What to the Muslim is Internet search: Digital Borders as Barriers to Information

In today’s digital age, searching for information online is considered a ubiquitous task that can be accomplished in just a few moments using various web-based technologies. Yet, information seeking has geopolitical burdens for users who are racialized and marginalized by the nation-state and other structures of power. In our paper, we conducted a qualitative interview study with 15 Muslim participants, mostly of South Asian origin, living in the US with varying citizenship or (non)immigration status about their information needs and concerns around privacy as a Muslim, and the resulting restrictive patterns of information seeking on various Internet platforms. We argue that our findings on the barriers faced and strategies employed by Muslim residents toward information access suggest a broader pattern of digital manifestations of border imperialism. We posit that HCI researchers should pay attention to how “digital borders" have epistemic implications for people marginalized by geopolitical boundaries.


INTRODUCTION
In orientalist views of the West [64], noted Palestinian-American academic and political activist Edward Said, Muslims are reduced to an other, and Islam to a particularly "narrow and constricted" semantic feld that is "to be looked at with a very special hostility and fear" [65].However, since 9/11, Muslims as a group have been uniquely racialized not only in the United States (US) [73], but around the Western world, making them hypervisible [71].In this paper 1 , we explore how Muslims as a racialized category experience information communication technologies (ICTs) when seeking information.
Although previous works in information studies discussed information search strategies [45,56] and have also highlighted some barriers to information seeking (such as language problems, institutional barriers, lack of social and economic capital [69]), they do not consider the hostility of the environment in which information exists, and concerns about racialized state surveillance as barriers to accessing information even when it is available online.
It is well-documented that Muslims living in the US have been subjected to increased scrutiny, a target of hate and harassment from individuals, and surveillance by the government [15].The US is generally among the countries with the highest surveillance of its residents [8], a governance tactic that is amplifed by surveillance capitalism [87].In this context, then, Said's warnings ring true to this day; except the West now has the enhanced ability to literally and fguratively "see/watch" its Islamic others, reducing their diversities and identities into one racialized, semantic category: Muslims.Our choice to describe the participants as such is rooted in this history.We acknowledge that "Muslims" are not a homogeneous category, neither is Islam; more importantly, neither being a Muslim nor practicing Islam is experienced as a homogeneous identity/practice by our participants.Furthermore, the majority of the participant sample of our study identifed as South Asian in terms of their ethnicity -this is possibly a refection of our relational networks that we were able to recruit from.At the same time, our fndings also suggest how their "Muslimness" often felt "seen" more than other aspects of their identities.Although we do not claim to have captured the lived experiences of Muslims of all ethnic, socioeconomic and geographical backgrounds, in order to highlight and question the ways of seeing performed by the United States [64,65], we will refer to our participant group as Muslims in the US.
Much of the information needs of Muslims in the US are similar to those of any user of the Internet regardless of religious belief or racial category.There are also some unique needs.For example, practicing Muslims often need information on prayer times, Qibla 2 direction, instructions on routine matters, Quranic texts, and interpretations.Islam has a vast body of literature on Quranic interpretation, issues of personal dealing rights and responsibilities concerning one's family, parents, spouse, children, relatives, business, fnancial matters, etc.With this qualitative interview study, we answer the following research questions in the context of Muslims residing in the US: • What are some needs and patterns, practices and barriers experienced while using ICTs for information seeking?• How do experiences of surveillance and immigration infuence their practices and perceptions of web-based information seeking?

Overview and Contributions
In our work, we have interviewed 15 participants (13 of them of South Asian origin) who self-identifed as Muslims of varying ages, number of years lived in the US, and citizenship or immigration statuses.Our fndings contribute an overview of the various information needs, practices, and barriers for Muslims in the US, lived experiences of surveillance and the ways in which they infuence information practices, ongoing conceptualization of the datafcation of Muslim lives, and fnally methods and tactics of mitigating these information barriers.We argue that our fndings indicate a pattern of internalizing bordered governance [41,43] -i.e.migrant subjects (in the Foucauldian sense [20]) of the state internalizing and self-imposing logics of border imperialism to survive [83].We construct this argument by mapping our fndings to key concepts in border studies such as racialization, surveillance, subjective illegality, and subsequent criminalization.As Foucault's theories of power [28] and governmentality [41] as well as the concept of border internalization would suggest [43], systematic racialization and criminalization have inscribed in Muslim migrant subjects a habit of self-policing, particularly in the form of self-censorship as part of their information practices.While such manufactured lack of access to information has been theorized as epistemic injustice in other contexts [31,57] seen primarily through the lens of settler colonialism [2,76], we argue that the implications of restrictive information seeking due to internalized boder governance are also a form of epistemic injustice.In that, we conclude that contemporary bordering regimes have stretched the borders to the Internet, making it necessary for HCI and social computing researchers to interrogate the epistemic implications of digital borders.By introducing conversations from border studies on border imperialism to HCI, we contribute to a long line of interdiscliplinary research on race, gender, religion, caste, class, and other social implications of ICTs.

