Pablo J Boczkowski. Handbook of New Media: Social Shaping and Consequences of ICTs. Editor: Leah A Lievrouw & Sonia Livingstone. Sage Publications. 2002.
Print papers are one of the oldest, most ubiquitous and most highly standardized elements in the modern media landscape. Newspapers have changed considerably since Timotheus Ritzsch established the Einkommende Zeitung—‘incoming news’—in Leipzig, Germany, in 1650, which according to Smith (1979) was the first daily publication. Furthermore, these changes have been part and parcel of some of the most significant societal transformations in the past centuries from the rise of nation-states to the emergence of mass production and consumption, and from the development of large-scale communication infrastructures to the advent of modern urban life (Anderson, 1991; Blondheim, 1994; Schudson, 1978; Strasser, 1989). However, amidst these changes in format, content and production processes, one feature has remained remarkably stable: information has been delivered via ink on paper, an element intimately tied to a complex ensemble of discourses, practices and artifacts regarding the construction and appropriation of information.
Newspaper firms tinkered with various alternatives to newsprint during the twentieth century, from the Telefon Hirmondó phone-based newspaper in Hungary at the beginning of the century (Marvin, 1988), to the radio-based facsimile editions of the Buffalo Evening News, Dallas Morning News, Miami Herald, New York Times and St Louis PostDispatch in the United States in the 1930s and 1940s (Shefrin, 1949), to videotex newspapers starting in the United Kingdom in the late 1970s (Schneider et al., 1991). However, none of these endeavours moved far beyond the experimental domain until the newspaper industry appropriated the World Wide Web circa 1995 to create the first non-print alternative to achieve widespread development and use. The evolution of online papers on the web has made visible taken-for-granted aspects of print’s production culture, provided a window into the emergence of new regimes of content creation, and allowed the examination of broader patterns in the construction of information on the Internet.
Given the significance of this challenge to the dominance of ink on paper, it is not surprising to find an emerging but already profuse body of literature on the production and consumption of online papers. Despite such a burgeoning of scholarly activity, little has been done to assess what has been found so far, an essential element in critically planning the direction of future work. In this chapter I contribute to filling that void by offering the first comprehensive review of research on online newspapers. Whenever appropriate I will also address work concerning contemporary efforts by broadcast media to extend their news franchise in the newer information environments. Moreover, as I will discuss in the final section, since the boundaries separating print and broadcast—as well as other key categories in our understanding of media—have been very much at issue in the development of online papers, this essay also aims to contribute to work on the larger transformations currently shaping the media industries.
Despite the existence of past relevant phenomena and secondary sources, current research predominantly focusing on online newspapers has tended to overlook historical matters and possible linkages between past and present findings. To overcome this limitation, and despite the more contemporary focus of this review, in the next section I will briefly assess the results of inquiries about the most immediate antecedents to current developments—those initiatives beginning with the advent of videotex and teletext in the 1970s. Then, I will examine studies about contemporary phenomena, dividing them into three sections: production, use and interactivity. I will conclude by addressing the existing literature’s limitations, and suggesting directions for further research.
A caveat. Current research about online newspapers—much like the object of study in itself—is evolving rapidly and in many directions. This exacerbates the challenges of a review essay, from difficulties in locating the relevant literature to the risk that any given assessment could be inaccurate had a yet not identified piece been taken into account. However, it is my contention that more reviews like this one are needed precisely because both object and inquiry are far from being stable: the more unstable the phenomena and their analyses, the more effort should be put into mapping the territory before embarking upon any particular journey—as well as while it is taking place. A relatively low degree of cross-citation among contemporary studies, and the even less sustained discussion of the historical record, have turned a significant part of current research into a collection of monologues. My hope is that this essay will contribute to begin transforming them into a field of interconnected conversations.
Before the Popularization of the World Wide Web
The 1970s witnessed considerable activity in the creation and deployment of electronic means for dissemination of text and graphic information. Most developments focused on two technical alternatives: videotex and teletext. The first, originally generated at the British Post Office early that decade (Campbell and Thomas, 1981; Sommer, 1983; Wilkinson, 1980), consisted of ‘computer-based interactive systems that electronically deliver screen text, numbers and graphics via the telephone or two-way cable for display on a television set or video monitor’ (Aumente, 1987: 14). Teletext—also developed first in the United Kingdom, at the British Broadcasting Corporation (McIntyre, 1983)—was ‘a one-way system for the transmission of text and graphics via over-the-air broadcasting or cable channels for display on a television set’ (1987: 19).
These developments took place at a time when the ‘information society’ rhetoric was popular in both the press and scholarly works. In a supposedly emerging era of information overload, ‘the basic premise of videotex and online publishing is that the information-rich society has unmet information needs’ (Neuman, 1985: 8). Thus, these new technologies became tool and symbol of an epochal change: if print had been an integral part of the industrial society, videotex and its electronic cousins were thought to be constitutive of the upcoming information society (Case, 1994). According to Sigel, ‘since its commercial introduction in 1976, videotex has become a metaphor for a new world of information dissemination … [in which] videotex stands for the future, whereas older methods of transmitting information—mainly, print on paper—represent the past’ (1983a: 1).
Coupled with such an information society rhetoric there were two factors contributing to newspaper firms’ exploration of electronic publishing. The first was a growing trend towards computerizing production and distribution functions across the newspaper industry since the 1960s, which meant that information was ready for electronic redeployment (Dozier and Rice, 1984; Marvin, 1980; Smith, 1980; Tydeman et al., 1982; Weaver and Wilhoit, 1986). The second was increasing concern about the long-term prospects for ink on paper as a delivery vehicle triggered by developments such as rising newsprint costs, less homogenized consumer tastes challenging mass advertising, and the growth of competitors in the markets for news and advertising (Albarran, 1996; Baer and Greenberger, 1987; Compaine, 1980; Picard and Brody, 1997; Pool, 1983; Smith, 1980; Stone, 1987).
