Corporate Involvement in Free and Open Source Software
* Book: Birkinbine, B. J. (2020) Incorporating the Digital Commons: Corporate Involvement in Free and Open Source Software. University of Westminster Press
URL = https://www.uwestminsterpress.co.uk/site/books/10.16997/book39/
"This study complicates and extends theorisations of commons-based peer production by investigating sites where the idealism of FLOSS production meets with the material realities of capitalism."
Description
1.
"The concept of ‘the commons’ has been used as a framework to understand resources shared by a community rather than a private entity, and it has also inspired social movements working against the enclosure of public goods and resources. One such resource is free (libre) and open source software (FLOSS). FLOSS emerged as an alternative to proprietary software in the 1980s. However, both the products and production processes of FLOSS have become incorporated into capitalist production. For example, Red Hat, Inc. is a large publicly traded company whose business model relies entirely on free software, and IBM, Intel, Cisco, Samsung, Google are some of the largest contributors to Linux, the open-source operating system. This book explores the ways in which FLOSS has been incorporated into digital capitalism. Just as the commons have been used as a motivational frame for radical social movements, it has also served the interests of free-marketeers, corporate libertarians, and states to expand their reach by dragging the shared resources of social life onto digital platforms so they can be integrated into the global capitalist system.
The book concludes by asserting the need for a critical political economic understanding of the commons that foregrounds (digital) labour, class struggle, and uneven power distribution within the digital commons as well as between FLOSS communities and their corporate sponsors."
2.
"The purpose of this book was to demonstrate how one such technology – free and open source software – is dialectically situated between the commons and capital. To illuminate the ways in which these forces struggle over free and open source software, my task was to ‘identify and challenge’ the ‘real agencies’ of free and open source software as commons under capitalism. In doing so, I identified the specific ways in which capital incorporated the forces of commons-based peer production into capitalist enterprises, the motivations for doing so, and the ways in which communities of free and open source software developers cope with unwanted interference in their projects. Moreover, I approached this study historically, paying close attention to the historical forces that enabled both the rise of commons-based peer production as well as the incorporation of those forces into capitalist production."
Excerpts
Commons Circuits of Value
By Benjamin J. Birkinbine:
"By combining systems theory (Luhmann, 1995), cybernetics (Maturana and Varela, 1998) and Marxist-feminist political economy (Marx 1906; Dalla Costa and James, 1975), De Angelis’s task is to demonstrate how the commons can be understood as a system capable of bringing about a social revolution through ongoing iterations of commoning activity that are reproduced over time. Rather than arguing that such a revolution is imminent, however, he takes an epochal approach by focusing on how an emergent alternative value system like the commons has the potential to bring about a change in social relations. Just as capitalist social relations and subjectivities emerged in the feudal era, De Angelis views the commons as a similarly emergent value system responding to the excesses and exploitative tendencies of capitalism. In the analytical portion of this work, De Angelis (2017) attempts to analyse the commons in the same way that Marx analysed capitalism. This leads him to develop a circuit of commons value, which accounts for the component parts of commons value systems.
In the circuit, an association of people (A) claims collective ownership of their commonwealth (CW), whether the sources of commonwealth are material, immaterial, commodity (C), or non-commodity (NC). This dual relationship between the association – as subjects – and their commonwealth – as objects – constitutes the commons (Cs). Then, through the activity of commoning (cm), which is derived from Linebaugh’s (2008) definition of the term, the commons are reproduced over time. Framing the commons this way not only adds to a growing corpus of scholarship that makes similar claims (Dyer-Witheford, 2006; Hardt and Negri, 2009; Ryan, 2013; Gutierrez-Aguilar, 2014; Singh, 2017), but it also adds critical weight to commoning practices by demonstrating how those activities are capable of bringing about a postcapitalist future.
