Extended Browsing: The Affordances and Platform Politics of Activist Browser Extensions
Master’s Thesis
New Media and Digital Culture University of Amsterdam Supervisor: Dr. Lonneke van der Velden
Second Reader: Dr. Marc Tuters Date: 28-06-2019
Abstract
This thesis investigates the use of browser extensions as an under-researched form of online activist engagement. The relevance of the study lies in the ubiquity of the web browser as a media tool and the increasing power of browser platforms in deciding how web content is delivered to the users. In response to this observation, the thesis seeks to identify the ways in which activist extensions redirect user’s attention from web interfaces to social issues, serve to promote user agency, and inform the mutual relationship between browser platforms and extension developers. In terms of theoretical framework, the thesis is informed by notions such as transduction and affordance, the historical practices of the Situationist International, and the conceptualization of platform along with the platformization of cultural production. These theoretical considerations are integrated with in-depth case studies of several activist extensions, including the conducting of discursive interface analysis, discourse analysis of user comments, and semi-structured interviews with two developers. The thesis concludes by establishing that activist extensions operate according to the mechanism of reminding and provide users with a coping mechanism in relation to a particular social issue. Meanwhile, browser platforms necessitate a process of infrastructural alignment for developers and consolidate their official extension store as the only legitimate channel for product distribution.
Keywords
Table of Contents
Introduction 1
1. Activist Extensions: A Brief Overview 6
2. Theoretical Framework 12
2.1 Interfaces and Users: From Transduction to Affordance 12 2.2 Users and Platforms: The Renegotiation of User Agency 14 2.3 Platforms and Developers: The Unfolding of Platform Governance 18
3. Methodology 20
3.1 Comparative Analysis and Conceptual Analysis 20
3.2 Discursive Interface Analysis and Discourse Analysis 21
3.3 Semi-Structured Interviews 22
4. Affordance, Agency, and Politics: An Analysis of Activist Extensions 24
4.1 A Transductive Understanding of Browser Extensions 24
4.2 The Reminder as an Extended Mechanism of Affordance 25
4.3 Extended Web Pages as Instances of Cybersituations 29
4.4 User Agency Reconsidered against Embedded Social Norms 34
4.5 The Platform Politics of the Extension Ecology 38
Conclusion 46
List of Illustrations 49
Acknowledgement 50
1 Introduction
Détournement not only leads to the discovery of new aspects of talent; in addition, clashing head-on with all social and legal conventions, it cannot fail to be a powerful cultural weapon in the service of a real class struggle. The cheapness of its products is the heavy artillery that breaks through all the Chinese walls of understanding. (Debord and Wolman 18)
The Situationist International, with its insurrectionary style and its subversive cultural weapon of détournement, ferociously resisted the society of the spectacle since the late 1950s. More than six decades later, the media-saturated spectacle has become all the more engrossing and ensnaring as it rapidly migrates to the virtual space. As perhaps the most ubiquitous media tool that constantly renegotiates the amorphous boundary between public domain and private space for the users, the web browser, with its interminable stream of web pages, seems to crystalize the “social and legal conventions” of our time. What then could be the détournement of the 21st century that not only discovers “new aspects of talent” but also clashes head-on with old conventions? As will be argued in this thesis, the browser extension, for all the “cheapness of its products,” proves itself to be the “heavy artillery” that could break through the walls of the new spectacle.
As the name suggests, browser extensions, also known as plug-ins or add-ons, are “small software programs that customize the browsing experience. They enable users to tailor [the browser’s] functionality and behavior to individual needs or preferences” (Google Chrome 2017). With the help of browser extensions, web pages cease to be a fixed terminal where web content as an end product is accessed and consumed. Instead, extensions grant users and developers the ability to customize the browsing tool, rendering the web page in a state of perpetual contingency, always subject to future manipulation on an interface level. In the meanwhile, however, a growing number of extensions have forgone the enrichment of functionality and instead assumed an activist mantle, seeking to draw attention to highly contested social issues. Without introducing significant changes to the functions of the browser itself, such “activist extensions,” so to speak, strive to redirect user attention towards social issues that often lie beyond the content of the web pages. What they have in common is the process of repurposing the intended use of the web page through annotation, with added or altered textual, visual, or auditory elements. This distinct type of activist extensions that “allow us to envision alternate realities” (Arenyeka) is the object of study of this thesis. Within the existing literature on such activist extensions, one of the most frequently investigated examples is Benjamin Grosser’s Facebook Demetricator, which removes all the metrics from the
2
Facebook user interface (Figure 1) and emphasizes the extent to which our contemporary social media are dependent on and conditioned by quantification (Grosser 2014). For instance, Christiane Paul examines his work through the perspective of internet art as a form of public art, considering the way in which it illustrates “how the boundaries between the public and the private have been redrawn in the age of social media and networked platforms of data aggregation” (212). For another, Leah Lievrouw regards Grosser’s work as an example of design activism, highlighting “a distinctive design philosophy and approach that embraces the hacker ethos” (70). In a similar vein, to identify how “the processes of gamification and datafication [can] be disrupted or rendered non-valid or non-reliable” (237), Daphne Dragona observes what she calls “counter-gamification” techniques, among which Facebook Demetricator exemplifies a practice that “both de-gamifies and de-datafies” social media (243). While the above arguments contribute to an in-depth analysis of quantification on social media, the characterization of Facebook Demetricator as a browser extension is limitedly descriptive, serving only to specify the medium of the work as does an exhibition label. In other words, existing research about Grosser’s work focuses primarily on the issue of quantification on social media rather than the browser extension itself as a unique media object. Considering the fact that few existing studies have examined activist browser extensions as a distinct category, this thesis takes Facebook Demetricator as a point of departure and attempts to venture beyond any specific activist issue in order to generalize the conditions on which most activist extensions operate.
3
To begin with, it may be useful to take into account existing research on the annotative capacity of browser extensions in general and other types of media objects. For instance, R Lyle Skains regards annotation, enabled by digital tools including browser extensions, as a contemporary practice of digital marginalia, which constitute “a form of academic discourse” (3), suggesting the potential discursive power browser extensions are endowed with. This discursive potential is crucial to the operation of activist extensions, as the redirection of user attention towards existing social issues entails the establishment of certain online discourse. Therefore, the use of activist extensions can be seen as the creation of a social commentary, or a narrative overlay, on top of the browser’s user interface. This interpretation of activist extensions can also be likened to similar practices in digital cartography. For example, Jason Farman describes how some advanced users in the Google Earth community have created visual overlays that “recontextualize and subvert ‘master representations’ of visual media within the authorial structure” (869, emphasis in original). Cartographic overlays, according to Farman, function as “rhetorical devices,” suggesting that “maps are not simply static visual facts to be received, but instead flexible signs that can be engaged in free play” (879). Whether it is academic discourse or rhetorical devices, what Skains and Farman both indicate is the narrative capacity inherent in the creation of an interface overlay, which is exactly what activist extensions utilize so as to highlight the social issues their creators have in mind. The difference, however, is the relationship between the overlay and the interface it annotates. For Skains, digital marginalia supplement academic publications as do their analog counterparts on printed books (3), whereas, for Farman, many user-generated information overlays in Google Earth are intended to repurpose, reimagine, and reconfigure dominant representations (882). This comparison is very important, since activist extensions, unlike those aimed at extended functionality, serve precisely to repurpose, reimagine, and reconfigure the web user interface they annotate and in so doing recontextualize and subvert the master representations delivered on the web page.
