What has defined us so far has been the configuration of our end product: the book. That has seemed to be quite enough definition. Only within some professional book publishing organizations do journals sometimes co-exist. But publishers of books very seldom morph or segment into publishers of consumer or professional magazines. Their retailers may increasingly sell recorded music, but the publishers know music is a different business. Videos don't figure in a book publisher's thinking. Audio does only in a sporadic and very ancillary way.

Despite this clarity, from time to time during my time in this business, the leadership of many book publishing organizations have, sometimes simultaneously, lost their defining focus. In the late 1970s and early 1980s, several large publishers ventured into the field of computer software development, with universally disastrous results. In the mid-1990s, many publishers started to deliver content-based CD-ROMs, which failed as badly as the software.

But the key to success can't be articulated simply as "stick to books", as if that advice could possibly be applied anyway into the future which is already in sight. At the height of CD-ROM-mania, we suggested that the publisher's defining capabilities had to do with its unique understanding of "content and markets". That, to us, put both category and channel specificity into the publisher's DNA. Viewing the business that way, it became easier to see that developing products that strayed far afield from the publisher's content and markets knowledge base was dangerous business, likely to fail.

But defining capabilities are not the same are core competencies. Publishers need a command of content and markets to envision, develop, and effectively sell their products. But they also need some skills: command of the language, mastery of the techniques for physical book development, the ability to execute the complex tasks involved in the marketing and distribution of their books. Being proficient at these core activities does not assure success if the content and market equations aren't right. But being incapable of these core activities does assure failure, and the companies that don't competently perform the critical tasks of publishing usually end up being owned by competitors that do.

Content and market knowledge necessarily varies with the publisher. Core competencies are universal. And the required core competencies for publishing from now on can be expressed in two words: databases and networks.

We have had databases and networks forever, of course, but technology has gradually, and now suddenly, elevated them to primary importance. Long before computers, databases were "lists" and networks were "webs of contacts". The difference between then and now is like the difference between a series of woodcuts and a blockbuster feature film, the difference between a horse-and-carriage and a jet plane. The woodcuts can tell the same story as the movie; the horse-and-carriage can take you from New York to Chicago just like the jet.

Well, not just like the jet. And that's the point.

The power to manage databases through computers today, combined with the enabling capabilities of digital networks, completely change all of our practices and push back our horizons to distances we never would have dreamed possible. How effectively the full power of databases and networks are harnessed will separate the living from the dead in book publishing in the very near future.

Please forgive a highly personal anecdote that demonstrates, I think, how quickly times are changing and how even savvy practitioners can find it necessary to revise the way they operate.

My firm, The Idea Logical Company, provides a lot of content to a Web site called CBS Sportsline about baseball. One of the features we create is a weekly batch of 18 trivia questions, which we have done for us in true virtual fashion by an ex-staffer who now has another job and does this assignment as a freelance in his off-hours. These trivia questions are a very small element of our overall content provision and buying them constitutes a very small expense item for our business.

About a year ago, I got all the questions we had provided Sportsline put into one file just to see what we had. It turned out we had 120 pages of material; enough to propose a book. So we made a deal with Sterling Publishing and, with some minor editorial revision they ordered up from us, they repackaged our material into what they called The Little Giant Encyclopedia of Baseball Quizzes, which they put into their catalog for presentation at last December's sales conference.

A few days before the Conference, Sterling's VP for Sales and Marketing, Charles Nurnberg, who is also my friend and who is also the guy who bought the material and repackaged it, showed me the cover and catalog roughs. It read very clearly: "The Little Giant Encyclopedia of Baseball Quizzes by Mike Shatzkin."

Now, the material on our Web site doesn't say anything about Mike Shatzkin. All of our work is attributed by name to The Idea Logical Company. Since we expected to promote the book heavily through the Web site, and also, frankly, because I personally had NOT written the book which was assembled by my staff, I suggested to Charlie that we change the author to the company. And being an obliging guy as well as a smart sales and marketing professional, he said "yes".

So now I tell him: "Charlie, you let me make a big mistake!"

As of the time the book actually first shipped to stores in early May, EVERY online mention of it carried "Mike Shatzkin", not "The Idea Logical Company", as the author. And since many of the online resources also chose to list, or perhaps required, a shorter title, most of them say some varient of "Baseball Quizzes", with hardly a mention of Encyclopedia, or Giant, let alone Little Giant Encyclopedia.

If you wanted to order "The Little Giant Encyclopedia of Baseball Quizzes" by The Idea Logical Company and you saw a listing for "Baseball Quizzes" by Mike Shatzkin, would you figure you'd found the book you were looking for? Possibly not. Maybe even probably not.

