Accepted by previous editor Maria Bakardjieva
Creativity and Play in Digital Milieus
Public Displays of Play: Studying Online Games in Physical Settings†
Version of Record online: 10 JAN 2014
© 2014 International Communication Association
Journal of Computer-Mediated Communication
Volume 19, Issue 4, pages 763–779, July 2014
How to Cite
Taylor, N., Jenson, J., de Castell, S. and Dilouya, B. (2014), Public Displays of Play: Studying Online Games in Physical Settings. Journal of Computer-Mediated Communication, 19: 763–779. doi: 10.1111/jcc4.12054
- Issue online: 18 JUL 2014
- Version of Record online: 10 JAN 2014
- Manuscript Accepted: 4 JUL 2013
- Manuscript Revised: 15 MAY 2013
- Manuscript Received: 2 DEC 2012
- Computer games;
- Multi-user Virtual Environments;
- Online Games
As research on virtual worlds gains increasing attention in educational, commercial, and military domains, a consideration of how player populations are ‘reassembled’ through social scientific data is a timely matter for communication scholars. This paper describes a large-scale study of virtual worlds in which participants were recruited at public gaming events, as opposed to through online means, and explores the dynamic relationships between players and contexts of play that this approach makes visible. Challenging conventional approaches to quantitatively driven virtual worlds research, which categorizes players based on their involvement in an online game at a particular point in time, this account demonstrates how players' networked gaming activities are contingent on who they are playing with, where, and when.
This paper reports on a mixed-methods study carried out across 20 public gameplay sites, including large- and small-scale local area network (LAN) parties, fan culture conventions, Internet cafés, and pub-based gaming nights. The purpose of this fieldwork was to document and compare the games-related and games-based forms of communicative activity, particularly with regards to play in Massively Multiplayer Online Games (MMOGs), that different public venues make possible. This work merges qualitative research methods (participant observation and interviews) with quantitative data collection and analysis (via a survey administered to each participant), enabling a mixed-methods, multisite exploration of game-based computer-mediated communication in public settings. The sites at which we met participants to observe and discuss their play involved multiple and different configurations of social, technological, material, and economic resources – in other words, different virtual and material “assemblages” (Taylor, 2009) for public play. Participants included classmates playing at an Internet café after school before heading home for the evening; hobbyists brought together for a day of card, computer and tabletop gaming at fan culture events; “clanmates” commuting for hours, and sleeping in tents and cars, to spend time with one another at massive, multiday LAN events; and GLBTQ gaming enthusiasts congregating for gaming gatherings at a neighborhood pub.
We situate this project alongside other current accounts of both public gaming and MMOG play, and we detail an approach that merges qualitative methods for studying players in public sites with a concern for generating quantitative accounts and analyses of MMOG-based interactions. The central question guiding this investigation is as follows: what happens to our understandings of the forms of sociality supported by networked digital games when the research sites are not individual online gaming environments, but rather the physical settings in which many players gather, publically, to play? We suggest that this marks a departure, both methodologically and ontologically, from the ways quantitative studies of MMOGs have conventionally been carried out, insofar as our research contexts allowed us to observe players' interactions, both computer-mediated and face-to-face, through direct observation instead of through survey self-report and/or in-game actions. We then briefly describe each of the public gaming contexts we visited, pointing to the ways in which the specific conditions of each structured different possibilities for sociality, communication, play, and fieldwork. After this sketch of the different public gaming sites involved, we present a quantitative snapshot of the population of each; then, drawing on stories from our qualitative fieldwork, we highlight ways this stable and somewhat predictable demographic picture of a gaming population was shaped and produced through shifting fieldwork situations. Building from recent studies that have applied science and technology studies (STS) notions of “assemblages” (Taylor, 2009), “cybernetic circuits” (Giddings, 2007), and “actor networks” (Taylor, 2011) to accounts of digital gaming, we discuss the challenges our approach offers to current quantitatively driven understandings of the role and significance of MMOGs in the lives of players.
The analysis we undertake here pays critical attention to the ways in which both ‘participants’ and ‘research sites’ are defined and delineated in research on networked digital gaming. In beginning from an account of the material conditions of play in public sites, and a consideration of what kinds of communicative activity (both in terms of play and fieldwork) these conditions enable, the aim is to provide an alternative grounds to the study of MMOG play that resists the easy (but limiting) dichotomy between online and offline practices and settings.
This work builds on two areas of scholarship on digital gaming: quantitative studies of MMOGs, and qualitative studies of how play is performed in public sites.
Studies of Online Gaming
A growing body of research uses the tools and techniques of ‘big data’ to examine gaming communities formed within specific MMOGs. In-game data on avatar behavior, gathered through automated scripts, via online avatar databases (c.f. the World of Warcraft Armory), or obtained directly from the game's publishers, is typically combined with self-reported survey data in order to characterize ‘life online’ in these virtual spaces, and to generate claims about the apparent motivations, dispositions, and offline characteristics of players. The breadth and scope of these studies is impressive: Yee et al (2011), for example, draw from online surveys and corresponding World of Warcraft (WoW) armory data from over 1,000 players, while Harrison and Roberts (2011) draw from WoW armory data on close to 15,000 avatars. Williams et al (2011) make use of 7,000 online surveys, with access to over 30,000 player logs, in their study of Everquest II, which also uses server-side data made available by Sony Online Entertainment, the game's publisher.