Commoning, therefore, includes the reproduction of both the objects that comprise the commons as well as subjectivities in which mutual aid, care, trust, and conviviality are reproduced over time. For De Angelis, this commons circuit can couple with capital circuits through the commodity form. His argument is not that these two can and ought to peacefully coexist, but that they do exist. For example, when commoners must interact with the money form of capital, they do so only as a medium of exchange to gain access to the materials necessary to reproduce the commons and themselves over time. As this relates to the digital commons, a free software contributor or user still needs to have access to a computer to code the digital commons or to have access to them. In addition, the programmer will also need to have access to food, water, shelter, and all those things necessary to reproduce her own capacity to code the digital commons over time. These goods may be provided by the welfare state or one’s family but, in the absence of such provision, one would need to intersect with capital circuits to obtain them. However, the extent to which commoners engage with capital circuits is left up to the community of commoners and will vary depending on the specific needs of the community. This framework is useful for understanding the ways in which FLOSS communities relate to their digital commons."
Cases
Red Hat
Benjamin J. Birkinbine:
1.
"By selecting cases in which capitalist firms are incorporating commonsbased peer production, this study was able to yield a novel insight into how intellectual property is used both within the FLOSS community and corporations. Specifically, the case of Red Hat demonstrated how a firm is able to profit from intellectual property that is covered by the GPL and, therefore, not amenable to enclosure by traditional copyright. Because Red Hat cannot exclude others from using its source code by relying on copyright, the company uses its trademarks to prohibit competitors from making a direct use of its products. However, Red Hat’s trademarks cannot prevent someone from using the underlying source code, which is protected by copyleft. Indeed, this was the case with CentOS, which was designed as a functionally equivalent operating system to that offered by Red Hat Enterprise Linux, Red Hat’s core commercial product. Similarly, Red Hat controls the types of licences that can be included in its Fedora Project, which is the FLOSS project that generates the code included in its commercial offerings. The ways in which Red Hat controls the intellectual property included in its commercial offerings complicates the claims made about the productive autonomy within FLOSS communities.
..
The issue of trademark is an often-overlooked feature of software development. Red Hat uses trademark protections to circumvent the permissive nature of the GPL and the other licences that do not allow it to claim exclusive ownership of the code used in its core products. Although Red Hat is just one example and, perhaps, an exceptional one, the case serves as a contradictory example to the overarching claims made about the degrees of freedom, democracy, and autonomy within FLOSS production.
..
Contributor Licensing Agreements within FLOSS production, particularly when a project has a corporate or other institutional sponsor. While these agreements are not uniform across all FLOSS projects, the organisations that issue them rely on these agreements to maintain control over their projects. However, control is achieved in at least a couple of different ways. The CLAs may ask contributors to surrender the rights to their submissions so that the organisation can defend itself from intellectual property claims. Similarly, the CLAs may be used to control the types of licences that are allowed into the code base. This was seen in the Red Hat case study, whereby Red Hat wanted to guarantee to its customers that they would not be in danger of intellectual property infringement suits. A common theme running throughout the Red Hat chapter was the extent to which copyright separates authorship from ownership. In this sense, the current project contributes to this critical understanding of copyright by demonstrating how FLOSS labourers are forced to abandon claims to ownership of their work in order to contribute directly to certain FLOSS projects."
2.
"In the case of Red Hat, which still maintains a relatively good relationship with the FLOSS community, the company was able to harness (which is to say, centralise) the collective labour power of the FLOSS community and transform it into a profitable business strategy. Red Hat was created with the intention of providing a formalised institution that could bring the power of free software to the market. However, since the underlying source code for free software was protected by the GNU General Public License (GPL), Red Hat was unable to rely on using copyright protection to exclude others from providing similar software and services.