This is NOT an easy matter to fix. In a recent survey our office did for another purpose, one top sales executive told us that database corrections are THE most discussed subject between Amazon and publishers, and that it is very hard to get things corrected. We'd speculate that the prevalence of these discussions has a lot to do with the authors' propensity to check out online listings, with great sensitivity to error. That they're hard to correct demonstrates one of the challenges of database management: if you own a complex database, you can't just let anybody who thinks you've got something wrong to go in and change it. That would be a recipe for chaos, and for the introduction of whole new batches of errors.

The time has already come when every sales department needs a Manager of Customer Database Integrity: one person charged with riding herd over all the online resellers as well as the databases of wholesalers, to make sure that errors are corrected promptly. Automation tools can be employed to help attack this problem, but there is no automated solution yet available. Some human is going to have to take the responsibility.

The lesson I took away from this experience is this: in today's environment, once you've put out the metadata on a book, don't change it unless you absolutely have to. My professional marketing judgement to capitalize on my company's name association was a sound one by the standards of two or three years ago. But the real working conditions of today say: no, it isn't worth the inevitable confusion in the electronic marketplace.

There is probably a window for each particular piece of book metadata: the title, the trim size, the price, within which it doesn't matter when the decision is made. But the window shuts, and making some decisions too late or changing them too late comes only at a real cost. This has always been true, but new circumstances require thinking it through all over again.

A company that masters database and network automation so it can easily correct problems like this will have a competitive edge in the marketplace. It will be able to change titles and authors and trim sizes and prices and jacket art without fearing that it is in jeopardy of erecting a barrier in its fastest-growing sales channel.

Demonstrating that inadequate mastery of databases and networks today is costing publishers real sales today would, one hopes, grab any publisher's attention. But these tools are so dynamic that mastery of them will absolutely transform every aspect of every publishing business. The competitive advantage that these competencies can deliver is transparent at the ecommerce cash register; it is no less critical throughout the business.

Of course, publishers have actively discussed digital content repositories for years. The proliferating opportunities to sell books as digital files are suddenly focusing a great deal more attention to the importance of this particular database. These opportunities are being enabled by the increased development of networks. The ebooks all distribute their product through the Internet; the print-one technology is so powerful because it puts the book "in stock" at Internet booksellers. A even more decentralized print-one solution to come may scatter the capability at bookstores and small digital printshops in every community, again with content travelling by Internet to its destination.

This is conceptually very attractive to publishers. Getting sales without an inventory investment naturally would be. But many publishers are discovering that the databases they have don't have what they need for the purpose. Whatever they think they have, they don't actually have databases of the content that will enable them to deliver it in formats that work for all these new applications.

It has become obvious today that every book should be made in a file that can be flexibly altered to deliver what each commercial manifestation of the file requires. Exactly when it became obvious may be open to debate, but we are hereby declaring that it is obvious now. Publishers who can cost- and time-effectively deliver the files to make the offset-printed book, the digitally-printed book, the Web presentation, and the format for Rocket Books, Softbooks, and their proliferating competitors, will make sales at the expense of publishers who cannot.

The publisher who did not prepare files by this standard as they made books in 1996 or 1997, or even in 1998, failed to see the future as well as they might. The publisher who is not preparing files as they make books that way right now is open to valid questions about the company's mastery of a core competence.

The ability to digitally manage the content to enable distributing content through proliferating format options also highlights the same needs for managing rights. This is another complicated, database-and-network driven challenge. The difficulties that are arising in this area, and the ways around them, also suggest the powerful paradigm-shifting nature of the new technology.

Of course, publishers must have databases that tell them what rights they have and the status of deals and negotiations for various categories of rights. In the electronic rights area particularly, but in others too, the author or the agent is often involved in the sale and perhaps required to approve it. That means that author and agent contact information, and perhaps the capability to reach them through networks, need to be part of this database.

In fact, the day is not far off where publishers making their rights availability accessible through online networks will enable rights sales to "just happen". Of course, it will happen first for those rights which are accessible first.

We are now at an interesting moment regarding clearing rights for ebooks. The economics of a world of books without inventory investment is so different that many agents are reluctant to negotiate royalty rates or percentage splits for fear that the industry will learn collectively in a year or two -- or maybe even in a month or two -- that fifteen percent royalty or a fifty-fifty rights split or any other formula has been rendered obsolete.

