This paper examines how and why Google in developing Android, the popular mobile operating system, has strategically adopted particular open source practices, but ignored others. It explores how through these practices, the corporation has been able to cultivate and control a vast mobile Internet ecology. This ecology involves large telcos and equipment manufacturers, as well as a mass of third party application developers and hundreds of millions of mobile Internet users. It allows Google to use and build on the contributions of independent programmers. More importantly, it facilitates the harvesting of valuable metadata of Android users, crucial for the development of new location–specific services and advertising. This paper shows how these corporate strategies, which combine intricate technical, legal, and political–economic maneuvering, shape the rapidly growing mobile Internet, and, consequently, have far–reaching economic and cultural consequences.Contents
Introduction
Open source critique
Android as techno–legal assemblage
Android’s political economy
Conclusion
Android, the Linux–based operating system for mobile devices, is marketed as an open source project. It is advertised as:
an open–source software stack for mobile devices, and a corresponding open–source project led by Google. We created Android in response to our own experiences launching mobile apps. We wanted to make sure that there was no central point of failure, so that no industry player can restrict or control the innovations of any other. [1] How Android is presented and developed by Google through the Open Handset Alliance (OHA), a consortium of 84 companies, is of crucial interest, as Android is quickly becoming the dominant mobile operating system (OS). By the end of 2010, three years after its launch, Android was already the best–selling Smartphone platform worldwide. In December 2011, Andy Rubin, senior vice president of Google responsible for the development of the OS, claimed that 700.000 Android devices were activated every day [2]. In light of Android’s extremely rapid growth, it is vital to understand how the OS is developed and organized at a technical, legal, and commercial level. In turn, this has a major impact on the political economy of the mobile Internet.
This paper examines how and why the ideology and some of the practices of the open source movement have been adopted by Google and the OHA in developing Android. The examination considers, on the one hand, how Android’s business model is strategically developed in collaboration and confrontation with a wide variety of actors in the mobile Internet arena. These actors range from large telecoms and original equipment manufacturers (OEM) to a mass of small programmers. On the other hand, we need to understand how these strategic relations are organized, considering Android’s rapidly evolving technological architecture and the legal framework in which it is embedded.
It is highly questionable whether Android, in the light of the ideals of open source, can in fact be characterized as an ‘open source project’. Yet, the analysis of how Google and OHA adopt certain open source practices, but negate others, provides key insights in Android as a techno–legal assemblage. Moreover, it helps to understand the workings of the broader political economy of the mobile Internet, in which it is aggressively positioned. The next section will shortly discuss the history of the open source movement, as well as the critical debate triggered by this movement. This debate and its history form the necessary backdrop for examining Android’s current open source practices.
What open source software is, and how it is, or should be, developed, has been the subject of intense debate from the moment the ‘open source’ label was adopted by a key group of people in the free software movement in 1998. The term was supposed to encompass the activities of the Free Software Foundation led by Richard Stallman, as well as the wide variety of freeware, shareware, open software, and public domain software (Berry, 2008; Kelty, 2008; Moody, 2002; Weber, 2004).
Eric Raymond’s article “The cathedral and the bazaar,” first presented at the Linux Kongress in 1997 and published in 1998 in First Monday, can be considered as the founding text of ‘open source’. In this paper, Raymond, a theorist and programmer, contrasted two models of software development against each other: the cathedral and the bazaar model. He maintained that the cathedral builder’s approach, most used in commercial software development, is based on “a priori, centralised mode of production”. By contrast, the bazaar approach, which he connects with open source, draws expertise from “a thousand eager co–developers” [3]. Raymond asserted that the centralised model of experts who control a product’s development can be replaced by a multitude of co–developers — multiple eyeballs with diverse perspectives, who all contribute to a product’s development and can stimulate innovation on a much larger scale (Raymond, 1998).
In subsequent years, various theorists have celebrated the practices and ideals of open source and free software. Lawrence Lessig (2005), for example, maintained that “the issues of open source and free software are fundamental in a free society” [4]. In turn, Benkler (2006) asserted that “free and open–source software” development constitutes a “new modality of organized production”, which he called “commons–based peer production” [5]. He described this modality as:
[R]adically decentralized, collaborative, and nonproprietary; based on sharing resources and outputs among widely distributed, loosely connected individuals who cooperate with each other without relying on either market signals or managerial commands. [6] Central to free software and open source are, in Benkler’s mind, “values such as autonomy, self–reliance, gift–giving, collaboration, active participation, liberation and creativity in motivating participation” [7].
These idealistic portrayals of open source have drawn critique from different sides. One of the first critics was Richard Stallman himself, who was quick to point out the differences between ‘open source’ and the tradition of free software. Stallman (1998) insisted that open source altered the core values of free software. Such a ‘pragmatic approach’, according to him, legitimises itself by emphasising freedom from central control, but at the same time denying or restricting user freedom. Stallman (1999) argued that free software, in contrast to open source, is based on the freedom to:
a) ‘run the program, for any purpose’, b) ‘have the freedom to modify the program to suit your needs’, c) ‘redistribute copies, either gratis or for a fee’, and d) ‘distribute modified versions of the program, so that the community can benefit from your improvements.’ [8] Stallman maintained that open source’s pragmatic approach actually obscures the importance of these values.
For Raymond, the practices of free software correspond with the logic and demands of the market. He argues:
The Linux world behaves in many respects like a free market, or an ecology, a collection of selfish agents attempting to maximize utility which in the process produces a self–correcting spontaneous order more elaborate and efficient than any amount of central planning could have achieved. [9] As Christopher M. Kelty (2008) points out, the open source label was introduced by Raymond and his collaborators with the objective to shift the focus from “resistance to proprietary software hoarding”, as propagated by Stallman, to the “economic value” and “cost savings” that free software could potential offer [10].
