4.2 The Organizational Structure of Peer Projects
The intricacies of self‐governance have been a prime focus of research into peer production (Arazy et al., 2019; Auray, 2005; Dafermos, 2012; O’Mahony & Ferraro, 2007; O’Neil, 2009, 2014; Pentzold, 2018). For example, Wikipedia has variously been described as “anarchic” (Reagle, 2005), “democratic” (Descy, 2006), “polycentric” (Mindel et al., 2018), and “meritocratic” (Bruns, 2008). It has also been called a “hybrid of different governance systems” (Holloway et al., 2007), a “self‐governing institution” (Spek et al., 2006), a form of “collective governance” (Aaltonen & Lanzara, 2015), an “adhocracy” (Konieczny, 2010), and an “ethical‐modular organization” (O’Neil, 2015). From a historical perspective, peer production projects can be likened to self‐run organizations such as cooperatives and kibbutzim. However, their most clear antecedents are what were previously known as “voluntary associations” and “collectivist organizations.” During the 1960s, the rejection of traditional societal institutions as well as of colonialism and imperialism led to the rise of countercultural groups in the Global North, which explicitly rejected what Max Weber called “traditional” and “legal‐rational” forms of authority (Weber, 1947). Social‐scientific interest in communes and cooperatives accordingly increased. The workings of collectivist organizations were analyzed by Rothschild‐Whitt (1979), who defined them as alternative institutions which “self‐consciously reject the norms of rational‐bureaucracy” (p. 509). Collectivist organizations are groups in which decisions become authoritative to the extent that all members have the right to full and equal participation. There are no established rules of order, formal motions and amendments, or votes, but instead a “consensus process, in which all members participate in the collective formulation of problems and negotiation of decisions” (pp. 511–512).
Following on from the 1960s concern for more inclusive and participatory forms of activism and politics (Kaufman, 1969), the necessity of rules was recognized in collectivist groups in order to avoid what feminist activist Jo Freeman (1972) called the “tyranny of structurelessness”: the absence of explicit rules facilitates power being monopolized by informal cliques who manipulate communications (by approving the declarations of their fellows and ignoring or disparaging those of others) and decisions (by deliberating secretly). In line with this concern – how to democratically organize cooperation amongst volunteers – researchers have examined rule‐making in self‐organized peer projects (Auray, 2005; O’Mahony & Ferraro, 2007; see also Karp et al., this volume; Pentzold, this volume). Authors who examined the emergence and evolution of FOSS communities such as Debian (Coleman, 2012; O’Mahony & Ferraro, 2007; O’Neil, 2014) and FreeBSD (Dafermos, 2012; Jørgensen, 2007) focused on the “succession problem,” that is, on the evolution from an informal mode of legitimacy, organized around the figure of the charismatic founder, to a more formalized and democratic mode.
Online collectivist organizations comprise typically bureaucratic mechanisms such as the maintenance of archives of all decisions, and precise rules (O’Neil, 2009), which are nowadays increasingly formalized in Codes of Conduct. In order to introduce a bureaucratic basis of authority into a community form, members must design democratic mechanisms to limit that basis of authority (O’Mahony & Ferraro, 2007). In Debian the power of the project leader is limited in four ways: leaders must defer to the project; they have limited authority over technical matters; members can recall leaders; and the authority of leaders is counterbalanced by that of a Technical Committee (see also interview with Zacchiroli, this volume). Wikipedia editors who demonstrate their willingness to work for the common good can become “administrators,” “stewards,” or “bureaucrats” and hence exercise varying degrees of control over other participants. They can be replaced by other Wikipedians: meritocracy and the separation of roles and persons are other key characteristics of bureaucratic systems, in contrast to charismatic and traditional leadership (Weber, 1947). The difference with traditional bureaus is that rules are supposed to be generated transparently and democratically (O’Neil, 2009; Haider & Sundin, this volume). Wikipedia is thus both a formal bureaucracy with a structured system of rules and roles and a “deliberative bureaucracy” as decisions are made through “consensus with universal and equal participation, and care is taken over time to create mutual understanding and agreement” (Joyce, Pike, & Butler, 2013).
In sum, protection from abuse in peer production is provided by the community, rather than by state‐backed contracts, so recourse procedures may be unstable; nonetheless, Kreiss et al.’s (2011) assessment that peer organizations offer no protection from unjust domination is incorrect. It remains the case that the legal protection of (for example) the GPL applies to the result of labor, not to workers themselves. Peer organizations are not legally responsible for the welfare of participants, so do not offer the same level of support as formal bureaucracies do (O’Neil, 2015).
How do contemporary collectivist organizations fit into the wider political economy? Table 1.1 maps contemporary configurations of work according to their extrinsic or intrinsic logic (vertical axis) and to the degree of control workers are afforded over their labor (horizontal axis).
Organizations where workers sell their labor for wages and where decision‐making rests with a restricted leadership are situated in quadrant A. In contrast to employees of capitalist‐centralized firms, independent workers such as informal networks of tradespeople in quadrant C have much greater latitude as to whether to work on a particular job. Quadrants B and D comprise communal forms of labor. The community validation and self‐fulfillment of domestic labor are lesser than those accrued by participants in voluntary/collectivist associations, but they exist nonetheless. Indeed, Marxist feminist scholar Kylie Jarrett (2016) draws a comparison between the affective or reproductive work of traditional housewives and the unpaid labor performed by contributors to digital platforms. But where workers in quadrant D have control over their production, workers in quadrant B who freely engage in consumption work (or its more interactive variants “prosumption” and “co‐creation”) must accept that their contributions to product development or networked communication are subject to an external authority's approval or disapproval, with no possibility of redress: this distinction is perhaps not sufficiently drawn out in Jarrett’s (2016) otherwise excellent book, which conflates FOSS labor with that of Facebook users for example.
Table 1.1 Organizational governance and logic: a typology.
Source: O’Neil, M. (2015). Labour out of control. The political economy of capitalist and ethical organizations. Organization Studies, 36(12): 1627–1647. © 2015 SAGE Publications.
|
|
Capitalist logic (“alienated” labor) | Ethical logic (“communal” labor) |
|---|---|---|
| Centralized governance | A private firms, public administration, non‐governmental organizations | B consumption work, co‐creation, prosumption |
| Modular governance | C independent workers, freelancers, contractors | D domestic labor, voluntary/collectivist organizations |
In contrast, maintainers in ethical free and open source software projects must take pains not to antagonize contributors, who may not only vote with their feet and exit the project, but have the capacity to use all produced resources and therefore reproduce and “fork” the project into a different direction. Actors may move from one quadrant to another: hobbyists can strike it big and grow into large centralized firms (D=>C=>A), as in the case of computer operating systems or search engines developers. Domestic labor can be
