Pagination and the newsroom: A question of time

Pagination and the newsroom: A question of time

Russial, John T

To a publisher, time is money; to an editor, time is quality.

In newsrooms today, technological change has made these simple equations more complex. Editors whose basic tools once were blue pencil, scissors, glue pot and proportion wheel now use an ever-expanding array of computer equipment to perform their jobs. VDTs, on-line library systems, pagination systems, electronic photo desks and commuter graphics terminals fill newsrooms.

Some of these innovations save time for editors. On-line library systems, for example, enable editors to search through news clips efficiently and to check background information easily. Data networks and Macintosh computers make it possible to produce and modify informational graphics quickly on site. But some computer-based systems have placed new time constraints on them. Many editors and reporters despised the optical character reader systems–the infamous scanners of the early ’70s–because OCR turned rewriting and editing into time-consuming and cumbersome chores. VDTs were a dramatic improvement for reporters as well as editors, but they added to the bundle of tasks editors were required to perform(1) and left them less time for copyediting and headline writing.(2) Sooner or later, many papers found that they needed to add positions, particularly on copy desks.(3)

Another computer innovation–pagination–began to appear in newsrooms in the early-to-mid 1980s. Pagination transfers the work of cutting and pasting to editors, who create page dummies on computer screens, then cut and paste electronically. It shifts control of the page-production process to the newsroom, creating what has been called a second VDT revolution. (4) Many editors and publishers saw it as the capstone of newspaper computerization, fulfilling predictions made as early as 194 that “the computer will remove a great deal of the drudgery that exists in the newsroom today and free all people in the editorial department to be far more creative.”(5)

Throughout the 1980s, advertisements, trade journal articles and conference presentations promoted pagination as the latest cost-cutting, timesaving device. But anecdotal indications began appearing that pagination was difficult to learn and manage and might be adding to the burden of editors.


Pagination did not sweep the country in less than a decade, as VDT systems had done in the 1970s. Year after year, predictions that many, if not most, papers soon would be paginating proved to be too optimistic. The reasons commonly cited were high cost, the difficulty of integrating pagination with other newsroom systems and problems with handling photos.

Information about the adoption of pagination by U.S.

dailies has been sketchy from the beginning. It consists chiefly of updates in the trade press, reports of technological advances and anecdotal articles on the experiences of individual newspapers. The first daily to completely paginate the Pasadena Star-News–began paginating in 1981. By the late 1980s, a substantial, though still rather small fraction of U.S. dailies paginated. From information in Editor & Publisher Yearbook, trade journals, pagination vendors and other sources, 90 dailies were using proprietary pagination systems in 1989.(6) Other papers, mostly small ones, used off-the-shelf, desktop publishing hardware and software to paginate. According to one estimate, as many as 200 U.S. dailies used pagination of some sort by late 1988.(7) Many more newspapers use pagination today, but it still is difficult to say just how many. A recent survey of the Society of Newspaper Design membership reported that nearly two-thirds of responding newspapers did at least some electronic pagination.(8)


Pagination, as it is used in most daily newspapers, is a process by which elements are placed one by one on electronic pages. Headlines, captions, stories, photos and informational graphics (or space allocations for photos and graphics) are handled as individual elements in pagination.(9) This electronic element handling appears to be the clearest example of the part of pagination that is, by nature, production.

To be sure, copyeditors and copy chiefs dealt with stories, headlines and captions as individual elements before pagination. A copyeditor might have worked on a single headline, then sent it to the slot, who would have checked it over, perhaps modified it, and then set it in type. The back shop took it from there, trimming the excess phototypesetter paper and pasting the headline onto a page in its proper location, properly aligned. With pagination, the copy desk role is roughly the same, but now somebody in the newsroom (whether it is a copyeditor, a news editor or a paginator) handles each element again by electronically placing it on a page and perhaps making an adjustment or two.

Pagination, it is generally acknowledged, saves time in the total page production process. Typesetting a Page in one burst is quicker than setting the stories, headlines and captions for a page one at a time and pasting them together later. The process saves time in the back shop, and it enables papers to cut positions there. Whether pagination also saves time in the newsroom is a different question, one that has significant implications for editorial quality. If pagination takes considerably more time and more editors are not hired to make up for that loss, quality will likely suffer.

