strings of legible text (with a minor visual distortion), and strip a connecting word such as a preposition? The user is then prompted to enter the connecting word. Synonyms could be matched. For example: "I can't stand all ____ your lies" (of) or: "This is all too ____ for me to handle!" (much) Sure, some…

Finding dropped words in coherent sentences is trivially easy given a large enough corpus of the language. Just looking at the occurrence probabilities of the surrounding words would be enough for most cases.

Re: Captchas Are Becoming Ridiculous

#108

Post by pjscott »

Shenglong wrote: post #103

I have no expertise in this area, but how about patterns? Since we're already parsing so much text, why not take strings of legible text (with a minor visual distortion), and strip a connecting word such as a preposition? The user is then prompted to enter the connecting word. Synonyms could be matched. For example: "I can't stand all ____ your lies" (of) or: "This is all too ____ for me to handle!" (much) Sure, some…

What stops captchabots from just brute-forcing it with common words? Or, if they wanted to get fancy, something like a Markov model of English text could really help narrow down the likely missing words.

Re: Captchas Are Becoming Ridiculous

#109

Post by monsterix »

Shenglong wrote: post #103

I have no expertise in this area, but how about patterns? Since we're already parsing so much text, why not take strings of legible text (with a minor visual distortion), and strip a connecting word such as a preposition? The user is then prompted to enter the connecting word. Synonyms could be matched. For example: "I can't stand all ____ your lies" (of) or: "This is all too ____ for me to handle!" (much) Sure, some…

Tweaking that a bit, I think Captchas really need to change. But not with cut-off of grammar, which can be probabilistically (brute-force) broken.

One can instead start asking simple questions presented over an image. For example, "how many blue dots on this pattern?" or "what animal is this?"(with a photo of camel or something). That's certainly going to be more manageable for humans and prevent us from behaving more like bots, and them like us. Of course, until the bots evolve a little more to circumvent this.

It is so true that one often meets destiny on the path taken to avoid it.

TLDR; Use image based 'A for Apple, B for Boy' aka the 'Kindergarten training' to filter out bots from humans. Ask questions that are human, to figure out humans in other words.

Re: Captchas Are Becoming Ridiculous

#110

Post by IvoDankolov »

Shenglong wrote: post #103

I have no expertise in this area, but how about patterns? Since we're already parsing so much text, why not take strings of legible text (with a minor visual distortion), and strip a connecting word such as a preposition? The user is then prompted to enter the connecting word. Synonyms could be matched. For example: "I can't stand all ____ your lies" (of) or: "This is all too ____ for me to handle!" (much) Sure, some…

Well, surprisingly, that would only be as strong as doing OCR to get the words.

Why is that? Well, if you deliberately remove words that are easy to put back into the sentence, it's an easy task not because we humans are incredible at "out of the box" thinking, but because one or two word choices are orders of magnitude more likely than all the rest.

And when it comes to statistical guessing, computers are very good at that. Get a large enough corpus of the English language, and I can assure you that the probability distribution of the missing word, given that it's preceded by "I can't stand all" and followed by "your lies" is overwhelmingly in favor of "of".

Calculating these conditional probabilities is not a very hard task to program at all, though it does require quite some computational power and memory for longer chains.

Post reply on HN