By design, these quantitative, game-based studies on MMOGs leave largely untouched questions of whether and how the physical contexts of online play affect players' computer-mediated interactions. Differentiated levels of access (whether to machines, game software, extended periods of leisure time), prior experiences with a particular game or genre, and the specificities of players' material conditions are either disregarded, or are treated as murky terrains that can be partially illuminated through inferences drawn from in-game behavior (see, for example, Mahmassani et al, 2010). Rather than view this as a limitation, these studies claim that virtual environments are sufficiently analogous to ‘real life’ settings to warrant comparison and inference from virtual contexts to physical ones, but with the added benefit that games can enable far more comprehensive and unobtrusive means of gathering data on players' “natural” behaviors than is possible (or ethical) in offline contexts (see, for example, Yee et al, 2011, p. 2).1 Leaving aside ethical considerations regarding the use of automated techniques to gather data on MMOG players without their explicit consent (Chee, Taylor and de Castell, 2012), this work is based on the (arguably under examined) premise that networked play can be studied and understood without recourse to the offline contexts and conditions of its production.
Studies of Public Play
With regards to the study of digital play in public sites, scholars have examined Internet cafés, “PC Bangs,” competitive tournaments, or LAN events, both small- and large-scale, exploring the sociality (Jansz and Martens, 2005; Swalwell, 2006) and aesthetics of public LANs (Simon, 2007), the ways young people fit public play into their everyday lives (Chee, 2006), the role of tournaments in the growth and development of e-sports (Taylor, 2011; Taylor, 2012), and the cultural positioning of public gaming as an increasingly mainstream practice (Taylor and Witkowski, 2010). Through observations and interviews with players, spectators and organizers/workers, these studies document participants' behaviors, practices and orientations with regards to games and public gameplay.
In addition to documenting how play is performed in different public sites, a number of studies look at how the materiality of public gaming sites affects the play practices of attendees. Though not writing specifically on gaming, Nina Wakeford's (1999) ethnographic work on the “gendered landscapes of computing” in a London-based net café explored the ways the physical contexts of computing in public spaces – including the layout and configuration of machines, the semiotics of décor, and rules and protocols for behavior – shape the kinds of communicative action made (im)possible in these sites. Wakeford's particular focus on the gendered dynamics of Internet cafés has been taken up by Jo Bryce and Jason Rutter (2005), and Holin Lin (2008) in their analyses of how gender is enacted and performed in public gaming sites. Bart Simon (2007) focuses more closely on the materiality of machines at LAN parties, calling attention to the practice of ‘case modding’ as an example of embodied sociotechnical practice that undermines attempts to understand gameplay purely as a matter of what happens on screen. In a similar vein, TL Taylor and Emma Witkowski (2010) emphasize the multiple, heterogeneous practices undertaken by attendees at Dreamhack, the world's largest LAN party held annually in Sweden; they remark that at an event ostensibly devoted to gaming, much more than gaming happens, both online and off. In what follows, we build from and extend these insights into the connections between networked online play and the physical contexts in which it unfolds. As we will demonstrate through an account of the spatial and temporal arrangement of different public gaming events and of our research activities at each, the ways players perform and describe their involvement in networked digital play – and even whether they consider themselves ‘MMOG players’ - varies according to particular and contingent conditions: what they are playing, with whom, and where, and crucially, who is asking the questions about their involvement, and how.
Reassembling the LAN
The work we report on here is the first large-scale study of online gamers in which they were sought out in public settings. As a theoretical orientation to this study, we draw from the work of qualitative games researchers employing actor-network theory (ANT), primarily as articulated by Bruno Latour (2005). Whether applied to studies of MMOG play (Taylor, 2009), single-player gaming (Giddings, 2007) or competitive first-person shooter tournaments (Taylor, 2011), actor-network approaches to digital gaming interpret play as the production of coconstitutive relationships between human and nonhuman actors, including the game's hardwired rules, the arrangement of online and offline play contexts, legal structures regulating play, hardware inputs and platforms, graphical user interfaces, and so on. This perspective works against a tendency in MMOG studies, as outlined above, to view gaming environments as ‘natural’ settings in which to observe player interactions – a perspective that largely ignores the tremendous role played by these material and technological systems in not simply mediating, but structuring players' communicative actions.
While actor-network theory provides a broad theoretical perspective for this work, sociocultural activity theory offers a productive means of schematizing the methodological differences between this study and other large-scale studies of MMOGs. Developed by Engeström (1990) and adapted to the study digitally-mediated practices by Nardi (1995), activity theory views sociotechnical practice as the product of intentional, goal-directed activity carried out through interconnected sets of mediators, including technologies and their specific affordances and constraints; organizational, institutional, and technical rules, protocols, and policies; other individuals engaged in the same pursuit; and members of the broader community or organization. Adapted from Engeström's schema showing the interrelationship of elements in a given “activity system” (1990), Figures 1 and 2 depict the differences in MMOG research carried out through our study of play in public settings and the majority of quantitative MMOG studies to date that are based on (and in) particular games. As we will demonstrate, the key difference between these two approaches, as depicted in these diagrams, concerns the “division of labor” in each. Interacting with players face-to-face and in some cases, multiple times enabled us to get a sense of how participants moved from game to game over the course of multiple site visits, a single weekend, or even a single observation, revealing the limitations in organizing and making claims about participants wholly based on their involvement ‘within’ a particular online game.