It is clear that editors who paginate spend some amount of time doing electronically what compositors did with Xacto knives. The question is whether that amount is trivial or substantial. This study, conducted in 1988 as part of a wider examination of the impact of pagination on newsrooms, sought to answer that question.(10) Timing of pagination for a variety of pages under deadline conditions at 12 daily newspapers showed that the process took editors additional time –about 15 minutes per page, on average. However, a study of Pagination, or of any other newsroom technology is inevitably a snapshot of a moving target. This study is a picture of pagination at a dozen newspapers in the late ’80s. As systems improve, pagination may take less time, but it is unlikely that the makeup burden that has shifted from the back shop to the newsroom will ever be erased by technological improvements.


The author selected 12 papers that used proprietary pagination systems(11) to represent a cross-section of U.S. dailies based on circulation size, length of pagination experience and type of system. They ranged from the North Penn Reporter in Lansdale, Pennsylvania (circulation 18,640), to the Philadelphia Inquirer (508,496 daily). Most of these paginated all pages every day; several did all but certain pages, and one, which had just begun pagination, did only several pages a day. They represented the range of pagination experience in the country at the time, from several months to about seven years. They also represented the most commonly used systems. Several paginated photos as well as text; most paginated all text and pasted photos in afterward. Choosing two geographic clusters of papers–seven from the East and four from the far West, as well as the Daily Oklahoman–provided a way to meet these criteria at reasonable travel cost.

The papers were visited on weekdays in spring 1988 for at least one full production cycle at each. To assess how much time editors spent paginating, the pagination process for a variety of pages was timed. The intent was to measure how long it took to perform the more mechanical tasks of pagination–creating an electronic page dummy, placing the elements of type and in some cases graphics and photos onto the page, making necessary adjustments and outputting the page.

An initial goal of this study was to compare the page layout process at papers that paginate with the same process at those that did not. However, visits to several newspapers without pagination systems made it clear that such a comparison would be impossible. It was difficult to disentangle the creative process of designing a page from the physical act of drawing the page. It also was difficult to differentiate page layout work from other desk work, such as scanning the wires, selecting stories and chatting with co-workers. At pagination sites, on the other hand, it was relatively easy to measure the time spent paginating. Pagination was done at a separate workstation–a page layout terminal. It was easy to see interruptions and to stop timing during them.

Different page-production practices among papers in the sample, and in some cases within papers, offered a way to make a similar comparison. At some of the 12 payers, pages were designed by editors, who then gave hand-drawn dummies to a pagination terminal operator. That individual, sometimes called a design assistant, sometimes a paginator, would then perform the mechanical aspects of pagination. At other papers, editors designed the pages on pagination terminals and performed all of the mechanical functions themselves. Some payers used both approaches. Whether a paper paginates or not, its pages still must be designed. The internal page-production differences at paginating payers offered a way to distinguish pagination time from page-design time–the distinction necessary to assess the impact of the pagination process itself on an editor’s time.

In the study, 66 pages were timed with a stopwatch. Paginators were not told they were being timed. Pagination took place only at special pagination (page layout) terminals and it was clear when a paginator was working on a given page. Timing presented no substantive problems–except at the two largest papers, where the type of system and intricacy of copy flow made timing a variety of pages difficult.(12) Also, the Phoenix Garette paginated blocks of material already coded with area composition formats. This shifted the cut-and-paste work to the copydesk, and thus was difficult to measure.

Pages represented a mix of size and section–from full pages to one-column inside pages. As many as possible were timed at each site. Limitations were the number of people paginating at a given point in the news cycle and the time required to complete other parts of the study, which included extensive interviews with editors. Fifty-two of these pages were used for analysis–the most comparable in terms of tasks done at a pagination terminal. They do not include about 10 pages for which copy was edited on pagination terminals, because this nearly doubles the per-page time spent at such terminals. They also do not include several Pages that posed other comparison problems, such as a section front designed and paginated by an editorial artist. Such pages tend to take much longer than pages laid out by news editors.