In addition, by redirecting user attention from virtual interfaces towards real-world social issues, activist extensions mobilize their annotative capacity and can be seen as creating what Steven Best and Douglas Kellner term “cybersituations,” a form of oppositional social practice, inspired by Guy Debord and the Situationist International (SI), that offers “expanded opportunities for resistance and democratization” (129-130). The group was active mainly from the 1950s to the 1970s, but its subversive orientation and the technique of détournement (misappropriation or hijacking) remain strikingly pertinent to the operation of activist extensions today. Indeed, as Lievrouw observes the hacker ethos in Facebook Demetricator, Best and Kellner regard hacking as the cyber equivalent of détournement (150). Following this Situationist perspective, the thesis investigates the critical, subversive, and resistant dimension of activist extensions, considering their ideological roots and
4
how they can serve to “promote oppositional activity aiming at progressive change and alternative cultural and social forms” (Best and Kellner 149).
Apart from the subversive nature of activist extensions, another crucial observation about current web pages is, as mentioned earlier, the perpetual state of contingency caused by the manipulation of browser extensions. This observation connects the use of extensions to a broader debate about the platformization of cultural production, in which cultural commodities become increasingly contingent. David B. Nieborg and Thomas Poell, for instance, explain such contingency through two interconnected perspectives. First, the notion of contingency is concerned with a heightened platform dependence, suggesting that “cultural production is progressively […] dependent on a select group of powerful digital platforms” (2). This trend is best exemplified by Microsoft’s recent adoption of Google’s browser engine Chromium for the development of its own browser Microsoft Edge, leaving Mozilla Firefox and Apple’s Safari as the only “Chromium-free browsers to influence future web standards” (Tung). Second, the concept of contingency also highlights the emergence of “contingent commodities,” which are “malleable, modular in design, and informed by datafied user feedback, open to constant revision and recirculation” (Nieborg and Poell 2). This second aspect is directly tied to the contingency-inducing nature of all browser extensions, as web pages cease to be an end product and are becoming malleable and subject to revision and recirculation. Considering Chrome’s dominant position in the browser market and the contingency of web user interfaces caused by the annotative capacity of extensions, the focus on activist extensions in this thesis can be seen as responding to Nieborg and Poell’s call to “develop in-depth case studies of how platformization unfolds in particular […] instances of cultural production” (14). In their interpretation of platformization based on the fields of business studies, political economy, and software studies, the two authors emphasize the critical interplay among market structure, platform governance, and media infrastructures (14). These issues are crucial to the understanding of activist extensions, which similarly operate in close connection with different market players (users and developers) and the browser platform on which the extension ecology depends. Thus, the notion of platformization also attests to the potential theoretical contribution of investigating activist extensions as an under-researched object of study.
As an annotative overlay operating on the web user interface, activist extensions not only exhibit a conspicuous subversive orientation that dates back at least to the historical SI, but they also evolve according to a symbiotic relationship with the browser as a sociotechnical platform that carries profound social implications for both users and developers. Situated at the junction of these two
5
considerations, this thesis asks the question: how can the use of browser extensions constitute an emergent form of online activist engagement? Informed by Nieborg and Poell’s conceptualization of platformization, the thesis will analyze activist extensions based on the interwoven relationships among interfaces, users, platforms, and developers. Accordingly, the main research question can be divided into three subquestions: in what ways do activist extensions 1) redirect user attention from virtual web interfaces to real-world social issues, 2) serve to promote user agency with the capacity for activism, and 3) inform the relationship between browser platforms and developers through different means of platform governance? By addressing these critical questions, this thesis strives to contribute to the field of new media studies by foregrounding activist extensions as an under-researched media object, shedding new light on the conditions on which they operate, the impacts they have on users, and the relational dynamics they occasion between platforms and developers. Focusing on new forms of activism while instantiating the platformization of cultural production, this research project also seeks to contribute to the creation of a theoretical bridge between existing activist practices and the web browser, not only as one of the most essential tools for users within the contemporary media landscape but also as an important cultural intermediary that constantly defines the fluid boundary between public domain and private space online.
In order to address the above research questions, the remainder of the thesis is divided into the following parts: the first chapter will start with a brief overview of activist extensions, including a precise definition of what an activist extension is, a list of major extension categories in terms of what they do on the interface level, and a description of all the activist extensions to be discussed later; the second chapter will outline a comprehensive theoretical framework to introduce relevant concepts and arguments based on the existing literature; the third chapter will describe the main research methods used to answer the research questions; the fourth chapter will bring into dialogue the theoretical framework and specific activist extensions as objects of study to analyze how their use, creation, and distribution can articulate the mutual relationship between web interfaces, users, browser platforms, and extension developers; the thesis will conclude by revisiting the research questions and their relevance, presenting major findings, and suggesting potential future research regarding activist extensions.
6 1. Activist Extensions: A Brief Overview
At this moment, there is no established definition of activist extensions in the existing literature. In fact, the query for “activist (browser) extensions,” or such variant forms as plug-ins or add-ons, in most search engines generates few relevant results. Nonetheless, among the limited number of relevant entries, an artist has compiled a comprehensive list of what she calls “activist” themed browser extensions and suggests that “[w]hat makes a browser extension ‘activist’ […] is when its creator starts with a problem they see in the world and makes the extension as a way to address it and share a new point of view” (Arenyeka). Based on the types of the social issues creators hope to address, Arenyeka observes several major categories of activist extensions, such as “Rethinking Social Media,” which includes those that address issues of datafication or user privacy on social media platforms, “Call It By Its Name,” which has its roots in the callout culture and reveals how things should really be called, “Turning the Tables,” which focuses on the struggle of marginalized social groups, and “A Different Type of Erasure,” which applies different filtering mechanisms to assist consumer boycott. Building on Arenyeka’s existing taxonomy, I would like to propose to rename the last category as “Selective Filtering” and add a new group named “Informed Annotation.” A summary of the resulting five categories and examples of each group is listed in Table 1.
