Politics of Code in Web 2.0
Essay: Mapping Commercial Web 2.0 Worlds: Towards a New Critical Ontogenesis. By Ganaele Langlois, Fenwick McKelvey, Greg Elmer, and Kenneth Werbin. Fibreculture Journal, Issue 14.
There must be a middle way between those who simply assert that Web 2.0 is a plot to exploit free labor, leaving its emancipatory potential on the wayside, and the cyber-utopians who believe Web 2.0 participation is automatic. Such an approach, which I believe is the one we hold at the P2P Foundation, uses participatory technologies to strenghten sharing communities active on Web 2.0 platforms, while also building our own autonomous infrastructures, so that we can engage in more peer to peer relationships and production and prepare a post-capitalist world. Such an approach hinges on the agency of peer production and a critical communion with other social forces that build such infrastructures. It’s success also depends on a solid understanding of the invisible architectures that influence social behaviour on networks, i.e. protocollorary power.
The above essay gives extra insight to such necessary code politics.
Here is the aim of the essay:
“Our approach focuses on understanding the code politics of user-generated content within commercial Web 2.0 spaces. We rely on software studies (Fuller, 2003, 2008) in order to understand the mutual shaping of software, hardware and cultural practices. Software studies invites us to focus on the cultural and communicational changes brought by software, and to pay attention to the ways in which online communication is not simply a human activity, but a set of practices negotiated through complex dynamics between software architectures and different categories of users (i.e. software engineers, citizens, activists, etc.). As a topic stemming from the broader field of software studies, code politics seeks to understand the conditions of code and software in relation to power, capital, and control. Studying code politics means studying how actors have ‘literally encoded the Web for their political purposes’.”
1. Web 2.0 as universal platform
The authors stress we should see Web 2.0 as a universal platform:
“For example, Grusin (2000) explores the code politics of the computer when he relates an operating system’s desktop with physical real estate that corporations compete to control. He argues that,’whenever you boot up your computer, you are engaging in a commercial transaction in a mediated public space which is being increasingly contested by Microsoft, the USA Government, and inevitably other governments and corporations as well’ (Grusin, 2000: 59). Even the matter of your default web browser has tremendous value for corporations. With user-generated content and commercial Web 2.0, a code politics approach requires moving beyond analyzing the content of the user interface to locating how that content articulates visible and hidden processes. A discussion group on Facebook, for instance, raises questions not only about the content of a discussion, but also about how information about content and users is captured and recirculated by software through data-mining and marketing, about how the presence of commercial forces is facilitated both at the software level and at the economic level of commercial partnerships, and about how users’ participation is rechanneled as marketable data. In short, a code politics approach to user-generated content requires paying attention to the articulations between the user, the software and the interface: to explore how the cultural experiences and practices available at the user-interface level (Cramer & Fuller, 2008; Fuller, 2003; Grusin, 2000; S. Johnson, 1997; Jørgensen & Udsen, 2005; Turkle, 1997) are articulated and coexist with the processes through which software encodes user input according to material (Hayles, 2004; Manovich, 2002), ideological (Chun, 2005), and legal (Grimelmann, 2005; Lessig, 2006) logics. In so doing, a code politics approach seeks to understand the connections that enable and shape the traffic and trafficking of information, data, immaterial labour and subjectivities online.
A code politics approach is hardly new, and actually has been a concern not only in the field of software studies, but also in the development of Web methodologies writ large. Understanding the relationship between the circulation and production of content and the shaping of the communicative possibilities of the Web has been a constant in developing methodologies that take into account the specificities of the Web environment. The problem does not lie in justifying the need for such an approach, but in adapting it to the specific dynamics of commercial Web 2.0. This requires a shift away from the protocological approach that has been dominating Web studies. Protocol refers to the ‘language that regulates flow, directs netspace, codes relationships, and connects life-forms’ on the Internet (Galloway, 2004: 74). Protocol points to the technical conventions that enable computers to communicate in a decentralized network such as the Internet. By extension, protocols are the sites through which possibilities for control and resistance on the Web and the Internet are defined. Studying the protocols that regulate the formation of computer networks thus offers a way to examine the power relations at stake on the Internet in general, and on the Web in particular. While Galloway (2004) focuses on TCP/IP as the protocols that enable the Internet, other Web studies approaches have focused on HTML/HTTP as the protocols that enable communication on the World Wide Web. While HTTP is about the transfer of information, the HTML language is a protocol that encodes content within a specific hyperlinked context. Web studies first started focusing on the hyperlink as a unit of analysis to understand, for instance, how the flow of information through hyperlinks can yield clues as to the communicational and social dynamics of the actors involved in a hyperlinked network (Garrido and Halavais, 2003). Other Web methodologies seek to examine how discussions on issues of common interest can be studied through looking at the evolution of hyperlink networks (Rogers and Marres, 2005). Still other approaches seek to understand the relationship between the protocological aspects of the Web and the circulation of content. Web sphere analysis (Schneider and Foot, 2005), for instance, does not only examine hyperlink networks, but also textual content and website features (e.g. email, message posting) in the case of election campaigns. More recently, efforts have been made to include not only hyperlinks to analyze the shaping of informational dynamics, but also other Web protocols such as metatags, cookies and robot.txt (Elmer, 2006). Common to all these approaches is a focus on the informational dynamics of single protocols.