Privacy and information seeking
Information practices include activities of information seeking, sharing, and use [68].Within information practices, we focus on information seeking which is a subset of information behavior, with the former being 'purposive' while the latter can include passive attainment of information as well [13, p.92].Information behavior is the "totality of human behavior in relation to sources and channels of information, including both active and passive information seeking, and information use" [86].This includes active communication with others and passive reception of information from sources in the environment without actively seeking information.Information seeking behavior on the other hand is "purposive seeking of information as a consequence of a need to satisfy some goal" [86].Seeking can be further broken down to searching behavior, which is the "'micro-level' of behavior employed by the searcher in interacting with information systems of all kinds"[ibid] including interaction with the system, or intellectual process of making assessments about the information retrieved [ibid].Information systems can be seen as the "complicated array of social and cultural practices and the political and technical infrastructures required for information tools to 'work' " [48].
To search and retrieve information, users need to implicitly or explicitly share some information about themselves such as search words, location, IP address, email addresses, etc. leading to a costbeneft analysis by the seeker, where they disclose personal information if perceived benefts "exceed the current or future risks of disclosure" [21].Morton et al. [47] studied the information seeking behaviors of users with respect to privacy cues of the technologies they were using.Users evaluate risks of disclosing information compared to benefts received, but time and cognitive constraints force users to rely on limited cues to minimize uncertainty.Newell argues that surveillance is also an information practice as it requires the retrieval of information to assess people for some regulatory purposes [49].Depending on the political climate, this then changes the nature of the cost-beneft being done by the user about disclosing information for seeking information.Surveillance has been shown to change the information behaviors of individuals [1,11] where users can sometimes self-censor to manage how they present themselves to others online in the absence of appropriate privacy controls [75].
Monitoring has become ubiquitous, combined with powerful analysis techniques, threatening privacy and autonomy of consumers and citizens.In their review of privacy behavior in the age of information, Acquisti et al. [3] contend that users face high uncertainty when it comes to tradeofs about privacy because of asymmetries of knowledge about what and how much data companies and government are collecting on them and its manner of use making it difcult to decide how much information to share.Moreover, privacy preferences are context-dependent as users take cues from their environment to decide preferences.Privacy depends on location, context [3] and culture [1,11] and afect the privacy needs and concerns of individuals according to their intersectional identities.For example, the privacy needs of Muslim women in US arise from their intersectional identities as American, Muslim and women [5].Another challenge of privacy preferences is their malleability and infuence by entities that capitalize on user data to derive fnancial benefts.They are much more sophisticated and capable of exploiting users to behave in ways which lead to higher disclosures [3].We contribute to the discussion on privacy in information age by demonstrating how concerns about surveillance by state and corporations alter the search and information behaviors of users of racialized and marginalized groups.

Enacting Governmentality and Surveillance with ICTs
Foucault conceptualized government not just as management by the state but also as problem of self-control [41].Governmentality is the 'conduct of the conduct', how governments conduct people towards certain behaviors.It includes governing the self and governing others, thus showing the codetermination of modern sovereign state and modern autonomous individual.Technologies of governmentality include disciplinarian power and control mechanisms.
Although disciplinarian powers lead to certain consequences, control mechanisms are less explicit and nudge people towards certain desirable behaviors while avoiding undesirable ones [41].People's perception and meaning of the state are shaped by their experiences with bureaucratic processes in daily life, providing them with insights into the micropolitics of the state [72].Walter [84, p.5] suggests that migration researchers take mid-range concepts like antipolicy (e.g., anti-trafcking, anti-terrorism, anti-poverty, etc.) to connect specifc aspects in the Focualt's conceptualization of Governmentality to migration governance.Antipolicies are campaigns originating from state or civil society organizations across a range of problems to combat and eliminate bad, drawing their legitimacy by mobilizing a polarization within the public sphere by asking people to take sides, such as with the good or against the bad [71].In racializing Muslims as others and labeling them as terrorists, such anti-policies work both to enforce migration controls in an overt manner of disciplinary power but also covertly exercise governmentality by creating a dichotomy of good/moderate and bad/extremist Muslims, forcing Muslims at once not only to mobilize and vocalize against the 'bad' but also to ensure that their behavior aligns with the 'good' [71].
The surveillance of citizens can be a mechanism of exercising governmentality.Communication scholar Bryce Clayton Newell posit that the three components of surveillance include: human beings as the most common objects of surveillance, implicit or explicit capturing of information about human beings, and as having "a purposive element, most frequently tied to regulating or governing human behavior" [49].Haggerty et al. demonstrated that the surveillance setup no longer consists of discrete systems, but 'surveillant assemblages' [35].These assemblages consisting of people, corporations, government contractors, states, agencies are used by governments -both democratic and authoritarian -to extensively surveil their citizens [42] to maintain social order, prevent terrorism, and distribute welfare.In case of Muslims in the US (and in several other nation-states), experiences with the state include targeted surveillance of their communities [4], technologies [18,19], and creation of databases for controlling entry and exit at the borders by the state [79].
Mediated through surveillance capitalism, technologies of daily use have become a means of monitoring any and all citizens not only by the nation-state [39] but also by corporations.Zubof defnes Surveillance Capitalism as "the foundational framework of surveillance economy" that "uses human experience as free raw material for hidden commercial practices of extraction, prediction and sale" [87].Zubof marks 9/11 as one of the catapult events in creating surveillance exceptionalism that allowed companies to include surveillance practices as the government and corporations collectively aimed for total certainty, practices that would have been outlawed under previous legislative discussions on privacy [87, p.115].Surveillance capitalism operates through asymmetries in knowledge and power as the corporations that extract behavioral surplus know everything about their users but the users do not know everything about them [87, p.11].A consequence of online government surveillance is the "chilling efect" or deterrance among people from engaging in certain legal activities because of a fear of unfavorable legal consequences and mistrust of the legal system to protect innocence.Disclosures about the US government surveillance is shown to have chilling efects on information seeking behaviors among people in US in general [55].Stoychef et al. [77] showed that the chilling efects of online mass surveillance among Muslims in the US deter them from legal behaviors like online political participation and disclosures of religious identities.
Jacob Scott ofered the bureaucratic perspective in managing people by reducing complex social and natural phenomenon by oversimplifcation in an attempt to make them legible and manageable by the state [70].Singh & Jackson [74] introduced the data infrastructure by showing how data representation of citizens in infrastructures used for provision of government services can render certain citizens as high resolution, making more data available about them, vs low resolution citizens.High resolution citizens beneft from government provided services but at the cost of loss of privacy and increased surveillance compared to low resolution citizens.We expand the discussion on surveillance as a means of exercising governmentality within border governance by demonstrating its presence within the online cyberspace.