Given these discursive, technical and socioeconomic trends, it is not surprising that newspaper firms in many industrialized nations took part in videotex and teletext efforts, especially between the mid 1970s and the mid 1980s. The most common scenario—prevalent in nations such as Canada, France, Germany and the United Kingdom—was that these firms participated in state-sponsored initiatives, providing content and editorial expertise (Charon, 1987; Desbarats, 1981; Marchand, 1987; Mayntz and Schneider, 1988; Miles, 1992; Sigel, 1983b; Tyler, 1979). In a minority of countries—most notably the United States—where the state had very limited involvement and the private sector took the front seat, several newspaper organizations were very active players in videotex and teletext endeavours, doing everything from hardware and software development to marketing and commercialization (Aumente, 1987; Baer and Greenberger, 1987; Branscomb, 1988; Criner, 1980; Davenport, 1987; Johansen et al., 1980; Mantooth, 1982; Noll, 1980; Sigel, 1980).
Regardless of the type of involvement pursued, these attempts by newspaper firms to experiment with consumer-oriented electronic publishing provide valuable information about the immediate precursors to contemporary online papers, and an important window into the dynamics of innovation in new media in general. To begin, usage of news content was lower than both other types of information and what newspaper people expected. For instance, during a joint project between the Associated Press and CompuServe to publish videotex editions of a dozen print papers between 1980 and 1982 (Hecht, 1983; Laakaniemi, 1981; Mantooth, 1982; Patten, 1986), ‘news accounted for less than 10 per cent of the average test family’s time on CompuServe network … [and] this usage was heavily skewed toward a few respondents; one out of ten households accounted for half of all news reading’ (Blomquist, 1985: 423). Furthermore, usage dropped dramatically after a period of initial enthusiasm, suggesting that the services did not add much value to already existing counterparts: ‘although electronic news may have been available sooner than its printed counterpart, it was no more comprehensive, contextualized or process-centered than any other form of news’ (Ettema, 1989: 111).
This leads us directly to a third characteristic of consumer-oriented electronic publishing during the first half of the 1980s: the reproduction of print newspapers’ content into the new platform. In her dissertation study of the Associated Press and CompuServe project, Mantooth found that ‘the news stories were for the most part word-for-word duplicates, and aside from the headlines which had to be rewritten from the print version, the only really different face of the electronic newspapers was … the long-term storage of such items as reviews and recipes’ (1982: 90). Brown and Atwater undertook a content analysis of the three largest electronic newspaper initiatives in the United States—Knight-Ridder’s Viewtron, Times-Mirror’s Gateway, and Field Electronic Publishing’s Keyfax—concluding that ‘wire services and newspapers were the sources of news stories on all three services. No evidence of stories originating from the videotex staffs was found’ (1986: 558). This was also consistent with Weaver’s conclusions about developments in the European scene: ‘journalists working for teletext systems in the United Kingdom and the Netherlands do almost no independent reporting. They rely primarily upon what others are writing in various wire services and newspapers’ (1983: 53). According to Overduin, ‘For the videotex journalist, the news of the day is “given” and the main journalistic activity is that of “massaging” or editing and rewriting that news to locate it in the database and fit it onto the screen in printed form’ (1986: 231).
Another research finding was users’ interest in employing videotex capabilities to communicate among themselves about a wide array of issues. The contrasting fate of two major videotex projects highlights the importance of this matter. The French government sponsored a nationwide videotex initiative, Télétel, since the late 1970s. The system was originally conceived as a way of providing users with an array of useful data; hence its initial configuration did not prominently feature one-to-one and many-to-many message flows. However, in a 1982 trial in Strasburg partly managed by a local paper, Les Dernières Nouvelles d’Alsace, a ‘fortunate accident’ happened:
The newspaper’s technical manager developed an application to correspond on line with users so as to help them or to monitor their uses. Service users quickly discovered that it was also possible to communicate in real time with each other. This application soon became so popular that it attracted users from all over France although it was only available through a local telephone number. Later it was reproduced by many other information providers. (Vedel and Charon, 1989: 101)
Analysts have suggested that the capacity to enable communication among users was key in the comparative success of the French videotex system vis-à-vis its contemporaneous European counterparts (Schneider et al., 1991). A different path unfolded at Viewtron, a joint venture of Knight-Ridder and AT&T in the United States between 1980 and 1986, which consisted of news, information and services such as shopping and banking, transmitted over regular phone lines to a dedicated terminal equipped with a keyboard. Viewtron afforded a multiplicity of message flows, and users showed interest in communicating among themselves. However, immersed in a media culture where one-to-many flows dominated, and unable to move away from it, Viewtron officials failed to fully exploit users’ preferences. For instance, a member of the Viewtron team reflected about this issue in the following terms:
In retrospect, the interviews and usage data clearly revealed that access to databases of general news, information, and advertising was less exciting to subscribers than the ability to easily communicate with other subscribers. But that was not what anyone was prepared to hear at this time. Nearly everyone involved in the trial saw Viewtron as an advertiser-supported electronic newspaper. Its potential role as an interpersonal communication medium was considered secondary. (Fidler, 1997: 148)
Such a disregard for many-to-many information flows was widespread among American newspapers engaged in videotex efforts during this period, which let their potential be exploited by the then nascent online services—much of whose ulterior success was predicated upon user-authored content. To Aumente,
Publishers steeped in newspaper traditions, seeking to transfer the same indexes of news, weather, sports, features, and commentary to the screen, paid dearly for a blindsided approach, while the growth of the personal computer and telecommunications segments stressing interactivity, messaging, and innovative communication modes took root. (1987: 114)
Another feature of newspapers’ videotex and teletext developments during this period was the disparity between technical and commercial performance: whereas most services functioned reasonably well, none of them achieved significant market success. This lack of commercial results was attributed to factors as diverse as ignoring users’ preferences, the high cost, low speed and limited capabilities of reception devices, and the fact that the services did not add much value to existing alternatives (Cameron et al., 1996; Carveth et al., 1998; Davenport, 1987; Ettema, 1989; Greenwald, 1990; Patten, 1986). The commercial fate of these initiatives ‘convinced the publishing industry that they faced no threat from electronic text’ (Baldwin et al., 1996). Some analysts attributed the termination of many videotex and teletext services by the second half of the 1980s to a defensive approach that most newspaper firms took regarding the consumer-oriented electronic products: they tinkered with them more to protect their existing position in the marketplace than to conquer new ones. To Carey and Pavlik, ‘many newspapers developed videotex services not as a positive step forward but out of fear that videotex might replace their core business. When it became clear that the perceived threat was illusory, they retreated quickly’ (1993: 165). Thus, Ettema has argued that ‘interest in the technology on the part of newspaper firms … was probably more defensive than offensive … The failures of videotex ventures were, then, at worst a mixed blessing for the newspaper industry’ (1989: 108).