This shift in focus turned out to be successful. In fact, Raymond’s article directly inspired Netscape’s decision in 1998 to release the source code for its Web browser. In the mid–1990s, the Netscape Navigator had been the dominant browser in terms of usage share. However, by 1998, during the first browser war, it was quickly losing users to Microsoft’s Internet Explorer. By releasing the source code of Navigator, Netscape hoped, corresponding with Raymond’s bazaar model, to receive help from the mass of outside programmers in developing its browser. This decision, however, neither paid off financially, nor led to a superior product. It did greatly enhance the resonance of the open source concept way beyond the circle of free software programmers and hackers.
While Stallman criticized Raymond for trying to make free software more attractive to corporate interests, a variety of scholars have pointed out that these free software practices were already at the time in correspondence with commercial interests. From this perspective, neither the release of the source code of the Netscape browser, nor Raymond’s article, were revolutionary. They were rather poignant signs of a culture in which the capitalist economy and gift economy were already deeply intertwined.
One of the first scholars to highlight this interconnection was Richard Barbrook. In “The high–tech gift economy”, Barbrook (1998) described the new Internet economy of the 1990s as a mixed economy. He emphasized that the free circulation of information between users, which is central to the gift economy of open source and free software, depends upon the capitalist production of computers, software, and telecommunications. Moreover, he stressed that the profits of commercial Internet corporations rely upon increasing numbers of people who participate within the hi–tech gift economy.
A similar observation, but from a more critical perspective, has been made by Tiziana Terranova (2004). Reflecting on the open source movement and its attempt to accommodate corporate interests, she maintains that
rather than representing a moment of incorporation of a previously authentic moment, the open–source question demonstrates the overreliance of the digital economy as such on free labor, free both in the sense of “not financially rewarded” and of “willingly given”. [11] This free labor not only includes the work of open source programmers, but also of amateur Web designers, mailing list editors, and online community leaders. Terranova insists that “especially since 1994, the Internet has always and simultaneously been a gift economy and an advanced capitalist economy” [12]. From the perspective of Barbrook and Terranova, the idealism espoused by Lessig, Benkler, and Raymond, and even Stallman’s attempt to distinguish free software from open source, seem to obscure the broader political economy in which free software has developed.
According to Terranova (2004), the reliance of big business on free labor is not just a specific feature of software development and the Internet, but is typical of late capitalism, which is characterized by under compensation and exploitation in the cultural economy at large (see also Söderberg, 2008). In a similar way, Matteo Pasquinelli (2010a) has criticized the apologists of open source and free software. He maintains “reading authors like Stallman and Lessig, a question rises: where does profit end up in the so–called Free Society?” Reflecting on this question, Pasquinelli characterized the relations in the Free Society as “parasitical”. He emphasizes that the peer production celebrated by the “freeculturalists” is certainly not improving the life conditions of the “last digital generations” [13].
In light of this critique and the history of open source and free software, Android presents an intriguing case. On the one hand, it appears as a continuation of the close relationship between open source and corporate interests; a relationship that has been traced in detail in the studies of Kelty (2008) and Weber (2004), and has been critically interrogated by Terranova (2004) and Pasquinelli (2010a). Yet, at the same time, Android’s particular use of open source also introduces a number of new, highly problematic, practices that require a renewal of open source critique, or more specifically of the political economy in which Android’s open source practices are situated.
Whereas the general features of this political economy have been theorized, it is still unclear how and why, in the context of this economy, Google and the OHA have adopted particular open source practices. Reflecting on the parasitic nature of late capitalism, discussed by Terranova and Pasquinelli, studies on today’s Internet have made clear that large Web 2.0 companies, such as Google and Facebook, first and foremost make money by gathering metadata on user activity to precisely profile these users, which, in turn, makes it possible to develop targeted advertising and services. To facilitate the collection of the necessary metadata, social media platforms, such as Facebook, Google+, and YouTube, are developed, in which users are invited to express and share their personal interests and identities (Allen, 2008; Cheney–Lippold, 2011; Fuchs, et al., 2011; Langlois, et al., 2009; van Dijck, 2009; Petersen, 2008).
Building on these insights, the following sections of this paper will untangle how Google and the OHA have, in the context of this political economy, strategically adopted particular open source practices while ignoring others. The legal licenses and the technological specifications documented online, as well as the disputes and lawsuits surrounding Android, are the foundation of this exploration. First, Android will be examined as a techno–legal assemblage, to understand how Android’s specific open source practices are embedded in technological specifications and legal frameworks. This part of the analysis has specifically been inspired by critical software studies research (Chun, 2006; Fuller, 2008; Galloway, 2004; Langlois, et al., 2009). Subsequently, in the second section, it will be demonstrated how Android’s techno–legal constitution allows Google to play a central role in the political economy of the mobile Internet. This part takes up a number of key concerns elaborated in the tradition of critical Internet studies and, more broadly, critical political economy (Dean, 2009; Dyer–Witheford, 1999; Fuchs, 2011; Terranova, 2004).
While Android was publicly introduced as a project aimed at preventing any “industry player to restrict or control the innovations of any other”, within the Android ecology Google clearly has control over the other involved actors. As will be discussed in the next section, through a combination of legal and technological specifications, it effectively sets the parameters of freedom and control in the Android ecology.