Internalizing borders to self-discipline online and ofline
Typically when we think of border, we think of the lines which demarcate the end of one country's territory and the start of another.Border studies have interrogated the concept of borders in detail and have already highlighted the move of borders beyond the physical borders to include technological spaces, politics, discourses etc. Johnson et al. invite scholars to rethink of borders in terms of place (of enactment or materialization), performance (of border work), perspective (of those performing borderwork) and politics of borderwork [40].For instance, Rumford introduced the concept of borderwork to refer to "envisioning, constructing, maintaining and erasing borders" [63] and argues that borders have generalized in the society as a whole and borderwork is being performed not only done by the state but also by ordinary citizens and organizations through 'rebordering' and 'debordering'.While borderwork might be performed by actors other than the state, it still remains an apparatus of oppression used for imperial state formation.Walia interrogates the "formation and function of borders as a spatial and material power structure" and argues that "borders are less about a politics of movement per se and better understood as a key method of imperial state formation, hierarchical social ordering, labor control and xenophobic nationalism" [83, p.2]. Extending this idea, Walia introduces the concept of Border Imperialism which further emphasizes the nature of borders as artifcial and political, an apparatus of colonization, othering [82].Border imperialism hence refers to "the processes by which the violences and precarities of displacement and migration are structurally created as well as maintained" [83, p.2].
Driven by neoliberal and racial logics, borders afect diferent groups diferently.Borders are not fxed lines but are elastic, "productive regimes which are produced by and reproduce racial social relations further imbued by gender, sexuality, class, ability, and nationality" [83, p.78].Contemporary border and immigration controls fnd their roots in anti-black violence, imperial expansion and indigenous elimination.Anti-Black logics aimed at regulating and punishing black movement and continue to be part of border controls aimed at controlling mobility, by keeping out the undesirables.Furthermore, the history of border formation in the US lies in the imperial wars against crime, terror, and drugs at home and abroad [83, p.38-60].The war on terror led to the racialization and othering of Muslims from diverse regions and races [73] lumping them into one group, positioning them as the enemy of the nation-state at both global and domestic levels and fusing overseas war on terror with domestic immigration policy.Walia describes the relation between the surveillance, deportation and detention of Muslims within the US and the military aggression in the form of air strikes and carcerality of citizens of Muslim majority countries as globalized imperial racism [83, p.55].
To understand borders as a mode of hierarchical control, it is important to understand how they are operationalized.In this vein, four border governance strategies discussed by Walia include exclusion, territorial difusion, commodifed inclusion and discursive control [83, p.77-92].Territorial difusion remains most relevant to our purposes and refers to the internalization of the borders such that the border does not end when a person crosses a physical border but "can be enforced anyhwere in the nation-state" [83, p.84].While Walia's analysis of border imperialism is primarily concerned with the physical technologies of control, adjacent work in border studies augment Walia's thesis and further argues that the modern borders rarely need to be enforced by a specifc institution.Instead, the migrant subjects of a nation-state often internalize this governance mechanism to self-regulate and self-discipline their existence.It is in this way that border internalization "creates a ubiquotous regime of bordering enacted through racialized surveillance, instilling fear of the everyday" [83, p.84].Along the same lines, Mahmoudi introduced the notion of 'digital periphery' to show "how urban migrant environments and subjectivities are commodifed and 'datafed'" [43] advancing racial capitalism beyond the typical spatial and physical limits of borders.Commodifed control is a border governance strategy applied to 'legal' migrant workers with defated legal power to ensure capital accumulation with an ever present risk of deportation making them susceptible to exploitation [83, p.85].Villa-Nicholas introduced the term data borders to include "layers of information technologies" involved in creation of not only the physical border infrastructure at the US-Mexico border but that "extend around the US through data and digital surveillance".The later includes data from various government and private sources through cooperation between Silicon Valley companies and US Government.Collectively, they expand the border to the whole of US via digital surveillance dragnet "with data as the underlying valued source of capital in this phenomenon" [81, p.13].
In this paper, we bring the works of Walia, Mahmoudi, and others in conversation with HCI.Our fndings show patterns of technology use (or non-use) that emerge when borders are internalized by Muslim users.We make sense of these patterns in the backdrop of border imperialism as formulated by Walia.

METHODS
In this section, we frst discuss the positionality and refexivity of both authors of this paper in 3.1.Section 3.2 discusses the interview protocol.Section 3.3 describes the recruitment process, participant profles and data collection methods.Next, in section 3.4, we discuss data analysis.Section 3.5 discusses the ethical considerations that we made during the study.Finally, section 3.6 discusses the limitations of our study.