To carry out this fieldwork, we visited over 20 public sites: Internet cafés, LAN (local area network) parties, and other public gaming events in, Toronto, Vancouver, and England between Summer 2010 and April 2012. We observed players' activities in multiple games and solicited, from 25 of these participants, an online “travelogue”, a visual journal of their everyday play in the MMOG of their choice (Taylor, Jenson and McArthur, 2012). We administered a survey to participants at each event (378 in total), that we asked them to complete on-site so we could be on hand to answer any questions. We also interviewed 14 players face-to-face about their gameplay preferences, play practices at home, over-arching MMOG career, and their reasons for attending the public site where we met them. Where possible, we visited sites multiple times to see who returned over the course of days and weeks. This represents what we believe is a first in studies of either online gaming or offline, public contexts of play: a comparison of players and player populations that, because of its documentation of different kinds of gameplay settings globally, is both more latitudinal and, by virtue of our multiple visits to multiple contexts over time, more longitudinal than most accounts of online play/ers.2
While an exhaustive account of the ‘lan-scapes’ of these events is beyond the scope of this paper, we want to briefly characterize each of these events in order to give an impression of what kinds of play, socialization, and research were possible across each. In keeping with an activity theory framework, we pay particular attention to the temporal and spatial arrangements of these events in order to draw attention to how these shaped our fieldwork.
Insomnia LAN Festivals
Between August 2010 and April 2012, we attended four large-scale LAN events, held in England (two in Newbury and two in Telford) and run by the organization Multiplay as part of their “Insomnia” series of gaming festivals. The Insomnia series (or i-Series) of LAN parties take place three times annually: in late April/early May, mid-August, and mid-November. Of the four events we attended (i40 and i41 in Newbury, in August and November 2010, respectively; i43 and i44 in Telford, in August and November 2011, respectively), the least-attended event (i41) attracted approximately 1000 attendees, while i43 attracted over 2500. Events were held at a horseracing facility in Newbury and then relocated to a convention center in Telford. Each event, regardless of location and number of attendees, involved a mix of activities besides day-to-day (and night-to-night) play in the ‘BYOC’ (Bring your own computer) area: tournaments for popular e-sports games (Counterstrike, Team Fortress 2, Starcraft, and League of Legends primarily); a small area for table-top gaming; a Pub Trivia Night; and exhibitor booths. These events offered outdoor campgrounds for attendees to spend the night, large event floors to set up their computers in long rows of tables, and competitive gaming tournaments.
Fieldwork at Insomnia LANS involved navigating the erratic and overlapping tournament schedules, and the even more erratic schedules of players as they moved from one activity to the next, taking breaks for meals, sleep, drinking, or just hanging out. Alcohol, and the relatively lax rules around public alcohol consumption in England (compared to Canada), also played a role in this fieldwork; being able to have a beer with participants as they did the survey, chatted with us, and showed us their MMOG play was often key in establishing a ‘fluid’ and easy rapport with them. By the third and fourth events we attended, we were recognized by many participants who directed their friends, clanmates or peers our way to participate in the study.
Canadian LAN events
We attended two LAN events in Canada: one in Vancouver (LANcouver, July 2011) and one in Toronto (ZeroPing, July 2010). These events were much smaller in scale than the Insomnia events in England (the LANcouver event had 250 attendees, the ZeroPing event closer to 70), but similarly featured attendees playing on their own machines in a large event space, organized along long tables. Additionally, the LANcouver event featured console-based fighting game tournaments and an area for tabletop gamers. These differed from the England-based events in terms of the scale, location, and rules governing attendee behavior. At the Insomnia event, for instance, alcohol consumption was allowed, if not encouraged; the Toronto and Vancouver events had a no-alcohol policy. Furthermore, because the Insomnia events drew large numbers of players from all over the UK and Europe, the large majority attendees – including the researchers – had no home to return to in the evenings (or early mornings), meaning that attendees stayed at the event for much long periods of time. In contrast, because the Canadian events were much smaller in scope and primarily attended by participants from the host city, there was a notable drop in attendance during the mid to late evenings, and the window for conducting recruitment and fieldwork was comparatively smaller.
2010 Toronto FanExpo and 2012 Vancouver FanExpo
In August 2010, we attended the Toronto FanExpo, a 3-day fan culture event, where we recruited participants by circulating on foot through the crowd and chatting up attendees participating in tabletop gaming. Opportunities for fieldwork were limited, given we did not have dedicated space on the event floor (as we had at LAN parties). There was also a lack of emphasis on digital gaming compared to other events. In April 2012 at the FanExpo held in Vancouver, we arranged an exhibitor booth where we set up a station of laptops for attendees to take part in the study. For both events, the majority of the large convention floor was devoted to promotional booths for comic, manga, and graphic novel creators; video game manufacturers; and sci-fi, fantasy and horror television productions. Each site also had space for tabletop and card gaming.
Internet cafés in Toronto and Vancouver
Over the period of 4 months, between July and October 2010, we visited over 20 Internet cafés in both Toronto and Vancouver, coming at multiple times during different days of the week and times of day, but were only able to recruit participants from eight of these sites. Even when we got the approval to conduct research from café managers, recruitment proved difficult and time-consuming; with few exceptions, most attendees who visit Internet cafés to play games do so in a way that makes efficient use of constrained and often brief periods in their days (coming home from school, on their way to or from work). What made recruitment even more challenging was the typical physical orientation of PC stations: Often separated by partitions, they screened attendees' activities from one another's view. This meant researchers had to approach potential participants from behind and ‘shoulder tap’ in order to strike up a conversation: In such an individualized setting, this approach was often unwelcome. Moreover, as we learned from observation and from discussion with participants, much of the clientele use Internet cafés as a temporary reprieve from school-, work-, or home-based accountability and responsibility and therefore solicitation was often regarded as an intrusion. In reference to the activity structure of fieldwork in public sites (Figure 1), these spatial and temporal conditions created very limiting ‘rules’ for interacting with café attendees.
