Lack of Excellent Display Technology Means There’s Still no Paperless Office

When Johannes Gutenberg thought up the printing press in 1450, he had only one application in mind: automating the manufacture of indulgences, the get-out-of-damnation cards then sold to anxious sinners by the church. As it turned out, the indulgence business had a scaling problem?mass production turned what had been a tolerated irregularity into the perception of complete corruption. The cultural reaction wrecked not only the market but very nearly the church itself. However, his printing system sidestepped this unpleasantness, spreading within a matter of decades to every corner of civilization. By the 19th century the printing press became an essential arm of commerce, responsible for everything from forms to catalogs to money.

In the Oct. 1, 1992, issue we ran an article covering the implications of the most recent twist in our relationship with print. More and more enterprises were starting to think of information technology as organized around decentralized groups of PCs. It seemed only natural to manage printing the same way, which meant moving the function out of central, often remote, departments and into workgroups. The growing number of multifunctional devices (which could print, copy and scan) on the market reinforced the idea by making it possible, at least in theory, to move copying out of copy centers and onto the edges of the enterprise as well. If computing was moving closer to users, didn’t it make sense for printing and copying to follow?

That’s how it seemed to us and to the many vendors whose products we examined in our piece. The actual realization of this vision has been strikingly slow. According to Bruce Dahlgren of printer manufacturer Lexmark, the large, central printing center remained a standard until very recently. There are numerous reasons why that might have happened. Centralization brought economies of scale and the convenience of single points of management. Software that supported network-based printing proved hard to write. Moving printing and copying to the edge of corporate networks sometimes required tinkering with the backbone of the organization (for example, moving copying from the facilities department to IT), and those changes are always slow.

However, the same arguments had been trotted out during the transition to microcomputer-based networks, and that movement took off like a prairie fire. A more subtle possibility for lingering print centralization is that in the early to mid-’90s everyone knew in their bones that networked PCs were the future, whereas most managers’ bones fell silent when they were asked about printing. Even if you didn’t believe in a paperless office, the role of printing in the 21st century enterprise was very cloudy. Given that murkiness, it just made sense to keep pushing decisions about the technology to the bottom of the pile.

During the past 18 months or so the situation has changed. A consensus on the future of printing has finally begun to emerge: Printing does have a future, limited in one way, but nearly boundless in another. Printing has historically had three jobs: information storage, distribution and display. At least in the business context, display has always been the poor sister; changes in printing technology were bought and sold on the promise of improvements in storage and distribution efficiencies (like microfiche).

Today the huge improvements digital technology has made to information storage and distribution are obvious. Yet no one has developed display technology that can rival, let alone improve on, paper’s convenience and comfort level. As a result, information display has become printing’s core competence and responsibility.

At the same time, IT-led enhancements in storage and distribution have increased the load on display, since the more information computer networks can remember and deliver, the more users end up printing. One illustration of this effect is the recent success of color printers. We devoted a lot of lines to color in our 1992 article, and for all the sales that followed in ’93 and ’94 we might has well have saved the ink. It was not until the Web started routinely delivering graphics and color (such as Microsoft PowerPoint presentations) that color-printing sales accelerated. According to IDC (a sister company to CIO’s publisher), the fraction of color printers shipped to U.S. businesses quadrupled between 1996 and 1999.

In just a few years enterprise printing has changed from a store-and-forward batch process?responsible for creating large numbers of relatively few, long documents to be stored, distributed to many people and then stored again?to one expected to produce large numbers of short documents, whenever needed, at or close to the point of consumption, that will be read by one or two people and then discarded.

This emphasis provides a new and more stable rationale for our premise in the 1992 article: that it makes sense to treat printers like PCs, as devices distributed over a network. It also erases the lines between copying, printing and faxing, since from a display point of view, all these technologies raise the same issues (speed, resolution and price). A device that prints digital files doesn’t care which technology generated them or whether they were transmitted across a network or created internally by a built-in scanner.

There is one question still up in the air: What is the optimal ratio of printing devices per employee for a given enterprise? Xerox believes that at least with current technology, economies of scale and versatility still matter. “Smaller printers are cheap, but the cartridges are not,” says Art Smith, vice president of Worldwide Software Solutions at Xerox Office Systems Group. (He estimates that a large machine can print at half the cost per page of smaller models.) Companies such as Lexmark represent the other end of the spectrum. They imagine devices similar to a distribution of devices so dense that it approaches that of fax machines today. From an IT manager’s point of view, both ends assume a high degree of network integration, with features such as automated management of driver libraries, networkwide access to status messages, a faxlike system for matching jobs to locations, and potentially, support for wireless inputs and kiosk-based printing. Tools to manage these issues, like Novell’s iPrint, are beginning to emerge.

It may be only a coincidence that the reemergence of paper-based printing coincides with the collapse of tech stocks. But given the digital triumph that many of these formerly high-flying stocks embodied, it is certainly fitting.