Researcher Positionality and Refexivity
In recent times, HCI has turned to the feminist social sciences [24,26,44] and indigenous research [2,76,85] for guidance on how to bring researcher subjectivities into the methods followed in empirical work [7,22,23,32].At the heart of this turn is the urge to ensure a relational ethics is followed in all stages of research planning and execution -formulation of research questions, data collection, data analysis, theoretical engagements, and presentation of fndings [44].In what follows, we act upon their recommendation and discuss our own positionalities in relation to the research topic .We also share some examples of how self-refexivity was applied as a method in various stages of this work.
The frst author is a Muslim woman with family origin in the territory of Kashmir, born, raised, and educated in Pakistan who moved to the United States to pursue her doctorate.She grew up listening to discussions about Kashmir between India and Pakistan, Afghan refugees in Pakistan displaced by Soviet war [38] and later the US war on terror [16] demanding Pakistan to prove itself as an ally in the 'with us or against us' stance of the US or be 'bombarded back into stone age' [14], US drone attacks on Pakistani Tribal regions [30], military operations, mass displacement of tribal people [60], terrorist attacks within Pakistani cities [58], and portrayals of Pakistan in Western and Indian mass media as a facilitator of terrorism.She also experienced the US visa and immigration process, saw and heard experiences of young people being denied visas at US embassies, experiencing the highest security checks while traveling from Pakistan to US, 'random' baggage inspections, anxieties of traveling to US and about the Muslim ban, and concerns of surveillance while living in the US when accessing religious information, using Muslim prayer and dating apps, ordering a copy of Quran online etc.
The second author is a South Asian person born into a (nonpracticing) Hindu family of Brahmin caste [80].Growing up in India in a family of freedom fghters and political activists [66], the second author was exposed to the history of British colonialism [66] and the Partition [12,53], and simultaneously developed a critical consciousness around the ethnonationalistic ambitions that in the contemporary setting are driving the Indian nation-state to be (re)produced as Hindu-Rashtra (i.e. the nation of the Hindus) [54].The second author migrated to the US for education -an experience that further exposed the institutions of border imperialism and their impact on individual's conception of safety, security, and belonging.
While the research questions we chose for the study are transferable to other context of migration, the project holds personal/political signifcance for both researchers as non-immigrants and as individuals whose ethnic and religious identities have been forced against one another for the service of ethnonationalism and border imperialism.In fact both authors have had their lives shaped by one of the most profound act of bordering and its aftermath -the Partition of India and Pakistan [12].
Both researchers' positionality as non-immigrants, and of the frst author as a Muslim, made them vigilant around any concerns participants might have had about the storage and use of data.Questions around data security, accidental leaking, surveillance, government requests, as a result, came to us from an understanding of research ethics that is grounded in embodied experiences of harm and fear.
Furthermore, frst author's experiences as a Muslim helped develop a deeper understanding during participant interactions and helped build rapport.Sharing her experiences that motivated the study helped her 'move from the distanced observer to the feeling participant' and learn about participant experiences and emotions diferent from and similar to hers [25, p.12].This interactive way of interviewing made the process 'less a conduit of information from informants to researchers that represents how things are, and more a sea swell of meaning-making in which researchers connect their own experiences to those of others and provide stories that open up conversations about how we live and cope' [25, p.15].

Interview protocol and preparation
The interview protocol was developed collaboratively by the authors, applying self-refexivity in its many iterations.In each iteration, we took into account potentially leading questions, bias, and language and/or phrasing that may be deemed as triggering or intrusive by participants.The authors also checked in incrementally, in the beginning after 1-2 interviews, with particular attention given to the possible efects of the research on the participants, research ethics, possible bias in the protocol, and the impact of this work on the frst author who also shares a social location with many of the participants.Overall, the topics discussed during interviews included information needs fulflled using the Internet (web searches, websites, online communities, social media, apps), use of the Internet for religious information, feeling of (un)safety while seeking information online, changes in usage of online information sources due to surveillance and privacy concerns, topics considered sensitive, expectations about privacy as a Muslim living in the US, impact of surveillance on their current lives and concerns about potential use of collected data on their future lives.

Recruitment, Participants and Data Collection
Participants were recruited from the frst author's social circle, through referrals from Muslims and non-Muslim friends and colleagues, by posting calls for recruitment to Muslim Student Associations (MSAs) in two US-based universities, and snowballing from existing participants.Participants signed up for interviews by flling a recruitment form asking for their age, education, religion, years lived in the US, immigration or citizenship status and whether or not they had any data privacy concerns when seeking information online as Muslims living in the US.We screened for individuals who self-identifed as Muslims, had concerns about the privacy of their data and had been residing in the US for more than two years regardless of their immigration or citizenship status.It was important for us to do this work through relational networks so that researcher(s) can foster accountability for any post-interview thoughts/concerns (e.g.distress, concerns around data storage) among participants given the potentially sensitive nature of the work.We interviewed ffteen participants of varying ages, years lived in the US, and citizenship or immigration statuses.Participant demographic summary is provided in Table 1.Most of our participants belonged to South Asian origin (see limitations section 3.6) and were college educated.Two participants inquired about the topics of discussion prior to deciding on participation that we shared with them beforehand.
The frst author conducted interviews with 15 participants on Zoom that ranged between 50 to 75 minutes in English or Urdu as per participants' preference.We sought consent from participants for recording the interviews.We informed participants that they can choose not to answer any question or ask to stop the recording at any time.

Data analysis
The interview transcripts and notes were open coded to come up with 213 initial codes in the frst round of coding [67, p.100].These codes were then combined and reduced to 55 codes using axial coding [67, p.218] and sorted into 11 code categories.Some examples of code categories include online search workarounds, expectations about who is tracking, concerns about future.Some example codes in each category include 'avoiding search topics', 'conceptualizing the watcher', 'power imbalance in surveillance', 'identity based targeted prosecution'.The interview data was inductively coded.For instance, themes on border and comparison of information ecologies appeared as a result of inductive coding of participants' conversations.The frst author coded initial interviews to come up with codes which were discussed with the second author for clarity and context.Both authors further thematically organized the codes into the fnal presentation of the fndings.

Ethical considerations
This study was approved by the Internal Review Board of the authors' institution.In our experience, participants' levels of (dis)comfort with their voice being recorded varied (the study protocol did not necessitate a recording).Some participants asked to stop recording and shared discomforting experiences of encountering information about being profled because of their Muslim identity but did not want it to be on the record.
For participants who opted out of recording, we shared the interview notes.However, one participant shared that knowing the data privacy policies of documents saved on the cloud, looking at the sensitive terms they mentioned written on the document made them uncomfortable.Moreover, as part of the discussions on the study topic of online information seeking, data privacy and surveillance, participants shared concerns about being careful as to not jeopardize their immigration or future employability while seeking information as all online activities leave a trace.While participants did not share any concerns regarding participation in the study and wanted security accommodations, as researchers we further assessed the connection between unknowingly sharing potentially problematic information and the need for data privacy in study protocols.To address this, we paused our data collection process and revised the protocols for data recording, storage and analysis to enhance data protection.We, therefore, increased the data security level for research data and resubmitted our IRB protocol.Upon IRB approval of this re-submission, we recorded data from interviews on analog recorders and transcribed using ofine locally installed software.We transcribed the frst four interviews using Otter.ai(a cloud based service) but later changed the study protocol to keep all the data ofine, and transcribe interviews using Whisper transcription software locally.