Two key exceptions to this, one in Vancouver and one in Toronto, were hobby shops that held regular Magic: The Gathering card tournaments. While the tournaments took place in areas with no computers, the seating arrangement along benches, with players looking across the table at each other, was more similar to LAN events, and more conducive to striking up conversations with attendees, than the typical Internet café. Similarly, the tournament format meant that we could talk to and recruit attendees during the gaps between rounds.
GLBTQ Gaming Community
Finally, between September 2011 and February 2012, we attended three gaming events put together by an association of GLBTQ gamers at a pub in the Toronto area. These events were primarily (though not exclusively) focused around console games. The material contours of this sitewere unique to the gaming events we visited. Gaming machines were not as numerous as other gaming parties; aside from the four research laptops we brought to these events, there were only three consoles attached to large flat screen televisions. In contrast to the prevalence and centrality of gaming machines in other locations we visited, the consoles were pushed against the walls, creating an open space in the middle where participants could chat, watch, play, and drink.
This section provided a sketch of the sociotechnical configuration of each event, outlining how conditions ‘on the ground’ affected our ability to carry out fieldwork. In particular, this sketch indicated some of the ways that the ‘divisions of labor’ involved in this fieldwork (Figure 1) were shaped by the arrangements of bodies, technologies, and spaces. In the following sections, we provide quantitative and qualitative accounts of these study populations, drawing attention to the ways that our qualitative stories unsettle an otherwise straightforward quantitative picture, and demonstrating how the material and embodied conditions of play and research are central to our understanding of online sociality and communication.
The demographic picture below organizes participants by site rather than by MMOG, in order to emphasize the significance of local and contextual conditions in this study of networked play. For the purpose of generating a demographic profile of players, we grouped research sites into the following categories: i) LAN events in England (including all four Insomnia LAN events); ii) Fan culture events in Toronto (including the GLBTQ gamer events and FanExpo); iii) Internet cafés in Toronto and Vancouver; iv) LAN events in Toronto and Vancouver.3 From this we were able to generate a comparative snapshot of the player populations at each site (Table 1).4
|LAN events, England (n = 183)||LAN events, Toronto and Vancouver (n = 57)||Internet cafés, Toronto and Vancouver (n = 49)||Fan culture events, Toronto and Vancouver (n = 93)|
|Sex (%)||Female: 12||Female: 16||Female: 12||Female: 14|
|Male: 88||Male: 84||Male: 88||Male: 86|
|Race (%)||Caucasian: 90||Caucasian: 63||Caucasian: 63||Caucasian: 66|
|Asian: 5||Asian: 33||Asian: 12||Asian: 26|
|Other: 5||Other: 4||Other: 25||Other: 8|
|SES (%)||Lower middle: 13||Lower middle: 27||Lower middle: 16||Lower middle: 13|
|Middle: 42||Middle: 31||Middle: 37||Middle: 30|
|Upper middle: 32||Upper middle: 31||Upper middle: 34||Upper middle: 19|
|Other: 13||Other: 11||Other: 13||Other: 38|
|Highest level of education attained (%)||Secondary: 31||Secondary: 45||Secondary: 55||Secondary: 32|
|Trade school: 29 Undergraduate: 20||Trade school: 21 Undergraduate: 21||Trade school: 19 Undergraduate: 15||Trade school: 15 Undergraduate: 32|
|Graduate: 20||Graduate: 6||Graduate: 2||Graduate: 14|
|Primary/Other: 7||Primary/Other: 9||Primary/Other: 7|
At first glance, player populations across these sites do not look very different from each other, and do not differ greatly from the populations reported on by other larger-scale studies (for example Williams, Yee and Caplan, 2008). They are all overwhelmingly male-dominated; the mean age is between 22.5 and 25.5 years old; and over 85% of players in each site claim to be lower- to upper- middle class (that is, there are no extremes of socioeconomic status at an of the events). Key differences are with regards to race, with almost all Insomnia participants describing themselves as either Caucasian or European, compared to much higher numbers of Asian-identified players in the Toronto and Vancouver sites.
Do you currently participate in virtual worlds?
Because we were recruiting participants from different public sites, rather than via online gaming forums or within MMOGs, we wanted to ascertain how many participants were actively engaged in MMOG play at the time of their participation in the study (Table 2). Survey responses show an apparent inconsistency in players' answers regarding their current MMOG play. Although one quarter to one half of participants at each type of site answered “no” to the question, “Do you currently participate in virtual worlds?” a large portion of these nonvirtual world players listed one or more avatars in response to the survey item asking them to list the gender and level of their virtual world characters.5 Rather than looking at this as an inconsistency in self-reporting, we can instead draw on our face-to-face interactions with players as they filled out the survey, where they explained that they had either recently given up, or were on temporary hiatus from, a particular MMOG but nonetheless wanted to include their avatars from that game. In other words, they demonstrated an allegiance to their MMOG avatars even when they were on hiatus from play, a practice that was only fully clarified through face-to-face interactions with players.