Category Example Behavior
Rethinking Social Media
Twitter Demetricator Hide all metrics on Twitter
Data Selfie Analyze user data based on Facebook activities
Call It by Its Name
Stop Normalizing Alt Right Replace “alt right” with “white supremacy”
You Mean Racist? Replace “racially charged” with “racist”
Turning the Tables
Jailbreak the Patriarchy Genderswap pronouns and other gendered words
Re-Search Make image search of professions more gender balanced
Selective Filtering
DropUnited Hide United Airlines from flight search results
BuyBlack Find products and services that support black economy
Informed Annotation
Grab Your Wallet Create alerts for websites on consumer boycott list
Coincidence Detector Encase Jewish names and organizations in parentheses
7
While such a classification system can be very useful in revealing the broad scope of activist agendas that have adopted browser extensions as a media tool, these categories are not exhaustive or strictly mutually exclusive. More importantly, in order to derive a working definition of activist extensions, it is also important to identify what their behaviors have in common. For instance, all of the above examples can be seen as playing around the notion of (in)visibility. The deceptive acts of social media platforms remain invisible due to the lack of readily accessible information for users; the struggling of socially disadvantaged groups remain invisible because of long-standing structural discrimination on the societal level. In either case, to play with the notion of (in)visibility is an attempt to engage users with certain social issues that exist and persist beyond what the web pages are originally intended for. Therefore, a central aspect of activist extensions is the disrupting or dismissal of the intended use of the web page. In the meanwhile, the manipulation of (in)visibility is realized through the modification of user interfaces, usually in the form of hiding, replacing, or providing information for certain textual, visual, or auditory elements of the web page.
In view of the above considerations, activist extensions in this thesis can be defined as browser extensions that modify textual, visual, or auditory elements of the web user interface in order to disrupt its intended use and redirect user attention towards highly contested or otherwise ignored social issues. Within such a definition, an implicit “indiscriminateness” of activist extensions must be emphasized. That is to say, despite its disruptive capability and widespread use, the browser extension, as a type of media object, is not inherently or by default geared towards the promotion of social progress. Instead, the same annotative capacity can also be and has already been exploited for insidious causes. This is perhaps best demonstrated by Coincidence Detector, an innocuously named extension that encases Jewish names and organizations with a set of multiple parentheses so that they become the target of anti-Semitic rhetoric and online harassment (Figure 2). In contrast, Uncover Harassers highlights celebrities who have been accused of sexual harassment and offers a short description about the accusation (Figure 3). Despite their markedly different motivations, the two extensions can be considered as following a similar logic of “informed annotation,” as both are supported by a human-curated database that decides which names will eventually be marked on the web user interface.
8
Figure 2. Coincidence Detector, screenshot, retrieved May 2019.
Figure 3. Uncover Harassers, screenshot, retrieved May 2019.
It is also worth noting, based on the above definition of activist extensions, that this thesis will not focus on popular browser tools such as ad-blockers or anti-trackers. While these tools indeed grant users more control over their browsing experience and usually advocate activist issues such as data security or user privacy, they nonetheless fall out of the scope of this thesis, since such extensions operate on web user interfaces precisely for the purpose of preventing user attention from being redirected away from the web page. In other words, the “non-disruptiveness” of most ad-blockers and anti-trackers situate them in the broad category of “utility” extensions, as opposed to those operating at the expense of the web page’s intended use. As an extension developer admits, ad-blockers are generally not activist in nature, although they may have influenced public opinion on such issues as data privacy indirectly and unintentionally (Ghostwords). That said, in the broader
9
spectrum of existing ad-blockers, AdNauseam is undeniably one that acts and is treated like an activist. According to the extension’s creators, “AdNauseam quietly clicks on every blocked ad, registering a visit on ad networks’ databases. As the collected data gathered shows an omnivorous click-stream, user tracking, targeting and surveillance become futile” (AdNauseam). While its radical approach does not align itself with most other ad-blockers, it does not fit the proposed definition of activist extensions either. However, the point here is to emphasize that the objects of study in this thesis have been chosen because they all exhibit a particular pattern of activist behaviors, not that other types of extensions are or should be, by exclusion, considered non-activist. As for access and dissemination, users can find and install most activist extensions in a browser’s official extension store, as they do with most utility tools. In addition, experienced users can also enable the “developer mode” of a browser and manually load extension scripts that are usually available through third-party hosting sites. This is particularly the case for extensions that have been banned in the official store or fail to be updated in accordance with the browser’s latest technical requirements. AdNauseam is one case in point. Similarly, Coincidence Detector has also been removed from the Chrome Web Store, Google’s official extension marketplace, “for violating Google’s hate speech policy” (Fleishman and Smith), while Dark Side of the Prism, an extension that plays a Pink Floyd playlist to remind users of the NSA’s massive surveillance scheme, is now labelled as a “Legacy Extension” on Firefox and no longer supported by recent versions of the browser that feature new security protocols (Mozilla Support). Considering the screening and moderating role of a browser’s official extension store, the ways in which activist extensions are, or fail to be, accessed through such walled gardens will also be examined as an important aspect of the browser’s platform governance.
Finally, several other objects of study in the thesis will be introduced as follows. One of the earliest examples of activist extensions is Google Alarm, created in 2010 by Jamie Wilkinson from The Free Art and Technology Lab. Similar to Dark Side of the Prism, it triggers audio and visual alerts on websites where Google secretly collects user data. In addition to the Demetricator series, Grosser has also created many other extension-based works, including Textbook and Safebook. The former removes all the images from the Facebook interface, prompting users to consider the role played by images not only in the way one reads Facebook but also in the proliferation of fake news, viral propaganda, and clickbait (Grosser 2017). The latter follows a similar logic but, more radically, removes all the images, texts, videos, and audios from the site, revealing how the Facebook interface is deeply ingrained in users and how its content may impair mental health, privacy, and even democracy (Grosser 2018). Another common type of activist extensions targets political figures,
10
the most prominent examples of which, incidentally all about Donald Trump, have been created by popular news satire television programs from the United States. For instance, The Daily Show with Trevor Noah has created Make Trump Tweets Eight Again, which changes the font and color of Trump’s tweets so that they look like the crayon scribbles of an eight-year-old child (Figure 4), and Trump Toad Extension, which replaces the “T” in “Trump” with a mushroom emoji (Figure 5), referring to the adult film star who claims to have had an affair with Trump and compares his private parts to a toadstool (McCarthy). For another example, Last Week Tonight with John Oliver has created Drumpfinator, which mocks the then presidential candidate by replacing all instances of “Trump” with “Drumpf” (his family’s ancestral name) and “Make America Great Again” with “Make Donald Drumpf Again” (Last Week Tonight).