Limitations
Even though we aimed for a diverse sample of participants, frst authors' identity and recruitment through snowballing led to 10 out of 15 participants being of Pakistani origin, frst or second generation immigrants or nonimmigrants.Most participants shared more than religious aspects of their identities with the frst author such as race, nationality, career, education etc., making her an in-group researcher.Muslims in the US belong to varying races, locations, educational and career backgrounds, and gender identities, all of which can vary their experiences and perceptions.This study does not represent the experiences of all of them instead its fndings are refective of the situated experiences of Muslims of South Asian origin in the US.This study took place in the US that has a unique combination of privacy laws, surveillance capitalism, Islamophobia and racism.The experiences of Muslims of South Asian origin in the US might not be generalizable to other Western or non-Muslim countries where Muslims are in minority.There was self-selection among participants considering the sensitive nature of the topic and the concerns associated with privacy.

FINDINGS
In this section, we present our key fndings on the following: 1) information needs and barriers experienced by Muslims in the United States (section 4.1), 2) their perceptions around surveillance and datafcation of Muslim lives and the impact it has on everyday information practices (section 4.2), 3) ongoing conceptualization and sensemaking of their datafed selves (section 4.3), and fnally 4) ways of mitigating these information barriers (section 4.4).

Information practices and barriers for Muslims in the US
Our participants shared a variety of information needs for which they commonly used information communication technologies (ICTs), and their sociotechnical practices to satisfy these information needs.They also shared their perceptions around the barriers they face on a daily basis exacerbated by the roles played by the state, the makers and marketers of the ICTs they use, and sometimes the ICTs themselves.
4.1.1Changed information ecology due to change of place.For participants who were born and raised in a Muslim-majority country [61], a change of place often led to a change in the information ecology on which they relied.This change was more pronounced upon arrival but still experienced to date by all participants who had moved to US as adults and had established information ecologies in home country.This resulted in an increased reliance on ICTs for their everyday information needs, particularly evident in the case of information seeking for religious practices.P2 explains: "If I am fasting, I want to know what time is Iftar [time for breaking fast] today.I need to look up that information online....because I don't have that information readily available around me.Back home, this information was readily available [on TV].So I never really had to use the Internet for these purposes.(P2, Non-immigrant)" In recalling their experiences with information seeking, our participants often compared their current practices with how they sought information in their home country.They refected, often with nostalgia, the ease with which religious information is more readily available in religious scholarship or through relational networks.As P2 describes: "Let's take the example of diferent duas [prayers] that I now look up online, back home we do have a lot of diferent small booklets with all these duas.So if I wanted to look up a dua [prayer] for a certain purpose, I would just open that book, and now that we don't have that information in a booklet form or in a physical form, we have to go to the Internet to fnd that information.(P2, Non-immigrant)" Another example of a shift in information ecology upon arriving in the US or Western countries is following Halal3 food regulations.When living in their home countries, most of our participants did not have to consider whether a certain food meets their Halal dietary restrictions, as all food is processed in compliance with Halal food standards.However, within the US, they have had to constantly seek out Halal food outlets including grocery stores, restaurants, brands, etc.They frequently used the Internet to not only search but also remain up-to-date about Halal food spots as change in suppliers and costs keeps changing Halal food availability.
Moreover, the criterion for Halal food is linked with how it is processed [37]; particularly meat, and the US has meats available that have been processed in a variety of ways and sometimes people have varying opinions about which process qualifes them as Halal.