|“Do you participate in virtual worlds?”||No / Yes||Percentage|
|LAN Events, England (n = 183)||No||27|
|LAN Events, Toronto and Vancouver (n = 57)||No||36|
|Internet cafés, Toronto and Vancouver (n = 49)||No||26|
|Game culture events, Toronto and Vancouver (n = 93)||No||51|
Our rationale for recruiting participants who may not have been playing MMOGs at the time of their participation in the study is twofold. First, we wanted to get a sense of how great an overlap there is between people who play digital games in public, and people who play in MMOGs. Researchers studying MMOGs have suggested that they act as virtual analogs of, and proxies for, real-life meeting places (for example, Steinkuehler and Williams, 2006); the underlying notion is that as opportunities to participate in ‘real world’ forms of civic life diminish, MMOGs become sites for new kinds of sociality and civic participation. We wanted to see whether this notion that online play is supplanting embodied, colocated forms of sociality holds weight, by getting an approximation for how many attendees at various public gaming events are also MMOG players. Second, we wanted to explore the ways that players' affiliation with a particular MMOG, or with MMOGs as a whole, and their ability or willingness to self-ascribe as “MMOG player,” shifts over time and place. In other words, describing oneself as an MMOG player is not a simple “yes” or “no” question. As we saw at all the sites we visited, but particularly at LANcouver (June 2011), i43 (August 2011) and i44 (November 2011) for instance, many participants either hesitated to self-identify as MMOG players or simply did not, even though they have been involved with MMOGs in the past and may be waiting eagerly for new ones (at the time, it was Star Wars: The Old Republic and Guild Wars 2) to arrive. That is, they were MMOG players in the past, and were likely to be MMOG players in the near future, but felt uncomfortable labeling themselves as “virtual world users6” on a survey. Clearly, at least with regards to these players whose participation in a community of gamers stretches across multiple games and includes both online and face-to-face contexts, the notion that online games somehow supplant ‘real world’ sociality is inaccurate and limiting.
Stories from the Field
We now turn to what this demographic picture does not capture, by offering stories of our interactions with participants that illustrate the ways in which they elude (and often explicitly resist) any kind of straightforward categorization based on which MMOG they are playing at a given point in time. We also demonstrate our own agency, as researchers shaping data gathered from fieldwork at multiple public gaming events into a single (semi) coherent account, in constructing silences, gaps, and partial understandings of these player populations.
A Time and Place to Play
As discussed above, for many participants we met, their self-ascription as players of a specific MMOG, or as “MMOG players” more broadly, was highly contingent upon both time and place. One participant we encountered at an Internet café in downtown Toronto, for instance, was a committed WoW player at the time we met him. The owner and manager of the sparsely furnished café, he told the researcher that he had WoW open on his computer whenever he worked, since his job consisted almost exclusively of sitting at his computer at the café's front desk, receiving, supervising and occasionally chatting with customers. He boasted that he played “96 hours of the week,” since he worked 4 days a week in 24-hour shifts, and pointed out to the researcher while completing the survey that his weekly investment in WoW is literally “off the charts”.7 He also explained that he ran a WoW guild comprised mostly of socioeconomically disadvantaged young men that frequented his café, and acted as a kind of father figure to them by serving as their WoW mentor and by providing them with a safe space to spend time - a relationship that makes any distinction between ‘real’ and ‘virtual’ sociality untenable. And yet, this participant clearly and repeatedly stated to the researcher that he only plays WoW at work; “when I go home,” he remarked, “I don't touch that shit.” With such a clearly demarcated division between work/WoW life and home life, this participant's MMOG play is bound to the temporal and spatial boundaries marking his role as Internet café manager.
While this participant viewed his WoW play (and his ability to come ‘out’ as a WoW player) as strongly bound by time and place, we saw other ways in which participants shifted allegiances and affiliations to particular MMOGs in different settings and at different times, becoming quite literally different players – and different people – in the process. One younger male participant whom we met twice at the Insomnia events (at i40, in August 2010, when he'd just turned 18, and 15 months later at i44, in November 2011), was at drastically different points in his MMOG career at these two events. When we first met him, he was in a competitive first-person shooter (FPS) clan, sponsored by a computer peripherals manufacturer; sporting short hair and glasses, he wore his clan's soccer-style jersey featuring the name of the sponsor in large letters on the front. When speaking to us, he described himself as a committed WoW player who was taking his first tentative steps into EVE Online, at the urging of other clanmates. When we met him over a year later, he sported long hair, a lip ring, no glasses, and a tight t-shirt instead of a clan jersey. Since last meeting him, his clan had disbanded for loss of sponsorship, he'd started university, moved out with another ex-clanmate, and started playing EVE Online almost exclusively. He not only looked different; partially by virtue of his shifting involvement in different gaming communities, he had changed as a player and as a person.
We met similar attendees at the Insomnia events who, like the café owner in Toronto, only played MMOGs when they were at events where they could interact with their guild, clan or corp-mates in face-to-face settings, as well as numerous players in Internet cafés who only had the time and opportunity to play MMOGs in these sites. Categorizing these players according to what game they played would mean disregarding the highly dynamic circumstances under which they not only play MMOGs, but self-identify as “MMOG players.”
In addition to these instances demonstrating the contingency of their MMOG play, a number of participants complicated the categorization of their gaming activities as either ‘online’ and ‘offline,’ using one data collection tool in particular – a virtual world travelogue – to show the interconnectedness of their MMOG play and their everyday relationships and embodied experiences. Building on the use of journals allowing ethnographic participants to document their experiences (Denzin and Lincoln, 2011), the travelogue was a web-based form that presented players with a series of questions about their activities in the MMOG of their choice, to which they appended both textual and visual responses. The travelogue was designed to get a glimpse into participants' everyday/everynight play, giving them control of what they choose to present to us as researchers, and in that way, providing a visual record from their play.
