First Monday

Inventing the dark Web: Criminalization of privacy and the apocalyptic turn in the imaginary of the Web by Thais Sarda, Simone Natale, and John Downey



Abstract
This paper examines how the deep Web, i.e., Web sites that are not indexed and thus are not accessible through Web search engines, was described and represented in British newspapers. Through an extensive content analysis conducted on 833 articles about the deep Web published between 2001 and 2017 by six British newspapers, we demonstrate that these technologies were predominantly associated with crime, crypto markets and immoral content, while positive uses of this technology, such as protecting privacy and freedom of speech, were largely disregarded. The consistent association by the British press between the deep Web and criminal and antisocial behaviors is exemplary of a recent “apocalyptic turn” in the imaginary of the Web, whereby Web-related technologies are perceived and portrayed in more negative ways within the public sphere. We argue that the use of such negative concepts, definitions and associations engender distrust about uses of the deep Web, propagating user stereotypes that reflect what we argue to be an overall criminalization of privacy.

Contents

Introduction
Struggles of meaning: New media, rhetorical inventions and the imaginary
Methodology
Constructing the dark Web in the British press
Hidden, encrypted, secretive: Describing the “dark” Web
Conclusion: Darkness and the power of metaphors

 


 

Introduction

Since the early years of the Internet, metaphors such as the highway, the frontier, the library, the archive and the cloud have been mobilized by various social actors to convey what the Web is and what it could be used for (Wyatt, 2021). Rather than being “mere” concepts, each of these metaphors alluded to specific dimensions of the Internet — the frontier, for instance, suggested its libertarian potential, while the idea of the highway implied that the Web was a suitable object for state intervention. Research has also shown that such discursive constructions have significant impact on public debates and ultimately on the governance of the Internet (Crawford, 2007; Mansell, 2012). Following the rhetorical inventions and diffusions of concepts and metaphors related to the Web helps shed light on wider shifts in representations and perceptions of online media that inform wider governmental approaches (e.g., HM Government, 2019).

Research in Internet studies and digital media studies has increasingly acknowledged the importance of metaphors and other forms of representation (Mager and Katzenbach, 2021). However, the study of how different concepts are adopted and used in regard with the Web is still characterized by two significant limitations. First, many existing studies draw on anecdotal evidence, while studies drawing on the analysis of significant bodies of data to explore representations of the Web are still relatively uncommon (Bory, 2020). Second, there is a need to consider not only wide concepts and metaphors that become associated with the Web (Hu, 2015) but also how narrower areas and more specific uses of the Web are represented and discussed. Mützel (2021), for instance, recently unveiled how the emergence of digital payments technologies and practices was accompanied by a range of sociotechnical imaginaries and metaphors that contributed to the banking industry’s move toward becoming payment platforms.

Tackling both these limitations, this paper draws on extensive primary research to unveil an important element in the representation of the Web in an age in which online surveillance has become ubiquitous and it is increasingly difficult to ensure privacy for Web users (Bauman and Lyon, 2013). Through empirical research conducted over a wide sample of newspaper articles, we investigated the recent emergence of a range of concepts used to describe Web sites that are not indexed and thus accessible through search engines such as Google, as well as the software technologies that provide access to them. Drawing on a content analysis conducted over a sample of 833 newspaper articles published in the United Kingdom between 2001 and 2017, the article examines how different denominations and attributes, such as the “dark Web,” have emerged across time to identify and designate a range of technologies and phenomena that have become increasingly prominent to escape surveillance in the Web. The results of this extensive empirical research show that within this timeframe, news media developed different and competing terminologies and attributes to label and describes such technical and social phenomena [1].

The emergence of concepts such as the “dark Web” and the “deep Web” represents a case of a “rhetorical invention” (Simonson, 2014) — a sociocultural process in which novelties are constructed by language and connected to a specific discourse and ideology. Such discursive constructions are never neutral, each carrying with them a range of connotations and meanings that contribute to shape wider media imaginaries and representations (Sturken, et al., 2004). Overall, our analysis shows that British newspapers’ uses of these concepts encapsulated specific visions of the relationships between surveillance and privacy on the Web, consistently linking privacy-enabling technologies to criminal activities and behaviors. Much like the sociotechnical imaginaries of electronic payment underpinned the equivalence of data and money (Mützel, 2021), news media’s representation of the “dark” Web brought forward a vision that associated the use of tools to evade surveillance in the Web with malicious endeavors. We argue that such a vision signals what we propose to call a criminalization of privacy that is affecting on how privacy and surveillance on the Web are framed and publicly discussed in the public sphere.

Beside discussions on the representations and the imaginary of the Web, surveillance and privacy, this article also contributes to existing studies on the “deep Web” — a term used here since it is widely considered an appropriate definition by experts and scholars (Bergman, 2001; Shestakov, 2008) — and to ongoing debates on this subject (Bradshaw and DeNardis, 2018; Davies, 2020; Gehl, 2016; Jardine, 2018a, 2018b; Lindner and Xiao, 2020). Despite being a recent phenomenon, the deep Web has already attracted considerable scholarly attention. Yet while the technical, material and social dimensions of the deep Web have been explored by a growing number of studies (Bellare and Rogaway, 2005; Devine and Egger-Sider, 2014; Dingledine, et al., 2004; Jardine, 2018a; McCoy, et al., 2008; Moore and Rid, 2016), the discursive dimension of this phenomenon has not yet been explored in depth. Although Gehl (2016) states that ‘most journalistic work on the dark Web presents it as composed of illicit activities in need of policing,’ empirical research is still lacking to provide evidence for this claim and to enable more sophisticated analysis and contextualization of the patterns of representation and perceptions surrounding the deep Web. By filling this gap, this paper corroborates widespread concerns about the demonization of deep Web-related technologies and expands insights into how technologies, activities and users of the deep Web are portrayed by the mainstream media, and specifically by the press.

 

++++++++++

Struggles of meaning: New media, rhetorical inventions and the imaginary

As noted by Gitelman and Pingree [2], when a new medium emerges, it often goes through a phase of ‘identity crisis,’ which is overcome by adapting technology and its uses to public understanding. One of the ways in which this crisis becomes manifest is uncertainty about definitions and names. New media are often described by making use of existing categories: what today is generally called radio, for instance, was initially described as wireless telegraphy, taking up the name of a medium that already existed, i.e., the telegraph (Balbi, 2015). Terms and definitions, however, are also informed by wider cultural meanings, as Simonson (2014) has insightfully shown about the emergence of the concept of “mass” medium. The rhetorical invention of specific terms and attributes, Simonson argued, helped anchor the novelty of technologies and social phenomena to existing discursive and ideological formations, and in turn shaped wider understanding and approaches to technology.

Similarly, the tradition of conceptual history research has shown that representation is connected to concepts, i.e., names for things that need to be explained or codified to make sense in the world (Loocke, 1998) and signify general knowledge about experiences in the world (Bolton, 1977), which can evolve and change over time (Koselleck, 2002). Conceptual history allows us to comprehend lived experiences and how historical change link to the emergence and reworking of concepts (Müller, 2014); these are intrinsically connected to interpretations (Kelley, 1996) and to ideological dimensions (Melton, 1996). As mentioned earlier, representations and definitions of media are never neutral. The general discourse about technology in the press and other media is often shaped through hopes and fears related to societal, economic and political aspects (Sturken, et al., 2004). The case of the Web provides an apt example of how representations include hope surrounding positive uses, and the fear of negative ones (Malbreil, 2007; Mosco, 2004).

Even if the history of the Web is relatively short, in the about 30 years since the invention of the Web the imaginary surrounding this medium has undergone very significant changes, moving from a quasi-unanimous enthusiasm to a range of representations and assessments that underline not only the benefits but also, and often especially, the problems of online practices and technologies (Bory, 2020). In the 1990s and 2000s, the emergence of the Web stimulated enthusiastic, techno-utopian claims envisioning new opportunities in areas such as access to information, direct democracy and individual freedom (Flichy, 2007). While potential risks and problems related to the Web were identified relatively early (e.g., Brunton, 2013), in its early history the Web was represented and perceived in predominantly positive tones (Mosco, 2004; Streeter, 2011). Yet increasingly in the last decade, public perception of Web-related technologies underwent a significant shift, opening to the emergence of dystopian assessments of the Web’s role, from both left- and right-wing perspectives. From the ubiquity of online surveillance (Bauman and Lyon, 2013) to the dominance of a few huge corporations (Vaidhyanathan, 2011), from the concerns over social media’s impact on disinformation and fake news (Creech, 2020) to preoccupations about the psychological and cognitive consequences of permanent connectivity (Turkle, 2011), from the criminalization of online anonymity to alarms about online spam and phishing (Brunton, 2013), a veritable “apocalyptic turn” has changed the way in which the Web as a technology and a social phenomenon was perceived and portrayed within the public sphere (Balbi, 2018). Today, an unprecedented range of negative representations characterize the “imaginary” of the Web, i.e., the discursive dimension of this medium.