Figure 4. Make Trump Tweets Eight Again, screenshot, retrieved May 2019.
11
Throughout the thesis, activist extensions mentioned above will be discussed at varying lengths to support the addressing of the research questions raised in the introduction. As a brief summary, all the activist extensions (including their form of distribution) to be discussed in the thesis are listed in Table 2.
Extension Distribution
BuyBlack, DropUnited, Drumpfinator, Facebook Demetricator, Grab Your Wallet, Jailbreak the Patriarchy, Make Trump Tweets
Eight Again, Re-Search, Safebook, Stop Normalizing Alt Right, Textbook, Trump Toad Extension, Twitter Demetricator,
Uncover Harassers, You Mean Racist?
Chrome Web Store (or Firefox add-on store)
AdNauseam, Coincidence Detector Dedicated sites with installation instructions
Dark Side of the Prism, Google Alarm Scripts available on third-party hosting sites
12 2. Theoretical Framework
Since activist extensions represent a relatively under-researched type of media object, an essential task of the research project is the establishment of an extensive theoretical framework, along with the introduction of pertinent concepts and arguments, that can be operationalized to inform the understanding of activist extensions, including their use among users, creation by developers, and distribution across browser platforms. As is mentioned in the introduction, activist extensions distinguish themselves from other media objects due to their disruptive orientation and their dependence on the browser as a sociotechnical platform. Therefore, the theoretical framework of the thesis will be divided into the following sections that focus, respectively, on the redirection of user attention from virtual interfaces to real-world social issues, the promotion of user agency, and the relational dynamics between platforms and extension developers.
2.1 Interfaces and Users: From Transduction to Affordance
As browser extensions alter the textual, visual, or auditory elements of the web user interface, they first of all change the relationship between various objects that reside in the interface and mutually impact each other. Therefore, an object-centered approach to interfaces is a helpful starting point for the theorization of browser extensions in general. For instance, James Ash regards interfaces as “sets of objects that continually encounter one another and generate particular qualities that are partially dependent on these encounters.” This technical encounter can be seen as a process of transduction, “the selective emergence of particular qualities via an encounter between technical objects with some end goal or effect in mind” (28). Here, the web interface should not be regarded as a singular object but a composite one, consisting of different sets of smaller objects operating with each other inside its perimeter. For Ash, “understanding all object relations as transductive can account for how non-human processes come to have tangibly human effects on the user of that interface” (31). Notably, the process of “selective emergence” is precisely why Ash’s interpretation is relevant to the operation of browser extensions, which introduce selective encounters between technical objects so as to extend the functionality of the web page. Following the same logic, activist extensions attempt to disrupt this very process of selective relation, creating alternative encounters that produce new qualities unintended by the web page.
Following this transductive approach, Ash suggests that we understand interfaces as environments in which objects interact with each other as well as with users (29). In other words, Ash considers interface environments as “assemblages of objects that are positioned and spaced in relation to one
13
another in order to transduce qualities for both other objects in the interface and the user engaging with that interface” (31). This ecological interpretation of interfaces is a perfect gateway to the inclusion of users in the discussion. As users browser the web pages, they start to make contact with and become a part of the interface environment. Therefore, departing from Ash’s object-centered approach, we also need to consider the relational aspects of interfaces that foreground how the transduced qualities unfold in practice for the users, as well as when activist extensions intervene to disrupt the intended unfolding process.
To examine the transduced qualities produced by activist extensions is to identify what the manipulated web interfaces allow users to do or prevent them from doing. Therefore, the concept of affordance provides an ideal conceptual basis for our discussion. Emerging from the field of ecological psychology, the notion of affordance is originally conceived by James J. Gibson as “what [the environment] offers the animal, what it provides or furnishes, either for good or ill” (127, emphasis in original); more broadly, Jenny L. Davis and James B. Chouinard define affordance as “the range of functions and constraints that an object provides for, and places upon, structurally situated subjects” (241); primarily concerned with affordances in the field of interaction design, H. Rex Hartson divides the concept into four categories (Table 3): cognitive, physical, sensory, and functional affordances (323).
Type Description
Cognitive Affordance Design feature that helps users know something
Physical Affordance Design feature that helps users perform a physical action in the interface
Sensory Affordance Design feature that helps users sense something
Functional Affordance Design feature that helps users accomplish work
Table 3. Four affordance types according to Hartson’s (323) conceptualization.
Hartson’s more nuanced approach of affordance is highly relevant to the understanding of activist extensions, as this categorization further dissects the notion of affordance in terms of a web page’s intended use. Specifically, most browser extensions are intended to help “users accomplish work” mainly to facilitate or expand the functional affordances of the web page. In contrast, activist extensions disregard and disrupt the intended functions and implement instead design tactics that primarily revolve around sensory and cognitive affordances. It is worth noting that, as browser extensions operate on virtual interfaces, the notion of physical affordance does not apply to our
14
discussion here. More importantly, the four types of affordances are not independent from each other but are rather structured within a hierarchical relationship. For instance, Hartson argues that sensory affordances play a “critical supporting role” and should be seen as an attribute of physical or cognitive affordances (322). This hierarchical structure is highly relevant, as it is precisely this critical supporting role of sensory affordances that activist extensions attempt to appropriate. While Hartson’s arguments contribute to an in-depth understanding of affordance that centers on the media objects, the inherently relational aspect of this notion requires similar efforts to avoid undue homogenization of the users or the usage of such media objects. Davis and Chouinard, for instance, have observed a general “failure to account for diverse subjects and circumstances” (241) in the existing literature on affordance. In response, the two authors attempt to “demarcate the mechanisms of affordance – as artifacts request, demand, allow, encourage, discourage, and refuse – which take shape through interrelated conditions: perception, dexterity, and cultural and institutional legitimacy” (241). In particular, the mechanisms of affordance address the question of how artifacts afford in practice, while the conditions of affordance explain for whom and under what circumstances artifacts afford. As none of the mechanisms mentioned above applies to the redirection of user attention, the investigation of activist extensions potentially contributes to the expansion of the two authors’ existing theoretical model. In the meanwhile, the conditions of affordance are more closely related to the browser’s platform governance over activist extensions. How users become aware of the activist extensions (perception), whether they have the necessary technical know-how to use the extensions (dexterity), and how these two aspects may change over time as more users become familiar with extensions or as platforms intervene to remove some but not other extensions (cultural and institutional legitimacy), these are all relevant questions that can be addressed by analyzing the conditions of affordance.