To summarize, research about consumer-oriented electronic publishing attempts by print papers during the 1970s and 1980s has shown that (1) consumers were not highly interested in the news content, (2) usage dropped considerably after the initial novelty wore off, (3) most news content available came from existing sources rather than being created originally for the new media, (4) users tended to employ interactive tools to communicate among themselves, and (5) most newspapers failed to fully exploit user-to-user information flows. Finally, some scholars have argued that online newspapers resulted from a defensive innovation approach more geared towards preserving print’s status quo than building new electronic markets.
The World Wide Web Takes Centre Stage
Consumer-oriented electronic publishing efforts by newspapers decreased during the second half of the 1980s. However, the first half of the 1990s saw a resurgence of activity on this front, triggered by a host of larger societal and technical factors ranging from an increase in the adoption of personal computers at work and home, to the development of the web in 1989, and its first popular browser Mosaic four years later, to changes in telecommunications policy and infrastructure at a global scale (Abbate, 1999; Baldwin et al., 1996; Campbell-Kelly and Aspray, 1996; Castells, 1996; Ceruzzi, 1998; Mulgan, 1991). Although most resources and expectations were focused on various forms of videotex, newspapers simultaneously explored other technical alternatives such as audiotex, CD-ROM, fax and portable digital assistants (Carveth et al., 1998; Dizard, 1997; Molina, 1999; Thalhimer, 1994; Willis, 1994). However, 1995 saw newspapers settling on the web as their electronic publishing environment of choice (Beamish, 1997; Carveth et al., 1998; Garrison, 1997; Martin and Hansen, 1998; Molina, 1997). A count by the Newspaper Association of America showed more than 175 US dailies publishing on the web at the end of that year, a number that grew to over 750 three years later (Newspaper Association of America, 1998; Editor and Publisher, 1996). As Molina wrote: ‘de facto, the Web has become the prime arena for multimedia newspaper developments in the near and medium-term future’ (1997: 208). In a survey of US online newspapers, Peng et al. (1999) have found that the three top reasons for publishing on the web versus other options were (1) availability of a large number of readers worldwide, 57 per cent, (2) ease of publishing, 27 per cent, and (3) superior graphical presentation, 15 per cent.
It should then come as no surprise that most research on online newspapers has also focused on the World Wide Web operations. In what follows I will review this literature by structuring it in three areas: production, use and interactivity. The last of these is the single most examined issue, and one that has blurred the boundaries that more neatly separate production and use in traditional media.
Production
Some authors have called attention to how different characteristics of the web as an information environment relate to features of online journalism. In their study of a large metropolitan paper’s online division, Riley et al. (1998) focused on what they considered to be a tension between the communitarian ethos of the Internet and the steps taken by this firm to exploit its online products commercially. They have argued that ‘while the Web is clearly about blending boundaries between media … the strategy of The Paper is clearly not of opening space but of confining it’. Thus, they described a series of steps this online newspaper took to ‘colonize’ the web, forming what they called a ‘virtual geographic space’. Their study raised a concern that ‘the commodification of information arising from the commercialization of the net and the expanding vertical market architecture is anti-community and anti-knowledge’. From a different angle, Lee and So (1999; 2000) examined transformations in the basic functions of newspapers when they move from ink on paper to pixels on the screen, concluding that some of the main changes relate to the fact that the former mediation role played by print papers is morphing into a stance of more direct participation. Regarding economic issues, this trend is manifested in the fact that for online papers ‘their role of advertisement deliverer is expanded to include electronic trader’ (2000: 22).