In fact, the meaning of digital technologies is often subject to significant shifts, as shown by the example of the debate on the so-called filter bubbles. When algorithmic personalization of the Web was becoming more and more sophisticated, many scholars worried about the risk that search engines and social media would deliver only content that agrees to each user’s personal opinions and believes (Pariser, 2011). There was a fear that these filter bubbles would promote radicalization and intolerance, undermining social progress and democracy. Since then, the power of filter bubbles has been consistently refuted by empirical research and criticized in academic environments for its technological determinism as well as irrelevance (Bruns, 2019). Dahlgren (2021) suggests that filter bubbles are not as influential as once imagined; overestimating them means simplifying human behavior and interactions, since new forms of online communication are more likely to expose people to multiple networks, creating a routine of confrontation with distinct views. Reflecting about technological innovations outside of the Western logic, Makhortykh and Wijermars (2021) even conclude that algorithmic personalization can provide more pluralistic discussion environments in countries with non-democratic media systems and limited press freedom. This demonstrates that research about the meaning and impact of filter bubbles has changed significantly in about 10 years: from the fear of destroying democracy (Pariser, 2011) to the hope of encouraging countries to reach a democratic status (Makhortykh and Wijermars, 2021).

We argued that the emergence of the concept of the “dark Web” also represents an extraordinary case in point to study and better understand how public discourses about the Web have shifted in recent years. The dark Web is one among a range of denominations usually employed to describe Web sites that are not indexed and thus not accessible through Web search engines such as Google. Before examining the data collected through research, it is useful to consider how the use of concepts such as deep Web and dark Web emerged and developed beyond the British press. In academia, the first concept widely employed about these technologies was the “invisible Web” (Ellsworth and Ellsworth, 1994), defining the entire body of Web content outside traditional platforms. This term is still occasionally used (e.g., Devine and Egger-Sider, 2014), although it is criticized for being overly generic (Shestakov, 2008) or inaccurate (Bergman, 2001). Since the early 2000s, the “deep Web” concept (Bergman, 2001) has emerged, opposing the idea of an “invisible Web,” since this content is hidden but still visible. In addition, terms such as “hidden Web” and “deep Web” are considered interchangeable, in that using one or the other depends on preference (Shestakov, 2008).

A similar dynamic is observed on Google Trends, a service provided by Google that allows people to collect data about searches regarding terms, locations and dates. Examining the searches on the topic since 2004 in the United Kingdom (Figure 1), there was a growing interest in the “deep Web” [3] until August 2015, when searches fell consistently. Although with consistent occurrences over time, the term “invisible Web” [4] peaked in March 2004 and became less relevant after November 2006. “hidden Web” [5] followed the same trend.

 

Frequency of searches on Google between 2004 and 2017
 
Figure 1: Frequency of searches on Google between 2004 and 2017 (U.K.).

 

Searches for the term “Dark Web” [6] have consistently grown since February 2012, becoming the most searched term from the beginning of 2016. This shows a pattern within social imaginaries that helps create meaning and sense around technological innovations (Mansell, 2012): from the deep to the dark Web. Using one or the other concept is part of a chain that moves from the wider imaginary in relation to technology to underpin specific interpretations and uses that remain embedded in the technology’s social and cultural construction. Although terms such as “deep Web” and “invisible Web” have dominated academic discussion, the increasing attention of popular media towards privacy-enabling technologies has led to the use of a different term: “dark Web.” As a matter of fact, 2015 was an important year for this concept, since it was shortlisted for “word of the year” by Oxford Dictionaries, although the emoji known as “face with tears of joy” [7] gained the title in the end. As the data discussed in this paper will show, “dark Web” has been the most used concept in the newspapers under examination. While existing research has shown that users protect their identities for multiple reasons (Jardine, 2018a), our research demonstrates that the British press has highlighted mainly negative appropriations of the deep Web.

 

++++++++++

Methodology

This research focuses on the analysis of newspaper content to understand how the British press represent deep Web technologies. Print media in general, and specifically newspapers, are a relevant source of information for adults in the context of the United Kingdom (Ofcom, 2019). While news media are only one of the many actors that contribute to shape discourses and imaginaries about the Web and digital technologies, news media framing of the deep Web provides a useful entry point to understand how it was represented in the public sphere. Moreover, the analysis helps unveil how news about the deep Web were related to news about crime, which might encourage personal and shared senses of fear and security (Banks, 2005). Therefore, analyzing the news media portrayal of the deep Web offers important insights on the circulation and development of social meanings about these technologies.

As argued by Chadwick, et al. [8], British newspapers are generally divided between “quality” and “tabloid”, so to pursue a more comprehensive view of the British print media, this research encompasses both types of newspapers in the analysis. Much existing research has shown that British tabloid newspapers tend to frame everyday life situations in a sensational way, focusing on extreme social experiences and behaviors as well as folk stories through rhetorical patterns such as dramatization and exaggeration (Conboy, 2006). Meanwhile, quality newspapers — also called broadsheets — have a more balanced approach. Taking up the two classes of newspapers in the sample, therefore, was important to ensure that an appropriately sampling, able to represent the overall British media, was operated.

Beside the distinction between tabloid and quality, the selection of newspapers for this analysis considers two additional points to assure the representativeness of the sample and a broader scenario of the U.K. print media: daily reach (PAMCo, 2019) and political affiliation (Wring and Deacon, 2010). After identifying the top 10 newspapers in the U.K. according to daily reach (PAMCo, 2019), we disregarded both newspapers that are currently freesheet, i.e., free of charge, namely Metro and Evening Standard. In addition, as this list has only three quality newspapers, we also selected three tabloid newspapers, to have a final sample of six newspapers (Table 1).

 

Table 1: Selected UK newspapers by political affiliation, nature and daily reach (June 2019). Source: PAMCo, 2019; Wring and Deacon, 2010.
NewspaperDaily reachNature
The Sun2,818,000Tabloid
Daily Mail2,675,000Tabloid
The Times1,113,000Quality
Daily Mirror1,032,000Tabloid
Daily Telegraph864,000Quality
The Guardian653,000Quality

 

In addition to the inclusion of both tabloid and quality newspapers, the sample includes diverse political and ideological views, although there is a general predominance of conservative newspapers, as they are the most read in the country (Wring and Deacon, 2010). Regarding daily reach, this sample includes newspapers that vary from 2,675,000 (Daily Mail) to 653,000 (The Guardian).

A preliminary survey helped uncover a range of terms usually related to deep Web technologies among printed articles published by the selected newspapers and led to finding the keywords listed in Table 2 to be included in the data collection. This survey was made through an exploratory investigation of newspaper articles and academic research to identify the most common denominations used to define the deep Web. The research then included all these terms in the data collection. These 27 keywords were each found at least once referring to deep Web technologies.

 

Table 2: Keywords selected for data collection.
Keywords
“dark internet”
“dark net”
“dark side of the internet”
“dark side of the web”
“dark web”
“darknet”
“darkweb”
“deep internet”
“deep web”
“hidden internet”
“hidden web”
“internet dark side”
“invisible internet”
“invisible web”
“non-indexable internet”
“non-indexable web”
“silk road”
“silkroad”
“tor browser”
“tor network”
“tor system”
“undernet”
“underweb”
“underworld of the internet”
“underworld of the web”
“web dark side”
“web underworld”

 

Regarding the time frame for data collection, this research includes all articles mentioning one or more of these keywords following the first appearance of articles up to 31 December 2017. The starting date selected for this study was 1 January 1989, symbolizing the year in which the Web was invented; in practice, though, the first articles about the deep Web technologies date from the 2000s, as shown in the analysis which follows. Data collection was enabled through the Lexis-Nexis Database. Using Lexis-Nexis can create methodological issues concerning information validity and reliability; as noted by Deacon [9], however, the database is a compelling resource if adopting precautions such as ‘checking for “false positives” and duplicated items, scanning the titles and periods sampled for any high-level omissions in data, and checking items for inconsistent unitization.’

The first phase of data collection included 25 terms used interchangeably to define the deep Web and selecting the entire period of this research, from 1989 to 2017 (see Table 3).

 

Table 3: Parameters of search on Lexis-Nexis (Phase 1).
Time frameKeywords
1 January 1989 to 31 December 2017“dark internet” or “dark net” or “dark side of the internet” or “dark side of the web” or “dark web” or “darknet” or “darkweb” or “deep internet” or “deep web” or “hidden internet” or “hidden web” or “internet dark side” or “invisible internet” or “invisible web” or “non-indexable internet” or “non-indexable web” or “tor browser” or “tor network” or “tor system” or “undernet” or “underweb” or “underworld of the internet” or “underworld of the web” or “web dark side” or “web underworld”

 

In a second phase, an additional search was conducted by employing keywords related to the crypto-market Silk Road. Considered the most famous market on the Tor Network, Silk Road was launched in February 2011 and ran on Onion Services for over two years, until its closure by the FBI in October 2013 (Aldridge and Décary-Hétu, 2014). The search for the two remaining keywords — “silk road” and “silkroad” — applied only to the period between 1 January 2011 and 31 December 2017 (see Table 5). This division into two phases proved itself helpful in reducing the number of false positives, since Silk Road is also the name of complex terrestrial and maritime routes crossing Asia and Europe, used for trade in ancient times.

After these two phases, the data collection sample reached 2,948 articles (see Table 4). After collecting the newspaper articles, the following step was to disregard duplicates and false positives. Data for coding were considerably reduced to 833 articles from the initial 2,948 articles, thereby disregarding 2,115 duplicates and false positives. This process was done manually through a complete reading of each article before disregarding duplicates and false positives. In addition, duplicates were usually identified by the SPSS software during the completion of the coding, as the system flags cases with the exact same coding. False positives were quite evidently related to a different subject, for instance, a recurrent false positive was the expression “deep web of lies”, which had no relation to technology, but it was applied to describe plots of movies and even some political situations.