Figure 1 shows how long it took to paginate pages.(Figure 1 omitted) The difference between the first two bars of Figure 1 (between all pages dummied by hand and all those dummied on a page layout terminal) is about 8 minutes. Assuming that the process of thinking about what a page will look like varies little whether one draws it by hand or electronically draws it on a pagination terminal, this 8 minutes can be interpreted as page design time.(13) This part of the process would appear to take relatively little time. This is generally true for pages with little news space available. The next two bars show that when pages are of substantial size (between 4 and 6 columns), the difference is greater (between 15 and 16 minutes on average).

Based on newsroom practices, it is reasonable to adjust this average page time of 8 minutes slightly. An additional several minutes should be added to allow for creative adjustment of the design on a page layout terminal. This is essentially part of the design process.(14) The remaining time, on average between 20 and 25 minutes per page, is spent on other tasks, primarily electronic page-makeup tasks. These include inputting page geometry, placing elements on the electronic page, adjusting type and other elements and copyfitting.

On non-paginated payers, assuming that the creative process takes about the same time, that a few minutes are spent sketching out the page, and that a small additional amount of time, say 5 minutes per page, would be needed to make trims in the composing room, it would take editors about 15 minutes less time, on average, per page. It is difficult to accurately measure the time an editor spends checking a given page in the composing room. Editors commonly look over several pages, of ten going from one to another to make sure photos are in the right places, headlines are spelled correctly, etc. Time spent in the back shop often includes other activities, such as trading insults with compositors. The biggest difference between editors’ back-drop duties on paginated vs. non-paginated papers is in trimming stories. On paginated papers, stories are trimmed to fit in the newsroom either by the paginator or by another editor. Observations of editors at paginated papers as well as at six matched non-paginating papers indicated that editors who paginated spent nearly as much time in the back shop as their counterparts on non-paginated papers of comparable size did. In some cases, editors on paginated payers said they spent more time, because job reductions left too few compositors to finish pages efficiently.

In any case, few editors spent as much as 30 minutes per edition in the back shop, and in general, editors had multiple pages to oversee. The actual difference is probably less than 5 minutes a page, but even if it were as much as 10 minutes, which is very unlikely, the extra time spent paginating would still be 10 minutes a page.

Figures per average page are given because the page is a commonly used unit of analysis in daily newspaper production. It is possible, however, that the average suggested by the first two bars of Figure 1 could be misleading. For one thing, the average page figure of 20-25 minutes might be a little high, because the ratio of all (six-column) pages timed is slightly higher than it was in those papers on average.(15) Different news organizations vary somewhat in how they distribute columns of editorial space throughout the paper. Some have proportionately more six-column pages than others, and on some days they have more of these full pages than on others. The average pagination time of 20-25 minutes, based on Figure 1, reflects a ratio of full pages to total pages of about 1:3. In general, that ratio for weekday issues is probably a bit lower, but no lower than about 1:4, based on the papers in the sample.(16)


These average page figures indicate that the act of putting the elements on the electronic page (the makeup job transferred from the back shop to the editorial departments) is far from trivial as far as an editor’s time is concerned. It takes longer than the creative process of designing a page, about twice as long, on average.

Information gathered from interviews with editors supported that interpretation. The design director of one of the large papers estimated that in a week paginating took 320 hours and page designing 112 hours.

At this paper, a department of 18 designers and assistants paginated between 1,000 and 1,400 pages a week. The design director estimated that 60 percent of the work his design desk did was work that once had been done in the composing room. The paginating time was almost triple the page-designing time, perhaps because this paper paginated photos as well as text, and this takes longer.


On papers of all sizes, few editors at any level would say they have as much time to do their jobs as they’d like. To those in production-oriented editing jobs, such as copyediting and layout, time is a particularly crucial commodity. As deadline approaches, time forces a continuing accelerating need to reassess editing priorities. Is there enough time to check this fact, to rewrite a convoluted passage, to move an important paragraph higher in the story, to redummy a page to give a certain story better play? Could a few minutes spent rewriting a headline be better spent editing a last-minute story, or making a call to the city hall reporter to double-check a name or to challenge an interpretation? Editors face a hundred such decisions every day, and much of their value to newspapers lies in their ability to make the mix of decisions that will have the best payoff in editorial quality.