Rather than only append screenshots of in-game MMOG play, seven of the 24 participants who completed the travelogue also included pictures of offline experiences, primarily in response to the prompts, “What do you do when your friends are not online?” and “Where do you and your friends hang out?” While most participants uploaded images of in-game environs where they gathered with other players or undertook solo activities, these seven participants uploaded pictures of offline leisure activities and relations - mountain biking, sitting on a beach with a romantic partner, asleep at a desk next to a pile of empty energy drink cans, to name a few. In the context of a tool explicitly geared towards documenting in-game activities, these participants seem to be resisting the clear distinction between game-based forms of sociality and other leisure activities. As MMOGs continue to be framed as virtual laboratories analogous to ‘real life’ social practices (Huh and Williams, 2010; Keegan et al, 2010; Xanthopolou and Papagiannidis, 2012), these travelogue responses show the ways in which, for these participants, MMOG play is deeply intertwined with their everyday experiences.
Recruiting Women: Populations of Inconvenience
In addition to the contingencies associated with whether and how participants in our study address the seemingly simple questions of whether they play in MMOGs, and what they do when online, we also encountered conceptual and practical challenges in generating a representative sample for our study – particularly when it came to recruiting female participants. Across all of our sites, women made up less than 15% of our study participants. While this figure appears low – particularly in comparison to studies of the proportion of female players in MMOGs (see, for example, Williams, Yee, & Caplan, 2008) and among more general populations of players (ESA, 2012) – it reflects our attempts to deliberately recruit female attendees at these public events. We often went out of our way to seek out and solicit female attendees who were massively underrepresented in these spaces. We find this gap between rates of female participation as reported on among other (online) gaming populations, and our own oversampled group of female participants in this study of public gaming sites, instructive: It demonstrates the extent to which the gaming in public remains a male-dominated endeavor, at a time when the rate of women involved in video game play is reported to be on the rise (ESA, 2012).
In conjunction with the low rates of female participation, it was difficulty in many cases to recruit female attendees to the study. Particularly at the LAN events, it was easy for male researchers to meet and recruit to the study groups of men; in contrast, we rarely saw women either on their own or in groups of other women. We observed none of the all-female clans that often appear at competitive gaming tournaments (Kennedy, 2006), and only on one occasion, at one of the Insomnia LANs, did we observe and meet a lesbian couple. Generally, female attendees were rarely unaccompanied by male partners, friends or relatives, meaning that they often came as part of a ‘package deal’: They participated in the study only because the men they were with participated. This is similar to findings reported by Williams, Yee, and Caplan (2008), regarding female participation in the MMOG Everquest 2.
Not only were women clearly a minority across these sites, but the sociospatial organization of their attendance often made it difficult to approach them. At LAN events in particular, female attendees who were part of larger, male-dominated groups (usually the girlfriend, wife or relative of one of the male attendees) were typically positioned at the end of that group's row of computers, often playing a single-player game while the other (male) group members played in multiplayer games together. Both in terms of their physical and virtual positioning, these women were marginal to their group's activities. While this would seem to make it easier to solicit their involvement, it often felt as if we were singling them out (of course we were), which seemed to elicit their discomfort on several occasions, regardless of whether they were approached by male or female researchers.
These examples illustrate the interconnected assemblages of participants, researchers, games and gaming platforms, and the spatial arrangement of public gaming sites, through which we constructed a demographic look of online play in public sites. What we have tried to illuminate are the ways in which research is actively produced, not just by the researchers, but by the material and technological contexts of research, and by the participants themselves.
If the work of sociology is, as science and technology scholars have argued, to “reassemble” the relationships between actors in a given domain (Latour, 2005), it is crucial that scholars interested in observing and documenting MMOG play take seriously the processes and tools involved in this reassembling – particularly as MMOGs are gaining increasing attention in educational (Ketelhut et al, 2007), business (Reeves and Malone, 2007), and military domains (Bonk and Dennen, 2005). The questions we have begun to address in this work are as follows: what happens when we generate a quantitative body of data on MMOG players that is driven by the tools of face-to-face qualitative research? What are the implications, and challenges, of this methodological and ontological shift in how players are reassembled?
Materialities in Play
The primary implication of this analysis is that time and place matter deeply to participants' ability to engage in networked play, and to our ability to study it. In her look at a local gaming club in Sydney, Australia, Melanie Swalwell (2006) argues that attention to public play sites can help disrupt the compelling but limiting notion that online games offer alternatives to (or analogues of) offline sociality, without the messiness or friction of embodied interaction. Part of what makes this ontological distinction so plausible (and convenient) is that the technological infrastructures constructing virtuality make our embodied identities invisible. Invisible does not, and should not, mean irrelevant. The materiality of the particular contexts we visited - what Simon (2007) calls the “social contextures” of gaming – shaped the kinds of activity that both researchers and participants were able to undertake. At most Internet cafés we visited in Toronto and Vancouver, where participants were separated by cubicles and verbal or physical contact with other patrons was discouraged, this mostly consisted in intense bouts of play – similar to the stereotypical image of the gamer as lone (and most often male) isolate. The location of most of these cafés, in large commercial and urban centers and in close proximity to schools and transit lines, meant that most attendees were “on the clock” as soon as they arrived, making use of a short window in their day (or night) to game, catch up on e-mail, or surf the web in settings that were ostensibly ‘public’ but in practice solitary and private. This assemblage, as noted above, made approaching and interacting with participants difficult.