 

Table 4: Data collection on Nexis by newspaper.
NewspaperNumber of articles
Phase 1Phase 2Total
Daily Mail16341204
Daily Mirror20554259
Daily Telegraph190234424
The Guardian569313882
The Times544327871
The Sun26147308
Total1,9321,0162,948

 

Beside the different denominations used to define the deep Web, the content analysis also identified a range of attributes through which the deep Web was described in the corpus of newspapers. In terms of methodology, we collected the first term or adjective used in the definition of these technologies delivered by the article. Among the data, there were attributes highlighting distinct aspects of these technologies: for instance, “anonymous” focused on the fact that users can hide their identities, “encrypted” was connected to the technical dimension and the layers of protection against online surveillance, while “secretive” connoted the mystery that surrounds these systems; and so on.

Content analysis was selected as the methodology since it allows working on a wide range of sources through quantitative methods (Riffe, et al., 2005), thus providing a comprehensive assessment of the representations of deep Web-related technologies. A codebook was developed and tested through an intercoder reliability test, which involved two independent coders applying all variables of the codebook to 10 randomly chosen articles, with a total of 440 codes. This test demonstrated an agreement of 89 percent, with coefficients Krippendorff’s Alpha, Scott’s Pi and Cohen’s Kappa calculated as .0876. Coding was then implemented manually on the full corpus by using the software IBM SPSS Statistics 25. The codebook comprises 44 variables: six are identification variables such as date of publication, newspaper and type of article; and the remaining are content variables that identify, for instance, which term was used, if the term was present in the headline, if there was mention to a criminal or antisocial behavior in the headline, what sources were included, and so on. Please see the Appendix for the complete codebook. The process of coding was time-consuming, as initially expected, because of the high number of newspaper articles: 833 in total. Content analysis requires very thorough work, which in this case was conducted manually; reading and coding one by one each of the 833 articles, with an average of 579 words each, or a total of 482,448 words.

 

++++++++++

Constructing the dark Web in the British press

Contemplating the relevance of metaphors to indicate how technologies are represented, the most common metaphor associated with these technologies in the selected newspaper was “dark.” This attribute was used in 77.3 percent of the headlines, when considering 159 headlines from the six newspapers that nominally mentioned these technologies and in 91.6 percent of the articles, from a total of 833. Variations of this metaphor include the terms “dark Web,” “darknet,” “dark side” and “dark Internet.”

As studies in journalism show, headlines aim at giving an overall picture of the content and at attracting readers, which leads sometimes to questionable uses of attention-getting words (Reah, 2002). In fact, headlines have the persuasive function of not only establishing the general view of a newspaper about a topic, but also manipulating and influencing opinions through the choice of words and which aspects are emphasized or not (Reah, 2002). The preference for using the attribute “dark” was evident: in a total of 159 cases in which newspapers used one of the concepts in headlines, this attribute was chosen 123 times (Figure 2).

 

Frequency of terms used in headlines
 
Figure 2: Frequency of terms used in headlines (total).

 

Considering the use of these concepts over time, the establishment of a trend in headlines was also clear. While the research included articles from 2001 to 2017, relevant articles were rare at the start of this timeframe, and only from 2013 did news on the topic see a substantial increase in numbers. Considering that articles about this technology were not frequent in news media, the term “deep Web” occurred more often than any other in 2012 (Figure 3). After 2013, however, “deep Web” was never mentioned again in headlines, so the lifetime of this concept was brief in the newspaper context. Although the term “darknet” was present in headlines from 2013, consistently growing until 2015 and declining thereafter, it also was never largely adopted by the British press.

 

Frequency of terms used in headlines over time
 
Figure 3: Frequency of terms used in headlines over time (total).

 

The choice to use the term “dark Web” in newspaper headlines was particularly significant if we consider that in other kinds of publications, such as academic research, competing concepts such as the “deep Web” had been dominant across the same period. Since 2012, the number of occurrences had consistently grown every year (2016 looks like an exception at first sight, but it was a year in which the overall number of articles about this topic was lower). In fact, “dark Web” or “darknet” were the preferred terms, but in 2017 the term “dark Web” established clear dominance, as it was used 45 times out of 47 cases. Considering that newspapers propose an understanding of the news through headlines (Bignell, 2002), and the important role of metaphors in defining and directing meaning (Lakoff and Johnson, 1980), British print media were framing this technology through a pejorative — dark, with no light — meaning.

According to Wyatt [10], “metaphors describe one thing in terms of another, because they help to describe something novel or for poetic and rhetorical effect”. As Osborn (1967) has influentially shown, the concepts of “light” and “dark” represent two key archetypal metaphors that are recurringly called upon to distinguish positive and negative associations. Light is not only associated to visibility, by which humans can escape danger and take control of the environment, but also to the warmth and the nutrition provided by the Sun. In contrast, darkness evokes “fear of the unknown, discouraging sight, making one ignorant of [their] environment” as well as coldness, “suggesting stagnations and thoughts of the grave” [11]. The extent to which this applies to contemporary discourses can be evidenced in the sharp negative tones of the dark metaphor evidenced in areas such as politics, film and foreign policy (e.g., Jarosz, 1992; Halverson, 2003; Forceville and Renckens, 2013). Finally, terms such as “light” and “dark” also present a layer of racism that contributes to the already supremacist logic identified in the overall conceptualization of digital technologies, as argued by Katz (2020). Katz analyzed ideologies in the development of artificial intelligence to conclude that there were intrinsic connections to ideas not only of white superiority but also male dominance and Global North hegemony.

To further investigate how rhetorical inventions were advanced in the researched newspapers, we also compared the use of attributes mentioned in headlines, grouping distinct terms that recurred in relation to the same metaphor (as seen in Table 5).

 

Table 5: Concepts used to define the Deep Web, based on attributes.
AttributeConcept
DarkDark Internet
Darknet
Dark Web
Dark side
DeepDeep Internet
Deep Net
Deep Web
UnderUndernet
Underweb
Underworld
HiddenHidden Internet
Hidden Web
InvisibleInvisible Internet
Invisible Web

 

Including the variations “dark Internet,” “darknet,” “dark Web” and “dark side,” the attribute “dark” is responsible for 98.5 percent of the cases in the tabloids (Figure 4). The attribute “deep” was used only once by tabloids in headlines, a total of 1.5 percent. The case is the article “Drugs, guns, assassins, jet planes ... All for sale on secret Deep Web,” published by Daily Mirror in September 2012. Although the use of the attribute “deep” instead of “dark” could possibly indicate a more neutral approach to the topic, the article in question can be summarized as a strong statement against the deep Web, which reduces all content outside of traditional search engines and browsers to crypto markets trading drugs, documents and child pornography, among other illicit uses (Palmer, 2012).

 

Attributes used in headlines (tabloid)
 
Figure 4: Attributes used in headlines (tabloid).

 

In quality newspapers (Figure 5), although there was more variety in relation to attributes — including those shown in Table 5 — the same preference for “dark” seen in tabloids was also underlined by its occurrence in 87.5 percent of the headlines. This means that the attribute “dark” was used 56 times and the others, combined, only eight times.

 

Attributes used in headlines (quality)
 
Figure 5: Attributes used in headlines (quality).

 

The use of the attribute “invisible” in quality newspaper headlines, nevertheless, was rare: it happened only once and in the very early stages of media addressing this topic in an article published in The Guardian in September 2001. Entitled “Search for the invisible web: There are more websites than those seen with the naked eye”, Sherman (2001) explained how thousands of databases were hidden from mainstream search engines. The attribute “hidden” was also rarely applied by quality newspapers; an example of this use appeared in The Times (2017). The same article applied the term “darknet” in the text.

Although there was a strong preference for associating these systems with the sinister idea the attribute “dark” connotes in headlines of both tabloids and quality newspapers, the use of concepts in the text of the article demonstrated a more nuanced representation of these systems. Although the articles also preferred the term “dark Web,” in 58.0 percent of the cases, many terms were included in the text but ignored in headlines (Figure 6), i.e., newspapers use 18 different concepts in the text, compared to 10 in the headlines. Some of these concepts were mentioned only once in the text, though. An example was “undernet,” used by The Guardian in October 2008 when comparing the surface and deep Web: “A trusted, controlled ‘overnet’ for commercial and business use, and an ‘undernet’ where anything goes” (Schofield, 2008). It was also the case for “underweb,” which was applied also by The Guardian in November 2002: “Other sites, part of what Leach calls the ‘underweb,’ will go to great lengths to ensure they are not found by uninvited guest” (Clayton, 2002).

 

Frequency of terms used in the text (total)
 
Figure 6: Frequency of terms used in the text (total).

 

Considering the use of attributes (as seen in Table 5), the pattern was consistent with the headlines, with the choice of the attribute “dark” in 91.6 percent of the cases (see Figure 7).

 

Frequency of terms used in the text by attribute (total)
 
Figure 7: Frequency of terms used in the text by attribute (total).

 

The preference for the concepts “dark Web” and “darknet” in the text of the articles was clear, as seen in Figures 8 and 9, for tabloid and quality newspapers.

 

Frequency of terms used in the text (tabloid)
 
Figure 8: Frequency of terms used in the text (tabloid).

 

 

Frequency of terms used in the text (quality)
 
Figure 9: Frequency of terms used in the text (quality).

