A book about manufacturing may seem an odd place to begin a disquisition about how to read a newspaper, but I have long thought that among the most penetrating accounts I have ever read about how an industry actually works -- any industry -- is Eric von Hippel's The Sources of Innovation.

The hallmark of our times is the seemingly endless stream of new goods and services whose emergence bedazzles and bewilders us. The conventional wisdom about where these new goods come from, at least as it usually is told, is that the stork brings them.

Thus manufacturers are said to perceive a need for new products. White-coated scientists in research parks figure out how to make them work. Engineers develop manufacturing processes. Marketing staffs then gauge potential demand and begin to stimulate it. Only then is the new product finally presented at the sales meeting and, in due course, to the waiting public.

Von Hippel, a management professor at the Massachusetts Institute of Technology, presents a much less linear view. For years, he talked to executives in a number of high-tech industries (semiconductors, steel, scientific instruments, plastics, pultrusion machines, tractor shovels), gained their trust, and discovered a result that will surprise no one who operates in the real world.

Users and suppliers do much of the innovating, not just their manufacturers. The pattern differs from industry to industry and, from company to company within industries, as detailed studies make clear.

But innovation in von Hippel's story is always and everywhere a distributed process. And somewhere near its heart is a process he calls "informal know-how trading," a practice of informal cooperative R&D especially common in rapidly growing industries.

Typical are the "user groups" associated with complicated new software programs -- for bank ATMs, for example, or inventory management systems, or nearly any other kind of fancy new machinery, digital and/or mechanical. Tech-reps from many competing companies meet regularly with the vendor -- all together, usually in a swank hotel or conference center.

There, they swap "patches," invent new applications and discard old ones, develop wish-lists and otherwise cooperate, generally with a view to making the product work better for more people. In the process they learn some of each others' secrets. "How much is exchanged depends on what the other guy knows," explained one executive of a high-tech steel company. "It must be reciprocal."

Of course these user groups serve as informal job fairs, too, insuring that knowledgeable persons are continually vetted and available to be hired, more often than not. Similar institutions operate in many industries -- probably most -- in the form of standards committees, working groups, task forces, prize panels, gypsy training staffs and so on.

What's this got to do with newspapers? It is to say that paper mills, computer software vendors and newstand dealers participate somehow in the ongoing publishing enterprise? They do to some extent, at least on a well-managed paper. But that is not at all what I have in mind.

I am thinking about the consumers and producers of the news itself -- that's to say newspapers' readers and the vast array of people whom reporters call their "sources."

It is a commonplace among news reporters that they operate in an informal economy of favors. Participation in newsgathering, for example, is usually based almost entirely on the (usually shrewdly) calculated rate of gain. Government officials in particular are experts at deciding which phone calls to return and which to ignore.

A name in the paper, a theory floated, a pet project boosted, a rival quietly gored, a scrap of gossip, even a fistful of clippings from the newspaper's "morgue" (of greatly diminished value now that the files have long since become digital databases!) -- these are some of the favors that newspaper reporters are free to dispense, or attempt to dispense, as they go about their business. They do not always succeed.

For it is a commonplace, too, among editors among their most important duties is to serve as gatekeepers to the myriad persons trying to use the newspaper every day to achieve narrowly personal ends. Maybe a reporter has a personal axe to grind. Maybe a politician is pursuing a vendetta. Maybe the dime-dropper has a grudge. Maybe the tipster is a short-seller.

The point is that the linear model in news is as misleading as it is in manufacturing. Like innovation, news can be precipitated at almost any node in a very complicated web. It is true that editors meet daily, that publishers make their wishes known, but the process is highly interactive and open to the "users" of newspapers up and down the line. Hire a PR man, leak a secret, sell some stock, spin an announcement, float a rumor, and you can manipulate the system to your advantage -- maybe. Then again, maybe not.

A particularly good case study of this process would be the story of the rise and fall of Enron. It will be years before the thing can be seen whole, but an important piece of it has just become available. I am half-way through a remarkably interesting book by the Wall Street Journal reporters who broke the story of its fraudulent balance sheet and so put the company on the road to dissolution. "It took 16 years for Enron to grow into a trading colossus," write Rebecca Smith and John R. Emshwiller in the aptly-titled 24 Days. Twenty four days is how long it took for the company to come apart after its most embarrassing secrets began to unravel in The Journal.

Indeed, it was reading their book that reminded me of von Hippel's "Sources of Innovation," for what is news if not a fleeting form of innovation? I've read enough of "24 Days" to feel that Smith and Emshwiller illuminate the process of journalistic discovery in much the same way as did James Watson describe scientific discovery in "The Double Helix" in 1968.

And if their story is not as racy as Watson's classic yarn about how the structure of the DNA molecule was solved, and if Enron didn't exactly "destroy faith in corporate America" as the book's publishers assert (or, for that matter, even greatly damage America's faith in its business press), then "24 Days" still casts a good deal of light how reporters use and are used by their sources.

Their story is all the more enlightening because The Wall Street Journal and The New York Times were highly competitive on the Enron story, and authors do not shrink for explaining how and when. There is something more to say about this, for it is clear that reporters for each paper enjoyed certain distinct advantages because readers of their papers saw their stories and volunteered to help. But I'll wait until I have finished the book.

Sometimes stories happen so fast that they have a nearly immaculate conception -- the recent electricity blackout in the northeastern United States was a good example. But even then readership is destiny. The Wall Street Journal had much better stuff on the blackout, and sooner, than did The New York Times. Industry sources knew what the paper wanted to know, trusted its reporters and were prepared to share their knowledge. Indeed, I think that just plain readers of The Journal are few and far between. Almost all are potential sources, and they know it.

So the way to read a newspaper is to think of it as a giant software system, in which you are embedded by use, to which you may occasionally be called upon to contribute by advice, or at least by daily consent. This may account for the feeling of club membership, of nearly complete identification with one paper or another, which many readers feel.

Yet the U.S. is much better off for having six independent national papers rather than one -- The Wall Street Journal; The New York Times; The Washington Post; the Chicago Tribune and its associated papers, notably the Los Angeles Times; the Financial Times; and Gannett's USA Today. (If you don't think the Gannett entry qualifies, take a look at its recent political coverage.)

The daily papers are intended, by their proprietors and all the others who use them, to serve as guiding systems for society, or at least a part of society, and that is pretty much how they function. All the more reason that their performance should be compared, against a variety of benchmarks. Regularly? Occasionally? It is not done much now.