Other authors have concentrated on how the web’s information architecture is intertwined with changes in the character of journalism. By contrast to print—with its high distribution costs—the expense for an online newspaper to have its content and applications being accessed by users located 1 or 10,000 miles away is the same. This potentially has the dual effect of both opening up non-local markets to local papers, and having their local dominance challenged by non-local competitors. Chyi and Sylvie (1998) have argued that—as opposed to the dual information and advertising submarkets of print papers (Lacy and Simon, 1993)—the online newspaper market is structured in four submarkets: (1) the local information market, (2) the long-distance information market, (3) the local advertising market, and (4) the long-distance advertising market. Thus, they have recommended a product differentiation and niche-oriented strategy because ‘the World Wide Web offers a seemingly infinite list of choices and renders a substitution marketing fairly useless in the global market’ (1998: 16). A later study by Chyi and Larosa has shown that ‘most people still read the local newspaper in the ink-and-paper format, while national newspaper sites were gaining more ground online’ (1999: 12). Despite this stated preference for local print news, in interviews with online news managers Chyi and Sylvie have found that ‘while larger online newspapers seem to have greater ambition and more confidence in seeking business opportunities at larger geographic levels, most online newspapers operate within the geographic boundaries defined by their print counterparts’ (2000: 9).
Another important research focus has been the study of how the structure of message flows on the web may affect traditional journalistic roles. According to Newhagen:
The hourglass shape of mass media architecture facilitates a power imbalance between message producers and receivers that may not exist on the Internet. In a mass media system such as television or newspapers, journalists are in a position of power over their clientele simply because of their position at the narrowest point of the system’s architecture—the point at which messages are produced. The Internet is a web of interconnected nodes; any user is equally likely to be a message receiver or sender during any given communication cycle. This architecture, then, may lend itself to more parity between communication participants. (1998: 117)
Some scholars have argued that in such an environment ‘the emphasis shifts from “allocution” to “consultation”’ (Bardoel, 1996: 287); thus the usual function of mediating between events and audiences is expanded to include tasks more geared to facilitating information search, creation and exchange by and among users (Kawamoto, 1998; Lee and So, 1999; Li, 1998). In particular, some researchers have called attention to the effect that these changes may have on the gatekeeping process, one of traditional journalism’s paramount characteristics (Huxford, 2000; Newhagen and Levy, 1998; Singer, 1998). As Williams has put it, ‘in a world where everyone can be a publisher, journalists are vulnerable to losing their franchise as gatekeepers of news’ (1998: 34). In a study of a project enabling non-profits to create free websites within an online newspaper, Boczkowski (2000) has shown that the editorial function got expanded to include work processes centred upon the facilitation of content creation by a vast network of heterogeneous production agents, and coined the term ‘gate-opening’ to capture such a modified editorial role.
This leads us to larger issues about the organization and practice of online newsrooms. A survey taken by Jackson and Paul (1998) in 1997 has found that one of the major challenges faced by personnel in online papers was a constantly changing publishing environment and that ‘the rapid pace does little to facilitate reflection or planning, but demands movement and action’. A study by Singer et al. (1999) has shown that these newsrooms are staffed by people younger and with more diverse professional and educational backgrounds than their print counterparts, and receiving financial compensation comparable to that prevalent in traditional media—‘tradepress reports of significantly higher salaries for online journalists may largely be wishful thinking’ (1999: 43). These workers predominantly produce what is known as ‘shovelware’: taking information generated originally for a given print paper’s edition, and deploying it virtually unchanged onto its website (Eriksen and Sørgaard, 1996; Martin and Hansen, 1998; Neuberger et al., 1998; Palmer and Eriksen, 1999; Ross, 1998; Tankard and Ban, 1998). This resonates with what happened in the case of videotex and teletext newspapers in the 1980s. Despite scholarly and anecdotal evidence suggesting that such content reproduction was not well received by users of those services, a decade later online newspapers on the web—at least in their first years—have been following the same path.
Several scholars have specifically looked at two sets of work relationships: between print and online newsrooms, as well as between editorial and advertising and marketing personnel within online papers. Regarding the former, in a survey by Endres (1998), editors of online publications complained about lack of cooperation from their colleagues in the print newsroom. In his examination of editorial routines at newsrooms of three online papers, Huxford has shown ‘significant differences in the cultures in which the print and online journalists operate’ (2000: 7). This was an issue that could have contributed to the dynamics Martin and Hansen have described in their observations of work practices at a handful of online papers:
Coordination between the main newsroom and the online newsroom at the three sites was not ideal. While the newsprint version of a story might include a tease that led the reader to the online version for more in-depth information, there was still a dearth of contact between those reporting and writing stories for the newspaper and those working for the online service. (1998: 117)
Finally, scholars and commentators have also focused on the relationship between the editorial and marketing and advertising personnel of online newspapers (Brill, 1999; Harper, 1998; Pavlik, 1998; Williams, 1998). More specifically, they have raised concerns about the erosion of the separation between these so-called ‘church and state’ components of traditional media. One study that has looked at this issue in detail has been Borum’s (1998) ethnographic research of an online paper. Among other findings, she has argued that:
The content and layout decisions of the online news site were influenced in both subtle and more overt ways by financial concerns … The most important characteristics of this process, as they relate to news content, are (1) a general organizational tone of acceptance, relative to traditional print journalism, about the profit motive’s impact on shaping content, (2) the fact that the online medium itself encourages this influence, and (3) the organizational structure that fails to encourage boundaries and ‘protection of editorial territory’ against the invasion of advertising concerns. (1998: 91)
To summarize, research about production matters has suggested that the commodification of online papers may not be congenial with the supposedly more communitarian ethos of the Internet. It may also include playing a more direct commercial role than the usual advertising vehicle function of the press. Moreover, despite the low distribution costs of information on the web, online papers have mostly focused on local markets. In addition, because of the web’s network information architecture, the dominance of traditional journalistic roles such as gatekeeping may get challenged by the emergence of an editorial function centred on gate-opening processes. Furthermore, according to one study, online newsrooms have tended to employ younger people with more diverse educational and career backgrounds than those in print, while paying them similar salaries; another study suggested that the volatile character of the web was a critical factor in the operation of online papers. Finally, these journalists have mostly produced shovelware, have not had an easy time working with their print counterparts, and have had a less stark boundary with their advertising and marketing colleagues than the more usual situation in print papers.