 

The third most used term in the articles, with a total of 39 appearances, “Silk Road,” deserves special discussion. First, “Silk Road” in this case related to a crypto market mainly known for commerce in drugs, and it was a metaphor related to the actual Silk Road [12], the network of routes connecting Asia and Europe that operated thousands of years ago. This road was developed over two thousand years ago for the purpose of trading Chinese silk, but it was also used for the exchange of other goods and even knowledge and ideas among Asian and European cultures over time. In the case of the crypto market, users from distinct countries were connected, thereby allowing illegal substances to cross borders and circulate between continents through the postal service. Second in this regard, a metaphor was commonly used by newspapers to define this Web site ny calling it an “eBay for drugs,” in reference to the e-commerce which facilitates consumer-to-consumer and business-to-consumer trades, or a variation “Amazon.com for drugs,” also in reference to an e-commerce site (Ahmed, 2012). Aldridge and Décary-Hétu (2014) pointed out that comparing it to eBay or Amazon was inaccurate: after an extant analysis of transactions, it was concluded that most of the trade on Silk Road was business-to-business, with a recognized flow from big drug dealers to local street sellers.

In summary, the findings show that although newspapers in the United Kingdom had different and positive alternatives available, they mostly adopted terms that were conceptually related to the negative uses of privacy-enhancing technologies, with little difference between tabloids and quality newspapers. As a sharp example of this, the term “dark Web” was the most common choice for journalists discussing the phenomenon over time. This suggests a sharply negative representation of these technologies in the British press, a finding that was corroborated by an analysis of secondary attributes, carried out in the next section.

 

++++++++++

Hidden, encrypted, secretive: Describing the “dark” Web

Considering the multiple word associations used to describe the deep Web, Table 6 presents a list of the most frequent attributes, i.e., those with more than two occurrences, used by newspapers in their definitions of these technologies.

 

Table 6: Occurrences of attributes associated with the deep Web by newspaper.
AttributeNewspaperTotal
Daily MailDaily MirrorThe GuardianDaily TelegraphThe SumThe Times
Hidden13129372165
Encrypted132477740
Secretive1065110436
Anonymous41603923
Specialized01242918
Underground03513618
Black market06322417
Shadowy32117216
Illegal33204214
Criminal11400612
Illicit41201210
Notorious4012119
Sinister0111339
Untraceable3120039
Invisible1101137
Sophisticated0111137
Lawless0111126
Unregulated4000026
Harder2001014
Murky2001104
Inaccessible2000013
Restricted1010013
Secure0020013
Unpoliced1001013

 

The most common attribute associated with these technologies and used by newspapers was the term “hidden,” with 65 occurrences. The first time that this attribute was applied was by The Guardian in November 2009. In the article, there was a script to access these systems, with the aim to “Enter a previously hidden online world” (Beckett, 2009). Since then, this attribute has been commonly used: in December 2017, for instance, the Daily Mirror described them as “a hidden part of the internet used by criminals” (Daily Mirror, 2017). Interestingly, these examples point to different perspectives of the same attribute: in one case, “hidden” was used to attract curiosity about content; it could also be translated as a hideout for felons. Considering each newspaper separately, the most used attributes were “hidden,” with 21 cases in The Times, 12 cases in the Daily Mirror and nine cases in The Guardian; “hidden” and “encrypted,” both with 13 cases each, by the Daily Mail; “encrypted” with seven cases in the Daily Telegraph and “secretive” with 10 cases by The Sun. Although these attributes highlight distinct dimensions, they all contributed to the general idea of the deep Web as something unknown, secret and inaccessible, giving opacity to these technologies. There are multiple forms of accrediting a dimension of opacity to technologies (Burrell, 2016), such as stressing an intentional element secrecy in technological development, highlighting digital illiteracy and therefore a lack of knowledge about how a given technology works. The fact that the British press favored the opacity of the deep Web becomes clearer when grouping the previously mentioned attributes by topic (Table 7, which also included attributes with only one or two occurrences).

 

Table 7: Attributes used by newspapers by topic.
TopicAttributes
AnonymityAnonymous 
TradeBazaarBlack market 
TechnologyComplex
Encrypted
Inaccessible
Not indexed
Restricted
Route
Safe
Sophisticated
Specialized
LegalityIllegal
Illicit
Lawless
Unpoliced
Unregulated
RelevanceComprehensive
Interesting
New phenomenon
Notorious
Popular
Resilient
Treasure
SizeBooming
Huge
Large
Major
Vast
CriminalityAbnormal
Chilling
Criminal
Dodgy
Drug dealing
Evil
Grey
Grim
Infamous
Nasty
Noxious
Sick
Ugly
Vile
SecrecyBuried
Chaotic
Dark
Harder
Hidden
Invisible
Murky
Mysterious
Not viewable
Parallel
Secretive
Shadowy
Sinister
Uncharted
Underground
Underside
Unlisted
Wonderland
Counter surveillanceCovert
Difficult
Free
Private
Refuge
Secure
Shield
Solution
Under the radar
Undetectable
Unreachable
Untraceable

 

First, the table shows that most of the attributes used in these cases were neutral — for instance, “anonymous” and “restricted” — or negative — in the cases of “criminal” and “sinister.” In opposition, rare examples of positive ones occurred, such as “solution,” used by The Guardian (Krotoski, 2012). Second, when looking at the attributes organized by topic (Figure 10), words connected with an idea of secrecy, and therefore attributing opacity to the technology (Burrell, 2016), were in the majority in British newspapers, 42.2 percent of articles. Associating a dimension of mystery to the deep Web, newspapers not only instigated the curiosity of readers, but they also shaped their perceptions of these technologies through images of the supernatural or extraordinary.

 

Attributes used in newspaper articles by nature (total)
 
Figure 10: Attributes used in newspaper articles by nature (total).

 

Considering that the main reason for tools such as Tor — open-source software enabling anonymous communication online — is to provide privacy, newspapers highlighted the relevance of these technologies in avoiding surveillance only in 6.9 percent of cases, using attributes such as “untraceable” and “undetected.” Yet, pointing to the problem of surveillance does not imply that privacy-enhancing technologies were represented positively. When a newspaper focuses on the fact that the users cannot be traced while using Tor, there is a preoccupation related to how it may be used by criminals to avoid persecution. The attribute “untraceable,” for instance, was so used by The Times in September 2014: “Miss Patel had bought abrin, a controlled substance under the Terrorism Act, from an illegal website called Black Market Reloaded (BMR) in the US using untraceable software and £900 of bitcoins” (Keate, 2014). This example shows how the idea of being beyond authority»s radar was commonly associated with crime.

In summary, on the one hand, newspaper articles consistently repeated attributes that were related to neutral or negative aspects of the deep Web, with little space for positive perspectives. And on the other hand, even when the attributes were not negative, they constantly gave opacity to these systems, increasing the distance between the technology which had positive uses (e.g., Hoang and Pishva, 2014; Jardine, 2018a) and regular Internet users.

 

++++++++++

Conclusion: Darkness and the power of metaphors

Conceptualization is a process that involves language, which per se is social information (Spolsky, 1998), and the power of words affects social structures in ways that are transformed over time (Koselleck, 2004). As Lakoff and Johnson (1980) insightfully show, metaphors are not only a common language resource and part of the conceptual system in which we attribute meaning to things in the world, but they are also a pervasive component of everyday realities because they are related to concepts and terms and, on a different level, to thoughts about a specific concept and actions that are taken in reaction to it. Metaphors likewise help to guide the imaginary about a concept, according to Langer [13], for whom “in a genuine metaphor, an image of the literal meaning is our symbol for the figurative meaning, the thing that has no name of its own.”

This can be exemplified by the terms that act as the core of this research: “deep Web” relates to content that extends far from the surface; “dark Web” highlights the part of the Internet that has little or no light upon it; “hidden Web” focuses on what is kept out of sight and so on. The choice of using one or another concept is not neutral but associated with specific interpretations and affects: “deep Web” and “hidden Web” are related to users looking for further content or knowledge, while “dark Web” is linked to illegal uses (Gehl and McKelvey, 2019). Although Wyatt (2021) recommends that social science scholars researching digital societies should dispense with metaphors and be literal, this logic is in practice ignored. Academic literature at least acknowledges a distinction between the concepts of the dark Web and the deep Web (Bartlett, 2015; Bradbury, 2014) but the bar is much lower in news media. This research shows that these terms are used interchangeably in the British press. Although newspapers struggle to differentiate between these concepts, this research shows that, overall, these technologies are framed in predominantly negative ways.

Overall, the analysis demonstrates that the “deep Web” was largely presented by the British press in a one-dimensional fashion: a secretive tool used for negative (or at least questionable) purposes. Readers of the most popular British newspapers were thus given little or no evidence, depending on which title they read in everyday life, that the deep Web had positive as well as negative uses (Jardine, 2018b; Moore and Rid, 2016). Considering that media representation can be achieved using concepts (Bolton, 1977; Loocke, 1998; Spolsky, 1998) and metaphors (Langer, 1954), the selection of terms used to describe the deep Web helps to understand the meanings that newspapers attribute to these technologies. As the empirical findings of this research demonstrate, British press coverage focused emphatically on negative implications of these technologies and undesirable associations, contributing to a criminalization of privacy-enhancing tools online.

Ultimately, the emergence of the term “dark Web,” a concept that was linked to illegalities and immoralities (Gehl and McKelvey, 2019), could be contextualized within a broader shift in an overall representations of the Web; from a situation in which it was presented as a harbinger of positive change (Mosco, 2004) to the emergence of negative views. While affirmative ideas of the Web were prevalent during the initial narrative (Flichy, 2007; Mosco, 2004; Streeter, 2011), discussions about the dark Web and its overall undesirable uses were part of a recent shift toward highlighting the impact of the Web in negative, even “apocalyptic,” tones (Balbi, 2018).