Android as techno–legal assemblage
A crucial element in governing the Android ecology, are the software licenses that define its legal framework. As explained on the Licenses page of the official Android platform Web site, the preferred license for Android and its source code is the Apache Software License 2.0 (APL2.0). The page goes on to say that “there may be exceptions to this, which will be handled on a case–by–case basis. For example, the Linux kernel patches are under the GPLv2 License with system exception, which can be found on kernel.org” [14]. Crucial is that the APL2.0 defines Google’s legal relationship with hardware manufacturers and other actors, who adopt Android’s source code.
The APL2.0 is a permissive free software license, authored by the Apache Software Foundation (ASF). It allows other parties to freely use and adapt the Android OS for their own purposes. However, the license is not copyleft, a concept developed by Stallman to ensure that free software remains free. Under copyleft an author first copyrights a work, but also adds distribution terms. These terms give everyone the rights to use, modify, and redistribute the program’s code. Vital to copyleft is that it ensures any program derived from this free software, even a small portion, to be distributed on the same terms. Copyleft, as Moglen (1999) points out, provides the legal context in which the labour of free software is mobilized. It prevents corporations from building on and distributing open source software under proprietary means (Stallman, 2002). The GNU General Public License (GPL), through which the Linux kernel is distributed, embodies par excellence these copyleft principles.
Reflecting more generally on the different legal frameworks for open content, David Wiley (2007) has argued, however, that copyleft principles can also obstruct the development of open content. Advancing this argument, he focuses on the GNU Free Documentation License (GFDL) and, more importantly, the Creative Commons (CC) project, which has broadened the concept of the Free Software license to other types of content [15]. Wiley maintains that “while the CC and GFDL copyleft clauses guarantee that all derivative works will be ‘open,’ they also guarantee that they can never be used in remixes with the majority of other copylefted works”. The problem being, that derivatives from a copylefted work need to be licensed in exactly the same way as the original. And since there are different copyleft licences in use, copyleft effectively prevents remixing.
Similar concerns appear to have played a role in Google’s decision to choose the APL2.0 instead of GPL. On its license page, Google explains:
Android is about freedom and choice. The purpose of Android is to promote openness in the mobile world, but we don’t believe it’s possible to predict or dictate all the uses to which people will want to put our software. So, while we encourage everyone to make devices that are open and modifiable, we don’t believe it is our place to force them to do so. [16] Using GPL libraries would clearly “force them to do so”. The APL2.0, which does not include such requirements, obviously accommodates the interests of large corporate partners, OEMs and telecoms, better than the GPL. For some of these partners it is critical that they can charge license fees for their products, and keep their code secret to ensure a competitive edge. Moreover, it is critical that they can combine code with different licenses attached. The APL2.0 facilitates this.
By the same token, the APL2.0 allows Google to use and build on contributions by others. How this works, is clearly illustrated by the ‘Corporate Contributor License Agreement’ and the ‘Contributor License Agreement for Individuals’. These licenses organize the legal relationship between Google and software and hardware developers who actually contribute to the Android open source project. They give Google, in legal terms the ‘Project Leads’, the right to use contributions to Android in any way it wants.
You hereby grant the Project Leads and the recipients of software distributed by the Project Leads a perpetual, worldwide, non exclusive, no-charge, royalty–free, irrevocable copyright license to reproduce, prepare derivative works of, publicly display, publicly perform, sublicense, and distribute Your Contributions and such derivative works. [17] Thus, the APL2.0 gives Google a lot of flexibility in developing the Android ecosystem. It allows Android partners to develop modified versions of the OS in a proprietary way. Yet, at the same time, Google retains the right to use and build on the contributions by others.
Compatibility
It is especially through technological specifications that Google directs the Android ecology. To understand how this works, it is important to see that the corporation does not fully embrace the ‘bazaar’ model of software development, as described by Raymond, and which Netscape experimented with in the late 1990s. Instead, Google largely follows the ‘cathedral’ model of software production; only the source code of completed versions of the OS is released. Because it fully controls the development of the OS, Google can determine the technological specifications to which Android partners must abide.
How these specifications, subsequently, allow Google to force industry partners to follow its rules can clearly be observed through the issue of ‘compatibility’. Compatibility is essential, since only a platform with a large user base will attract a critical mass of third–party application developers. Specifically addressing OEMs, Google points out: “by building a fully compatible Android device, you benefit from the huge pool of apps written for Android, while increasing the incentive for developers to build more of those apps” [18].
The compatibility issue is directly related to the problem of ‘fragmentation’. Thanks to its open source license, hardware manufacturers can customise the Android source code and develop their own user interface and unique applications [19]. The customised OS is then tied to particular hardware. This practice can lead to fragmentation in case each Android phone would need its own specific third party applications and other services. Such fragmentation does not comply with the promoted vision of Android as a ‘platform’, where each device has its own specific functionality, but is still compatible with other Android devices to ensure a large platform for third party applications.
Compatibility problems arising from fragmentation and forking are well known in the open source community. They have often been presented as the main weakness of open source projects (Kelty, 2008; Lerner and Tirole, 2002; Weber, 2004). Incompatibility is especially problematic in the case of mobile devices because not every device has the same hardware features, such as a camera, or GPS. This limits the compatibility of applications, building on these features, even further.
To prevent fragmentation, Google has developed the Android Compatibility Program [20]. This program revolves around the Compatibility Definition Document (CDD) [21]. The CDD addresses the requirements of Android’s hardware and software, and how the OS should function. It defines, for example, to a far–reaching degree how Android’s application programming interfaces (APIs) are developed and managed. APIs are technological specifications that enable third–party application developers to have controlled access to a device’s hardware features, such as a camera or GPS, or to the back–end of particular services, such as Google Maps.