2.2 Users and Platforms: The Renegotiation of User Agency
While notions such as transduction and affordance can help us understand how activist extensions operate on the interface level to redirect user attention and disrupt the intended use of the web page, it is also crucial to realize that the process of redirection and disruption is not only technical but also deeply political. With a preconceived intended use built in, web interfaces are in fact making normative claims as to how users should behave or respond. Through the manipulation of the web interface, activist extensions then have the potential capacity to respond to and resist such normative claims. In other words, while the web interface can be seen as “carriers of cultural logics and ideologies” (Ash 20), the use of activist extensions can be employed to promote counter-logics
15
and counter-ideologies. It should be pointed out that such a politicized interpretation of the interface is anything but new. For instance, Lev Manovich suggests that the interface “provides its own model of the world, its own logical system, or ideology” (76); More closely related to the web user interface, Mel Stanfill proposes what she calls “discursive interface analysis” as a critical tool to investigate the construction of social logics and norms through interface design, which offers users with “the normative or ‘correct’ or path of least resistance” (1060). Disrupting the intended use of the web page, activist extensions subvert this “path of least resistance” and in so doing seek to counter the social logics and norms that are, according to Stanfill, not only reflected but also non-deterministically reinforced through the user interface (1060).
As the phrase “discursive interface analysis” already suggests, Stanfill’s interpretation seeks to highlight the web user interface as a discursive space. As activist extensions annotate such spaces, Stanfill’s interpretation of the interface echoes Skains’ arguments, mentioned earlier in the introduction, about digital marginalia being a form of academic discourse. Both authors recognize the discursive nature of the web user interface, implying the discursive capacity that browser extensions possess. Likewise, Stanfill’s understanding of the interface also resonates with Farman’s observation about information overlays created in Google Earth that recontextualize and subvert master representations: just as maps are “flexible signs that can be engaged in free play” (879), so too are web user interfaces contingent discourses that can be engaged in narrative manipulation. Therefore, if interfaces indeed carry social logics and ideologies whereas activist extensions can provide equally political counteracts, such provisions would stem from the activist extensions’ inherent narrative capacity, the capacity to create a narrative overlay on top of the user interface that is ultimately discursive.
Despite the relatively short history of browser extensions, the provision of activist interventions against master representations that carry social logics and ideologies is not unprecedented. In fact, the use of activist extensions to foreground real-world social issues can be regarded as the creation of “cybersituations,” a term coined by Best and Kellner to characterize a novel form of oppositional practice that is aimed at expanded opportunities for social resistance. According to the two authors, Debord’s notion of the spectacle conceived in the 1960s refers not only to “a media and consumer society, organized around the consumption of images, commodities, and staged events” but also “to the vast institutional and technical apparatus of contemporary capitalism, to all the methods power employs, outside of direct force, to relegate subjects to passivity and to obscure the nature and effects of capitalism’s power and deprivations” (132). In response to this institutional and technical apparatus, the SI promoted, among other things, the technique of détournement, or “the reuse of
16
preexisting artistic elements in a new ensemble” (Situationist International 67). The French word means “deflection, diversion, rerouting, distortion, misuse, misappropriation, hijacking, or otherwise turning something aside from its normal course or purpose” (Knabb 480). Through “the creation of new forms and the détournement of previous forms of architecture, urbanism, poetry and cinema,” the central idea of the SI is to construct situations, which is “the concrete construction of momentary ambiences of life and their transformation into a superior passional quality” (Debord 2006 38).
In order to update Debord’s original version of the spectacle, Best and Kellner formulate what they consider “as the advent of a new stage of the spectacle, requiring new technologies and forms of oppositional practice” (129). In particular, the authors define the “megaspectacle” as a “significant escalation of the spectacle in size, range, and intensity,” examples of which include, as they wrote in 1999, blockbuster films like Star Wars, technologically mediated theme parks, and breaking headlines like the Simpson trial or the Clinton scandal, which “produce such saturation coverage that they define an era of culture” (135). More pertinent to the use of media technologies, the authors also observe the “interactive spectacle,” which involves “the creation of cultural spaces and forms that present exciting possibilities for creativity and empowerment of individuals, as well as novel forms of seduction and domination” (144). Again, written in the year when the term Web 2.0 was just coined, the article illustrates interactive spectacles only with reference to early versions of web television, webcam videos, video games, and prototypical virtual reality devices.
Just like the historical SI, Best and Kellner, in response to the new interactive spectacle, propose the construction of cybersituations, which “involves the appropriation, use, and reconstruction of technologies against the spectacle and other forms of domination, alienation, and oppression” (149). Here, what Debord and the two authors have in common is the advocacy of active oppositional subjects “fully participating in the production of everyday life, their own individuality, and, ultimately, a new society” (Best and Kellner 142). Consequently, in theorizing the (interactive) spectacle as a form of domination and manipulation while the construction of (cyber)situations as a demonstration of agency and resistance, the Situationist vocabulary is highly relevant in framing the manipulation of the web user interface as another cyber equivalent of détournement and the use of activist extensions as the creation of contemporary cybersituations.
In the meanwhile, Best and Kellner also claim that the spectacle generates a distinction between passivity and activity as well as consumption and production (142), suggesting active production, as opposed to passive consumption, as a key indicator of heightened user agency. However, José
17
van Dijck problematizes the understanding of user agency through bipolar categories and hybrid terms such as produsers, seeking instead to relativize the implied assumptions behind the notion of online participatory culture:
First, the concept of user is often bolstered by a deceptive opposition between the passive recipient, couched in the rhetoric of “old media”, and the active participant cast ideally as someone who is well-versed in the skills of “new” media. Second, participation refers to citizens and community activists as well as to people who deploy their skills and talents towards a common cause. Yet can terms such as “communities” and “(cultural) citizenship” be unequivocally transferred to internet communities? And third, now that citizens have become creators and arbiters of media content, what role do platform providers play in steering the agency of users and communities? (43)
These three aspects are all highly relevant to the investigation of activist extensions. First, for the deceptive opposition between passive recipient and active participant, van Dijck cites a 2006 survey that identifies six different levels of participation: active creators, critics, collectors, joiners, passive spectators, inactives (44). Users of activist extensions, however, do not fit into any of the existing categories, but rather exhibit a new form of online cultural participation that potentially expands the participation ladder. Second, van Dijck also highlights the ambiguity of the term “community,” which supposedly entails “the inclination of users to belong to a (real-life) group and be involved in a common cause.” In reality, however, she claims that most communities on user-generated content sites represent what she calls “taste community” or “brand community” (45). While this may be a valid observation, the socially engaging nature of most activist extensions in fact provides a clear common cause required for the user communities to transcend consumer or entertainment groups. Third, van Dijck foregrounds the “substantial role a site’s interface plays in manoeuvring individual users and communities,” which undermines user agency because of the platform’s algorithmic conditioning and human moderation. However, this is precisely why the examination of activist extensions should approach extensions as media objects entangled in the process of platform governance and highlight the relationship between platform providers and extension developers.