While the Web has reached a stage of maturity in most developed countries, its representations continue to shift with the emergence of new technical and social phenomena related to the Web as well as of new concepts and attributes describing its changing technical and social landscapes. In this context, the dynamics of representation of privacy-enhancing systems, which this article examines, shows a pattern by which the act of seeking privacy on the Web is associated with criminal endeavors in the British press. One of the implications of the dynamic representation highlighted by our research is that motivations of users for reasons of privacy are consistently questioned and undermined, and institutions and groups which fight for civil rights are silenced in the public discussion in the press and granted little or no significance in the news. Overall, our research shows that reports on the deep Web in the British press contribute to what we describe as a criminalization of privacy by which the pursuit of privacy on the Web is consistently seen in a negative light, with little consideration of the broader surveillance mechanisms on the Web that underpin and, in some cases, may actually justify the use of privacy-enhancing technologies. End of article

 

About the authors

Thais Sardá is a lecturer in the Department of Criminology and Criminal Justice at Nottingham Trent University.
Direct comments to: thais [dot] sarda [at] ntu [dot] ac [dot] uk

Simone Natale is an associate professor at the University of Turin (Italy).
E-mail: simone [dot] natale [at] unito [dot] it

John Downey is Professor of Comparative Media Analysis at Loughborough University.
E-mail: j [dot] w [dot] downey [at] lboro [dot] ac [dot] uk

 

Notes

1. It is important to underline that the technologies and systems to give access to the deep Web are diverse both in technical terms and in terms of their applications and uses. In considering the rhetorical invention of wide concepts such as “deep Web” or “dark Web,” this paper does not intend to argue that these can be seen as a unique field. What this study addresses, in fact, is not much the technology per se but its representation. As we will see, the news media under examination largely presented and narrated privacy-enabling technologies and Web sites accessed through them as part of a distinct phenomenon, conceptualized through different notions such as “dark Web.”

2. Gitelman and Pingree, 2003, p. xii.

3. Available at https://trends.google.com/trends/explore?date=all&geo=GB&lq=%22deep%20web%22, accessed November 2018.

4. Available at https://trends.google.com/trends/explore?date=all&geo=GB&q=%22invisible%20web%22 accessed November 2018.

5. Available at https://trends.google.com/trends/explore?date=all&geo=GB&q=%22hidden%20web%22, accessed November 2018.

6. Available at https://trends.google.com/trends/explore?date=all&geo=GB&q=%22dark%20web%22, accessed November 2018.

7. See https://www.oxforddictionaries.com/press/news/2016/9/2/WOTY, accessed December 2018.

8. Chadwick, et al., 2018, p. 427.

9. Deacon, 2007, p. 29.

10. Wyatt, 2021, p. 411.

11. Osborn, 1967, p. 117.

12. More information at https://en.unesco.org/silkroad/about-silk-roads, accessed 2 November 2022.

13. Langer, 1954, p. 113.

 

References

M. Ahmed, 2012. “It’s like Amazon, but for criminals,” The Times (3 April), p. 40, and at https://www.thetimes.co.uk/article/its-like-amazon-but-for-criminals-buying-illegal-items-online-zn0k8ww36s2, accessed 4 November 2022.

J. Aldridge and D. Décary-Hétu, 2014. “Not an ‘eBay for drugs’: The cryptomarket ‘Silk Road’ as a paradigm shifting criminal innovation,” SSRN (15 May).
doi: http://dx.doi.org/10.2139/ssrn.2436643, accessed 4 November 2022.

G. Balbi, 2018. “La ‘svolta apocalittica’ negli studi sul digitale: introduzione,” Quaderni di Teoria Sociale, volume 1, pp. 11–24, and at https://www.morlacchilibri.com/universitypress/uploads/files/QTS_1_2018/QTS_1_2018.1.Balbi.pdf, accessed 4 November 2022.

G. Balbi, 2015. “Old and new media: Theorizing their relationships in media historiography,” In: S. Kinnebrock, C. Schwarzenegger and T. Birkner (editors). Theorien des Medienwandels. Köln: Halem, pp. 231–249.

M. Banks, 2005. “Spaces of (in)security: Media and fear of crime in a local context,” Crime, Media, Culture, volume 1, number 2, pp. 169–187.
doi: https://doi.org/10.1177/1741659005054020, accessed 4 November 2022.

J. Bartlett, 2015. The dark Net: Inside the digital underworld. London: Melville House.

Z. Bauman and D. Lyon, 2013. Liquid surveillance: A conversation. Cambridge: Polity.

A. Beckett, 2009. “A walk on the darkside: A tiny fraction of everything online can be found via search engine,” The Guardian (26 November), p. 4, and at https://www.theguardian.com/technology/2009/nov/26/dark-side-internet-freenet, accessed 4 November 2022.

M. Bellare and P. Rogaway, 2005. &dquo;Introduction to modern cryptography” (11 May), at https://web.cs.ucdavis.edu/~rogaway/classes/227/spring05/book/main.pdf, accessed 4 November 2022.

M.K. Bergman, 2001. “White paper: The deep Web: Surfacing hidden value,” Journal of Electronic Publishing, volume 7, number 1.
doi: https://dx.doi.org/10.3998/3336451.0007.104, accessed 4 November 2022.

J. Bignell, 2002. Media semiotics: An introduction. Second edition. Manchester: Manchester University Press.

N. Bolton, 1977. Concept formation. Oxford: Pergamon Press.

P. Bory, 2020. The Internet myth: From the Internet imaginary to network ideologies. London: University of Westminster Press.
doi: https://doi.org/10.16997/book48, accessed 4 November 2022.

D. Bradbury, 2014. “Unveiling the dark Web,” Network Security, volume 2014, number 4, pp. 14–17.
doi: https://doi.org/10.1016/S1353-4858(14)70042-X, accessed 4 November 2022.

S. Bradshaw and L. DeNardis, 2018. “The politicization of the Internet’s Domain Name System: Implications for Internet security, universality, and freedom,” New Media & Society, volume 20, number 1, pp. 332–350.
doi: https://doi.org/10.1177/1461444816662932, accessed 4 November 2022.

A. Bruns, 2019. “Filter bubble,” Internet Policy Review, volume 8, number 4.
doi: https://doi.org/10.14763/2019.4.1426, accessed 4 November 2022.

F. Brunton, 2013. Spam: A shadow history of The Internet. Cambridge, Mass.: MIT Press.
doi: https://doi.org/10.7551/mitpress/9384.001.0001, accessed 4 November 2022.

J. Burrell, 2016. “How the machine ‘thinks’: Understanding opacity in machine learning algorithms,” Big Data & Society, volume 3, number 1 (6 January).
doi: https://doi.org/10.1177/2053951715622512, accessed 4 November 2022.

A. Chadwick, C. Vaccari and B. O’Loughlin, 2018. “Do tabloids poison the well of social media? Explaining democratically dysfunctional news sharing,” New Media & Society, volume 20, number 11, pp. 4,255–4,274.
doi: https://doi.org/10.1177/1461444818769689, accessed 4 November 2022.

N. Clayton, 2002. “Shop tactics,” The Guardian (28 November), p. 2, and at https://www.theguardian.com/technology/2002/nov/28/shopping.newmedia, accessed 4 November 2022.

M. Conboy, 2006. Tabloid Britain: Constructing a community through language. London: Routledge.

S.P. Crawford, 2007. &dquo;Internet think,” Journal on Telecommunications and High Technology Law, volume 5, pp. 467–486, and at http://www.jthtl.org/content/articles/V5I2/JTHTLv5i2_Crawford.PDF, accessed 4 November 2022.

B. Creech, 2020. “Fake news and the discursive construction of technology companies’ social power,” Media, Culture & Society, volume 42, number 6, pp. 952–968.
doi: https://doi.org/10.1177/0163443719899801, accessed 4 November 2022.

P.M. Dahlgren, 2021. “A critical review of filter bubbles and a comparison with selective exposure,” Nordicom Review, volume 42, number 1, pp. 15–33.
doi: https://doi.org/10.2478/nor-2021-0002, accessed 4 November 2022.

Daily Mirror, 2017. “Just Eat data crook in dock” (15 December), p. 28.

G. Davies, 2020. “Shining a aight on policing of the dark Web: An analysis of UK investigatory powers,” Journal of Criminal Law, volume 84, number 5, pp. 407–426.
doi: https://doi.org/10.1177/0022018320952557, accessed 4 November 2022.

D. Deacon, 2007. “Yesterday’s papers and today’s technology: Digital newspaper archives and ‘push button’ content analysis,” European Journal of Communication, volume 22, number 1, pp. 5–25.
doi: https://doi.org/10.1177/0267323107073743, accessed 4 November 2022.

J. Devine and F. Egger-Sider, 2014. Going beyond Google again: Strategies for using and teaching the invisible Web. Chicago: Neal-Schuman.

R. Dingledine, N. Mathewson and P. Syverson, 2004. “Tor: The second-generation onion router,” SSYM’04: Proceedings of the 13th conference on USENIX Security Symposium, at https://www.usenix.org/conference/13th-usenix-security-symposium/tor-second-generation-onion-router, accessed 4 November 2022.

J.H. Ellsworth and M.V. Ellsworth, 1994. The Internet business book. New York: Wiley.

P. Flichy, 2007. The Internet imaginaire. Translated by Liz Carey-Libbrecht. Cambridge, Mass.: MIT Press.

C.J. Forceville and T. Renckens, 2013. “The good is light and bad is dark metaphor in feature films,” Metaphor and the Social World, volume 3, number 2, pp. 160–179.
doi: https://doi.org/10.1075/msw.3.2.03for, accessed 4 November 2022.