In contrast, the temporal and material arrangement of LAN events made it possible for participants to engage in far more fluid and prolonged forms of play. The heterogeneous mix of activities we observed was similar to that reported on by Taylor and Witkowski (2010) regarding their fieldwork at DreamHack, the massive annual LAN party in Sweden. Attendees taking part in networked play, whether in MMOGs or in other online games, most often played with other attendees in physical proximity to them (usually their friends, romantic partners, or guildmates), emphasizing that for many of these players, the purpose of attending a LAN is to take part in both the mediated and unmediated forms of sociality they afford.
This work troubles the persistent notion that MMOGs are analogous to (rather than embedded in) real life contexts, and that they can act as virtual laboratories in which to study anything from drag (Huh and Williams, 2010) to leadership (Goh and Wasko, 2009) to criminality (Keegan et al, 2010). This framework continues to be hoisted in quantitative studies of online gaming, resulting in accounts in which, for example, game-based practices around the accumulation and selling of virtual currency (“gold farming”) are regarded as significant insofar as they illustrate how real world criminal networks operate (Keegan et al, 2010) – and not for the ways they re-entrench global systems of economic inequity (and attendant racialized discourses) through the proliferation of cheap developing world labor supporting developed world leisure (Nakamura, 2009). As our participants themselves demonstrated, in our observations and through their travelogue responses, what goes on in MMOGs is not analogous to offline contexts, as if players are either here in the ‘real world’ or there online, but are enmeshed in these contexts. And yet, this idea continues to shape how we understand the significance of MMOGs.
In contrast to this framework, and in keeping with an actor-network perspective, the attention we draw here to the material and temporal conditions of these different sites invites us to view networked play as production made possible by players' interactions within a network of human and nonhuman actors, encompassing both offline and computer-mediated activities. That, for example, the young man we met one year was a very different person the next time we encountered him, demonstrates with some force the precarious and often times overstated claims about MMOG players, when these claims may be more accurately described as simply “moments in time” that we as researchers might log and describe. Instead, constructed in the ways we have explored in this paper, one is not a “World of Warcraft player”; one is a player who may have played WoW for years and is now fed up with it, or used to lead raids but now only grinds out titles alone, or only plays at LAN events. These kinds of engagements illustrate the limitations of categorizing players based on the MMOG they currently play, and of reading that MMOG play as somehow analogous to but separate from their everyday life.
‘Real World Avatars?’
We suggest that in holding players to fixed categories - the “World of Warcraft player,” the “gold farmer,” the “role-player,” and so on – we deny them agency and dynamism as players and they essentially become, for the purposes of social scientific knowledge-building, avatars. Only they are not in-game avatars created by players themselves, but ‘real world’ avatars – stand-ins for living, breathing humans – built and deployed by researchers, in order to make the messy, meaty aspects of MMOG play more manageable. By contrast to the ephemeral, pliable avatars wireframed in the ‘data points’ selected by quantitative, online research based on a specific MMOG at a specific time, we have attempted to preserve in our account the lively, unpredictable activities and identities of our participants. Following from this, the challenge of communications research in studies of networked play, whether qualitative or quantitative, is to deploy methodologies that acknowledge that online gaming is produced through coconstitutive relationships between bodies, contexts, tools (including the tools we use as researchers), and games. That is, we need approaches that are as capable of accommodating contingency, multiplicity and transformation as players are themselves.
Having explored the implications of this work for our understandings of the contingent identities of MMOG players, what remains is to extend this framework to analyses of players' in-game activities. How do the material settings of play - including other cosituated players (and nonplayers), platform, Internet connection, and so on – affect how players communicate and interact in online game spaces? What, in terms of players' affiliations and practices, ‘travels’ as they move from one MMOG to the next? Given that we not only collected survey data for the purpose of this study, but also data collected from players in an instrumented, browser-based multiplayer role-playing game, our next step is to empirically document how participants become different players under shifting virtual and material conditions.
This research was sponsored by the Air Force Research Laboratory.
There is very little about the contexts in which play unfolds that is “natural” – games are, after all, thoroughly designed spaces, where players' interactions are mediated and enframed by environments enacted wholly through code. Similar can be said for the physical contexts of play spaces, as has been discussed elsewhere with regards to e-sports tournaments (Taylor, 2011) and as we will discuss here in relation to MMOG play in public sites.
This organization is based on the following rationale: we wanted to separate sites and events based on country (Canada vs the UK); the physical configuration of bodies and gaming machines (FanExpo and the GLBTQ pub nights vs Internet cafés and LAN events); and whether the sites were managed by businesses occupying a commercial space (Internet cafés) or by organizations temporarily making use of an event space (LAN events, FanExpo, and GLBTQ pub nights).
With regards to race, the top two responses for each group are displayed, with remaining responses grouped as “Other”; for socioeconomic status (SES), categories “Lower” and “Upper” are grouped as “Other”; and for education level, “Primary school” and “Other” are combined.
The survey did not make questions about virtual world use dependent on whether the participant had said they currently play in virtual worlds.
While we clarified to participants that for the purpose of the study, the term “virtual worlds” was interchangeable with “MMOGs”, it is important to note that many participants were initially confused by the term. They were either unsure of what constitutes a “virtual world” (Facebook? League of Legends?) or they regarded them as different from MMOGs. In most instances, we were able to clarify our use of the term with them on the spot, as they took the survey, a clear advantage of administering surveys face-to-face.
On the survey question, “Approximately how many hours do you spend each week in virtual worlds?” the highest possible answer is “30+”.
