2.1 PHYSICAL VS. DIGITAL
While the separation between “offline” and “online” or “virtual world” and “real world” networks and communities has become increasingly blurred, and arguments against the very existence of a division were proposed more than 10 years ago (e.g., Wellman, 2004), the qualities of interaction among people, information, and objects differ in physical vs. digital environments. The serendipitous sociocognitive microenvironment (Merton and Barber, 2004) needs, at the very least, some massaging when considering the relatively recent shift from physical to digital. Since the 1990s, research has examined how interactions with people and information differ in physical vs. digital environments and whether the positive, serendipitous aspects of physical environment may be mimicked or augmented through our interactions with digital environments.
Discussion relating to the shift from physical to digital and its implications on serendipity and how digital can either replace or enhance our physical interactions is evident, for example, in relation to the shift from physical to digital photography, in approaches to humanities research, and the development of collaborative work environments. Research has sought to find ways in which digital photographs may be remembered and shared as easily, or perhaps more so, than their physical counterparts (Frohlich, Wall, and Kiddle, 2013; Nunes, Greenberg, and Neustaedter, 2009). Findings relating to serendipity in the humanities suggest unease over the shift from physical to digital source materials, although Verhoeven and de Costa (2014) note that this may dissipate with time through new methodological approaches and technological changes. Currently, however, among historians, research suggests that while more and more artifacts and manuscripts are becoming digitally accessible, there is apprehension that the shift from physical books to eBooks may reduce opportunity for serendipity rather than support it (Quan-Haase and Martin, 2012; Martin and Quan-Haase, 2013, 2016). In serendipity research relating to work or enterprise, the emphasis is often on co-workers’ information-rich interactions with each other and how the phenomenon may be facilitated by enabling informal online communication (e.g., Guy et al., 2015; Whittaker, Frohlich, and Daly-Jones, 1994). We focus in the remainder of this section on the work or enterprise area of research where the motivation for serendipity research has a distinctly physical vs. digital perspective: Computer-Supported Cooperative Work (CSCW).
Early CSCW research (e.g., Whittaker, Frohlich, and Daly-Jones, 1994) was motivated by findings indicating that physical proximity has a significant influence on opportunistic collaboration among researchers (Kraut, Egido, and Galegher, 1988). Researchers at Hewlett Packard, for example, examined how informal communication, including communication prompted by “chance encounters,” could in some way be replicated online to support geographically distributed groups (Whittaker, Frohlich, and Daly-Jones, 1994). Jeffrey (2000) examined whether chance encounters occur in a “networked, virtual world with three-dimensional avatar representation” (p. 331) and found that chance encounters known to occur in physical environments can be reproduced in virtual environments. CSCW continues to examine how to support serendipity through, for example, updates on fellow employees’ social media activity (Guy et al., 2015) and the implications of serendipitous experiences in work environments such as enhanced communication and productivity (Brown et al., 2014).
While researchers and developers are forging ahead with the development of approaches to increase the potential for serendipity in digital information environments, there is a recognition that technological support for serendipity is not quite “there yet,” as evidenced from moves by technologically sophisticated companies such as Yahoo!, Google, and IBM to encourage face-to-face interactions among its employees (e.g., Lindsay, 2014; Silverman, 2013; Wolsen, 2013). Yahoo! made news in 2013 when CEO Marissa Mayer barred employees from working from home, a move widely held to be associated with the desire to increase productivity as well as the belief that serendipity, a driver of innovation, was more likely to occur through diverse, face-to-face interactions with colleagues than at one’s home office or through online communication (Wolsen, 2013). Because face-to-face interactions were credited with innovations at the search engine giant Google, including Gmail and Street View, the company designed its headquarters to ensure its employees could, according to a Google spokesperson, “collaborate and bump into each other” (Silverman, 2013, n.p.). Similarly, IBM’s Accelerated Discovery Lab, with its open space and dynamic concept, was designed to ensure “cross-pollination” among colleagues from different disciplines and teams and visitors to the lab would have opportunities to interact with each other and big data. Laura Haas, the lab’s director of technology and operations, noted
We call it cultivating “strategic serendipity.” It’s those “A-ha!” moments you have in the shower or often around the water cooler. We want to bring people together in a rich enough environment they want to play in it, and then create serendipity by leveraging the connections in the room, the connections in the data, and our ability to see what users are doing (Lindsay, 2014, n.p.).
Currently, without a better alternative, high-tech companies continue to recognize the need for face-to-face interactions to facilitate serendipity. Regardless, however, of the push to get colleagues in the same room together through company policies and architectural design, a significant amount of worker interactions with data, information, and knowledge now take place online, through email, social media, search engines, databases, and other digital information resources and sources. Therefore, the need to get serendipity “right” in digital information environments is critical and continues to be a prime motivation for serendipity research.
2.2 INFORMATION OVERLOAD
One of the main benefits of digital information environments is the plethora of dynamic, diverse, and hyperlinked information that those environments contain, with the potential to trigger serendipitous experiences. At the same time, some argue that this type of information-rich environment is just as likely to spur information overload as it is to trigger serendipity—arguably more so. This tension between the need to manage both the quantity and quality of information has been a key driver of serendipity research. How can digital environments provide a balance between manageable information exposure and drawing attention to information that may be considered unexpected but useful (i.e., serendipitous)? Relative to the information overload phenomenon, associated with enterprise time and money (Barta, 2014; International Data Corporation, 2001) as well as anxiety and stress (Erdelez, 1996; Yadamsuren and Heinström, 2011), a serendipitous digital environment must meet the demands of user experience like any other digital environment otherwise people will not stay or return (Åman et al., 2014).
Information overload is a term “often used to convey the simple notion of receiving too much information” (Eppler and Mengis, 2004, p. 326). Research across a variety of disciplines indicates that the quality of individuals’ decisions correlates with the amount of information received, but only up to a point. Once that threshold is reached, information overload ensues as information can no longer be integrated into decision-making (Eppler and Mengis, 2004). Eppler and Mengis describe the inverted U-curve associated with this relationship between decision-making and information load, first articulated by Schroder, Driver, and Streufert (1967). In serendipity research, information overload is often referred to in related terms; a similar U-curve schematic can be imagined in which “decision-making” is replaced by “serendipity.” The more information provided in a digital information environment, the more opportunity for serendipity—but still, up to a point. Figure 2.2 illustrates the relationship between serendipity and information load that is often articulated in serendipity research as a phenomenon to be wary of and to limit by design (e.g., Bellotti et al., 2008; Cleverley and Burnett, 2015a; Guy et al., 2015; Rädle et al., 2012).
Figure 2.2: Serendipity relative to information load; “decision making” replaced by “serendipity” (adapted from Eppler and Mengis, 2004).
Serendipity research