A nickel says I am human
CAPTCHA is a form of payment. It's the most useless payment there is: it is 100% wasteful. The user pays with genuine human attention, and the service provider gets absolutely nothing.
The ways of getting around CAPTCHA are well-known: employing a dedicated person to solve it, asking porn surfers to solve it, asking clever computer scientists to write a program to solve it. The marginal cost of breaking a CAPTCHA in each case is less than a nickel and probably less than a cent in bulk. So you can easily buy your way into services protected by a CAPTCHA.
On the other hand CAPTCHAs are annoying, especially as the arms race causes them to become more complicated for humans to solve. They are a huge user interface problem. And last, but not the least, they discriminate against disabled humans.
Why not just ask users to pay a nickel directly instead of solving a CAPTCHA? Because we don't have any good way to charge a Web user a nickel, or to verify their reputation. So let's fix the underlying problems—the Responsibility paradox and the Compensation paradox—and bury the ugly hacks. (This post is sloppy—please read the follow up)
posted by jrp at 12:22 PM
Subscribe.
3 Comments:
"Why not just ask users to pay a nickel directly instead of solving a CAPTCHA?"
Firstly, because it generally costs more than a nickel to receive payment of a nickel.
Secondly, because there are classes of people who would in fact find it quite hard to pay a nickel over the Internet. Teenagers who don't want their parents to find their Livejournal could make do if you could buy "AbandonTheWeb cards" over the newsagent counter, but since such cards couldn't really exist in denominations under $5, you'd still cause problems for quite valuable things like "The Homeless Guy Blog".
Thirdly, you could milk a couple of stolen credit cards almost indefinitely with nickel-sized payments, especially if people were used to making them all over the place -- nobody would be able to keep track of which ones were theirs anyway.
Fourthly, as with any micropayment system, you run into what I've just decided to call the "homework problem". When I was at school, my teachers would set what they believed to be a reasonable amount of homework that could be done in one evening. So the French teacher would set a reasonable amount of homework. The English teacher would set a reasonable amount of homework. Maths, etc.
What the student ends up with is "reasonable amount of homework" multiplied by six, which was occasionally verging on the ridiculous. You get the same problem with micropayments. 5c to post a blog comment might sound reasonable, but if _everything_ on the 'net starts costing five cents, you end up being nickeled and dimed out of participating in anything.
I had a related experience when I was at University. I managed to talk myself into a business account with my ISP, which meant I paid 19c per megabyte I downloaded. Which was good on one hand because, as a broke student, if I was having a poor month I could keep my Internet bills down to next to nothing. But on the other it meant that everything I did, from subscribing to a mailing list or trying out some new software, was a direct economic decision, and I missed out on a _lot_ of the cool shit that my friends could do on flat-rate plans.
Worse, I convinced myself I wasn't really missing out on anything, and the low-bandwidth, turn off automatic image loading, use lynx if you can get away with it Internet was the 'pure' way to go.
Or, to put it another way. Filling in the CAPTCHA attached to this form took me a fraction of one percent of the time it took to write the comment. Marginal cost to me, zero -- it's just part of the time I spend procrastinating before I go to work.
On the other hand, I'm not sure writing this comment would really be worth five cents to me. Maybe it would in isolation, but if it were being added to the big pile of five cents I'd been accumulating during my month of web browsing, probably not.
(Also, CAPTCHA isn't really a form of payment, but I'm going to treat equivocation as background noise from now on, otherwise I'd just spend my whole day arguing semantics)
Charles, you are right. This is a sloppy post. What I meant to write is this:
Today, you can pass the CAPTCHA test either by reputation (actually being human and not blind and not dyslexic) or by paying a nickel, but not to the service provider but to somene else.
What you really want is a test that you can pass either by reputation (not being a spammer) or by paying something directly (if for some reason you haven't built up a reputation yet.) That something can be cents, CPU cycles (hashcash), or some form of play money that even a homeless guy gets 10,000 units per month just for connecting to the system.
What is the right model for the reputation system and what is the right economic model, how to avoid the "homework problem"—these are all the right questions, and I have some idea on how to address them and I'll present those ideas on this blog and would love to discuss them with you. However, before the question of the right economic model can be asked, we need to make sure that overhead of payments of this new platform is essentially zero, unlike the overhead of payments on the Web. This would make charging 0.1 cent practical—and that would probably be enough to deter spammers.
My main point: if we are to design a new platform today, we should build the payment protocols into the infrastructure, and we should build a reputation system into the infrastructure. If both of these are done right, then CAPTCHAs will not be needed. Do you agree with that? (You can argue that it is impossible to get the payment system right in principle, in which case I'll present a specific proposal—with flat rate pricing—and you can shoot it down.)
Payment systems and reputation is what caught my eye. The current currency exchanges (whether that currency be paypal, credit cards, or e-gold) charge too much to use to be useful. I do not think flat rate pricing would work. There are too many variations in perception of quantity and quality.
An Exchange (like a stock, bond, or forex market) of internet currencies would work well I believe to solve the problem.
IGE has done so already for major games and their currencies.
to extend the model to the rest of the internet should be easy enough. (waving magic wand with grenade in my other hand).
Ebay nearly cornered online currencies with paypal, but now the time for an emerging currency exchange should bring greater flexibility and widespread use or payment systems on the 'net.
lemme know if any of this creap is re-hash
Post a Comment
<< Home