A final issue regarding the promotion of user agency with activist extensions is whether or how the awareness of social issues can be transformed into actual social actions. As Best and Kellner claim, the construction of cybersituations “provides potential articulations between cyberworld and the real world, while pseudo-interaction merely entangles one ever deeper in the matrices of escapism
18
and corporate entertainment” (152). For such articulations to be established, pseudo-interactions that entangle user attention must be disrupted or invalidated. Therefore, the redirection of user attention towards external social discourses is a crucial aspect of activist extensions. As a plethora of interactive spectacles compete for limited user attention, activist extensions also participate “in the attentional assemblages of digital media,” allowing users to become “part of social processes where paying attention triggers responses of imitation which shifts between the virtual form of a passing impression and the actual form of acts” (Terranova 7). In other words, “the economy of attention is […] also the economy of socialization of ideas, affects and percepts, and hence an economy of social production and cooperation” (8).
2.3 Platforms and Developers: The Unfolding of Platform Governance
Based on van Dijck’s last observation about online participatory culture, the role of platforms also needs to be investigated. However, while van Dijck’s considers the two-sided relationship between platform providers and users, examining activist extensions as objects of study requires additional attention paid to the role of extension developers. Therefore, an expanded theoretical framework that takes into account the multisided relationship among users, browser platforms, and extension developers is urgently needed.
Nieborg and Poell’s conceptualization of the platformization of cultural production provides one such approach. Triangulating platformization from the perspectives of business studies, political economy, and software studies, the two authors consider “how shifts in market structures, forms of governance, and infrastructures mutually articulate each other” (14). To begin with, the business studies perspective draws particular “attention to multisided markets as an increasingly dominant institutional configuration” (4), which can be mapped onto the extension ecology that is similarly organized around different groups of participants. Specifically, extension developers, based on this perspective, can be considered as platform “complementors.” Second, the field of software studies raises questions such as “how the sociotechnical features of platforms allow and prompt end-users to afford particular types of activities, connections, and knowledge” and “under what conditions platforms allow complementors access to the means of production and circulation” (6). Whereas the first question is more concerned with the relationship between users and platforms and can be addressed based on the examination of interface affordances, the second question is particularly relevant to non-utility extensions, since those deemed inappropriate by the browser platform, often on unilaterally determined terms, will be and have been banned from accessing its infrastructural services that are essential for production and circulation of the extension. The ancillary nature of
19
the extension ecology has created a lopsided power structure between platform providers and their complementors, which leads to the central questions of the third part of Nieborg and Poell’s model of platformization. Drawing on the discipline of political economy, the two authors ask “how digital platforms affect power relations in the cultural industries and the autonomy of complementors” and “how platform power is operationalized through platform governance frameworks” (11). In the meanwhile, despite the ostensibly overbearing position assumed by the platform providers, extension developers are not without agency. Due to the programmability of contemporary digital platforms, developers and advanced users have the capacity to bypass the technical and sometimes legal restrictions imposed upon unsanctioned extensions. In order to reconcile the programmable aspect of the browser as a technical platform with the affordances it provides users through its interfacial regime, Benjamin Bratton’s conceptualization of the platform logic provides a totalizing model. According to Bratton, “an essential logic of platforms is a reconvergence of architectural, computational, and political connotations of ‘program’ back into one,” in which architectural and computational programs can both be seen as an “intended organization of Interfaces in a particular arrangement so as to coordinate social contact and interaction (or prevent it),” whereas recognizing platforms as an institutional form “situates the convergence of its architectonic and computational forms in a more specific and fundamental way” (43-44, emphasis in original). The creation of a computational software, in other words, is comparable to the construction of an architectural space, and the highly situated institutional context is where this logic of platform unfolds in practice. Bratton’s conceptualization of the three dimensions of digital platforms is relevant not only because it articulates the multisided configurations of the browser as a sociotechnical platform, but, more importantly, it also accounts for the occasionally conflicting forces within the extension ecology that result in the simultaneous codependence as well as contestation between platform providers and their complementors. For Bratton, “platforms do not work according to detailed premeditated master plans; rather they set the stage for actions to unfold through ordered emergence. […] Platforms begin by fixing equally strict means but are strategically agnostic as to outcomes” (47). Here, the strict means specified by the platform is what has created its dominant position, but its strategic agnosticism towards outcomes, as will be demonstrated through the discussion of recent changes to Google’s browser platform, is what allows developers to devise roundabout strategies. All in all, Bratton’s conceptualization of platform and Nieborg and Poell’s platformization model will serve as the theoretical foundation on which this thesis seeks to reveal the relational dynamics between the browser as platform providers and extension developers as platform complementors.
20 3. Methodology
In order to investigate how the use of browser extensions constitutes an emergent form of online activist engagement, this thesis primarily adopts the method of case studies, in which I will first introduce different categories of activist extensions to establish a broad overview before examining a few cases in greater detail. In the meanwhile, the use of case studies also necessitates extensive comparative analysis among these extensions, not only focused on the (different and common) ways in which they operate on the interface level but also guided by the conceptual analysis of relevant theories introduced in the second chapter. Through extensive comparative analysis of the extensions’ behaviors, I would like to foreground the variety of potentials for activism afforded by extension-based interventions and the common mechanism through which activist extensions operate in order to realize such potentials. In addition, the case studies method adopted in the thesis also involves discourse analysis of user reviews for some extensions to be discussed as well as developer feedback made in response to recent changes of the Chrome browser’s infrastructural services. Finally, semi-structured interviews with two extension developers will be conducted to gather additional insight into how the process of developing and distributing browser extensions is controlled through various means of platform governance. Through these analytical steps, I hope to operationalize the previously established theoretical framework and supplement the conceptual analysis with more empirically oriented research work.
3.1 Comparative Analysis and Conceptual Analysis
Considering the under-researched nature of activist extensions within the existing literature, the identification of relevant objects of study and the conducting of comparative analysis first require the formulation of a clear definition of what an activist extension is. As has been established in the first chapter, activist extensions in this thesis refer to browser extensions that modify sensory or cognitive elements of the web user interface to disrupt its intended use and redirect user attention towards existing social issues. Thus, the comparative analysis approach will seek to draw particular attention to the types of interfacial modifications implemented by the extensions and the types of social issues they intend to highlight. More importantly, I hope to draw on this comparative analysis in order to address the first research question concerning the common mechanism through which the process of redirecting user attention unfolds in practice.