The transition from Web 1.0 to Web 2.0, however, has changed the language and protocols of the Web. HTML is no longer the dominant language on the web, but rather one among a multiplicity of co-existing and competing languages and protocols. The situation has changed from one dominant markup language on the Web to multiple code languages. As a consequence, hyperlinking is now but one feature of Web 2.0 spaces. Increasingly buttons replace linking and the logic of embedding Web objects on different Web 2.0 spaces is becoming more and more popular: if I like a video on YouTube and want to show it to my friends, I do not send a link via e-mail anymore, I push the Facebook share button on YouTube and the video is immediately embedded in my Facebook profile and visible to all my friends. Furthermore, whereas the production of hyperlinks in the Web 1.0, HTML-dominated environment was created by human users, hyperlinks in Web 2.0 are increasingly produced by software as tailored recommendations for videos or items of interest, suggested friends, etc. The technocultural articulations that regulate the production and circulation of hyperlinks are thus different in the Web 2.0 environment from the Web 1.0 environment, particularly with regards to the re-articulation of hyperlink protocols within other software and protocological processes. The production of an HTML tag linking to a personalized recommendation is the result of the algorithmic processing of a user’s profile correlated with other profiles and potentially commercial interests. While these processes are invisible from a user perspective of instantaneous communication, they nevertheless represent a central site of analysis in order to understand the new communicational and cultural regimes of user-generated environments.
The folding of the once dominant HTML language into a range of other protocols requires changing our assumptions about the Web as an object of study. The dramatic increase of code, operating systems, programs, languages and browsers in the Web 2.0 environment reduces the applicability of protocol to fully encapsulate the conditions and possibilities of Web 2.0. How then might one conceptualize the relationships between different protocols? The model put forward by Galloway starts with the assertion that ‘the content of every protocol is always another protocol’: HTML is encapsulated in HTTP, HTTP is dependent on TCP, TCP on IP, and all these protocols are encapsulated within ‘physical media’ (Galloway, 2004: 10-11). Galloway’s model is ambivalent in that the ‘encapsulating’ of protocols could be understood either as a reduction of all online activity to one protocol or as pointing out that different protocols co-exist, and that they have a hierarchical relationship based on the material requirements needed for them to exist. When the list of protocols was relatively simple - HTML, then HTTP, then TCP/IP - the articulation of protocols was quite straightforward and unproblematic. However, if, as Galloway suggests, protocol determines the limits of possibility, and the number of protocols increases dramatically, we need an answer as to how different protocols interact. One of the central technical characteristics of Web 2.0 is the reliance on APIs, on customized software programs that rearticulate protocols in different ways. Common examples involve Web services, such as Amazon web services, which allow third parties to access the Amazon catalogue, mashups of different Web 2.0 spaces (i.e. Google maps and Flickr images) and internal applications, for instance Facebook applications. APIs process and represent data in different ways, yet they are based on a common set of protocols. Therefore, we need an approach that interrogates the Web as an assemblage of protocols rather than the nesting of one protocol into another. In response, we suggest that protocols act as modular elements, assembled as part of different Web 2.0 platforms.
The modularity of protocols, as elements that serve to create different assemblages, necessitates a platform-based perspective. While the previous protocological approach conceptualized the Web as a carrier of information, the concept of the platform as ‘hardware and/or software that serves as a foundation or base’ points out that the articulation of modular protocols enable the possibility for Web 2.0 to run complex software similar to a desktop operating system. Web 2.0 actualizes the universal platform, a constructive space independent of hardware, imagined by the Java project (Mackenzie, 2006). Where Java failed to network enough of its actors to stabilize the platform, Web 2.0 has created a platform by drawing in a variety of standards and actors into its network.”
