 |
Jim CarnicelliGeneral Blog |
|
|
|
| |
Ç
10/11/2004 -
First Entry
Listen to an audio version
Welcome to my newest blog. It's not the first one I've kept, and it probably won't be the last. I'm creating it along-side my next-to-newest blog on the subject of artificial intelligence because I occasionally have thoughts I'd like to share that are not related to AI.
Ç
10/11/2004 -
How to End the Scourge of Spam
Listen to an audio version
I've figured out a way to largely end the scourge of "spam", those incomprehensible masses of unwanted email we all seem to get each day.
To be sure, I am not at all an advocate of laws to govern the use of email. There have been laws in existence since long before the advent of the Internet to prosecute fraud, coercion, and other crimes that can be committed using the Internet as well. In this regard, the Internet is nothing new at all.
The Internet is a wonderful tool we use to share information and improve our lives. The fact is, though, that it comes with its own issues and annoyances. Spam is one of them. And it seems to many people that it is a fundamentally insoluble problem, but it's not.
One of the best solutions to date has been technological: to enable "spam filters", programs that study each incoming email message in search of patterns that would suggest that a given message is spam. While this is a neat idea, there are people out there who work every day to figure out inventive new tricks to foil spam filters. And they'll never stop coming up with better solutions.
A better solution occurred to me just this morning. Ask yourself why your postal mail box rarely has more than a few pieces of "junk mail". The answer is quite simple: it costs too much for most companies to engage in direct mail campaigns. Email, by contrast, costs nearly nothing, and most of the costs of sending a message are borne not by the sender, but by those who provide the infrastructure all along the way once a message gets "out the door".
What if it cost 34 cents to send an email message, plus another penny for every kilobyte of data over 100KB? How many companies would be in a rush to pay for that kind of cost to send spam?
Laughably, the US postal service some years ago attempted to impose a postage cost on email, but I'm sure it was not aimed at ending spam. One may naturally ask whether the benefit of reducing spam would be worth the cost of having to pay for email oneself. Would it not dampen people's productive use of email?
Here's the twist that makes this problem virtually disappear. Imagine a new email network set up with a fee-based structure in mind. Each person with an address sets a postage rate policy on his own mail box. Every email going forward also has a brief policy statement that indicates how it should deal with postage costs. We'll call this the "stamp". The stamp says something like, "if the postage rate will be below $X, go ahead and deliver it. If it will be above that rate and below $Y, prompt the recipient if he'll be willing to rescind the charges, at least over $X. And if it's above $Z, cancel delivering this message." For an email that requires prompting, the user will see who it's from, how big it is, the subject line, and some other summary information, but the message will be unavailable until he agrees to rescind part or all of the fee. I assume a sort of culture will arise in which most people will always choose to cancel the charges on email they are expecting, from friends, family, clients, partners, etc. As for spam, they'll either get extra money for bearing with the spam or will simply not accept most spam. The senders of such spam will know with certainty that their messages are not received. And if they are, they'll also know because they get the bill for successful receipt.
This sort of system is actually not hard to implement. I would advocate a protocol that allows there to be multiple "postal services". They would be Internet services, of course, but simply host email using a shared standard. Since people would most often have to pay for accounts, their accounts would also be the way of queuing up monies for use in paying postage. Most people would never have to pay any money for sending email, because most recipients will cancel the charges. But otherwise, there would be micro-charges internal to this network of services.
At the risk of getting long-winded, I'll bring up the point one might raise about micro-payments being an inefficient sort of system that has been dealt heavy blows by Internet history. I argue that that's mainly because they've been done wrong for the most part. Some organizations are starting to get it right, though. Moreover, if you get the network of mail service provides set up right, money will rarely change hands among these providers. Instead of wiring money with each transaction that occurs, they will instead maintain accounts among themselves, perhaps through trusted third party financial institutions. By and large, they will transfer information, not money. The theory is that any given postal service provider will generate roughly as much outflow of fees as it takes in in fees to other service providers. In a given month, this network of postal services can automatically work out whether to demand that the balance of payments due by any given provider be simply kept as debt or cashed in using bulk bank transfers instead of lots of individual micro-payments.
Interestingly enough, a new email service can be used to serve other valuable goals, like finally getting everyone using the same encryption techniques, having a shared network of email virus detection and cleansing, modest identity verification, and so on.
I'm not necessarily advocating a dramatic shift in systems, here. But I thought it worthwhile at least to mention this concept. When you spend a lot of time deleting spam, it's hard not to get excited by the thought of getting rid of most of it forever. Or at least making some money for the dubious honor of receiving spam.
 |
Produced in cooperation with Carnell Information Systems, Inc. |