The selection of relevant cases for comparative analysis departs from Facebook Demetricator and its creator Benjamin Grosser, who has created several extension-based works. Subsequently, more
21
extensions have been added as potential objects of study based on the inclusion of Grosser’s works within the existing literature and other non-academic sources, including the list of activist themed extensions compiled by Arenyeka. The resulting five categories of activist extensions are listed in Table 1. Based on this classification, I will continue to examine one or two extensions from each category in greater detail so as to highlight not only the behavioral variety of activist extensions but, more importantly, the commonality in terms of how they redirect user attention away from the intended use of the web page.
Throughout this process, the comparative analysis of activist extensions is informed by extensive conceptual analysis that integrates relevant concepts and theories introduced in the second chapter, such as transduction and affordance (including its mechanisms and conditions) for the discussion of interfacial modifications, the Situationist vocabulary of (interactive) spectacle, détournement, and (cyber)situation for the consideration of user agency and resistance, as well as the platform as a logic of governance and the platformization of cultural production for the understanding of the browser platform’s power and control over extension developers. While many of these concepts are not necessarily directly related to the use or creation of activist extensions, they nevertheless serve as the theoretical underpinning of the thesis, as their focus on interface, resistance, platform is operationalized within new analytical contexts.
3.2 Discursive Interface Analysis and Discourse Analysis
Through the creation of a narrative overlay, activist extensions can be seen as operating on the web user interface that is decidedly discursive. This overall discursive nature not only manifests itself on the interface level, but it can also be examined based on the accumulated user comments on the extensions and developer feedback on a browser platform’s infrastructural services. Therefore, the investigation of this expansive discursive space to which user attention is being directed requires two distinct analytical approaches.
On the interface level, Stanfill’s method of “discursive interface analysis” can serve as a valuable analytical tool, which “allows examining the assumptions built into interfaces as the normative or ‘correct’ or path of least resistance” (1060). Crucially, the identification of social logics and norms embedded in the web pages, according to Stanfill, is based precisely on Hartson’s categorization of affordances. Therefore, the adoption of discursive interface analysis as a research method can be seen as a continuation of the aforementioned conceptual analysis centered around the notion of affordance. Disrupting the intended use of the web page and unveiling its embedded assumptions
22
about the users, activist extensions exemplify the critical value of discursive interface analysis as an analytical tool. Therefore, I will use this method to analyze how specific activist extensions serve to reveal and repurpose the intended affordances of the web page, along with the cultural logics and social norms that have been disrupted in the same process.
In the meanwhile, to understand the intervened web interface as a discursive space on the textual level, it is also useful to examine the growing collection of user comments posted in a browser’s online extension store as well as developer feedback in response to a browser platform’s changing policies and infrastructural services. Specifically, I will perform discourse analysis of user reviews to identify what activist extensions allow users to do, what kinds of user control they grant, and how they may serve to promote user agency. Considering the dominant market position of Google Chrome, I will focus on user comments posted on the Chrome Web Store. In the “Reviews” section of an extension’s home page, signed-in users can provide a rating based on a system of one to five stars and post text-only comments, to which other users may also reply. In addition, an optional “Support” section also provides users with the opportunity to directly contact the developers. These two sections then combine to provide the source of user comments as textual data, which will be collected and analyzed manually to shed further light on the issue of user agency. As for developer feedback, I will focus on comments made in response to the Chrome Manifest V3 proposal recently published by Google, which “uses manifest version as a method to restrict certain capabilities of certain extensions” (E. Carter). Some comments are posted on a dedicated extension developer forum thread managed by members of the Chrome team, while many more can be found in articles published on popular technology media outlets. I will combine both sources to analyze how the contestation of platform governance can be understood based on developer feedback about recent changes made to Chrome’s developer policies and infrastructural services.
3.3 Semi-Structured Interviews
In parallel with the discourse analysis of developer feedback, semi-structured interviews with open-ended questions will also be conducted with two extension developers to better understand the relational dynamics between extension developers and the browser platform, particularly regarding the creation and distribution of activist extensions.
The first interviewee is Benjamin Grosser, arguably the most prolific creator of activist extensions. Not only are his projects, especially the Demetricator series, frequently featured within the existing literature, but he has also written extensively on issues related to user agency, the impacts of social
23
media, and the use of browser extensions as an artistic strategy. Since many of Grosser’s works are the objects of study of the thesis, an interview with him will greatly facilitate the understanding of whether and how he considers his works as a form of activist intervention, what motivations and expectations he may have, and, more importantly, whether and how his creative efforts have been helped or hindered by the browser platform’s infrastructural services and configurations, such as developer policies regarding content creation and distribution.
The other interviewee is Alexei Ghostwords (who has asked to be addressed by his first name and his GitHub ID), who has previously worked for the privacy ad-blocker Ghostery and is currently a member of the Electronic Frontier Foundation, a non-profit digital rights group behind the privacy and security extension Privacy Badger. While the two extensions, as argued in the first chapter, do not fit the definition of activist extensions, the experience of a developer of two popular utility tools can nevertheless shed important light on the relationship between the browser platform and the developer community in general. Moreover, the decision to interview the developer of two ad-blocking extensions is also based on Google’s recent Chrome Manifest V3 proposal, which includes changes of the Chrome APIs that may threaten to stop most ad-blockers from working (Ghostery Team). In response, many in the ad-blocking community have vehemently protested against the proposed changes, with Ghostery even considering filing an anti-trust complaint against Google (Bradshaw). Since the Manifest V3 proposal signals a more restricted access to the infrastructural services of Chrome as a sociotechnical platform, the perspective of a developer whose previous and current work may be negatively impacted is therefore of great value to understand the unfolding of platform governance through policy and service changes.
For both interviews to be conducted for this thesis, I will first introduce to the interviewees the objectives of the research project and the purpose of conducting interviews. The interviewees will be provided with an information brochure detailing the research topic and an informed consent form confirming their voluntary participation in the research project. Specifically, no significant risk is expected for the participation of the interviewees, and all information gathered through the interviews will only be used for further analysis and publication of the thesis. In terms of interview setup, synchronous online interview is chosen to accommodate the two interviewees’ different geographic locations. Through initial contact with the interviewees, both of them have indicated that the synchronous online format is their preferred form of communication.