Challenges with quality and authenticity of information about
Islam in the US.Although the previous section indicates that the information required for religious practice is a common reason for web based information seeking, participants with varying citizenship and immigration statuses reported having faced serious issues with the authenticity of the information, which was experienced as a concern not only for practicing Muslims among our participants but for anyone who was trying to understand the Islamic perspective on any matter.Multiple participants alluded to the problem of misinformation about Islam and Islamophobic content on the Internet.
"If you are living in the West, you are going to get an Islamophobic view of Islam.For example, the Yaqeen Institute has a study and had a statistic, I think it is still on their website, where it says that there are millions of dollars spent every year towards [spreading] Islamophobia in the United States [62] where there is misinformation and its misguiding people about it.(P10, Citizen born))" Another related concern shared by participants was the accuracy of Islamic interpretations on the Internet.Islamic teachings require contextual interpretations, depending on the context of the particular individual looking for guidance.However, these interpretations require in-depth knowledge of Islamic jurisprudence or Fiqh [59] and are performed by qualifed scholars.P1 talked about the use of discussion forums or Q&A style websites used by people to seek advice on their specifc situation, but as P1 notes: "Sometimes people have questions and go to a forum and some aalim [scholar] answers them.And that is even vaguer because you don't know the authenticity of the aalim, and you don't know how accurate or how conservative or progressive that aalim is (P1, Nonimmigrant)" Moreover, the experiences and expectations around surveillance difered based on the participants' immigration status, as cited by themselves.Non-immigrant 4 participants had to share their social media credentials and email addresses as part of their visa application process and they expect the government to continue to monitor their accounts.P2 (non-immigrant) adds: "Suppose that you want to see what Muslims in a particular geographical region are talking about online, [...] the frst thing that they need to do is follow the names and surnames, and they can just fnd out what I've been doing because of my name.It's like very, very Muslim, especially my surname (P2, Non-immigrant)" Furthermore, in our interviews, when we explored the topic of identity and privacy in more detail, most of our participants thought that among all the identity markers they possessed, their Muslim identity afected their privacy the most.We observed that younger male (self-identifed markers) participants of Pakistani origin were particularly concerned about the impacts of surveillance.Women, while explaining their concerns, often cited their experiences with additional checks performed on their male relatives while traveling due to their Pakistani origin, age, or name.This experience was echoed by participants P1 (Non-immigrant) and P15 (Naturalized citizen) (both of whom self-identifed as women).P15 shared, "I wear a headscarf, I think that has something to do with [how I am targeted].I think being female actually is an advantage.I think males are probably targeted much more than I am.(P15, Naturalized citizen)" On a similar note, P14 shared that she thinks that her Muslim identity afected her online privacy more than her being a black woman in America.
"I am Black African but born and raised in the US.But I am not too concerned about being targeted online for my Black identity.Because although Black people are villainized in the media and police brutality is a thing, I don't think that really afects me in the same way that my Muslim identity afects me.Because my Muslim identity is on a diferent scale.Like when we talk about police surveillance, I don't think they are surveilling black protesters the way that they would probably surveil a religious Muslim fgure.Because the connection between Islam and terrorism is on a larger, more international scale than the concerns that I have as a black woman in America.(P14, Citizen born)" 4.2.1 Experiences at/about border.Participants, when refecting on their anxieties related to information-seeking on the Internet, cited several incidents or details of their visa application and interview process.Almost all of our participants reported at least one instance where they felt discriminated against as Muslims and faced privacy breaches while crossing the US border.
P15, who is a naturalized citizen, mentioned that they expected to have the same level of privacy as other US citizens, but that was not always the case in their lived experience.They cited two incidents at the US border, 15 years apart, where their family was subjected to added surveillance because they were Muslims.
"There was a registry of all Muslim men that were living in the States and were not US citizens at the time and so you had to register yourself every single time you would go out of the country and then register yourself back into the country when you come in.And so my dad didn't know about this rule because it was not well known... when we returned, the entire family was detained.(P15, Naturalized Citizen)." According to P15, excluding people not because of any criminal history, but simply because of their religious afliation, and asking them to register every time they enter or leave the country is a huge breach of privacy.
"It is sort of comparable to the list of Jews that the Nazis made in World War 2.You are basically targeting a certain group of people, not because of the crimes they committed, but just because of their ethnicity and because of their religious afliation.My dad was a clean-shaven dude at the time.He is a hardworking immigrant.He was not very religious.After that incident, I knew that our privacy is up to the political environment and at the whim of the government here."(P15, Naturalized Citizen) Another incident experienced by P15 as recently as 2022 further bolstered her beliefs about the criminalization of certain groups at the border.She was randomly stopped and searched at the airport when returning from Turkey.The search involved ofcers going through her phones, laptops, and social media profles.P15 adds, "If they need a warrant to get into your house, how is it possible that they don't need a warrant to access your technology?(P15, Naturalized Citizen) " Similarly, P10 shared how her knowledge of the TSA impacts her thinking on the role other ICTs play in enhancing these mechanisms.
"I know that there's also databases within TSA systems [that] stop people, based on your frst names, and most of the frst names you see are South Asian or Muslim.So imagine now AI as a powerful tool can be used to generate even more Muslim names that are just going to be added to the top results for airport tracking systems.Many of us inevitably share location online, whether we realize it or not, just through our IP addresses as well.So now imagine all these location data to be very easy to access.And then imagine being able to deploy AI services in these areas for surveillance.These are very dystopian.(P10, Citizen born)" 4.2.2Targeted surveillance of Muslims through technologies beyond physical borders.P2 (Non-immigrant) was warned by their friend to not use the Muslim Pro app 5 , which is simply used to fnd out prayer timings every day and fasting schedules in Ramadan, as it had been reported to provide data to the US government [19].P1 (Non-immigrant) also came across a news article talking about the Muslim Pro app, the article mentioned that the app routinely sold data to the US government.Information such as this was a signal to participants that there are attempts at targeted surveillance of Muslims in the US even beyond immigration technologies.They lost trust in the app and stopped using it.
"Eventually, they said, don't use any app, just look up information online.It's available on diferent Masjid (mosque) websites or Islamic centers' websites.So don't download anything on your phone.Recently, I think there was this news that US law enforcement is using Muslim apps to sort of keep tabs on Muslims and us.So I never used it.(P2, Non-immigrant)" 4.2.3Sharing data with the government(s).Our participants explicitly connected their mistrust of applications to their knowledge of the makers and marketers of these applications, and then ultimately to the market's relationship with (the US) government.P10 raises the point of government involvement in data collected through ICT applications.
"I took a computer science ethics course and I wasn't surprised to fnd out this information is used to decide criminal justice situations, whose name is more likely to appear in airport system, stopping people for inspection etc.I wouldn't be surprised that social media companies would sell this data to the government for such things.It's unfortunate but a reality that Muslims are more aware of than others because we've seen it post 9/11 with the Patriot Act.And with AI, it's occurring at a faster rate.(P10, Citizen born)" Several US-based applications, P10 further noted, share the data they collect with other nation-states.P10 refected on her knowledge of US-based applications that share their data with the government of Israel.
"I know that some of these apps sell data to [other] governments like the Israeli government which makes it more concerning.As a benign user, I'm going on a prayer time app, it's probably a fne app, right?Then you fnd out that they share information with other places.It's scary to think that it could be used for something that you do not even know about.(P10, Citizen born)" 4.2.4Searching for/on websites leading to prosecution.The targeted surveillance as perceived and experienced by our participants is not without serious consequences.P12, who has been living in US for 35 years, explained that some websites are considered problematic by law enforcement agencies but continue to be present on the web.When searching the web for any information as benign as the lunar Islamic calendar, someone might unknowingly end up on those websites through search engines and get in trouble because of increased monitoring and surveillance."You can be, for example, searching for moon sighting, right?And in return you get a few links of websites, right?So you click on a link and you don't know if that site is, being tracked, who comes to this site and does what, right?So you can be in trouble for just going to that site.(P12, Naturalized Citizen)" As a result, P12 was concerned about the risks such websites posed to the wider Muslim community.
"I do feel that there should be a better way to handle these things.The site should be tracked for what they do, not the people going to the website.They can just end up there accidentally.(P12, Naturalized Citizen)" 4.2.5 Fearing misinterpretation of search history.Multiple participants shared concerns about their search history potentially being misinterpreted or misrepresented not only because they are Muslims, but also because they are frst or second-generation Pakistani immigrants or non-immigrants.They were aware that Pakistan is perceived by the US in a particularly negative light.P1 shared her concerns about being misinterpreted due to her nationality."I'm not a native English speaker.So sometimes I see things and I don't perceive them as what they mean, but then in my typing, I realize [that] what I did now sounds like a weird thing.And then I rephrase it...It is defnitely about identity not only as a Muslim, and woman, but I think I also come from a conservative country that doesn't have a lot of goodwill and good perspective.The world doesn't think good of my whole country in terms of the actions that it has been associated with in the past.(P1, Non-immigrant)" Participants with technical backgrounds draw on their technical understanding of how web searches and web browsing might be monitored to make changes to their web search practices.P1 added "I'm from a tech background, I know that even incognito doesn't help, nothing helps, no term helps (laughs).But yeah, I will say, not for Islamic content, but generally, I have seen myself sometimes rephrasing words in my head, so that, you know, I hope I'm not triggering something, though.Because I feel like there is a tendency to misrepresent and misunderstand.(P1, Non-immigrant)" P5 shared similar concerns about the diaspora their parents belonged to and their identities as a source of further targeting, making them careful about the articles they open.
"My national origin is the US but my parents are from Pakistan.So I identify as second generation Pakistani.There is already scrutiny on Muslims in the US and there is extra scrutiny on Pakistanis in the US.So the combination is not great.But when I hear news about someone with Pakistani heritage doing something in the US in the name of Islam, I am concerned about looking it up on the internet.If someone sees that I looked it up, they might get the wrong idea." (P5, Citizen born) 4.2.6 Searching for sensitive topics leading to increased surveillance.Multiple participants, including citizens, permanent resident and non-immigrants, shared concerns or expected to be surveyed about looking for politically sensitive topics such as abortion, LGBTQ, Palestine, and terrorism.Concerns were raised about how this knowledge might be used by someone monitoring their search activity.The entities they expected to have access to their search activity included the US government, foreign governments with whom the US government shared their data, and IT network administrators in hotels or their organizations."Also, views about Palestine are another thing to think about too.Since the United States has a huge connection with the Israeli government, and so that's also a concern I can imagine for some people, because there will be repercussions most likely for speaking against Israel.(P10, Citizen born)" Due to the stigma attached to Muslims, participants generally stayed away from topics like terrorism, shootings, or guns.However, for P1 this was not an option, as such topics are directly related to her work.
"I work in a space where I have to search even on my work machine on a lot of fatalities, ofcer-involved shootings, 911 calls, and things like that.And while it's a part of my job, sometimes I think about how does it come across?Somebody I was talking to [for work] said Tennessee had a shooting.She was telling me about an incident I had no idea about.And then I searched for Tennessee school shootings.And then I was like what does it come across?(P1, Non-immigrant) " This concern was also shared by P15 whose husband searches for topics related to guns as an enthusiast, but it makes her uncomfortable because it can be used against a Muslim.
"I always fnd it uncomfortable.And so I was telling him, hey don't be searching for these things online.Because as a Muslim, we just can't.It's diferent if an American is searching for it versus when a Muslim is searching for it, you know, it can be used against you.(P15, Naturalized citizen)" P2 mentioned that they are sometimes curious about certain topics after watching historical movies but either avoid searching for such topics altogether or have them at the back of their mind as they search.
Many Muslims say they do not look up news about people who are Muslims in the US because they are afraid of how they might come across.Therefore, they cannot look for things that might be happening in their own community or in their home country.P2 explained that he did not open news articles on terror attacks in Pakistan.
"Before moving here, I used to consume news a lot.I am mindful of what news article I am clicking on now.I'm scared.So I used to read, and still read, the Express Tribune to read the news from back home.If there is any article on, for example, TTP attacked a particular area in Pakistan, I don't always open that article.(P2, Non-immigrant)" Participants P5, P14 and P15 (all of whom are US citizens) shared that they did not search for news about Muslims in the US involving terror charges.Like many other participants, P14 shared that she does not follow politics anymore, but in the past when she did, she would think about following certain cases she was curious about.
"I'm from Minnesota, and there was a case where a couple of young men were arrested for potentially going to Syria and joining some group there.I was really curious about this.So I would follow the case online.In those situations as well in the back of your mind, you're like, God, I hope this doesn't look suspicious to whoever's monitoring me.(P14, Citizen born)" This was further echoed by P15."There was some case online about some woman that had some terrorism charges against her and I was just curious about and I think I was doing like some searches around it and it was probably not a good idea to be searching so much into it.Or into some movie that had come out like about 9/11 or anything like that.(P15, Naturalized citizen)" P13 was worried about the efect of the search on their family's green card status.Her concerns come from her parents being told by community elders that the fnances of Green Card holders and Muslims are tracked closely.They were asked for fnancial information when applying for a green card creating an association in their minds between fnancial activity and green card status.
"I am hesitant in searching about anything money related because I'm worried if it would look like I'm trying to avoid taxes.Back in high school, I tried to search how much high schoolers can earn before they have to pay taxes.I remember being hesitant about it because I didn't want to seem like I was avoiding taxes.I was like, "What will happen if someone fnds this out and uses it against me?" (P13, Permanent resident)" P8 who had a generally optimistic outlook on things felt that for certain search topics, Muslims are more susceptible to being redfagged and put under the radar than any other person searching on the same topic.
"I mean really I feel safe but of course, I feel like with the whole environment, it's very possible if you search something about Islam or religion maybe you are being monitored.When searching for something you need to be careful with being Muslim and an immigrant.I think it's not as bad as people think or as Muslims think but it's better to be careful to some extent.(P8, Naturalized Citizen)"