R.W. Gehl, 2016. “Power/freedom on the dark web: A digital ethnography of the dark Web social network,” New Media & Society, volume 18, number 7, pp. 1,219–1,235.
doi: https://doi.org/10.1177/1461444814554900, accessed 4 November 2022.

R. Gehl and F. McKelvey, 2019. “Bugging out: Darknets as parasites of large-scale media objects,” Media, Culture & Society, volume 41, number 2, pp. 219–235.
doi: https://doi.org/10.1177/0163443718818379, accessed 4 November 2022.

L. Gitelman and G.B. Pingree (editors), 2003. New media, 1740–1915. Cambridge, Mass.: MIT Press.
doi: https://doi.org/10.7551/mitpress/5001.001.0001, accessed 4 November 2022.

C.M. Halverson, 2003. “Lifting the dark threat: The impact of metaphor in the war against terror,” UW-L Journal of Undergraduate Research, volume 6, pp. 1–6, and at https://www.uwlax.edu/globalassets/offices-services/urc/jur-online/pdf/2003/halverson.pdf, accessed 4 November 2022.

HM Government, 2019. “Online harms white paper,” at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/793360/Online_Harms_White_Paper.pdf, accessed 12 May 2020.

N.P. Hoang and D. Pishva, 2014. “Anonymous communication and its importance in social networking,” 16th International Conference on Advanced Communication Technology.
doi: https://doi.org/10.1109/ICACT.2014.6778917, accessed 4 November 2022.

E. Jardine, 2018a. “Privacy, censorship, data breaches and Internet freedom: The drivers of support and opposition to dark Web technologies,” New Media & Society, volume 20, number 8, pp. 2,824–2,843.
doi: https://doi.org/10.1177/1461444817733134, accessed 4 November 2022.

E. Jardine, 2018b. “Tor, what is it good for? Political repression and the use of online anonymity-granting technologies,” New Media & Society, volume 20, number 2, pp. 435–452.
doi: https://doi.org/10.1177/1461444816639976, accessed 4 November 2022.

L. Jarosz, 1992. “Constructing the dark continent: Metaphor as geographic representation of Africa,” Geografiska Annaler: Series B, Human Geography, volume 74, number 2, pp. 105–115.
doi: https://doi.org/10.1080/04353684.1992.11879634, accessed 4 November 2022.

Y. Katz, 2020. Artificial whiteness: Politics and ideology in artificial intelligence. New York: Columbia University Press.
doi: https://doi.org/10.7312/katz19490, accessed 4 November 2022.

G. Keate, 2014. “Woman forbidden from marrying ‘tried to poison her evil mother’,” The Times (23 September), p. 5.

D.R. Kelley, 1996. “On the margins of Begriffsgeschichte,” In: H.Lehmann and M. Richter (editors). The meaning of historical terms and concepts: New studies on Begriffsgeschichte. Washington, D.C.: German Historical Institute, pp. 35–40, and at https://www.ghi-dc.org/fileadmin/publications/Occasional_Papers/THE_MEANING_OF_historical_terms.pdf, accessed 4 November 2022.

R. Koselleck, 2004. Futures past: On the semantics of historical time. Translated and with an introduction by K. Tribe. New York: Columbia University Press.

R. Koselleck, 2002. The practice of conceptual history: Timing history, spacing concepts. Translated by Todd Presner, Kerstin Behnke and Jobst Welge. Stanford, Calif.: Stanford University Press.

A. Krotoski, 2012. “Is it more important to have a single persona or the freedom of anonymity?” The Guardian (20 April), p. 19, and at https://www.theguardian.com/technology/2012/apr/19/online-identity-authenticity-anonymity, accessed 4 November 2022.

G. Lakoff and M. Johnson, 1980. Metaphors we live by. Chicago: University of Chicago Press.

S.K. Langer, 1954. Philosophy in a new key: A study in the symbolism of reason, rite, and art. New York: New American Library.

A.M. Lindner and T. Xiao, 2020. “Subverting surveillance or accessing the dark Web? Interest in the Tor anonymity network in U.S. states, 2006–2015,” Social Currents, volume 7, number 4, pp. 352–370.
doi: https://doi.org/10.1177/2329496520919165, accessed 4 November 2022.

Philip R. Van Loocke (editor), 1998. The nature of concepts: Evolution, structure and representation. London: Routledge.
doi: https://doi.org/10.4324/9780203029596, accessed 4 November 2022.

A. Mager and C. Katzenbach, 2021. “Future imaginaries in the making and governing of digital technology: Multiple, contested, commodified,” New Media & Society, volume 23, number 2, pp. 223–236.
doi: https://doi.org/10.1177/1461444820929321, accessed 4 November 2022.

M. Makhortykh and M. Wijermars, 2021. “Can filter bubbles protect information freedom? Discussions of algorithmic news recommenders in eastern Europe,” Digital Journalism (15 September).
doi: https://doi.org/10.1080/21670811.2021.1970601, accessed 4 November 2022.

X. Malbreil, 2007. “About the Internet imaginary and its evolution,” Texto Digital, volume 3, number 1, at https://periodicos.ufsc.br/index.php/textodigital/article/view/1397/1093, accessed 4 November 2022.

R. Mansell, 2012. Imagining the Internet: Communication, innovation, and governance. Oxford: Oxford University Press.

D. McCoy, K. Bauer, D. Grunwald, T. Kohno and D. Sicker, 2008. “Shining light in dark places: Understanding the Tor network,” In: N. Borisov and I. Goldberg (editors). Privacy enhancing technologies. Lecture Notes in Computer Science, volume 5134. Berlin: Springer, pp. 63–76.
doi: https://doi.org/10.1007/978-3-540-70630-4_5, accessed 4 November 2022.

J. van Horn Melton, 1996. “Otto Brunner and the ideological origins of Begriffsgeschichte,” In: H.Lehmann and M. Richter (editors). The meaning of historical terms and concepts: New studies on Begriffsgeschichte. Washington, D.C.: German Historical Institute, pp. 21–34, and at https://www.ghi-dc.org/fileadmin/publications/Occasional_Papers/THE_MEANING_OF_historical_terms.pdf, accessed 4 November 2022.

D. Moore and T. Rid, 2016. “Cryptopolitik and the darknet,” Survival, volume 58, number 1, pp. 7–38.
doi: https://doi.org/10.1080/00396338.2016.1142085, accessed 4 November 2022.

V. Mosco, 2004. The digital sublime: Myth, power, and cyberspace. Cambridge, Mass.: MIT Press.
doi: https://doi.org/10.7551/mitpress/2433.001.0001, accessed 4 November 2022.

J.-W. Müller, 2014. “On conceptual history,” In: D.M. McMahon and S. Moyn (editors). Rethinking modern European intellectual history. Oxford: Oxford University Press, pp. 74–93.
doi: https://doi.org/10.1093/acprof:oso/9780199769230.003.0004, accessed 4 November 2022.

S. Mützel, 2021. “Unlocking the payment experience: Future imaginaries in the case of digital payments,” New Media & Society, volume 23, number 2, pp. 284–301.
doi: https://doi.org/10.1177/1461444820929317, accessed 4 November 2022.

Ofcom, 2019. “News consumption in the UK: 2019” (24 July), at https://www.ofcom.org.uk/__data/assets/pdf_file/0027/157914/uk-news-consumption-2019-report.pdf, accessed 10 December 2019.

M. Osborn, 1967. “Archetypal metaphor in rhetoric: The lightdark family,” Quarterly Journal of Speech, volume 53, number 2, pp. 115–126.
doi: https://doi.org/10.1080/00335636709382823, accessed 4 November 2022.

A. Palmer, 2012. “Drugs, guns, assassins, jet planes ... All for sale on secret Deep Web,” Daily Mirror (22 September), p. 18; revised version at https://www.mirror.co.uk/tech/what-deep-web-dark-web-1337131, accessed 4 November 2022.

PAMCo, 2019. “PAMCo newsbites,” at https://pamco.co.uk/news/newsletter-q3-2019, accessed 10 December 2019.

E. Pariser, 2011. The filter bubble: What the Internet is hiding from you. New York: Penguin Press.

D. Reah, 2002. The language of newspapers. Second edition. London: Routledge.

D. Riffe, S. Lacy and F.G. Fico, 2005. Analyzing media messages: Using quantitative content analysis in research. Second edition. Mahwah, N.J.: Lawrence Erlbaum.

J. Schofield, 2008. “It's every man for himself,” The Guardian (2 October), p. 1, and at https://www.theguardian.com/technology/2008/oct/02/interviews.internet, accessed 4 November 2022.

C. Sherman, 2001. “Search for the invisible web: There are more websites than those seen with the naked eye,” The Guardian (6 September), p. 1, and at https://www.theguardian.com/technology/2001/sep/06/internetnews.onlinesupplement, accessed 4 November 2022.

D. Shestakov, 2008. “Search interfaces on the Web: Querying and characterizing,” Turku Centre for Computer Science (TUCS) dissertations, number 104, at https://www.utupub.fi/bitstream/handle/10024/38506/diss2008shestakov.pdf, accessed 4 November 2022.

P. Simonson, 2014. “Reinventing invention, again,” Rhetoric Society Quarterly, volume 44, number 4, pp. 299–322.
doi: https://doi.org/10.1080/02773945.2014.938862, accessed 4 November 2022.

