Once Web information is treated as an economic good, it will become necessary to have explicit user interface representations for the value and the charges that are associated with Web pages and Web links. Users will be very upset if they see charges on their monthly statement that they didn't know they had incurred.
On the other hand, usability dictates simplicity in the user interface and a general desire to protect users from information overload. To reconcile these conflicting demands, I expect that the payment user interface will consist of multiple representations of information value, depending on the magnitude of that value. In general, the notion of richer attribute representations will be an important part of the next-generation content-oriented user interfaces that will replace the Macintosh-derived designs in about five years.
Users will be able to set individual preferences for the exact cut-off values to be used to determine when to use which user interface representation. A person of limited means would want to be protected against expenses at a level where a rich person would pay without blinking. The numbers listed here are my guess as to reasonable default preference settings that should satisfy many users.
Very cheap costs of maybe less than a cent per page would be invisible in the user interface. If it would cost half a cent to follow a link, then that link should probably be shown in exactly the same way as free links. The time to consider the payment would be more expensive than the payment itself, so the payment should be hidden for the user (except, obviously, for appearing on the monthly statement from the payment service).
Slightly more expensive costs of maybe 1-10 cents per page could be visualized by a simple glyph, a slight color change for the link, or by having the cursor show the cost as a pop-up when the user points to the link. It would also be reasonable for the user's computer to contact a reputation service to gather information about other users' experience with the link: If most other users felt that the destination page was not worth the cost, then a dialog box stating that fact should be shown if the user tries to activate the link. If the destination page has a good reputation rating, then it would be a waste of time to show the dialog box, and the user would be taken directly to the page if he or she activated the link.
Expensive pages costing more than maybe 10 cents would always require the user to click "OK" in a confirmation dialog before the cost was incurred.
Very expensive actions costing more than maybe $10 would require a different interaction technique than simply clicking an OK button since users often do so automatically without reading the warning text. An unusual interaction should be employed to make it clear to the user that a major expense was about to be authorized. The figure shows one of our ideas for a way of authorizing a payment: the system would show an image of a check with the cost and payee filled in. The user would authorize the payment by a sweeping gesture across the signature field , which would cause a digitized image of the user's signature to appear. Ideally, the sweeping gesture would be done by actually touching the user's hand to the screen, but it would also be possible to use a mouse gesture on systems without a touch screen.