Ongoing conceptualization and sensemaking of datafed and surveilled selves
Participants with technical educational and professional backgrounds and varying immigration and citizenship status reported on events of "being seen as data [74]."These included dealing with visa application disclosures, arriving in the US, and more generally, time spent working and living in the US.This conceptualization often required taking a view from the government's perspective of what data the US government might have or might be continuously collecting about participants, how they might come across to an entity or person monitoring their activity, how and why information on participants might be used in potentially harmful ways, and how to modify their behavior to not trigger biases in surveillance systems.P1 explained her journey of conceptualizing what her relationship with the US government is; what data it has on her, and how her online search activity could be construed.
"When I came to the US I thought about it a lot more.And I will say that not only because I didn't know about the country, but I think they also didn't know about me.I will say, I started of a lot more paranoid than I am today.(P1, Non-immigrant)" The latter part of thinking about the level of familiarity that the US government might have with Muslim people is echoed by several of our participants' lived experiences in the US regardless of their immigration status.The lack of familiarity or data the US government initially had on them made them 'paranoid' about how their online search history could be misinterpreted by someone monitoring it, as they do not know them as a person, and they will be reduced to negative stereotypes about Muslims.
Building on the conceptualization of data relationship, P1 also thinks of how much data the US government has on her diasporapeople with not only her specifc religion but also national identity.
She sees each diaspora member as a data point for the Government and considers that since her diaspora is not very large in the US, it has fewer data points and hence more room for misinterpretation.
"They don't have a lot of data points from that country.I have met a lot of other people from the neighboring countries but they don't have a lot of us here, which doubles the problem.(P1, Non-immigrant)" 4.3.1 Recalibrating expression of Muslim identity.Some of our nonimmigrant participants learned from their American friends that tolerance for their particular Muslim religious identity and race varies between diferent states in the US.While in some liberal states and environments (e.g.academia), people are more tolerant of their disclosure of religious identity, this might not be applicable to the rest of the US.Hence, they should be careful about disclosing or expressing their Muslim identity.
"Now I am comfortable talking about my own religion because I've seen everybody's OK sharing who they are.But everybody tells me that this is Washington and academia.Once I get out of that bubble [I will] still need to recalibrate myself.(P1, Non-immigrant)" Although this advice may have been about ofine interactions, participants have applied it to monitoring on public networks at hotels and universities.P1 tried to anticipate any issues that might arise because of someone monitoring their search history while staying in a hotel.She considers people's (in)tolerance in any new place towards her specifc identity.Utilizing her technical knowledge about networks, she considered the possibility of both 'bad actors' or unwanted intruders on the hotel network and hotel IT staf themselves as having visibility into her online search and browsing and their reaction to it."Let's assume that the hotel is just keeping an eye on what everybody's searching for, and I hope nothing is going wrong in the hotel.But if this hotel had an unsecured network, then there's defnitely a chance of bad actors or other things that are possible there.(P1, Non-immigrant)"

