Pay as you go. Today, attackers can send junk mail free of charge–and so they do. Resources that aren’t metered tend to be wasted. If they had to pay for each thing they sent, there would be fewer attacks. It would be harder to steal people’s accounts if it cost the victims money–if such thefts happened a lot, there would quickly be a vibrant market for better security products.
Stagger pricing. We can defend the Net without compromising its openness and innovation by introducing several classes of service. A premium service–a FedEx for the Net–could be built from dedicated circuits; a first-class service might be available if you identify yourself and pay, for which you’d receive a certain level of quality; a free and anonymous bulk service would still be available, preserving the innovative frontier of the Net but without delivery guarantees.
User ID. For all but bulk service, we will need to identify who is doing what; otherwise, how can we do accounting or have rules? Without identity, we will have no idea whose packets are flooding the server. We will put ownership marks on packets for the same reasons we put license plates on cars: to identify owners and enforce the rules of the road. And for the higher-paid classes of service we will need digital driver’s licenses to authorize and account for access to these resources.
Secure the hardware. Personal computers today are very often insecure. PCs were not designed with secure network use in mind. Fortunately, more secure Net access appliances and Net-enabled cell phones are coming soon. These “personal communicators” will be far more secure: they can use tamper-resistant identity chips, protected by a password or by biometrics, and will be suitable places for keeping digital money.