As a result, the company began offering customised versions of free software that could be packaged and protected under the Red Hat corporate logo. As such, the company’s products could be protected by trademark. The software that the company provides, then, is protected by the Red Hat trademark, and the company sells customised subscriptions for its software and services. However, Red Hat still needed a way to protect its customers against potential intellectual property infringement claims. Consequently, the company needed a way to control the types of licences allowed in its software offerings. To accomplish this, Red Hat first required all contributors to its software to sign an Individual Contributor License Agreement (ICLA), which would assign the rights to protect the code to the company. The ICLA later changed to the Fedora Project Contributor Agreement (FPCA), which served as a mechanism to control the range of possible licences that could be included in contributions to its Fedora project. Nonetheless, the consequence of controlling the commons was the same. From one point of view, Red Hat might be viewed as a pragmatic solution to the problem of organising commons-based peer production so that it can become conducive to the establishment of a capitalist enterprise. In effect, Red Hat serves as a formal organisation that can accept liability for the products and services it provides to other businesses. In other words, the problem of organising commons-based peer production under capitalism was solved by establishing a legally recognisable and formal institution that serves as a mediator between corporations and the commons. To accomplish this, however, Red Hat needed to find a way to control what types of code – or at least the types of intellectual property licences – were included in its software so that it could protect itself and its clients against intellectual property infringement claims. In this sense, Red Hat functions as a curator of the commons.
Just as a curator is responsible for collecting, organising, and interpreting artefacts for the purpose of public display, Red Hat performs a similar function for its subscribers. In each case, the curator charges a fee to the public for entrance to a purposefully organised and constructed display of artefacts that has been interpreted in a particular way. The key difference, however, is that Red Hat does not rely on the collection of artefacts exactly as they existed previously. Rather, Red Hat relies on commons-based peer production from its FLOSS project, Fedora, for inclusion into its customised distributions of Red Hat Enterprise Linux. Moreover, the contributions to Fedora are controlled by worker agreements that all contributors to the Fedora Project must sign. Importantly, however, because Red Hat is transparent about its intentions, the company has been able to enjoy a relatively good relationship with the broader FLOSS community throughout its history. Whereas Red Hat is situated as a mediator between corporations and the commons of free software production, the Fedora Project Board also serves as a boundary organisation (O’Mahony and Bechky, 2008) between the community of programmers who contribute to the Fedora Project and Red Hat. As such, it is here where the boundaries between Red Hat and the Fedora Project community are negotiated. Similar organisations exist in other FLOSS projects and serve as a useful mechanism for negotiating the boundaries between capital and the commons. Through these processes, as well as the mechanisms used by Red Hat to use FLOSS production as part of its business model, the Red Hat case study represents the ways in which the value of FLOSS production can move from the commons to capital.
Discussion
The Debate about the Copyfair Proposals of the P2P Foundation
Benjamin J. Birkinbine:
"DuLong de Rosnay and Musiani (2016) are not the only scholars wrestling with how to advance decentralised peer production forward to mount a challenge to capitalism. One such debate took place in a series of articles published in tripleC: Communication, Capitalism, and Critique in 2014. The debate stemmed from a proposal made by Bauwens and Kostakis (2014). Noting the contradictions of commons-based peer production being co-opted by capitalist firms, as well as the growing co-operative movement and worker-owned enterprises, Bauwens and Kostakis (2014) propose a convergence that they call ‘open co-operativism’ that would ‘combine Commons-oriented open peer production models with common ownership and governance models such as those of the co-operatives and the solidarity economic models’ (356). To facilitate such a movement, the authors suggest the creation of an alternative intellectual property licence that would require reciprocity to benefit the commons. They frame this as a shift from a ‘communist’ licence like the GNU General Public License (GPL), which allows anyone – including capitalist firms – to use the commons-based resource, toward a ‘socialist’ commons-based reciprocal licence which, they argue, is exemplified by the Peer Production License (PPL) as proposed by Kleiner (2010). Such a licence would allow for commercial use of the licenced resource, but would require reciprocity to the community. This means that licensing fees would be charged to for-profit companies that use the resource. This, then, would allow the community to establish a co-operative, which could