2. Web 2.0 as Constrained Worlds
Web 2.0 is not just a universal platform, but a ‘world-space’ in which users can exist:
” Web 2.0 spaces serve to establish the conditions within which content can be produced and shared and where the sphere of agency of users can be defined. With regards to understanding commercial Web 2.0 platform, this distinction draws a parallel with Maurizio Lazzarato’s argument that ‘in reversal of the Marxist definition, we could say that capitalism is not a world of production, but the production of worlds’ (2004: 96). Lazzarato’s exploration of contemporary forms of capitalism draws a distinction between the factory and the corporation. While the factory is about fabricating and producing objects, the corporation is about the creation of the world within which the process of manufacturing, distributing and consuming can take place (2004: 95). As Lazzarato further argues: ‘the corporation does not create objects (merchandises), but the world within which such objects can exist’. Furthermore, the corporation ‘does not create subjects (workers or consumers), but the world within which such subjects can exist’ (2004: 94). That is, the corporation creates flows of discourses, practices and materials that delineate a horizon of possibility and therefore work to create subjective norms that can be interiorized by individuals. The transition from Web 1.0 to Web 2.0 becomes more understandable if we follow the model provided by Lazzarato, especially when it comes to the commercialization of and capitalization on users and their content. While the type of commercializing processes at stake with Web 1.0 were primarily about transforming users and their content into commodities, Web 2.0 dynamics establish the conditions within which such processes of commercialization can occur through the promotion and harnessing of user-generated content.”
These worlds, argue the authors, citing Tiziana Terranova’s work, are alienating:
1. “Unpaid content production, then, is one aspect of a capitalist process focused on the ‘creation of monetary value out of knowledge/culture/affect’ (Terranova, 2000: 38). This new form of capitalism as expressed on the Internet and the Web is about nurturing, exploiting and exhausting the ‘cultural and affective production’ of the labor force (Terranova, 2000: 51). Terranova’s analysis is easily applicable to commercial Web 2.0 models, where processes of commercialization can take place at the level of users and their content. Establishing a new - and free - Facebook account lays the ground work for the growth of one’s social network. In exchange for this service, Facebook reserves the right to use information provided about users - who they are friends with, what their preferences are, what they read or consult - in order to share such information with third parties. YouTube likewise relies on freely and voluntarily produced videos to attract people and sell audiences to the advertising industry, among other marketing techniques, such as promoting partners’ videos. Commercial Web 2.0 spaces are, however, not simply technologies of content commercialization. There is clearly a qualitative difference between the America Online (AOL) chat hosts working for free that Terranova describes (Terranova, 2000: 33) and Facebook or YouTube users. The main difference lies in how the process of alienation inserts itself within a variety of online spaces. Terranova’s investigation into what can now be considered Web 1.0 describes a process where a capitalist machine ‘nurtures, exploits and exhausts’ user-generated cultural production. In that sense, there is a process of alienation of the user from their cultural production as the content produced serves to maintain and further an economic system. Such processes are far from being absent from Web 2.0 platforms, but the ways in which the platforms re-articulate processes of alienation to users is more complex. The alienation of users from their information first takes place through a technico-legal system whereby the implementation of surveillance software is accompanied by terms of service that authorize the platform to re-use user information.”
2. The “second characteristic of commercial Web 2.0 platforms is that processes of alienation are kept invisible to users. A case in point is Facebook’s Beacon. The Beacon was designed to enable more targeted advertising on the Facebook site through the sharing of users’ information with commercial third parties. The Facebook Beacon was quickly met with resistance because of privacy concerns and was changed to an opt-in system rather than a set feature of the website. This software change could be interpreted as a victory against the alienation of users from the information they provide, yet this is far from being the case. The Beacon, after all, was a visual representation of processes of commercialization that are still taking place on the Facebook platform. These processes, however, increasingly take place at the back-end level and because they are invisible to users, they meet with less resistance.
3. A third characteristic of commercial Web 2.0 concerns the re-articulation of the dynamics of alienation to make alienation disappear altogether. What Zimmer (2008) describes as a Faustian trade-off between augmented personalized exploration and surveillance and commercialization gives way to a dynamic whereby the process of commercialization is part of providing to users augmented cultural knowledge, affect and desire, to borrow from Terranova (2000). YouTube’s recommendations are designed to assist users in their quest for knowledge that corresponds to their interests. Facebook is about enhancing the personal socialized world of users and multiplying social exchanges through joining groups and sharing stories. Commercial Web 2.0 is about us - it is about re-presenting ourselves through the mediation of the platform. This where Web 2.0 platforms echo Lazzarato’s point that contemporary forms of capitalism is about the creation of worlds, which means the setting up of a horizon of possibilities. This also means that specific processes of subjectivation can be formulated as the crystallization of psychological, social, economic dynamics and factors that favour the formation of specific subject positions. These processes are present on Web 2.0 platforms and present us with the paradox of narrowing down the field of possibilities while creating, producing and enriching our experience of being on the Web. Commercial Web 2.0 platforms are attractive because they allow us, as users, to explore and build knowledge and social relations in an intimate, personalized way. In this dynamic, the commercialization of users and information is one of the central factors through which this enrichment takes place. As a consequence, alienation disappears, as in the Web 2.0 worlds there is no contradiction anymore between the marketing of user information and the subjective enrichment of users: what used to be two separate processes are now one in the augmentation of social and cultural factors. Third-party advertising is reinscribed as cultural capital produced by the platform for the user through personalized recommendations. The role of the platform, in that sense, is to set up the context, or world, within which such re-articulations can take place.”