B. Spolsky, 1998. Sociolinguistics. Oxford: Oxford University Press.

T. Streeter, 2011. The net effect: Romanticism, capitalism, and the Internet. New York: New York University Press.
doi: https://doi.org/10.18574/nyu/9780814741153.001.0001, accessed 4 November 2022.

M. Sturken, D. Thomas and S.J. Ball-Rokeach (editors), 2004. Technological visions: The hopes and fears that shape new technologies. Philadelphia, Pa.: Temple University Press.

The Times, 2017. “Unsafety Net: The policing of the hidden Internet cannot be left to vigilantes” (8 May), p. 21.

S. Turkle, 2011. Alone together: Why we expect more from technology and less from each other. New York: Basic Books.

S. Vaidhyanathan, 2011. The Googlization of everything (and why we should worry). Berkeley: University of California Press.

D. Wring and D. Deacon, 2010. “Patterns of press partisanship in the 2010 General Election,” British Politics, volume 5, number 4, pp. 436–454.
doi: https://doi.org/10.1057/bp.2010.18, accessed 4 November 2022.

S. Wyatt, 2021. “Metaphors in critical Internet and digital media studies,” New Media & Society, volume 23, number 2, pp. 406–416.
doi: https://doi.org/10.1177/1461444820929324, accessed 4 November 2022.

 

Appendix: Codebook for content analysis

This research aims to understand how six British newspapers represent the deep Web technologies in their news coverage. For that, a quantitative method approach of content analysis will be applied to a range of newspapers’ articles mentioning these systems and results will be interpreted to understand the overall media representation.

Sampling

This work uses LexisNexis databases to collect articles from the following British newspapers: Daily Mail, Daily Mirror, Daily Telegraph, The Guardian, The Times, and The Sun. The time frame is between 1 January 1989 and 31 December 2017.

Terms of inclusion

This codebook is suitable for articles mentioning any of the following terms: “dark internet,” “dark net,” “dark side of the internet,” “dark side of the web,” “dark web,” “darknet,” “darkweb,” “deep internet,” “deep web,” “hidden internet,” “hidden web,” “internet dark side,” “invisible internet,” “invisible web,” “non-indexable internet,” “non-indexable web,” “silk road,” “silkroad,” “tor browser,” “tor network,” “tor system,” “undernet,” “underweb,” “underworld of the internet,” “underworld of the web,” “web dark side,” and “web underworld.”

Unitization

Each newspaper article will be analysed separately and considered as a case. Different elements of these cases are then coded according to the variables and values set out below.

 