Use
How do people use online newspapers? Much like any other time in which a new option was added to the media repertoire, one issue that has interested researchers is the effect that usage of ‘new’ media has upon that of ‘old’ media, and vice versa. A study of the Blacksburg Electronic Village users conducted in 1994 has shown that ‘more than twice as many [people] reported a decrease in television viewing time—18 per cent—as said they spent less time reading a newspaper—7 per cent’ (Bromley and Bowles, 1995: 22). According to the authors, ‘this is consistent with the notion that early adopters of interactive computer networks are avid information seekers and thus more likely to reduce media use for entertainment than for news purposes’ (1995: 22). An examination of the early users’ preferences in an online newspaper has also found a relationship between usage and information seeking: ‘media use generally correlated positively with both the electronic preference and satisfaction variables. It should be the goal of electronic newspaper publishers to seek out heavy media users, since they represent the most lucrative initial market’ (Mueller and Kamerer, 1995: 12).
A survey of Internet use of over 500 students at a US university conducted a few years after these studies has painted a different picture, since respondents used the web ‘mainly as sources of entertainment and only secondarily as sources of news’ (Althaus and Tewksbury, 2000: 21). However, the authors have drawn similar implications to the previous studies: ‘our data suggest that while the Web supplements traditional news media, it may be in direct competition with entertainment programming on television … entertainment rather than news content is more likely to lose audience members to the Web’ (2000: 38). Although the three studies have argued that use of online news sites puts television at a ‘higher risk’ than print, they have explained the result employing opposed arguments, which is probably a signal of changes in who has been using these sites—and for what purposes. Research undertaken in 1998 and 1999 by Lewenstein et al. (2000) has concluded that:
Fewer subjects appear to be news junkies in the last two years compared to four years ago. At least this seems so as judged by their other news consumption habits: often they don’t subscribe to daily papers; they don’t do much TV news viewing; they do still listen to radio news programs; they have cancelled some magazine subscriptions. Not many have cancelled newspaper subscriptions, but that’s because some had given up subscribing before taking up online news reading. So online news has brought them back to the news reading fold. (emphasis in the original)
Thus, as the web has become more mainstream, the ‘typical’ user has switched from information seeking early adopters to entertainment-oriented latecomers. This may explain why the previous studies have different results but similar implications. Among other findings of the study by Lewenstein and her colleagues—the first scrolling-screen eyetracking investigation of its kind—were that (1) users initially look at text rather than graphics or other media, (2) banner ads do catch users’ attention for an average of one second, which is enough time to perceive an ad, and (3) users ‘read shallow but wide, while at the same time pursuing selected topics in depth’. This last finding resonates with an earlier study by Aikat (1998) in which he examined the traffic logs of an online paper from November 1995 to May 1997. He found that users ‘spent an average of fewer than 14 minutes per visit, viewing an average of seven pages each time’ (1998: 94). Other interesting results include that most users accessed the site ‘at their offices instead of their homes, and that introduction of new interactive features, particularly local content, may contribute to increased usage’ (1998: 108).
Other studies have used experimental designs to examine specific features of site use. Tewksbury and Althaus focused on how changes in presentation formats from print to online affected reading practices, and concluded that ‘as the online version presents fewer cues about the importance of events, it appears that people are more willing to use their own interests as the guiding criterion’ (1999: 24). Oostendorp and Nimwegen (1998) looked at how people locate information in online newspapers, especially the influence of different navigation strategies, concluding that ‘finding information for which scrolling down on a deeper hypertextual level was necessary took extra time and probably extra cognitive resources, leading to a lower recognition performance’. In a series of studies focusing on cognitive processes of online news use, Sundar and his colleagues have shown that (1) on news sites, quoted sources increased the credibility and quality of stories (Sundar, 1998), (2) similar criteria were used to perceive print and online news (Sundar, 1999), (3) readers of print articles remembered ad material more than users of websites (Sundar et al., 1999), and (4) multimedia materials increased memory for advertising, but decreased it for story content (Sundar et al., 2000).
To sum up, research focusing on how online newspapers have been used has shown that a few years ago people looked at them mostly seeking information. However, more recently they have predominantly used them for entertainment. This change may reflect a mainstreaming in audience composition and behaviour, which initially had an important presence of early adopters. Based on these findings, current speculation is that usage of online newspapers may be affecting more television viewing than reading of print papers—a trend that, if accurate, may still take a few years to manifest more clearly. Furthermore, users have tended to concentrate on text materials, to look at banner ads long enough to perceive them, to spend little time on sites but to view several pages, and to access online papers a lot from their offices—although this may have changed since the reported Aikat study ended in 1997 owing to increased home use of the Internet. Finally, experimental studies have shown that people (1) have increasingly used their own interest to guide content appropriation in the face of fewer editorial cues provided by online papers, (2) have located information more easily the less navigating and scrolling they have to do, (3) have employed similar criteria to perceive print and online news, (4) have remembered advertising material more in print than on the web, and (5) have recalled news more and ads less in relation to multimedia.