Pagination may be changing the parameters of such decisions, adding a variety of tasks that simply must be done if the paper is to be published. An editor cannot fail to perform the 10, 15, 20 or more minutes of keyboarding and graphic manipulation needed to create an electronic page, even if he or she feels that the time could best be spent on another editing task.

Pagination saves overall production time, but for the most part that time is not saved in the editorial department. Editors may save a little time in the composing room, but they spend more in the newsroom doing electronic makeup chores–at least 10 minutes a page more and probably closer to 15.

This added time is not trivial. At a medium-size paper whose editorial department paginates 50 pages a day, 10 minutes per page is 8-1/2 hours of work that had been done in the back shop. In practical terms, that is more than a full shift a day. For a large paper that produces as many as 200 pages a day with zoning and edition makeovers, it would be about five shifts a day. Using the 15-minute-per-Page figure suggested by this study, staff time would increase proportionately in this sample, according to interviews with managers, 9 out of 12 payers had added editorial staff because of pagination in numbers generally proportional to circulation size. Several of these did so, however, only after months or years of doing without added staff. Pagination, incidentally, is not marketed as a technology that typically leads to newsroom staff increases.

It is important to note that pagination is not simply an added technological burden on editors. Many editors interviewed during this study said they appreciated the increase in control over the page that pagination allows, and many felt it enhanced design flexibility and creativity. Many remarked that pagination enabled them to make better trims and to produce pages that looked straighter–when they had the time to use it properly. But they also said pagination increased their workload and reduced the time they had for traditional editing tasks.(17) It appeared to increase the proportion of must-do tasks, and, in particular, the proportion of such tasks that were relatively mechanical or routine. The addition of editorial staff members reduced that burden somewhat, but it remains to be seen whether that pattern will continue in a period when staff freezes and cutbacks are becoming commonplace.

This study is limited in several ways. Practical considerations ruled out random selection of papers to be studied or pages to be timed. Thus there is no way to say with any degree of probability that another sample of pages will produce the same result. Also, the measurement method is somewhat imprecise. It is impractical, if not impossible, to do a completely accurate time study of journalists at work, if for no other reason than much of the work is done inside the head. This was clear at the largest two papers, where the type of system in use and the type of copy flow setup made it difficult to disentangle the work of pagination from the more traditional work of editing.

One can argue that the extra time is a function of the newness of the technology. Any new technology will take people more time at first–until people learn to use it efficiently and until improvements are made. That is certainly true with pagination. There is a learning curve at individual papers and industry-wide. But pagination was not new to most papers in the sample, nor was it new to the industry at the time of the study. Several of the papers in the sample had been using the same pagination system since the early ’80s.

The direction of proprietary pagination development since this study was done has been toward systems that enable more of the workload to be shifted from paginators to other editors in the newsroom. Desktop publishing vendors are moving in this direction as well.(18) Does this mean pagination will no longer take more time? That is unlikely. It probably means that it will be more difficult to determine whose time it is taking.

This study may have to suffice as a snapshot of a time when the pagination process was more clearly differentiated in newsrooms and easier to measure. Though limited, it does offer a strong suggestion that pagination takes more than a negligible amount of an editor’s time. System improvements may reduce those extra minutes but it is unlikely that they will completely eliminate them. In the computer age, there’s no free lunch. And newspapers cannot afford to upgrade every time a better system comes along. Many of the papers in the sample are using the same technology now as they did in 1988.

It is difficult to quantify newspaper quality. To an editor, time is a reasonable surrogate for quality. Pagination, as well as other newsroom technologies, bears further study, because if it continues to take up those extra minutes and additional editors are not hired, quality will almost surely suffer.


1. William S. Solomon, Technological Change in the Workplace. Unpublished doctoral dissertation, University of California-Berkeley, 1985.