The authors further explicitate this third condition and conclude: Commercial Web 2.0 platforms help construct worlds and set up the subjectivation processes through which users can inhabit and explore these worlds.. Following Guattari, they then propose to analyze “the economic, legal, techno-scientific and semiotics of subjectivation”, which these worlds entail.
3. What needs to be done
This they conclude, is a task at hand for critical scholars, and they propose a working method:
“As the role of the platform is to enact and realize Web 2.0 worlds through the articulation of protocols, it becomes necessary to dissect these protocological articulations. Platforms perform connections via protocols among “objects”: users, pieces of content, pieces of information produced through recommendation and personalization software. These connections enable the formation of the relationships that populate Web 2.0 worlds. In other words, Web 2.0 platforms establish the channels through which information can circulate and the challenge lies in developing tools to track, map and visualize such channels, from the protocols that enable them to their effects on stabilizing specific modes of being online. Such an approach has roots in the critical aesthetics of software studies - for instance, Fuller’s Webstalker (2003) as an alternative Web browser was an important step in understanding how user perceptions of what the Web is are constructed through specific visual regimes at the level of the user interface. Such an approach also stems from info-visualization as the production of tools to render visible previously invisible processes.
This critical approach to Web 2.0 platforms is based on disaggregation. Disaggregation as a method through which to strip, parse and rip the platform into its components is a useful approach, albeit one that needs to be adapted to the Web 2.0 environment. Indeed, most disaggregation tools rely on Web 1.0 protocols accessible to users, such as hyperlinks, domain names, or metatags. There are user-accessible protocols in Web 2.0, and those can be understood as traffic tags. Traffic tags are pieces of identification that are attached to a Web object: a user, a video, a picture each have, for instance, a distinct ID number on Web 2.0 platforms. Traffic tags can be human-generated, such as the title of video, or the real name of a user as they appear on the user-interface, or the user tags that describe how an object belongs to a class of object (i.e. “X’s wedding” or “election 2008″). Traffic tags are also computer-generated: unique identification numbers are assigned to a YouTube video, as well as to users on Facebook. Traffic tags allow not only for the recognition of objects within Web 2.0 platforms, but also are used by protocols to allow objects to circulate across platforms. For instance, when a user presses the “Share on Facebook” button after watching a video on YouTube, the ID number of the video will reappear in the Facebook source code of the user’s page. The current challenge thus lies in identifying and following traffic tags associated with Web objects so as to see how information circulates within and across Web 2.0 platforms. This will give clues as to how cultural processes that are traditionally only visible at the level of the user-interface are dependent on the software interfaces. In turn, this disaggregation of the articulation of Web 2.0 protocols will serve to identify techno-scientific semiotics and the ways in which they are associated and articulated with legal and economic semiotics and semiotics of subjectivation.
Such an approach is a challenge for researchers accustomed to working at the level of the user interface. Yet, moving from the user interface to the software interface is promising in terms of not only analyzing the technocultural economy of commercial Web 2.0 worlds through the mapping of the unfolding of protocological assemblages, but also with regards to using commercial Web 2.0 in non-commercial ways. Of particular interest are the Application Programming Interfaces (API) which have become a central feature of Web 2.0 spaces. APIs allows software programs to connect to Web 2.0 platforms and databases and undertake specific tasks. APIs offer a way to access information and tags that bypass the limits of the user interface. Therefore, their potential as ways to develop critical methodological tools should be explored. The goal would be to create new visualizations, new geographies - a map of where objects flow, where they migrate, where they are reshaped and re-circulated. The mapping of the connectivities and disconnectivities of Web 2.0 platforms could thus make use of techno-scientific semiotics as an entry point for understanding the production of Web 2.0 worlds.
Critical interventions into commercial Web 2.0 platforms are needed if we are to recognize the cultural importance and critical potentials of Web 2.0. Many instances in code studies and software studies show that such interventions can take different forms, from radical ruptures to aesthetic experiments and methodological tools. It is crucial, however, to not to be too quick in formulating critical judgments, and to understand the power dynamics in commercial Web 2.0 as both repressive and productive.” (http://journal.fibreculture.org/issue14/issue14_langlois_et_al.htm)
A little critique (from Michel Bauwens):
The approach to recognize the protocollary power of platforms is both repressive/alienating and productive, fits very well with the general approach we have been following here at the P2P Foundation. But I do have an important question: is such an approach not too platform-centric. In other words, rather than to start from the analytical approach of existing networks, shouldn’t we start with what we want to do as peer producers first, and deal with these limiting factors on a pragmatic basis? After all, isn’t our role not just to interpret the world, but to actually change it?