Interactivity
This has been the single most examined issue of post-web online newspapers. A common approach has been to list a series of features that supposedly represent interactivity, then assess how interactive a given online paper is depending on the presence or absence—or varying degrees of realization—of these features. For instance, in an analysis of 135 sites conducted in February 1998, Tankard and Ban have found that most of them ‘were providing e-mail addresses for the editor or webmaster and search engines for archives, but that few were providing such other means of user interactivity as discussion forums, chat rooms, surveys or polls, customized news services, and interactive games’ (1998: 10). Other such studies have had similar findings (Kenney et al., 2000; Massey and Levy, 1999; Schultz, 1999; Tremayne, 1998).
Although valuable, this approach has the problem of making conclusions based on aggregate scores that lump together ‘apples and oranges’. The difficulty resides in that different attributes are linked to different dimensions of interactivity: sending electronic mails to a webmaster, voting in an electronic poll, participating in a chat room, among other actions, are part of various communication experiences tied to diverse production and use dynamics. This has partly resulted from the fact that:
The concept of interactivity in relation to news is often muddled and poorly defined. In fact, the term is used to describe two essentially unrelated characteristics of online media. In one sense, ‘interactivity’ is used to describe the process of empowering users with additional control over the sequence in which information is presented to them. This definition relates to increased interactivity with content. But the term also is used to describe an increase in the interaction news consumers can have with news producers, a definition relating to increased feedback. (King, 1998: 26)
Those studies which have gone beyond mere ‘counting’ have usually focused on the second type of interactivity: users communicating with online newspapers’ staff as well as with fellow users employing tools such as electronic mail, forums and chat rooms—the last two options available in one-third of the online papers surveyed by Peng et al. (1999) in February 1997. Those investigations that have examined staff-user communication have shown that journalists have not been very keen about it. In their interviews at a large metropolitan paper, Riley et al. (1998) have concluded that ‘most of the reporters interviewed were horrified at the idea that readers would send them e-mail about a story they wrote and might even expect an answer’. A sense of separation between staff and users resonates with what Schultz has found in his survey of The New York Times journalists: ‘12 out of 19 admitted that they do not even visit the Times’ own online forums. Only six claimed to visit the discussion sites “from time to time”. No one visited them regularly’ (2000: 214).
A related dimension of interactivity has to do with users’ expectations of online newspapers and their personnel. In her first-person account of creating the Washington Post‘s Digital Ink online service in the mid 1990s, McAdams (1995) has claimed that when a paper goes online its users see this process ‘as evidence that the paper now wishes to have a closer relationship with its readers, and they are eager to let their opinions be known—not just in public discussions, but in personal e-mail to specific individuals’. However, in a study about how users’ evaluated The Guardian‘s website forums, Light (1999) found that their expectations of a paper’s online edition were tied to their expectations of the print product. Thus, in the same way that they did not expect to interact with the print paper, they did not have that expectation online either: ‘there are major obstacles to innovative news practice on the Web, especially where there is identity with a paper product. There was a tendency for the evaluators to overlook or reject interactive elements on The Guardian site.’
Interactive environments such as forums and chat rooms have been often positively received by scholars since they have seemed ideal vehicles for increasing users’ participation. To Pride, ‘with embedded newsgroups, electronic newspapers could become a functional town square: news and views and criticism all together’ (1998: 148). Along this line, Jankowski and van Selm have suggested that the role of online news sites in fostering participatory democracy should be an ‘important focus of future research’ (2000: 99). Moreover, to Friedland, by combining forums with non-linear narratives and archiving capabilities:
A collection of views can be archived, reread, explored and connected in new ways that offer new models of problem solving that expand the narrative boundaries of traditional journalism. These models offer an alternative to the plebiscitary model of electronic democracy that is both deliberative and practical. (1996: 202)
So far, little empirical research has been conducted about what users actually do in these discussion contexts, what relationships they establish with online papers’ staff and with fellow users, and what implications their participation may have on the character of news and on larger issues of democratic culture. One exception is Light and Rogers’ (1999) study of a forum opened by The Guardian to discuss the British 1997 election. They have shown that people’s previous experience with electronic discussion spaces influenced their participation in the forum: for instance, newcomers were ‘less discursive, more likely to post in order to present opinions, and more likely to visit the forums again to read other people’s contributions’. They have also found that participants and non-participants had diverse interaction styles. On the one hand, participants ‘were more likely to have the view that Web culture should uphold freedom of speech’. On the other hand, non-participants ‘were more likely to support the idea that postings should be edited to make them more informative, rather than lengthy, repetitive exchanges’. Another exception is Schultz’s (2000) examination of participation patterns at the New York Times’ site forums. Among other findings, he has shown that users ‘not only refer to each other’s postings publicly on the forum, but in addition by personal e-mail … the more comments they sent, the more likely it was that they also got e-mail feedback’ (2000: 215-16). Finally, in another empirical study of user-authored content, Boczkowski (2000) has looked at the interrelated technological, editorial and organizational processes shaping a project in which non-profits built their own publications within an online newspaper. He has argued that such a regime of content creation that he calls ‘distributed construction’ is characterized by:
(a) An artifact that inscribes users as active content creators and that is domesticated partly in public, (b) work processes that center the editorial function around the facilitation of content production and exchange, (c) a network of information flows in which every node can be source and destination of messages of potentially both generalized and specialized character, and (d) an organizational form based on relationships of interdependence, distributed authority, and multiple rationalities. (2000: 22-3)
In sum, research on interactivity has sometimes argued that online papers are not very interactive, although these ‘scores’ may not be all that useful for they have been based on homogenizing too disparate sets of communication processes, experiences and technologies. A more fruitful approach has instead looked at the practices of interactivity. Some studies have found very little staff-user interaction, something consistent with previous research on traditional media. Other investigations have concentrated on the role of expectations, yielding conflicting results: whereas one study has suggested that users of online papers expect it to be more interactive than its print counterpart, other inquiry has found users expect it to be no more so—thus potentially limiting the realization of interactive potentials. Also, experience and degree of participation have been associated with different interactive behaviours and experiences, and one examination of forum dynamics has shown the presence of a densely interconnected flow of private and public messages among users. Finally, engaging users as content co-producers has been associated with technical, editorial and organizational transformations that depart markedly from the traditional dynamics of content production in print newspapers.
