Open Source Infrastructures
Discussion
Alberto Corsín Jiménez:
"The open source movement has drawn considerable attention of late. ‘Openness’ has become a favoured slogan for describing the epistemic pressures and transformations undergone by the political economy of knowledge and technoscience in the age of informational capitalism. Thus, whilst some authors have argued that the structure of digital information—in particular the negligible costs of reproduction—instantiates a de facto regime of superabundant or ‘open knowledge’ (Foray, 2006, pages 172–179), others have drawn attention to the enclosure of such informational commons by existing proprietary regimes (Boyle, 2008).
However, as Charlotte Hess and Elinor Ostrom have pointed out, it is worth remembering that access to information depends to this day on the maintenance and management of complex infrastructural facilities. Their suggestion is to think of information not as a superabundant commons but as a common‑pool resource (Hess and Ostrom, 2003): a resource that requires storage and preservation, and over which one must define rights and rules of access, extraction, exclusion, and alienation (Hess and Ostrom, 2006, page 7).
Whilst the economic and political underpinnings of open knowledge are on the whole well understood, there are fewer studies of the cultural practices and social organisation characteristic of such initiatives. In this regard, the aspects of the new economy of open knowledge that has received most commentary to date is its grounding in novel organisational forms, in particular so‑called peer‑to‑peer networks of collaboration (Benkler, 2006). The common view here is that peer‑to‑peer decentralised networks are blurring traditional distinctions between production, distribution, and consumption of informational forms. In this economy, users themselves become producers of content (so‑called ‘prosumers’), and cooperation becomes the economy’s main, if not only dynamo (Benkler, 2011).
Although the organisational qualities of F/OS software projects have been much commented on (Ghosh, 2005), in‑depth or ethnographic studies of F/OS cultures are much harder to come by. Chris Kelty’s (2008) historical and ethnographic account of the development of free software has already become a classic in the field. Kelty has suggested that communities of free software developers may be conceptualised as ‘recursive publics’: a form of public sphere where the architectural framework for debate and exchange is self‑ grounded in the very activity of writing, editing, patching, or recompiling the infrastructure (code) that programmers work with. In a recursive public, technology is deployed “as a kind of argument, for a specific kind of order: [free software developers] argue about technology, but they also argue through it. They express ideas, but they also express infrastructures through which ideas can be expressed (and circulated) in new ways” (Kelty, 2008, page 29, emphases in the original). The notion of a ‘recursive public’ offers, then, a very useful analytical framework with which to rethink the nature of politics when the infrastructures of participation are themselves open to (self‑)modulation.
Gabriella Coleman’s recently published ethnography of the ethics and aesthetics of hacking similarly draws on long‑term anthropological engagement with free software programmers (2012). Whilst Kelty draws attention to the structural innovation that ‘recursivity’ brings to public sphere theories, Coleman focuses instead on the cultures of liberalism that hacking enacts. According to Coleman, the toils and pleasures of hacking reveal a contradictory and tense relationship with (American) traditions of political liberalism. Hacker attitudes and work routines simultaneously challenge and take residence within liberal conceptions of freedom, labour, and intellectual property. Thus, whilst some hackers work towards protecting individual autonomy from intrusive corporate behaviour, others promote an experience of autonomy and freedom that rejoices in the virtues of sharing and learning. And yet other programmers take pleasure in the culture of transgression that characterises certain underground hacker practices (see also Coleman and Golub, 2008).
These innovative ethnographies of F/OS projects have shown how interventions in the domains of technology and property may also be conceptualised as interventions in the domains of collaboration and social and political invention; indeed, how these in fact subtend all politics as infrastructural politics. There is a historical lesson worth briefly rehearsing here, about the proprietorial and sociological frameworks and traditions that such infrastructural politics has contributed to exposing. According to Fred Turner (2006), digital utopianism, hacker ethics, and the political economy of peer‑to‑peer collaboration are long‑term developments from Norbert Wiener’s interdisciplinary postwar experiments in cybernetic philosophy (1989), and in particular from the do‑it‑yourself, environmental, and ‘homebrew’ computing countercultural movements inspired by cybernetics. In this reading, the philosophy of cybernetics is deemed responsible for laying the conceptual ground for talking of ‘informational exchanges’, virtual communities, or digital economies of cooperation more broadly. It has contributed to the understanding of open source information as a political technology.
It is hardly coincidental, therefore, that the original impetus and advocacy for free software as a nonproprietary technology took root in the context of such countercultural movements. For indeed what first sparked the creation of free software licences—and has remained the most important cultural and political innovation in free and open source (F/OS) projects over the past thirty years—was the status of ‘openness’ as a proprietary object. The creation of the GNU General Public Licence (GPL) by Richard Stallman in the 1980s, which is the original and archetypical licence on which F/OS software programmes run, marked the first time that copyright law was hacked by the invention of a ‘copyleft’ licence (Kelty, 2011). But the controversy surrounding its original development is echoed today in almost all debates concerning the legal status of digital objects (Gillespie, 2007).
These include discussions about the intangibility, materiality, and legal agency of software and digital objects; the distinction and/or breakdown of the idea–expression dichotomy; and the ensuing problem of adjudicating between creativity, collaboration, and shared practice in software development. What is a copyrightable object (code, firmware, binaries)? Who are the (collective, individual, relational) subjects of rights? And how do these various distinctions map onto political, social, and ethical categories (the public domain, freedom, the commons)? These are all controversial debates that have been sparked anew by the digital revolution (Biagioli et al, 2011).
The study of open source urban hardware projects is part of these debates, posing a formidable challenge to—and spurring considerable innovations in—property forms and law. For in most countries the licensing of designs in open hardware projects falls under the jurisdiction of patent law, not copyright law.
Unlike F/OS programmes, open source hardware projects produce tangible outputs— artefacts, devices, machines. According to the Open Source Hardware Definition and Statement of Principles, what makes a piece of hardware ‘open’ is its design process (OSHW Draft, “Definition of free cultural works”, http://freedomdefined.org/OSHW). It is the design, not the object qua object that remains open. This makes open source hardware fundamentally different from F/OS software, in that design and output do not coincide in the same object.
In software the code is the design and the infrastructure; in hardware, the tangible output is assembled from the design. In other words, the source of openness in open source hardware is the design. The design of an open source hardware project encompasses therefore both the design of its hardware and software and the documentary record of its design process. That is, one needs to make explicit how (in what formats, files, stages, languages) the process of design is described, documented, and published. Step by step such documentary registry spells a methodology of design. Hardware design documentation includes, for example, mechanical drawings, circuit‑ board layouts, photographs, and descriptive texts. There are many layers to a design, and open source hardware projects require that every native component of a design be rendered ‘open’, or that if a portion of the design is not released under a F/OS licence it be made clear.
An important consequence of the role played by the methodology of design in the standardisation of an open hardware project is its legal affordances. Open source hardware licences are generally limited to controlling the circulation of design documents and have no purchase over manufactured objects. In open source hardware, then, the method of design is turned into a social form with proprietary effects: that is, the organisation of the network of collaborators; the material qualities of components, formats, and layers (photographs, sketches, code); and the methods of description and documentation are all entangled in the social process of making or hacking property. Property is not the outcome or output of a design process, but the very design process itself. The question of what is open—and how the process of opening it up is carried out—when the underlying object is hardware rather than software, remains, therefore, a hotly contested issue, and one about which little is known to date." (http://www.envplan.com/abstract.cgi?id=d13077p)