Generally, APIs are developed to let third parties access already existing hardware features so they can use them for their own applications. With Android this is reversed. Rather than allowing third–party application developers to access features provided by hardware manufacturers, Android’s APIs direct which hardware features can be present. They effectively steer hardware developers’ innovation process. To facilitate this, the CDD states: “Device implementations MUST NOT omit any managed APIs, alter API interfaces or signatures, deviate from the documented behaviour, or include no-ops, except where specifically allowed by this Compatibility Definition” [22]. Device manufacturers are thus absolutely prohibited from changing Android’s APIs unless specifically allowed by the CDD, and thus Google.
The Skyhook case
Beyond using compatibility to steer hardware development, Google can use it offensively as a strategic weapon. This is clearly illustrated by the ‘Skyhook case’. In September 2010, Skyhook Wireless, a service that pinpoints a mobile device’s location via Wi–Fi and cell tower signals, sued Google for business interference and patent infringement. The case has not been settled, but a large number of unsealed documents, made available online by the Web site Business Insider, provide insight into Android’s approval process.
The documents mostly concern e–mail conversations between Motorola and Google, or internally within Google. They reveal that Google did not approve of Motorola’s choice to use Skyhook as their Network Location Provider service, which determines the geographic location of a mobile device. It aimed to ‘persuade’ Motorola to choose Google’s services instead. Location provider services are important for Google:
because we need wifi data collection in order to maintain and improve our wifi location services (...). Our wifi location database is extremely valuable to Google because it is not a competitive market, even worse than the map data market. Skyhook is the only other viable alternative and there would be incredible risk to depend on them. [23] The location provider services allow Google to collect valuable end–user’s metadata.
Google’s e–mail messages unveil how they considered a number of tactics to convince Motorola to use its services. First, it contemplated highlighting upcoming innovations to its mobile services, and using access to these innovations as leverage. One e–mail message read: “I think our chances of success may further be improved if we provide more education to Motorola on some of the innovations we have planned in the pipeline for Places, Latitude and GMM.” Furthermore, “to the extent we decide there is value, we could potentially give Motorola an informal ‘pole position’ to market those features sooner to their end users” [24].
In addition, Google considered a more aggressive approach by pushing Motorola to use its services through the introduction of technical restrictions. It specifically considered tying its Network Location Provider (NLP) directly to the Google Maps service. In that case, an internal Google e–mail message explained, “it would be fine for Maps to just use the Google NLP or GPS, and ignore any other [3rd party Location Provider]. Under such a technical restriction the OEM is free to pre–load a 3rd party Location Provider — we just won’t use it in Maps ;-)” [25].
In the end, Google decided to invoke the ‘compatibility’ argument. In an e–mail message to Motorola, Google argued the compatibility program “sets an expectation for app developers that they will get a GPS location with accuracy ranging from <10m that they can use for cases where they need a very fine–grained location, such as turn–by–turn navigation” [26]. It argued that the Skyhook service did not provide such accuracy and was therefore incompatible.
When Google convinced Motorola to refrain from using Skyhook, compatibility was also the main argument through which Motorola was to communicate its decision to Skyhook. Google literally suggested to Motorola to use this argument: “We think it’s better to explain that this is an issue of Android Compatibility.” The eventual e–mail message from Motorola to Skyhook stated: “As you know, Motorola has certain contractual obligations that generally require Motorola’s Android device to be Android Compatible Devices as defined by Google” [27]. Taken together, Google’s strategic use of compatibility issues demonstrates the politics of technological specifications. It shows how “the conditions of code and software relate to power, capital, and control” (Langlois, et al., 2009). It illustrates that Android’s open source model, with its accompanying concerns of compatibility and fragmentation, became a weapon to push Google services and steer the hardware development of Android devices.
The politics of technological and legal specifications have allowed Google to manoeuvre into a central position in the political economy of the mobile Internet. Reflecting on Android’s role in this economy, it is important to see that Google is, first and foremost, an advertising network. It generates the largest share of its revenue through the AdSense and AdWords services (Anderson, 2009; Levy, 2011; Vaidhyanathan, 2011).
AdWords is a pay–per–click keyword advertisement program through which advertisers place ads based on keywords related to their content. In turn, AdSense is an ad–placement program that utilizes Google’s search capabilities to determine the best positioning for ads purchased through the Google AdWords program. Adsense allows anyone publishing a Web site to generate income. It uses information about users who visit a site to better target its advertisements. Ultimately this information is used to “show ads that might appeal to (...) users based on their inferred interest and demographic categories” [28]. Thus, the more precisely Google can infer user interests and determine the demographic categories to which they belong, the better AdSense works. The necessary metadata for this type of profiling is harvested through the various Google services: Google Search, YouTube, Google+, Google Reader, Google Checkout, Google Books etc. Hence, for Google it is essential that as many people as possible use these services.
From this perspective, the rapid growth of the mobile Internet constitutes a challenge for the company. To sustain its leading position in online advertising and secure future growth, it is essential that its services are, and remain, easily accessible on mobile Internet devices. Moreover, it is vital that it can build a large user base for new mobile services, such as Google Offers, Google Wallet, and Google Shopper. And finally, it is crucial that it can gather key mobile user data from new and existing services. No longer limited to the online activities of clicking and searching, mobile Internet devices potentially allow Google to monitor the activities of its users even more closely, following their every step. Clearly, such location specific data allow Google to even more precisely profile users, predicting their specific interests at a particular place and time.