The Road Ahead
The research reviewed above has told us useful and important information about how online newspapers have been produced and used, before and after the popularization of the web. However, there is still much more that remains to be known about these matters. This is partly because of limitations in the inquiries undertaken so far, and—especially in the case of events in the second half of the 1990s—partly due to the newness of the object of study. In these final paragraphs I will point to several directions for further research by analysing shortcomings in the existing literature and suggesting some important areas still unexplored, in the spirit that such a critical examination will foster reflection about past accomplishments and design of future endeavours.
There has been a dearth of historical analyses about the evolution of online newspapers, and the relationships between past attempts and current developments. This is all the more surprising given the richness of events during the last decades and the availability of both archival material and actors who could be good candidates for oral history interviews. To Deuze (1999), recent studies ‘seem to lack a contextual and historical foundation’. More work is needed with historical focus and questions. Issues that this line of inquiry could address include (1) the effect of pre-web endeavours on more recent efforts, (2) the presence or absence of longitudinal trends regarding ethnical, class and gender variations in the production and use of online newspapers, (3) the ‘after life’ (Simon, 1999) of artifacts, like videotex and teletext, that despite being written off as failures have continued to exist in the margins, (4) the fate of technical alternatives—like audiotex, around since the mid 1980s and in many cases commercially successful since the mid 1990s—which have not moved centre stage, (5) the career paths of actors who played important roles in both pre- and post-web projects, and (6) the influence of corporate culture and strategy upon technological change, in particular why some firms have been very active since the late 1970s, while others have remained comparatively more passive, and yet others have switched their approach over time.
There has been a methodological pattern in the inquiries undertaken so far. On the one hand, most studies focusing on production issues have been either conceptual essays or what could be called ‘site analysis’—for instance, looking at what interactive features a set of sites has—or surveys. On the other hand, research looking at uses of online papers has primarily relied on experimental designs, and to a lesser extent on surveys and interviews. Although there have been a handful of examinations of production issues employing ethnographic methods (Boczkowski, 2000; Borum, 1998; Eriksen and Ihlström, 1999; Eriksen and Sørgaard, 1996; Huxford, 2000; Martin and Hansen, 1998; Molina, 1999; Riley et al., 1998), the limited breadth and depth of most of these inquiries, and the lack of such studies on usage matters, suggests that much more work remains to be done into employing the full spectrum of naturalistic methods—so crucial to understand the practice, meaning and experience of emerging media. Possible issues to be addressed using these methods include (1) the routines and values of online news-making, and their relationships to those of print and broadcast journalism, (2) the construction and reconstruction of occupational identities in the new work environment, (3) the negotiations among the different occupational groups as they influence the news-making process, and (4) the relationships among all agents of production—not just editorial workers, but also design, technical, advertising and marketing personnel, users when they become co-producers, and the artifacts the different actors use—that constitute, borrowing from Becker (1982), ‘the online newspaper world’.
The combination of little attention to historical matters with a low use of naturalistic inquiry has probably contributed to a third type of limitation of current research: a tendency to build analysis upon a usually taken-for-granted technologically deterministic matrix.15 When only short temporal sequences are examined, and when actors’ actual practices are not observed, analysts may be more inclined to think that ‘a technology enters a society from outside and “impacts” social life … [describing] a form of cultural lag, during which adaptive problems arise’ (Fischer, 1992: 12). To Bardoel, ‘modern communication technology will lead to new information services and to new journalistic practices’ (1996: 295). Focusing on information architecture issues, Newhagen and Levy have argued that ‘data concentration is unnatural in distributed network architectures that facilitate dispersed message production. Thus, the application of canons or standards produced to deal with mass media systems may be unnatural, unrealistic, and practically impossible to apply’ (1998: 16, emphasis added).
However, the history of developments in information and communication technology during the twentieth century is replete with events in which supposedly ‘unnatural’ or ‘unrealistic’ uses of new artifacts became not only ‘possible’ but also major commercial turning points. For instance, amateur users played a crucial role in transforming the radio from a point-to-point communication device into a broadcast medium (Douglas, 1988); residential users were key in turning the telephone from an instrumental and business device into a leisure and domestic tool (Fischer, 1988); and early adopters triggered a redefinition of the French minitel videotex system from a one-to-many service into a many-to-many communication space (Charon, 1987). Further research is needed that not only avoids technologically deterministic temptations, but also underscores the weaknesses of that kind of account. Issues particularly germane to this end could be (1) the presence of resistance to new technologies within organizations adopting them as well as among the audiences for their products, (2) the emergence of unexpected uses broadening the repertoire of activities consciously inscribed by producers in their design of media artifacts, and (3) situations in which the same technical potentials are realized in diverse ways, and different potentials are domesticated into relatively similar uses.
Other types of limitations of current research have to do with theory-building matters. When traditional media use of the web was in its infancy, Morris and Ogan urged scholars to ‘rethink answers to some of the central questions of mass communication research, questions that go to the heart of the model of source-message-receiver with which the field has struggled’ (1996: 39). Echoing this call, Boczkowski (1999) has suggested various ways in which computer-mediated theorizing could be exploited to study what has traditionally been the province of mass communication scholarship. Furthermore, Singer (1998) has argued that:
The questions about online journalism have many facets, and multi-disciplinary, wide-ranging approaches may work best in attempting to answer them … The issues raised by this new form of communication in general and journalism in particular invite us not only to make better use of what we already know but also to be open to new ways of asking those vital questions.
