October 20th, 2008
CAPTCHAs and Friends
As spam bots are getting smarter CAPTCHAs are getting harder and harder. We got to the point that it’s not funny anymore.
Image from Flickr
Even the name CAPTCHA which stands for Completely Automated Public Turing test to tell Computers and Humans Apart is complex.
I never understood why a website punishes their users instead of punishing the spammers.
The concept of white lists are always easier to implement than black lists. Fighting spammers by using black lists (IP tables, etc.) is practically impossible due to the dynamic nature of the Web.
On the other hand knowing what type of action is good and discarding the rest is much more easier from development point of view.
What type of CAPTCHAs we have so far?
- Test images
- Math tests
- Random numbers
- Shapes and signs
The problem is that each form of these human verification methods have to trigger a more complex action in our brains.
Image from Flickr
We could say that a click is a standardized action across computer (Internet) users. A click is a simple action, but adding two numbers or figuring out the distorted text from an image and writing it into a text field is more complex.
And there’s the question of different cultures. For example: Chinese symbols, numerals are different from most western letters and arabic numerals. I guess Chinese, Japanese, etc. people have a lot harder time using CAPTCHA ‘enabled’ online forms.
1. Test Images
I’m not writing it in details as we all know what they are.
Image from Flickr
2. Math test
These consists of adding, substracting, etc. usually two numbers and writing the result in a text field.
3. Random numbers
For example writing 9 random number in a text field. This gives us the flexibility to choose any number, but still it’s a complex process than a simple click.
4. Shapes and signs
Developers realized that plain old ‘text in images’ are boring. This resulted in beautiful shapes and signs. Where a user have to think that an ‘ovalish’ circle is in fact a circle or an oval or maybe a square. (?!)
Image from Flickr
Image from Flickr
The idea is that human verification should be as easy as possible (or even totally transparent) for humans and a lot harder for spam bots.
If you enjoyed this article then help spread the word and please follow us on Twitter or subscribe to our RSS feed.
Subscribe to our RSS feedFollow us on Twitter