The result has been that Ebook initiatives are being strangled in the cradle by very slowly-growing content libraries. Many publishers are trying very hard to cut this Gordian knot by securing these rights clearly in new contracts they initiate, but even that is difficult to do with the most powerful authors. And the big names are needed to provide the consumer appeal if Ebooks are to reach a mass market.

One solution that has not yet been tried is the 90-day license, which we believe would eliminate the impasse very quickly. A license this brief could never be practical in a world where printed inventory is maintained. But when there is no inventory investment and everything is managed by databases through networks, it really is no problem. And it gives the ultimate copyright holder the assurance that renegotiation can take place whenever the market shifts and requires it.

If the 90-day license were suddenly accepted as the magic bullet for ebook rights, each publisher's command of databases and networks would receive another sudden, and urgent, challenge.

This is really endless. Publishers will in a relatively short time be delivering sales and royalty information to authors on the Web. Publishers will need databases of authors they have signed and authors they'd like, and their agents, and the deals they and others have made with those authors and agents, in order to compete intelligently in the acquisitions marketplace. Databases and networks will be used to capture and act on new book ideas; to manage projects in development by teams in remote locations; to allocate paper and presstime, to maintain all the key contacts that are the lifeblood of any publishing company.

And other people's databases and networks will affect publishers' business and business relationships as well. Imagine a retailer making "sales by writer" accessible on their Web site. Or imagine the retailer providing authors with information about how promptly the publisher filled orders for their book, or how successfully they made it available at wholesalers.

Think of the world we're headed to. A big trade publisher's most critical marketing asset is a database of 62,356 reading groups. Or maybe it is ten times that many. Whatever it is, the database contains the email addresses of many of the people in each group, the titles of what they last read and what they intend to read, and each group's own rules and regulations for considering its next book, expressed in a way that permits automated action.

A database like this must be built from many sources: eps, stores, the publicity department, the company's Web site, and through various marketing efforts over a very long period of time. Some of the information will actually have come from smaller reading group databases of companies this big company has acquired. Building and maintaining it is a permanent preoccupation, and will require receiving new and updating information from a variety of sources every day.

When the publisher considers the acquisition of a new cookbook or a new mystery, the resources in that database could well be the single most carefully considered marketing asset. How many reading groups can we reach that might buy this? How many copies can we sell to them? Do we entice them with first chapters, extended excerpts, free downloadable files of the author's last book?

Can a publisher who has no such database compete in a world where all his competitors do? Obviously not.

Fortunately for all of us, the entirely changed world we will all someday occupy doesn't arrive in the next instant moment. And when it does arrive, the companies that will do best in it will not have discovered it that morning. There is both a lot of time to prepare and no time to waste.

It is emphatically not necessary, and also not desirable or possible, for every publishing employee to become a technical expert in databases and networks. It is, however, quite necessary for each publishing company to offer the support that permits a conceptual understanding of the power of databases to be translated into action quickly and easily.

One executive of a large consumer publisher that we talked to for this year's research pointed out how useful it would be if his company's editors could know before they met with an agent all about the house's business with that agency: how many books were under contract, how many dollars in advances were committed, and so forth. This kind of information could provide indispensable leverage, and certainly all of it is "known" to the house. "That's a good idea", we said, "what stands in the way of just getting it done?"

It turns out that the barrier was in communication between operations and IT. On the chance that an editor saw the value in this and initiated it, they would be led through a process by which IT would attempt to cost-justify it and probably attempt to modify the initial request beyond recognition.

This set of difficulties is aside from the scary issues related to the controls on proprietary information; there could be valid concern that a brand new intern could discover such information and commit some indiscretion with it which could damage the house. And, of course, one can't just dismiss the possibility that open networks of this kind, even if open only inside the company, could invite espionage or sabotage as well.

This is a prototype of the modern publishing management challenge. Neither the editor with the right idea nor the IT department really has the tools or perspective to evaluate the opportunity or to determine what controls need to be instituted along with it. It is typical of what arises when we try to do business better by harnessing databases and networks; change creates a set of corrolary complications.

Publishers today live in a period where they are forced to change the way they think about and see their business, because the business itself is changing at warp speed. Seeing the primacy of databases and networks, and envisioning them as a key organizing principle, helps interpret change in a practical way that leads to understanding what needs to change about what publishers actually do. Indeed, merely acknowledging as a fact that competent command of databases and networks defines a topic of concern for every top manager may be the next important change in some publishing houses. Only with this vision can they assure that the technical resources are at their disposal to address their needs as they arise and that top management is appropriately involved in defining the issues and how the business's planning and processes must adapt to them.