Android’s development should be understood in close connection to this business model. Android is the means through which Google tries to remain in control of how people access their services, and to make sure that it can harvest essential mobile user metadata. While the company also offers its services for iOS and Windows 7 users, it is uncertain whether these services will remain easily accessible through these mobile operating systems. Furthermore, it is also because of Google’s particular business model that Android’s open source distribution strategy works so well. For Google, it is most important that Android runs on as many mobile devices as possible, maximizing the user base of its services. Android’s open source strategy is particularly instrumental in achieving this objective. For industrial partners, the OS is not only attractive because it is free, but also because it can be easily customized, allowing manufacturers and telcos to put their own stamp on Android devices, while at the same time having access to a wealth of third party applications. In turn, for Google, as discussed in the previous part, the open source strategy and the APL2.0 allow it to push its services.
To be able to use Android as a platform for its different services, and as an instrument for harvesting user data, it is critical for Google to remain fully in control of the development of the OS itself. In a truly open source software development model, the ‘bazaar’ model in Raymond’s words, Google would have to involve independent developers directly in the coding of the OS. This would make it difficult, if not impossible, for Google to control how users obtain access to its services, and it would complicate the collection of user data. Moreover, it would no longer allow Google to give ‘early access’ to the source code of new versions of the OS to particular OEMs, one of its key governing strategies in relation with manufactures. Motorola, for example, was given early access to the source code of Android 3.0, the Honeycomb version of the OS. Hence, it is also in the light of these political economy considerations that Google’s decision to develop the OS following the so-called ‘cathedral’ model should be understood.
Thus, while Google strategically adopts particular open source practices, it also strategically refrains from fully embracing the open source model. Reflecting on the core principles and practices of open source, Steven Weber (2004) has pointed out that “open source is a way of organizing production, of making things jointly” [29]. It “is configured fundamentally around the right to distribute, not the right to exclude” [30]. On both accounts, Google has taken a different approach.
The Future of the mobile Internet
Google’s particular business model and open source practices directly conflict with the business models of other corporations producing mobile operating systems. Apple develops iOS primarily to sell Apple hardware, to which it is exclusively tied. The same pertains to RIM, which develops the Blackberry OS, and purchased the operating system QNX, in the Spring of 2010, to be able to create and sell its own tablet hardware, the Blackberry Playbook [31]. Unlike Apple and RIM, Microsoft does not sell its own hardware, but the company ties its Windows Phone 7 OS to hardware created by developers like Nokia and HTC. Microsoft first and foremost generates revenue through license fees. By contrast, Google has not developed Android to generate revenue through licensing or hardware sales. As argued above, its prime objective is to constitute a platform for the collection of user metadata, and for displaying targeted advertising.
By offering the Android OS for ‘free’ to OEMs, Google is effectively undermining the business models of other corporations that produce mobile operating systems. Not surprisingly, these corporations have responded aggressively, resulting in a series of lawsuits mostly concerning patent and copyright infringement of ‘open source’ code implemented in Android. In total there have been 37 Android related lawsuits [32]. Many of these lawsuits have been settled. A clear example of this is Microsoft’s agreement with HTC for a license fee per HTC phone sold, following a lawsuit over intellectual property infringement [33]. In a similar fashion, Microsoft has made patent licensing deals with 11 different Android manufacturers, including HTC, Samsung, Acer and LG [34]. Taken together, these lawsuits especially increase the ‘price’ of the Android OS, which can no longer be considered ‘free’.
While most of the lawsuits do not address Google directly, they do potentially undermine the development of Android. Apple, for instance, has accused HTC of infringing 10 Apple patents. Concerning this case, the U.S. International Trade Commission has ruled that Android–run HTC devices indeed violate one, easy to work around, Apple patent [35]. The lawsuit could, however, have had far–reaching consequences, as it also addressed a patent that concerned core features of the Android’s OS. If the court had ruled in favour of Apple, Google would have been forced to restructure the architecture of the OS, and possibly even the architecture of the underlying Linux kernel [36]. Another potentially damaging lawsuit has been launched by Oracle, the first one to directly accuse Google that Android is infringing its Java patents. To settle, Google has offered Oracle a percentage of Android’s revenue — if they could indeed prove patent infringement. Oracle declined this offer. In May 2012, a federal jury ruled largely in favour of Google. It ruled that Google did not infringe on Oracle’s patents when it used Java software in the Android operating system, leaving Oracle with a relatively small claim of copyright infringement [37].
To protect itself against these lawsuits, Google, in the Summer of 2011, acquired Motorola Mobility for 12.5 billion dollars. The objective of this acquisition is mostly to enlarge Google’s relatively small mobile patent portfolio and to “help protect the Android ecosystem” [38]. Given how much Google is willing to spend defending Android, it is clear that the stakes have become very high in these struggles over the political economy of the mobile Internet. This conclusion is further underscored by the fact that most of the lawsuits from either side concerning patent or copyright infringement have triggered ‘counter’ lawsuits.
Of course, the outcome of these struggles not just affects the corporations involved, but also has large consequences for hundreds of millions of mobile Internet users. So far, these consequences have mainly been considered in terms of privacy concerns. Especially in the popular press, much has been made of how Android Smartphones trace user activity. These phones are criticized for sending location data to Google. In addition, certain Android applications from third party developers are criticized for including “embedded information–harvesting code — APIs”, which are used “for gathering advertising and app usage metrics.” It is said that “detailed information about users, including their locations and unique mobile identifiers” is sent “to mobile advertising companies to track, profile, and personally identify users” [39]. This type of data collection has been called out for violating the privacy of users.
