Cheaper than printing it out: buy the paperback book.

On the counter of every American auto parts store sits a massive row of catalogs, a horizontal stack of pages as wide as a dump truck, spines down, page edges outward. Even from the other side of the Formica you can easily spot the dozen or so pages out of ten thousand that the mechanics use the most: their edges are smeared black by a mob of greasy fingers. The wear marks help the guys find things. Each soiled bald spot pinpoints a section they most often need to look up. Similar wear-indicators can be found in a cheap paperback. When you lay it down on your night table, its spine buckles open slightly at the page you were last reading. You can pick up your story the next evening at this spontaneous bookmark. Wear encodes useful information. When two trails diverge in a yellow wood, the one more worn tells you something.

Worn spots are emergent. They are sired by a mob of individual actions. Like most emergent phenomena, wear is liable to self-reinforce. A gouge in environment is likely to attract future gouges. Also, like most emergent properties, wear is communication. In real life "wear is tattooed directly on the object, appearing exactly where it can make an informative difference," says Will Hill, a researcher at Bellcore, the telephone companies' research consortium.

What Hill would like to do is transfer the environmental awareness communicated by physical wear into the ecology of objects in an office. As an example, Hills suggests that an electronic document can be enriched by a record of how others interact with it. "While using a spreadsheet to refine a budget, the count of edit changes per spreadsheet cell can be mapped onto a gray scale to give a visual impression of which budget numbers have been reworked the most and least." This gives an indication of where confusion, controversy, or errors lie. Another example: businesses with an efficiency bent can track what parts of documents acquire the most editorial changes as it bounces back and forth between various departments. Programmers call such hot spots of wheel-spinning change "churns." They find it useful to know where, in a million lines of group-written programming code, the areas of churn are. Software makers and appliance manufacturers would gladly pay for amalgamated information about which aspects of their products are used the most or least, since such explicit feedback can improve them.

Where Hill works, all the documents that pass through his lab keep track of how others (human or machine) interact with them. When you select a text file to read, a thin graph on your screen displays little tick marks indicating the cumulative time others have spent reading this part. You can see at a glance the few places other readers lingered over. Might be a key passage, or a promising passage that was a little unclear. Community usage can also be indicated by gradually increasing the type size. The effect is similar to an enlarged "pull quote" in a magazine article, except these highlighted "used" sections emerge out of an uncontrolled collective appreciation.

Wear is a wonderful metaphor for a commonwealth. A single wear mark is useless. But bunched and shared, they prove valuable to all. The more they are distributed, the more valuable. Humans crave privacy, but the fact is, we are more social than solitary. If machines knew as much about each other as we know about each other (even in our privacy), the ecology of machines would be indomitable.