receive the licensing fees as income that could then be used to maintain the commons. In effect, the goal is for the community to retain the surplus value of their production. The authors further argue that the goal of this project is to transform the mode of production toward the commons. Furthermore, they claim that without such a transition commons-based peer production ‘would remain a parasitic modality dependent on the self-reproduction through capital’ (Bauwens and Kostakis, 2014: 360). Meretz (2014) critiqued Bauwens and Kostakis’s proposal on a couple of fronts. First, Meretz critiques the ‘logic of exclusion’ embedded within the proposal for licensing. He argues that free software is not a commodity; it can be appropriated and used by everyone, but the GPL prevents its transformation into a commodity. Second, he critiques the authors’ use of ‘reciprocity’ by claiming that licences are never reciprocal. Rather, licences grant or deny access or use. Reciprocity must involve people who are reciprocal in a social relationship. Meretz’s own view is that social transformation is not possible by building a counter-economy for progressive social movements. In his words, ‘it is not possible to out-compete capitalism … to be better than capitalism on its own terrain in order to finally get rid of it’ (Meretz, 2014: 364). Rather, we need a new social logic of producing our livelihood, which will not be built upon existing logics of exclusion that mark commodity production. Indeed, capitalism must constantly open up spaces for new logics to emerge so that they can be exploited. In the end, Meretz views the proposal for a new socialist licence as a mechanism for accessing the economy rather than a means for societal transformation. Rigi (2014) offers his own views on these proposals by revisiting some foundational concepts from Marx’s work (i.e. value, profit, surplus value, and rent), then demonstrating how Bauwens and Kostakis fall short in their application of these concepts. His point is not to impose Marx’s own views on Bauwens and Kostakis, but rather to suggest that they offer concrete definitions for how they use these terms, which would aid in the development of a theory. In addition, Rigi agrees with Meretz’s claim that further engagement in the market economy on behalf of peer production communities would only lead to those practices being assimilated into capitalism. However, he also critiques Meretz for underestimating the communist nature of the GPL. Rigi’s point is that the GPL already requires reciprocity by stipulating that any derivative work produced with GPLlicenced code must also be made available under the same licence. In this regard, Rigi argues that the GPL abolishes knowledge rent, as there is no ‘owner’ of the commons who can charge rent for using the commons. Furthermore, Rigi points to companies like IBM who decided to release their proprietary code to the commons so that it could be integrated with Linux. In so doing, the scope of available commons-based code expanded through the specific mechanism of the GPL.
In the final section of his article, Rigi (2014) outlines his own vision for how radical social transformation is possible. His goal is to examine how it would be possible to use the principles and lessons from the production of digital commons to revolutionise material production. Rigi identifies two fundamental problems that must be overcome for this to be possible: territorialisation and automation. First, the production of Linux can occur regardless of geographic location, and contributions to the digital commons can be shared easily across space in very little time. This is because anyone with access to a computer (and the necessary coding skills) is able to contribute to Linux or another FLOSS project. The same cannot be said of material production. Noting both the transportation and ecological costs associated with moving material production across space, Rigi concludes that any attempt at applying commons-based peer production to material production must be geographically bounded so that the production site is in close proximity to the consumption site. Second, material production is increasingly automated, and the human contribution in this sphere is increasingly relegated to science, design, and software. Therefore, ‘a combination of a Linux mode of cooperation with automation will generalise peer production to all branches of production’ (Rigi, 2014: 400). However, certain spheres of social life will remain untouched by automation: symbolic activities (like artistic expression, knowledge, etc.), and care for humans and nature (education, ensuring ecological survival, etc.). Rigi concludes his article with some speculative proposals for how we might bring about some of these changes by specifically arguing for something he calls ‘revolutionary peer producing cooperatives’. I will revisit this proposal later in the conclusion, as it dovetails nicely with some of my own proposals. In the meantime, however, one can begin to imagine how a set of diverse and distributed communities could begin to implement practices associated with commons-based peer production. Indeed, we have already seen examples of this around the world, but these communities still need to be linked through common interests to mount a significant challenge to existing institutions. This is where De Angelis’s (2017) use of ‘boundary commoning’ becomes useful."