I AM NOT A ROBOT
I'm tired of always having to, "type the letters from the image above." It's waste of my precious time and it's annoying when I don't know what the flipping image is and have to do it again. God! There has to be a better way to verify if someone is a human or not.
"Annoying As Hell"
So I am going to propose a new system to the world. One that is much simpler and less annoying but still provides a high level of security. It's so simple, I'm surprised you eggheads never even thought of it...
See how simple it is? All you have to do is pick a "randomly" generated image, instead of doing all that typing! Ah you say, but won't the bots just scroll through the menu and pick each choice? Not so, because the image generated is based on a specific algorithm where the questions (images) chosen are based on a myriad of factors from the site user.
So with the question so variable, so will be the answer. Even then if you figure that out, it's still a 1 in 10 or a 1 in 5 chance of being right, depending on what you choose.
For example, say you have chosen 10 images for your quiz: circle, square, cube, triangle, star, diamond, rectangle, pentagon, dragon, person...
Each image will be assigned a random generation by factor taken directly from the user.
For example say my IP address is, 69.256.133.200.
The website will read my address
and select the images based on the numbers. If the amount of odd
numbers is greater than the even numbers the image chosen will be CIRCLE. If the
even numbers outnumber the odd numbers, the image will be SQUARE. If there is an
equal number the image will be CUBE.
The second image selection algorithm will be based on the user's time clock. If say it is 3:19 AM, the first number is odd, so the image might be TRIANGLE. If the second number is even it might be STAR.
The third image selection will be based on the date. For example, if Monday lands on an odd number the image might be DIAMOND. Or if it lands on an even number it might be RECTANGLE.
Yes, so you see, the question is
based on three factors: IP address, time and date clock.
Which will generate an almost truly unique question for each user, depending on
who they are (IP), what time it is and what day it is.
"AH HA!" you say again, but that doesn't matter the bots will still just scroll through the answers and choose one, one has to be right. WRONG. The image choosing algorithms will also be used to generate the mixing of the answers, so they will never be in the same order at any time. So depending on the date, time, and where you are from, it will mix it up differently each time. But here's the kicker, there will always be one answer that says I AM SPAMMING. A human will NEVER EVER choose this. But if a bot does it will be locked out (of the website) for any X amount of time. But in case a human does pick that, so he or she won't be locked out she can use the traditional word verification.
But can't a spammer just assume multiple identities? No, the website will record their IP address. What if they use a proxy and change their address all the time? The website will mark several common essential software objects within their computer and assign a stagnate identity (for example, when you right click and select properties on image files you can leave comments, this can be used for identification -- or maybe do REGEDIT) So it can't change (automatically). Of course you would be able to unmark it through the website with word verification (the marking of objects is also randomly assigned).
If they refuse to accept the transaction? Then they CAN'T use the the website.
Is this nuts? No, it's not malicious. It doesn't effect the users computer, just those who want to abuse the website. So is this nuts? Maybe...but I hate spammers.