- 2005). Massive multiplayer online gaming: A research framework for military training and education. Madison, WI: Advanced Distributed Learning Initiative. , and (
- 2005). Gendered gaming in gendered space. In J. Raessens & J. Goldstein (Eds.), Handbook of computer games studies (pp. 301–310). Cambridge: MIT Press. , and (
- 2006). The games we play online and offline: Making Wang-tta in Korea. Popular Communication, 4(3), 225–239. (
- 2012). Re-mediating research ethics: End-user license agreements in online games. Bulletin of Science, Technology & Society, 32(6). , and (
- 2011). The SAGE handbook of qualitative research, 4th ed. Thousand Oaks, CA: SAGE Publications. and (
- 1990). When is a tool? Multiple meanings of artifacts in human activity. In Y. Engeström (Ed.), Learning, working and imagining: Twelve studies in activity theory (pp. 171–195). Helsinki: Orienta-Konsultit. (
- Entertainment Software Association. (2012). 2012 Sales, Demographic and usage data: Essential facts about the computer and video game industry. ESA. Available online at: http://www.theesa.com/facts/pdfs/ESA_EF_2012.pdf.
- 2007). Playing with non-humans: Digital games as technocultural form. In S. de Castell & J. Jenson (Eds.), Worlds in play: International perspectives on digital games research (pp. 115–128). New York: Peter Lang. (
- 2009). Where's the leader? Identifying leadership candidates within virtual worlds. MG 2009 Proceedings, Paper 9. and (
- 2011). Using sequential observations to model and predict player behavior. FDG ‘11. Bordeaux, France, June 29-July 1. and (
- 2010). Dude looks like a lady: Gender swapping in an online game. In W. Bainbridge (Ed.), Online worlds: Convergence of the real and the virtual. Springer, New York. and (
- 2005). Gaming at a LAN event: The social context of playing video games. New Media & Society, 7(3), 333–355. and (
- 2010). Dark gold: Statistical properties of clandestine networks in massively multiplayer online games. SocialCom 2010. Minneapolis MN USA. Aug. 20–22, 201–208. , , , , and (
- 2006). Illegitimate, monstrous and out there: Female 'Quake' players and inappropriate pleasures. In J. Hollows and R. Pearson (Eds.), Feminism in popular culture (pp. 183 – 201). London: Berg (
- 2007). Studying situated learning in a multi-user virtual environment. In E. Baker, J. Dickieson, W. Wulfeck & H. O'Neil (Eds.), Assessment of problem solving using simulations. Mahwah, NJ: Lawrence Erlbaum Associates. , , , , & (
- 2005). Reassembling the social: An introduction to actor-network theory. Oxford: Oxford University Press. (
- 2008). Body, space and gendered gaming experiences: A cultural geography of homes, cybercafés and dormitories. In Y. B. Kafai, J. Denner, C. Heeter and J. Sun (Eds.), Beyond Barbie and Mortal Kombat: New perspectives on gender and computer games (pp. 54–67). Cambridge, MA: MIT Press. (
- 2010). Time to play? Activity engagement in multiplayer online role-playing games. Transportation Research Record: Journal of the Transportation Research Board, 2157, 129–137. , , , , (
- 2009). Don't hate the player, hate the game: The racialization of labor in World of Warcraft. Critical Studies in Media Communication, 26(2), 128–144. (
- 1995). Context and consciousness: Activity theory and human-computer interaction. Cambridge, MA: MIT Press. (
- 2007). Leadership in games and at work: Implications for the enterprise of massively multiplayer online role-playing games. Report prepared for IBM. and (
- 2007). Geek chic: Machine aesthetics, digital gaming, and the cultural politics of the case mod. Games and Culture, 2(3), 175–193. (
- 2006). Where everybody knows your (screen) name: Online games as "third places." Journal of Computer-Mediated Communication, 11(4). , and (
- 2006). Multi-player computer gaming: 'Better than playing (PC Games) with yourself'. Reconstruction: Studies in Contemporary Culture, 6(1). (
- 2009). LAN gaming groups: Snapshots from an Australasian case study, 1999–2008. In L. Hjorth and D. Chan (Eds.), Gaming cultures and place in Asia-Pacific (pp. 117–136). Oxon, UK: Routledge. (
- 2011). Play globally, act locally: The standardization of pro Halo 3 gaming. International Journal of Gender, Science and Technology, 3(1). (
- 2012). Virtual postcards: Multimodal stories of online play. Alt.chi ‘12 Extended Abstracts on Human Factors in Computing Systems. Austin, TX: May 6–10. , , and (
- 2009). The assemblage of play. Games and Culture, 4(4), 331–339. (
- 2012). Raising the stakes: E-sports and the professionalization of computer gaming. Cambridge, MA: MIT Press. (
- 2010). This is how we play it: What a mega-LAN can teach us about games. FDG ’10. Montery, CA, June 19–21. and (
- 1999). Gender and the landscapes of computing in an Internet café. In M. Crang, P. Crang, & J. May (Eds.), Virtual geographies: Bodies, space and relations (pp. 178–202). New York: Routledge. (
- 2011). Behind the avatar: The patterns, practices, and functions of role playing in MMOs. Games and Culture, 6(2), 171–200. , , and (
- 2008). Who plays, how much, and why? Debunking the stereotypical gamer profile. Journal of Computer-Mediated Communication, 13, 993–1018. , and (
- 2012). Play online, work better? Examining the spillover of active learning and transformational leadership. Technological Forecasting & Social Change, 79, 1328–1339. and (
- 2011). Introverted elves and conscientious gnomes: The expression of personality in World of Warcraft. Proceedings of CHI 2011, 753–762. , , , (