Case identification variables.
IDNameCode
(value)
Code labelCoding instructions
V01ID number  Newspaper number + Date (DDMMYYYY) + Number
V02Date  DD/MM/YYYY
V03Newspaper1Daily MailIn which of these newspapers the article was published
2Daily Mirror
3The Guardian
4Daily Telegraph
5The Sun
6The Times
V04Type of article1News articleSpecify the kind of article on the newspaper
2Editorial
3Comment, opinion or reader letter
4Review (book, movie, TV show, radio series and others)
5Interview
6Fiction
V05Length of the text  Number of words
V06Does the headline mention deep Web or related terms (listed aside)?1YesIf the headline of the article mentions any of these terms: “dark side of the internet” or “dark side of the web” or “dark internet” or “dark net” or “darknet” or “dark web” or “darkweb” or “deep internet” or “deep net” or “deep web” or “hidden internet” or “hidden web” or “internet dark side” or “invisible internet” or “invisible web” or “non-indexable web” or “non-indexable internet” or “tor browser” or “tor network” or “tor system” or “undernet” or “underweb” or “underworld of the internet” or “underworld of the web” or “web dark side” or “web underworld” or “silkroad” or “silk road”
2NoIf the headline doesn’t mention any of those terms
V07Which of these terms the headline mentions? (In the case of more than one, use the first one)1Dark Internet 
2Dark NetAlso “darknet”
3Dark sideReferring to one of these variables: “dark side of the internet” or “dark side of internet” or “internet dark side” or “dark side of the web” or “dark side of web” or “web dark side”
4Dark WebAlso “darkweb”
5Deep Internet 
6Deep Web 
7Hidden Internet 
8Hidden Web 
9Invisible Internet 
10Invisible Web 
11Non-Indexable Web 
12Silk RoadAlso “Silkroad” and “Silk Road 2.0”
13TorAlso “The Onion Router,” “Onion Routing” and “Tor Project”
14Tor Browser 
15Tor Network 
16Tor System 
17Undernet 
18Underweb 
19UnderworldReferring to one of these variables: “underworld of the internet” or “underworld of the web” or “web underworld”
20Deep Net 
99Not Applicable 
V08Does the headline of the article mention some of these activities? (Select the main one when more than one is mentioned)1Black marketHeadline mentions black market on Deep Web in a generic way (not specifying if it is focused on drugs, guns or stolen items, or mentioning multiple appropriations), Silk Road (and similar Web sites) or “trade”
2CybercrimeHeadline mentions any cyber or digital attack or threat, victim of leaking of personal (such as celebrity photo) or corporate information, stealing of identity, online scams, and/or computer viruses, when the word hacker or hacking is not used
3DrugsHeadline mentions specifically some kind of drug, poison, toxin, overdose, medicine addiction, death caused by drugs, legal highs, or drugs market available on Deep Web
4EspionageHeadline mentions any case in which Deep Web was used to commit espionage against any government
5FinancialHeadline mentions financial issues such as cloning of credit card, Bitcoin, currencies, financial fraud, money laundering, or another crime directly related to money and banks
6GamblingHeadline mentions gambling addiction and crimes related to gambling
7WeaponsHeadline mentions specifically some kind of weapons (including nuclear drones, bombs, biological, and others) or guns market available on Deep Web
8HackingHeadline mentions a specific person that is charged, arrested or persecuted for running a Web site on Deep Web, a person considered a hacker (such as Dread Pirate Roberts), or a hacking organisation, or a person or company that is victim of hacking — only when the word “hacker,” “hacking” or “hacktivism” is applied on the title
9KidnappingHeadline mentions kidnapping of a person, or related issues, such as hostage, ransom, specifying the reason (such as selling a person for sexual slavery) or not
10Mass murderHeadline mentions any kind of mass murder (when not considered a terrorist attack) that was planned or done using Deep Web systems
11Off-line CrimeHeadline mentions generically crimes that are committed off-line, such as murder, serial killer, rape, fake tickets or documents, environment crimes, and/or criminals (not specified in the other options)
12PaedophiliaHeadline mentions cases of paedophilia, and/or pornography, abuse or exploitation of children
13PornographyHeadline mentions cases of pornography, porn addiction, BDSM (excluding cases with children)
14TerrorismHeadline mentions terrorist attack, terrorist organization, jihadis, plans for a terrorist attack, or fight against terror (including states or organizations considered enemies on this fight)
15Whistle blowingHeadline mentions to whistle blowers, or the leaking of NSA files, mentioning Edward Snowden, WikiLeaks and software Prism (used by the U.S. government for surveillance)
99Not ApplicableHeadline doesn’t mention any crime
V09Does the article refer to surveillance practices?1On the headlineJust the headline mentions surveillance practices (including censorship, over watching, state control, vigilance, tracking, digital footprints, face recognition, electronic footprint, monitoring, cookies, spies)
2On the textJust the text mentions surveillance practices
3Both on the headline and on the text 
4Nor on the headline or on the text 
V10Does the article refer to privacy issues?1On the headlineJust the headline mentions privacy (including private life and communications, secrecy, hiding, personal data, personal details, encrypted communication)
2On the textJust the text mentions privacy issues
3Both on the headline and on the text 
4Nor on the headline or on the text 
V11Does the article refer to anonymity?1On the headlineJust the headline mentions anonymity (including anonymous communications, invisibility, hiding identity, namelessness, unknown person, alias, codename, pseudonym)
2On the textJust the text mentions anonymity issues
3Both on the headline and on the text 
4Nor on the headline or on the text 
V12Does the article refer to authoritarianism?1On the headlineJust the headline mentions authoritarian states and/or regimes and/or practices (including autocracy, despotism, dictatorship, fascism, monocracy, totalitarianism, tyranny) that go against the freedom of speech
2On the textJust the text mentions authoritarian practices
3Both on the headline and on the text 
4Nor on the headline or on the text 
V13Which Deep Web related term the article uses (referring to the first occurrence of the term on the text)?1Dark Internet  
2Dark NetAlso “darknet”
3Dark sideReferring to one of these variables: “dark side of the internet” or “dark side of internet” or “internet dark side” or “dark side of the web” or “dark side of web” or “web dark side”
4Dark WebAlso “darkweb”
5Deep Internet 
6Deep Web 
7Hidden Internet 
8Hidden Web 
9Invisible Internet 
10Invisible Web 
11Non-Indexable Web 
12Silk RoadAlso “Silkroad” and “Silk Road 2.0”
13TorAlso “The Onion Router,” “Onion Routing” and “Tor Project”
14Tor Browser 
15Tor Network 
16Tor System 
17Undernet 
18Underweb 
19UnderworldReferring to one of these variables: “underworld of the internet” or “underworld of the web” or “web underworld”
20Deep Net 
99Not Applicable 
V14Which Deep Web related term the article uses on second place (referring to the occurrence of a different term on the same text)?1Dark Internet 
2Dark NetAlso “darknet”
3Dark sideReferring to one of these variables: “dark side of the internet” or “dark side of internet” or “internet dark side” or “dark side of the web” or “dark side of web” or “web dark side”
4Dark WebAlso “darkweb”
5Deep Internet 
6Deep Web 
7Hidden Internet 
8Hidden Web 
9Invisible Internet 
10Invisible Web 
11Non-Indexable Web 
12Silk RoadAlso “Silkroad” and “Silk Road 2.0”
13TorAlso “The Onion Router,” “Onion Routing” and “Tor Project”
14Tor Browser 
15Tor Network 
16Tor System 
17Undernet 
18Underweb 
19UnderworldReferring to one of these variables: “underworld of the internet” or “underworld of the web” or “web underworld”
20Deep Net 
99Not Applicable 
V15Does the article use some of the Deep Web related terms with quotation marks (on the headline or text)?1On the headlineOne or more of these terms is written between quotation marks on the headline: “dark side of the internet” or “dark side of the web” or “dark internet” or “dark net” or “darknet” or “dark web” or “darkweb” or “deep internet” or “deep net” or “deep web” or “hidden internet” or “hidden web” or “internet dark side” or “invisible internet” or “invisible web” or “non-indexable web” or “non-indexable internet” or “tor browser” or “tor network” or “tor system” or “undernet” or “underweb” or “underworld of the internet” or “underworld of the web” or “web dark side” or “web underworld” or “silkroad” or “silk road”
2On the textAt least one of these terms is written between quotation marks on the text
3Both on the headline and on the text 
4Nor on the headline or on the text 
V16Does the article use the expression “so-called” to refer to Deep Web (on the headline and/or text)?1On the headlineThe expression “so-called” is associate with Deep Web and/or related terms on the headline
2On the textThe expression “so-called” is associate with Deep Web and/or related terms on the text
3Both on the headline and on the text 
4Nor on the headline or on the text 
V17Does the article offer an apposition (explanation or definition) for the term?1YesArticle explains what Deep Web systems are using apposition, definitions, concepts, comparisons, analogies, or other linguistic resource
2NoArticle takes for granted what the system is
V18In the case that the article offers an apposition, what is the source of the information?1AcademicWhen the source is specialist on the topic and connected to a university (which is mentioned on the article)
2BookWhen the source is a sentence from a book, so the title and/or author is cited on the article
3CorporateWhen the source is specialist on some topic and related to a company (which is mentioned on the article), or a corporative spokesperson
4GovernmentWhen the source is member of political parties, government department, NATO (excluding law enforcement and police)
5HackerWhen the source is connected to a hacker organization (such as Anonymous or Global Vigilance) or assumes itself as a member of the hacker community (which is mentioned on the article)
6Law professionalWhen the source is member of the justice system (such as judge, lawyer, persecutor), excluding police
7Civil society organizationWhen the source is connected with organizations that advocate for privacy rights and/or against surveillance (which is mentioned on the article), civil organizations, charity, activists
8PoliceWhen the source is member of police, military or intelligence agencies (such as Scotland Yard and FBI)
9SpecialistWhen the person is consulted because of professional knowledge, such as IT consultants, cyber security experts, researchers, psychiatrists, journalists, and others, when the company or institution is not mentioned on the article
10NewsWhen the source is a previous news article or report for TV, Internet or radio, made by the press
98Not SpecifiedArticle gives a generic definition without specified the source of the information
99Not ApplicableWhen there is no definition
V19Which attribute is associated with or used to describe Deep Web or related terms (considering the first one to be mentioned on the same sentence as Deep Web or related terms)?1AnonymousAlso “anonymising”
2Black market 
3CriminalAlso “crime facilitator”
4EncryptedAlso “highly encrypted” or “heavily encrypted”
5HiddenAlso “used to hide” or “a way of hiding”
6Illegal 
7Illicit 
8InaccessibleAlso “not accessible”
9Invisible 
10LawlessAlso “beyond the law” and “unlawful”
11Not viewable 
12Resilient 
13SecretiveAlso “secret”
14Secure 
15ShadowyAlso “shady” or “shadier”
16UnlistedAlso “not listed”
17Unpoliced 
18Unregulated 
19UnreachableAlso “beyond the reach”
20UntraceableAlso “that cannot be traced” or “without being traced”
21Parallel 
22Notorious 
23Murky 
24Noxious 
25Refuge 
26HarderAlso “hard-to-access” or “hard-to-track”
27Restricted 
28Booming 
29Undetectable 
30Treasure 
31Uncharted 
32Huge 
33SafeAlso “safer”
34Underground 
35Wonderland 
36Infamous 
37Sinister 
38Popular 
39Complex 
40Free 
41DarkAlso “darkest”
42Solution 
43Shield 
44Comprehensive 
45Interesting 
46Buried 
47Route 
48Sophisticated 
49Vast 
50SpecialisedAlso “special” and “specialist”
51NastyAlso “nastier”
52Drug dealing 
53Dodgy 
54New phenomenon 
55Sick 
56Ugly 
57Under the radarAlso “beneath the radar”
58Chilling 
59Difficult 
60Covert 
61Grim 
62AbnormalAlso “not normal”
63Grey 
64Bazaar 
65Large 
66Underside 
67Major 
68Mysterious 
69Chaotic 
70Not indexed 
71Private 
72Vile 
99Not applicable 
V20Which term is associated with people that use Deep Web systems on the article (considering the first one to be mentioned on the same paragraph)?1Buyer 
2Consumer 
3Coward 
4CriminalAlso “organised crime,” “people engaging in criminal activities,” “crime group” and “cybercriminal”
5FacelessAlso “faceless person”
6HackerAlso “hacking group”
7Outlaw 
8PaedophileAlso “paedos”
9Pervert 
10Seller 
11TerroristAlso “terror leader,” “terror network,” and “terrorist group”
12Troll 
13Pornographer 
14Abuser 
15User 
16DealerAlso “drug dealer” and “arms dealer”
17Crook 
18SuspectAlso “suspected of ...”
19Player 
20Fraudster 
21ResearcherAlso “person conducting research”
22Whistleblower 
23Drug user 
24Predator 
25Community 
26Connoisseur 
27Activist 
28Extremist 
29JihadistAlso “jihadis”
30OffenderAlso “sex offender”
31Addict 
32GangAlso “gangland,” “crime gang” and “gangster”
33Journalist 
34Government 
35State 
36Trafficker 
37Recruiter 
38Conspirator 
39Kidnapper 
40Peddler 
41Defendant 
42Ruler 
43Student 
44Costumer 
45SupplierAlso “drug supplier” and “arm supplier”
46Thief 
47Killer 
48YoungsterAlso “young people”
49Smuggler 
50Syndicate 
51Trader 
52Fiend 
53Fanatic 
54GeekAlso “computer geek”
55Member 
56Backer 
57TroubledAlso “troubled person”
58Rapist 
59Nut 
60Teenager 
61Arab 
62Group 
63ScientistAlso “computer scientist”
64BomberAlso “suicide bomber”
65Client 
66Isis memberAlso “Isis activist” and “Isis sympathiser”
67Gunman 
68Visitor 
69Sucker 
70Islamist 
71Exploiter 
72VillainAlso “cyber-villain”
73Al-Qaeda 
74Molester 
75Captor 
76Cryptographer 
77Reject 
78Operator 
79Mafia 
99Not applicable 
V21Source 1
(Code only those who are either directly quoted or paraphrased)
(Use the same criteria for Source 2 to Source 12)
1AcademicWhen the source is specialist on the topic and connected to a university (which is mentioned on the article)
2CorporateWhen the source is specialist on some topic and related to a company (which is mentioned on the article), or a corporative spokesperson
3GovernmentWhen the source is member of political parties, government department, NATO (excluding law enforcement and police)
4HackerWhen the source is connected to a hacker organization (such as Anonymous or Global Vigilance) or assumes itself as a member of the hacker community (which is mentioned on the article)
5Law professionalWhen the source is member of the justice system (such as judge, lawyer, prosecutor, justice minister), excluding police)
6Civil society organizationWhen the source is connected with organizations that advocate for privacy rights and/or against surveillance (which is mentioned on the article), civil organizations, charity, activists
7PoliceWhen the source is a member of police, military or intelligence agencies (such as Scotland Yard, NCA and FBI)
8SpecialistWhen the person is consulted because of professional knowledge, such as IT consultants, cyber security experts, researchers, psychiatrists, journalists, “a source,” and others, when the company or institution is not mentioned on the article
9Suspect or criminalWhen the source is the person that committed the crime, or is being accused of a crime, or was planning a crime, or someone talking on their behalf (including family)
10VictimWhen the source is a person that was impaired by some crime that is connected with Deep Web systems functionalities, or if someone is speaking in the name of this person (such as a spokesperson, family member or PR agency), or if the person is a witness of a crime
11CaseWhen the source is a neutral person used to exemplify something on the article but is not an expert on any topic
99Not applicable 
V22Does the Source 1 offer an opinion about Deep Web or related systems?
(Use the same criteria for Source 2 to Source 12)
1YesThe source gives an opinion about what he/she thinks or understands of the Deep Web or related systems
2NoThe source doesn’t give an opinion or talk specifically about Deep Web or related systems
99Not applicable 

 

 


Editorial history

Received 2 July 2022; revised 2 September 2022; accepted 4 November 2022.


Creative Commons License
This paper is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Inventing the dark Web: Criminalization of privacy and the apocalyptic turn in the imaginary of the Web
by Thais Sardá, Simone Natale, and John Downey.
First Monday, Volume 27, Number 11 - 7 November 2022
https://firstmonday.org/ojs/index.php/fm/article/download/12691/10745
doi: https://dx.doi.org/10.5210/fm.v27i11.12691