However, despite this and related statements, most theorizing about online newspapers has so far remained confined to the familiar landscape of notions and models such as gatekeeping, agenda-setting, uses and gratifications, and information processing—all of which have mostly been used within the field of communication and media studies. Although it is valuable to approach new technologies and their related social practices from ‘tried and true’ conceptual resources, it is also the case that the study of emerging media is a more fertile soil for novel theorizing than that of existing media. Thus, further research is needed that could lead to multidisciplinary theory building about online newspapers. Within communication and media studies, and given that actors have already been mixing mass and interpersonal attributes in the production and use of online papers, there seems to be a conducive environment for combining concepts and theories born out of the usually distanced camps of mass and interpersonal communication. Other fields of inquiry that could be brought to bear in a multidisciplinary theory-building effort include (1) sociology, history and economics of technical innovation, (2) sociology and anthropology of work, occupations and organizations, (3) policy studies, especially to examine the relationships between the often separate domains of media and technology policy, (4) cultural studies of technology, and (5) contextual approaches to knowing such as activity theory and distributed cognition.
As I showed above, studies looking at production matters have provided valuable knowledge about online newspapers. However, there are still some limitations common to most of them. First, although shovelware has been the dominant type of content, it is no less true that there has been a growth in original content in the last few years. Despite not constituting the majority of material available, original content may be a more appropriate ground to examine changes in the technologies and practices of newspapering as it moves from ink on paper to pixels on the screen. This relates to another shortcoming of existing research: the focus on interactivity seems to have been at the expense of looking at the appropriation of other technical features of the web, most notably multimedia. For all the talk about media convergence, there has been a dearth of studies examining the dynamics of producing multimedia content. Which leads us to a third limitation of existing research on production matters, a more conceptual one: little attention has been paid to the role of technology in the actual work practices of editorial personnel—something which is not surprising given the neglect of technology’s role in most studies of news-making in print and broadcast media (Cottle, 2000; Hansen et al., 1994; Schudson, 1997; Sumpter, 2000). Reflecting upon the initial analyses of online newspapers on the web, Martin and Hansen have argued that: ‘None of these studies … has focused on the way in which computer technology and reporting norms may affect news content as it is selected from the newsroom’s newsprint delivery traditions, prepared and moved to the electronic environment, and offered as online news’ (1998: 108).
Further research is needed to illuminate what goes on in the creation of original content, taking advantage of the web’s unique technical features—paying special attention to multimedia storytelling—and, more conceptually, examining the role of technology in news-making, a long-overdue theme in the sociology of news production. Potential issues to be addressed include (1) the differences and similarities in news-making routines for shovelware and original content, (2) the comparison between newsworthiness on the web and in traditional media, (3) the acquisition of new technical and narrative skills and their relationships to existing ones, as well as the decrease in use of existing skills, (4) the changes in gathering, processing and delivery of news content in relation to having multiple media for storing and conveying information, and (5) the presence or absence of tensions derived from the encounter of the different news-making cultures of print and broadcast journalism, as well as their relationship to computer-oriented cultures of information production.
As I described above, some studies have begun looking at how online newspapers are used. Despite the valuable information they have produced, they have been limited in various forms. The most important of these shortcomings has been the lack of naturalistic research about what people actually do with and through online newspapers. For all the useful knowledge that studies based on experimental, survey, interview and traffic log data can provide, they tend to fall short on shedding light on the habits and meanings that constitute the ‘flesh and blood’ of using online papers. Thus, more research is needed that examines the discourse and practice of users. Issues that could be examined include (1) the routines and experience of using online newspapers, (2) the influence of major socioeconomic factors such as race, gender and class, (3) the relationships with the use of other media and information artifacts, (4) the similarities and differences linked to diverse contexts of use such as home, workplace, libraries, cybercafés and various transportation means, (5) the factors determining why some users become content and marketing co-producers while others remain just consumers of what others create and promote, and (6) the emergence of resistance and uses unintended by online newspapers’ producers.
Beyond the specific contributions and shortcomings of the research reviewed in the preceding pages, I would like to conclude this essay by pointing out one image that comes to mind after surveying this growing area of inquiry as a whole: the emergence of a multifaceted process of boundary blurring, shaping the contours of traditional media’s forays in the new information environments. Where does the division between the spheres of production and use reside when new media sites feature vast quantities of user-authored content? What is the basis for distinguishing between interpersonal and mass communication in the chat rooms and forums that not only pervade the websites of news operations, but spill over onto television programmes, radio shows and print pages? What happens to the separation between print and broadcast journalism when websites of print papers make increasing use of audio and video files, and their counterparts of television and cable operations are filled with text? What is the rationale for dividing between the immediacy of wire services and the analysis of news outlets when the websites of print and broadcast operations routinely weave these two types of content? How can we make sense of the separation between media and other information industries when software manufacturers seek to seamlessly integrate content with code, and traditional media corporations invest increasing resources in technology development? Paraphrasing Geertz’s comment about the refiguration of social thought, this blurring of discursive, practical, material and organizational boundaries may represent, ‘or will if it continues, a sea change in our notion not so much of what knowledge is but of what it is we want to know’ (1983: 34). If that were the case, this would mean expanding the assessment of what research tells us towards a collective reflection on what we might want to know.