From a broader political economic perspective, the ascent of Android should be understood as a crucial step in the integration of the mobile Internet in what has been labelled, from different theoretical perspectives, ‘information capitalism’, ‘communicative capitalism’, and ‘cognitive capitalism’ (Dean, 2009; Fuchs, 2011; Pasquinelli, 2010b; Terranova, 2004). In this mode of capitalism, knowledge, which is here conceptualized broadly, has become a productive force. Knowledge production includes, as Fuchs (2011) has made clear, not only the production of intellectual property by corporations, but also “consumers of media who produce social meaning”, as well as “users of MySpace, YouTube, Facebook and so on who produce informational content” [40]. In informational capitalism, this knowledge is appropriated by capital. Pasquinelli (2010b) emphasizes that the “diagram of cognitive capitalism [...] is not simply an apparatus of surveillance or control, but a machine to capture living time and living labour and to transform the common intellect into network value”. This is precisely what Android introduces in the political economy of the mobile Internet, as it enables Google to capture mobile Internet user activity, profile these users, and sell them to advertisers.
Of course, it would be a mistake to exclusively link the development of targeted advertising and services to Android/Google. Android should rather be seen as a particular political economic configuration, which is specifically build on these practices and business model. In this configuration, revenue is not primarily generated through the commodification of intellectual property, which is precisely why the open source model works so well for Android, but through the transformation of “common intellect into network value”.
There are strong indications that this political economic configuration is rapidly gaining influence. In response to the ascendance of Android, other mobile Internet corporations have, beyond litigating, started to become active in the personalized advertising space as well. Apple tried to buy AdMob, the biggest mobile advertising company, but was outbid by Google. Eventually a smaller company, Quattro Wireless, allowed Apple to develop its own mobile advertising platform, iAd. Microsoft, in turn, is pushing its own advertisement platform by integrating an application store in Windows 8, and by developing a new ‘social advertisement solution’ called ‘People Powered Stories.’ Microsoft claims that “with it, advertisers can incorporate real peoples’ ratings and reviews about their products within a rich brand ad” [41]. Furthermore, besides the producers of mobile operating systems, there are a wide range of companies, most prominently Facebook, which are also in the business of profiling mobile Internet users to develop targeted advertising and services. From this perspective, it appears that Android is paradigmatic for the overall direction in which the political economy of the mobile Internet is moving.
Reflecting on Android, we have argued that open source critique needs to be renewed. Over the past decade, open source has been frequently criticized for facilitating the exploitation of the free labor of open source programmers by aligning free software with corporate interests. In the political–economic configuration constituted by Android and directed by Google, exploitation has not only become more pervasive, but also more encompassing and multifaceted. In this configuration, open source is effectively the techno–legal vehicle of cognitive capitalism. Android makes it possible for Google to systematically collect, analyze, and commodify mobile user data. The OS not only enables the profiling of user interests, but also the close tracking of user movement, which is crucial for the development of new location specific services and advertising. To facilitate this, particular open source practices have been strategically adopted. Through these practices, Google is able to cultivate and control a vast mobile Internet ecology, involving: large numbers of telcos around the world, various major equipment manufacturers, a mass of third–party application developers, a broad range of social media corporations, and hundreds of millions of mobile Internet users. In this ecology, Google can use the contributions of independent programmers, but also, and more importantly, the metadata of Android users.
In light of this massive concentration of corporate control over the political economy of the mobile Internet, it is crucial that governments start putting limits to this control. Over the past years, government regulators have especially shown an increasing concern over the corporate collection of user data. Various initiatives have been taken to act upon these concerns. On 25 January 2012, the EU released a proposal for a ‘General Data Protection Regulation’, which includes the individual’s right not to be subject to profiling, that is “automated processing intended to evaluate certain personal aspects relating to this natural person or to analyse or predict in particular the natural person’s performance at work, economic situation, location, health, personal preferences, reliability or behaviour” [42]. In February 2012, the Obama administration has presented the “Consumer Privacy Bill of Rights”. The first principle advanced by this bill is that “Consumers have a right to exercise control over what personal data companies collect from them and how they use it” [43]. In addition, as part of this new privacy protection initiative, a ‘Do Not Track’ agreement has been developed and signed by a number of major Internet companies, including Google and Microsoft, to allow users to opt out of targeted advertising, and block tracking cookies.
While the proposed government initiatives to some extent limit targeted advertising and user profiling, they do not make these practices purely opt–in options, which would entail a major restriction on how corporations such as Google can collect and commodify user data. Furthermore, these initiatives are primarily focussed on the protection of user privacy, and only partly address the enormous concentration of corporate control. The political economy of the mobile Internet is increasingly divided in a few closely–knit ecosystems, of which Android is rapidly becoming the largest one. If governments will not step in to regulate how these ecosystems are organized, we might well look at a future in which only a few corporations control the data that drive the mobile Internet economy, as well as the technological architectures through which we shape our relations with the world.
New media theorists can make an important contribution to the discussion on this issue. As this paper has tried to show, to understand how the political economy of the mobile Internet works and evolves, it is vital to combine a techno-legal and a political economic approach. The corporate power strategies of Google and its competitors, as discussed, combine intricate technical, legal, and political–economic maneuvering. By focusing exclusively on one of these aspects, whether it is legal frameworks, exploitative practices, or technological specifications, one is bound to misunderstand how these corporate strategies shape the development of the mobile Internet. And, how they, consequently, have far–reaching economic and cultural consequences. The challenge is to trace how techno–legal specifications and political–economic strategies mutually reinforce each other, and, subsequently, to think through how such complex configurations can be regulated.
About the authors
Kimberley Spreeuwenberg (M.A., University of Amsterdam) is a new media researcher and graphic designer. She lectures in the Department of Media Studies at the University of Amsterdam.
Web: www.kimmyspreeuw.nl
E–mail: kimely [dot] s [at] gmail [dot] comThomas Poell (Ph.D., Utrecht University) is assistant professor of New Media and Digital Culture in the Department of Media Studies at the University of Amsterdam.