2. For summaries of the impact of VDTs on reporters and editors, see Judee K. Burgoon, Michael Burgoon and Charles K. Atkin, The World of the Working Journalist. New York: Newspaper Advertising Bureau, 1982, pp. 100-105; and David Weaver and G. Cleveland Wilhoit, The American Journalist, 2nd edition. Bloomington, Indiana: Indiana University Press, 1991, pp. 146-159. Also, Copyeditors to Feel Automation’s Impact. Editor & Publisher, March 19, 1977, p. 16.

3. Nancy M. Carter, Computerization as a Predominate Technology. Academy of Management Journal, 1987, pp. 247-270.

4. Robert Radolf, A Second VDT Revolution. Editor & Publisher, June 23, 1984, p. 30.

5. Automation, The APME Red Book 1964. The Associated Press, p. 50.

6. These are hardware and software systems designed specifically by vendors to do pagination. Hastech and Harris were the two most commonly used. The number of papers using proprietary systems was probably greater than 90, but it could not be determined more accurately because some vendors would not release customer lists and the standard sources, such as E&P Yearbook, were not up to date for all papers.

7. Interview with Eric Wolferman, manager of pre-press systems for the Gannett Co., Dec. 2, 1988.

8. Pagination Survey, Editor & Publisher. April 3, 1993, p. 27. The 600-member survey had a 35 percent response rate.

9. The Phoenix Gazette made extensive use of area composition formats. Instead of placing headlines, captions and stories separately, the designers used formats created on the front end system to allow the copy desk to place a story and its headline and caption in a single file that could be imported into the pagination system and placed on the electronic page as a unit. This seemed to be a way to simplify the process, although it created additional work for the designers, whose job it was to calculate the wraps and put the length information into the formats. Editors at other papers using the same pagination system said they doubted that it saved much time.

10. John Russial, Pagination and the Newsroom: Great Expectations. Unpublished doctoral dissertation, Temple University, 1989. Besides issues of time, the study considered the impact of pagination on deadlines, creativity, staff organization, management, training and editorial control.

11. Systems represented were Hastech, Harris, Information International and Atex, the four dominant proprietary pagination systems in use in the country at the time, and Autologics, an unusual total pagination system in production at one U.S. daily (Morristown, New Jersey) at the time of the study. Many small dailies and weeklies did partial or total pagination with Macintosh computers and off-the-shelf software packages.

12. Few pages were timed at the two largest papers–at the Philadelphia Inquirer, because of the use of the Atex system, and at the Arizona Republic, because it had a more elaborate copy flow than most papers. At the Republic, stories that needed trimming often were sent from designers back through the copydesk and slot, sometimes more than once. Atex presented a special problem of timing. Because of the way the system is designed (as part of the front end system) and used, the added time (the added workload) can be distributed among copyeditors and layout editors.

13. Observations at the 12 paginated papers and six non-paginated papers also visited suggest this assumption is warranted. Several editors said they felt they could design faster on a page layout terminal: others said they felt they couldn’t. In any event, the difference, if any, should be negligible.

14. This would be comparable to a layout editor redrawing a page on a non-paginated paper.

15. Full pages are overrepresented because one focus of the study was to observe how each paper handled several pages of high complexity.

16. A recalculation for the same pages based on columns rather than pages showed that the difference was proportionately about the same on a per-column basis. This supports use of the average page figure. In papers in the sample, all weekday editions, that ratio varied from about 1:3 to 1:4. The ratio on Sunday papers, which tend to have proportionately more advertising, is probably lower.

17. This is based on open-ended interviews conducted with more than 50 editors who paginated. In a study of pagination and job satisfaction, Marlin Shipman and Gil Fowler reported a similar perceived increase in editors’ workload with pagination. Shipman and Fowler, paper presented at the American Newspaper Publishers Technical Exposition, Las Vegas, Nevada, June 1987.

18. Eric Wolferman, Pagination Systems Have Moved to Desktop. Editorially Speaking (Gannett publication), February 1993, p. 7.

Copyright Ohio University Winter 1994

Provided by ProQuest Information and Learning Company. All rights Reserved