24
4. Affordance, Agency, and Politics: An Analysis of Activist Extensions
Based on the list of activist extensions described as objects of study, the theoretical framework, and the research methods outlined in the previous three chapters, I will now proceed to bring the three components into critical dialogue with each other and present the analytical results as informed by the three research questions raised towards the end of the introduction. The first two sections will focus on the behavioral aspects of activist extensions and propose the common mechanism of their operation based on the notions of transduction and affordance. The section that follows will draw a historical comparison to map the Situationist vocabulary onto the activist extensions in order to reveal their critical orientation and ideological roots. This is followed by a critical reconsideration of user agency based on a nuanced understanding of online participatory culture and discourse analysis of user comments on some of the activist extensions. Finally, first-hand insights gained from the two interviews will be integrated with the conceptualization of platform governance to address the mutual relationship between the browser platform and extension developers.
4.1 A Transductive Understanding of Browser Extensions
As software programs, browser extensions are “zipped bundles of HTML, CSS, JavaScript, images, and other files used in the web platform” to be distributed as a single package file or to be published in the extension store of a browser platform (Google Chrome 2018). When we consider interfacial modifications caused by extensions, it is in fact these bundled files of the extension that interact with existing ones of the web page. Therefore, in order to understand how the interface operates to produce particular qualities for users, it is helpful to begin with an object-centered approach that theorizes the behavioral aspects of browser extensions in general.
According to Ash, interfaces are “sets of objects that continually encounter one another and generate particular qualities that are partially dependent on these encounters” (28). This process of selective relating is one of transduction that aims to achieve particular effects through the encounter between technical objects. Following this transductive approach to interfaces, we may consider, for instance, the texts elements of an image caption on a web page as selectively relating to the corresponding image object so as to produce the quality of explanation or clarification. Similarly, we may consider the metrics on social media sites as selectively relating to individual posts so as to convey the quality of activeness or influence. As Ash argues, “understanding all object relations as transductive can account for how non-human processes come to have tangibly human effects on the user of that interface” (31).
25
In order to generate such tangibly human effects for users, browser extensions tap into this transductive logic and create new selective encounters between files on the original web page and those bundled in the extension package. In a sense, the web user interface becomes an enveloping environment where selective encounter takes place. This understanding is also supported by Ash’s conceptualization of the interface as an environment, “made up of the arrangement of objects and the transductions that occur between these objects” (29). In this aspect, activist extensions follow the same operating logic. Grosser’s Demetricators remove the selective encounter between metrics and social media posts, invalidating the transductive process that creates the ostensible quality of activeness or influence; Uncover Harassers adds new selective encounter between celebrity names and their accusations of sexual harassment, facilitating the transductive process that promotes public awareness and supports the #MeToo movement; Make Trump Tweets Eight Again, while not removing previous or adding new objects, tweaks the existing selective encounter between Trump’s tweets and other parts of the Twitter interface, manipulating sensory elements such as color and font to create an alternative transductive process that mocks and trivializes the US president’s social media rhetoric.
The above instances of interface modifications demonstrate how activist extensions can introduce object-based selective encounters that transduce particular qualities other than those intended by the web page’s original creators. It is also worth mentioning that the technical composition of an extension is also important when we consider the specific types of interface modification that can be realized. In particular, HTML (hypertext markup language) is often associated with the structure of the web page, whereas CSS (cascading style sheets) usually manages the web page’s format and style. In other words, the type of technology used to create the extension can determine the type of the resulting interface modification, which can appear in the form of HTML-enabled restructuring (adding or removing web elements) and/or CSS-enabled restyling (changing element color, size, or font). Therefore, a transductive understanding of browser extensions entails the examination of browser interfaces and extensions as composite media objects, whose technical structures form the material basis of their behavioral patterns.
4.2 The Reminder as an Extended Mechanism of Affordance
Understanding the interface as a site of transduction or an environment facilitates the theorization of extension behaviors in general, but to comprehend how the creation of tangibly human effects unfolds in practice, it is necessary to adopt a relational, rather than object-centered, approach that
26
takes into account how users perceive and respond to these intended human effects. The notion of affordance provides one such perspective.
Since activist extensions seek to disrupt the intended functionality of the web page, Hartson’s classification of affordance is particularly helpful as it dissects this concept based on a design feature’s intended use. As a reminder, Hartson’s four affordance types are cognitive, physical, sensory, and functional, which, respectively, helps users know, perform physical actions, sense, and accomplish work (323). Based on this classification, the intended use of a web page ultimately depends on the delivery of its functional affordance, which in turn is contingent on the proper coordination between its sensory and cognitive affordances. Again, physical affordance is negligible here, as browser extensions operate only on virtual user interfaces. Therefore, by subverting the supporting role played by cognitive and sensory affordances, activist extensions eventually disrupt the delivery of the web page’s intended functional affordance.
To demonstrate how this disruptive process plays out, we may consider Grosser’s Facebook-based extensions. First, Facebook Demetricator removes all the metrics on the Facebook interface as part of its cognitive affordance. While a demetricated Facebook would still allow users to “accomplish work,” an essential part of the Facebook function that is designed to engross its users is stripped away. Then, Textbook (Figure 6) pushes the separation of cognitive and sensory affordances one step further, removing all the images on the Facebook interface. While a text-only Facebook would still fulfil some of its cognitive function, image-based sensory elements that most effectively capture user attention are now hidden. Finally, Safebook (Figure 7) pushes this logic to the extreme by completely removing all the content on Facebook, including image, text, video, and audio, leaving behind only “the empty containers that frame our everyday experience of social media” (Grosser 2018). With most of Facebook’s sensory and cognitive elements now eliminated, the content-less interface now fails to deliver any significant function, although, as Grosser points out, one may still be able to navigate this abstract space and identify the meaning of all those blue boxes and red circles simply by memory (2008). Comparing the three cases helps us understand how activist extensions can disrupt the process of users sensing and knowing something in order to ultimately prevent any work from being accomplished on the interface level. In other words, the three Facebook-based activist extensions, operating in accordance with the same logic of separation, demonstrate that tweaking sensory and cognitive elements of the user interface can eventually result in the subversion of a web page’s intended functional affordance.
27
Figure 6. Textbook (Grosser 2017).
Figure 7. Safebook (Grosser 2018).
The same analytical process can also be applied to other activist extensions. For instance, Google Alarm and Dark Side of the Prism both turn everyday web browsing into an auditory reminder of issues such as network surveillance and data privacy through the addition of a sound effect or soundtrack. Similarly, Coincidence Detector and Uncover Harassers both combine the effect of “knowing” (that someone is Jew or has been accused of sexual harassment) with that of “sensing” (a linguistic or visual signal that facilitates such a process of knowing). In the above cases, the manipulated user interface points user attention towards an existing social discourse (surveillance and privacy, sexual harassment allegations against celebrities, or anti-Semitic conspiracy theories)