Web: http://home.medewerker.uva.nl/t.poell/
E–mail: poell [at] uva [dot] nl
Acknowledgments
We want to thank José van Dijck, Florian Cramer, and Morgan Currie, as well as First Monday’s anonymous reviewers, for their useful comments, suggestions, and help.
Notes
1. http://source.android.com, accessed 3 May 2011.
2. http://techcrunch.com/2011/12/22/android-700000/, accessed 22 December 2011.
3. E. Raymond, 1998, p. 8.
4. L. Lessig, 2005, p. 350.
5. Y. Benkler, 2006, p. 60.
6. Ibid.
7. Y. Benkler and H. Nissenbaum, 2006, p. 411.
8. R. Stallman, 1999, p. 17.
9. E. Raymond, 1998, p. 20.
10. C. Kelty, 2008, p. 99.
11. T. Terranova, 2004, pp. 93–94.
12. T. Terranova, op.cit., p. 94.
13. M. Pasquinelli, 2010a, p. 289.
14. http://source.android.com/source/licenses.html, accessed 3 May 2011.
15. For a discussion on the relationship between free software and the Creative Commons see C. Kelty, 2008, pp. 269–300.
16. http://source.android.com/source/licenses.html, accessed 3 May 2011.
17. http://source.android.com/source/cla-corporate.html, accessed 12 July 2011.
18. http://source.android.com/compatibility/index.html, accessed 12 July 2011.
19. http://source.android.com/compatibility/overview.html, accessed 12 July 2011.
20. Ibid.
21. http://source.android.com/compatibility/2.3/android-2.3.3-cdd.pdf, accessed 23 March 2011.
22. Op.cit., p. 4.
23. http://www.businessinsider.com/google-skyhook-documents-2011-5, accessed 4 July 2011; http://www.businessinsider.com/google-skyhook-documents-2011-5 - a-bunch-of-the-good-stuff-lots-of-emails-from-inside-google-and-tofrom-skyhook-and-motorola-employees-1?, accessed 4 July 2011; http://www.businessinsider.com/google-skyhook-documents-2011-5#more-emails-and-the-documents-where-skyhook-is-requesting-documents-from-google-for-the-case-2, accessed 4 July 2011, p. GOOG_SKY_ST 00046598.
24. http://www.businessinsider.com/google-skyhook-documents-2011-5#more-emails-and-the-documents-where-skyhook-is-requesting-documents-from-google-for-the-case-2, accessed 4 July 2011, p. GOOG_SKY_ST 00017371.
25. Op.cit., p. GOOG_SKY_ST 00040970.
26. Op.cit., p. GOOG_SKY_ST 00027006.
27. Op.cit., p. SKY00000159.
28. http://www.google.com/adsense/support/bin/answer.py?hl=en&answer= 140378, accessed 30 May 2011.
29. S. Weber, 2004, p. 224.
30. Op.cit., p. 228.
31. http://www.eweek.com/c/a/Mobile-and-Wireless/BlackBerry-PlayBook-Tablet-a-Winner-Despite-Heavy-Costs-RIM-283057/, accessed 14 April 2012.
32. http://www.fosspatents.com/2011/03/infographic-37-android-related-patent.html, accessed 12 April 2012.
33. http://moconews.net/article/41 9-htc-will-pay-royalties-to-microsoft-for-android-phones/, accessed 14 April 2012.
34. http://www.techspot.com/news/47030-microsoft-strikes-another-android-patent-licensing-deal.html, accessed 12 January 2012.
35. http://www.businessweek.com/articles/2012-03-29/apple-s-war-on-android, accessed 14 April 2012.
36. http://www.ibtimes.com/articles/182074/20110718/android-apple-htc-itc-mobile-os-patent-infringement-oracle-motorola-oem-samsung-iphone-steve-jobs-li.htm, accessed 18 July 2011.
37. http://www.reuters.com/article/2012/03/28/oracle-google-lawsuit-idUSL2E8ESRKB20120328, accessed 14 April 2012; http://bits.blogs.nytimes.com/2012/05/23/google-cleared-of-java-patent-violation/, accessed 7 June 2012; http://www.businessinsider.com/oracle-google-android-lawsuit-legal-fees-2012-6, accessed 7 June 2012.
38. http://www.google.com/press/motorola/, accessed 14 April 2012.
39. http://www.huffingtonpost.com/2011/04/22/google-android-and-apple-track-your-location_n_852529.html, accessed 12 July 2011.
40. C. Fuchs, 2011, p. 280.
41. http://community.microsoftadvertising.com/blogs/advertising/archive/2012/02/14/people-powered-stories-social-advertising.aspx, accessed 14 April 2012.
42. http://ec.europa.eu/justice/data-protection/document/review2012/com_2012_11_en.pdf, accessed 25 January 2012; for a discussion of these regulations see http://fuchs.uti.at/789/.
43. http://www.whitehouse.gov/sites/default /files/privacy-final.pdf; http://arstechnica.com/tech-policy/news/2012/02/white-house-announces-new-privacy-bill-of-rights-do-not-track-agreement.ars, accessed 14 April 2012.
References
M. Allen, 2008. “Web 2.0: An argument against convergence,” First Monday, volume 13, number 3, at http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2139/1946, accessed 15 April 2012.
C. Anderson, 2009. Free: The future of radical price. London: RH Business Books.
R. Barbrook, 1998. “The hi–tech gift economy,” First Monday, volume 3, number 12, at http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/631/552, accessed on 24 April 2012.
Y. Benkler, 2006. The wealth of networks: How social production transforms markets and freedom. New Haven, Conn.: Yale University Press.
Y. Benkler and H. Nissenbaum, 2006. “Commons–based peer production and virtue,” Journal of Political Philosophy, volume 14, number 4, pp. 394–419.http://dx.doi.org/10.1111/j.1467-9760.2006.00235.x
D. Berry, 2008. Copy, rip, burn: The politics of copyleft and open source. London: Pluto Press.
J. Cheney–Lippold, 2011. “A new algorithmic identity: Soft biopolitics and the modulation of control,” Theory Culture & Society, volume 28, number 6, pp. 164–181.http://dx.doi.org/10.1177/0263276411424420
W. Chun, 2006. Control and freedom: Power and paranoia in the age of fiber optics. Cambridge, Mass.: MIT Press.
J. Dean, 2009. Democracy and other neoliberal fantasies: Communicative capitalism and left politics. Durham, N.C.: Duke University Press.
N. Dyer–Witheford, 1999. Cyber–Marx: Cycles and circuits of struggle in high–technology capitalism. Urbana: University of Illinois Press.
C. Fuchs, 2011. Foundations of critical media and information studies. New York: Routledge.
C. Fuchs, K. Boersma, A. Albrechtslund and M. Sandoval (editors), 2011. Internet and surveillance: The challenges of Web 2.0 and social media. New York: Routledge.
M. Fuller (editor), 2008. Software studies: A lexicon. Cambridge, Mass.: MIT Press.
A. Galloway, 2004. Protocol: How control exists after decentralization. Cambridge, Mass.: MIT Press.
C. Kelty, 2008. Two bits: The cultural significance of free doftware. Durham, N.C.: Duke University Press.
G. Langlois, F. McKelvey, G. Elmer and K. Werbin, 2009. “Mapping commercial Web 2.0 worlds: Towards a new critical ontogenesis,” Fibreculture, issue 14, at http://journal.fibreculture.org/issue14/issue14_langlois_et_al_print.html, accessed 21 May 2011.
J. Lerner and J. Tirole, 2002. “Some simple economics of open source,” Journal of Industrial Economics, volume 50, number 2, pp. 197–234, and at http://www.jstor.org/stable/3569837, accessed 7 November 2010.
L. Lessig, 2005. “Open code and open societies,” In: J. Feller, B. Fitzgerald, S. Hissam and K. Lakhani (editors). Perspectives on free and open source software. Cambridge, Mass.: MIT Press, pp. 349–360.
S. Levy, 2011. In the plex: How Google thinks, works, and shapes our lives. New York: Simon & Schuster.
E. Moglen, 1999. “Anarchism triumphant: Free software and the death of copyright,” First Monday, volume 4, number 8, at http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/684/594, accessed 15 April 2012.
G. Moody, 2002. The rebel code: The inside story of Linux and the open source revolution. Cambridge, Mass.: Perseus.
M. Pasquinelli, 2010a. “The ideology of free culture and the grammar of sabotage,” In: D. Araya and M. Peters (editors), Education in the creative economy: Knowledge and learning in the age of innovation. New York: Peter Lang, pp. 285–304.
M. Pasquinelli, 2010b. “Google’s PageRank algorithm: A diagram of cognitive capitalism and the rentier of the common intellect,” Pankov The Underground Research (16 March), at http://pankov.wordpress.com/2010/03/16/google%E2%80%99s-pagerank-algorithm-a-diagram-of-the-cognitive-capitalism-and-the-rentier-of-the-common-intellect/, accessed 22 June 2012.
S. Petersen, 2008. “Loser generated content: From participation to exploitation,” First Monday, volume 13 number 3, at http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2141/1948, accessed 24 April 2012.
E. Raymond, 1998. “The cathedral and the bazaar,” First Monday, volume 3, number 3, at http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/578/499, accessed 22 June 2012.
J. Söderberg, 2008. Hacking capitalism: The free and open source software movement. New York: Routledge.
R. Stallman, 2002. “What is copyleft?” In: J. Gay (editor). Free software: Selected essays of Richard M. Stallman. Boston: GNU Press, pp. 91–92.
R. Stallman, 1999. “The GNU project,” In: J. Gay (editor). Free software: Selected essays of Richard M. Stallman. Boston: GNU Press, p. 17.
R. Stallman, 1998. “Why ‘free software’ is better than ‘open source’,” In: J. Gay (editor). Free software: Selected essays of Richard M. Stallman. Boston: GNU Press, pp. 57–62.
T. Terranova, 2004. Network culture: Politics for the information age. London: Pluto Press.
S. Vaidhyanathan, 2011. The Googlization of everything: (and why we should worry). Berkeley: University of California Press.
J. van Dijck, 2009. “Users like you? Theorizing agency in user–generated content,” Media, Culture & Society, volume 31, number 1, pp. 41–58.http://dx.doi.org/10.1177/0163443708098245
S. Weber, 2004. The success of open source. Cambridge, Mass.: Harvard University Press.
D. Wiley, 2007. “Open education license draft,” iterating toward openness. pragmatism over zeal (8 August), at http://opencontent.org/blog/archives/355, accessed 8 June 2012.
Editorial history
Received 26 April 2012; revised 8 June 2012; accepted 14 June 2012.
This work is licensed under a Creative Commons Attribution–NonCommercial–NoDerivs 3.0 Unported License.Android and the political economy of the mobile Internet: A renewal of open source critique
by Kimberley Spreeuwenberg and Thomas Poell
First Monday, Volume 17, Number 7 - 2 July 2012
https://firstmonday.org/ojs/index.php/fm/article/download/4050/3271
doi:10.5210/fm.v17i7.4050