Rationalizing concerns (or the lack thereof).
In the course of our interviews, participants were observed to actively refect on the rationality of their concerns.In addition, a wide range of concerns were shared in these interviews.For example, participants P4, P6, P7 (Non-immigrants), P8, P12 (Citizens) rationalized their lack of concern about information seeking by noting that they rarely searched for religious information or sensitive topics, so being monitored is not a concern for them.They did not consider themselves religious and were not members of any clubs or organizations.
"Someone can watch me all day long because I don't do anything wrong.But this is personal information and the good thing is religion is not part of your personal information in the US.So I don't particularly have any concerns, except I do believe that everyone should have the right to privacy and their information should not be shared with anyone else without their permission.(P12, Naturalized Citizen) " Several of these participants also noted that they were not involved in anything that US law did not permit.As a result, they did not think they needed to worry.They even felt comfortable with surveillance because it helps catch the bad guys.
"We want all the bad people to be caught and handled the right way.Over-surveillance and going into... they shouldn't be crossing the privacy lines, but a certain level of surveillance is good for everyone, I think.(P12, Naturalized Citizen)" On the other hand, 10 (of 15) of our participants were greatly concerned.Some of them rationalized their own concerns by connecting with sociopolitical events that impact the Muslim community in the United States as a whole.P5 shared the potential use of data in the future if attention turns to Muslims for any reason.
"If the radar is turned against Muslims again then I would be very concerned because they have access to. . .[all this data].Some world events might cause the radar to be turned to Muslims.When 9/11 happened, we didn't have all these tech companies and their ability to collect so much data [on us].Even with the Patriot Act, it was focused on Muslims.In the future, if there is a focus on Muslims, I am more concerned about that.(P5, Citizen born) " This aligned with P10's views on things as a Muslim born and raised in the US.P10 found a reason to be concerned as any world events involving Muslims turned the focus towards Muslims living in the US regardless of their nationality or immigration status.Several other participants, particularly those who had faced breach of privacy and grew up in the US after 9/11, echoed P5 and P10's concerns referring to the Patriot Act as a breach of the privacy of US citizens.
Furthermore, many relied on their academic training to make sense of ongoing harms conducted by state or market-driven datafcation.For example, P5 identifed to have a comprehensive knowledge of data privacy and storage.They shared their concerns about the potential use of app data for prosecution by the US government, as is being done in the case of abortion law to target women."I will give you an example of why I am concerned.There is no telling when the government and its policy will change from time to time.When Roe vs. Wade was turned over, some of the women were concerned about their data on period tracking apps.For some reason, that particular data is not protected under HIPAA.There are some concerns around if someone wanted to prosecute these women using that data for getting abortion, they can do that.After 9/11 when Bush passed the Patriot act it gave the government a ridiculous amount of access to data.It was clearly targeting Muslims.Nothing is stopping the government from accessing it if they ever wanted it.Any govt or prosecutor can access this data any time.(P5, Citizen born)" The constant sensemaking also had a range of impacts on the emotional well-being of our participants.Some of our participants who are born or naturalized citizens shared that they did not fnd these concerns 'debilitating.' Non-immigrants noted to have accepted surveillance as a given when living in the US, something which also happens to US citizens, and there is no workaround.P5 (citizen born) mentioned, "I am not concerned to a debilitating point but concerned at the back of my mind where I am defnitely aware of it.(P5, Citizen born)" However, several of our non-immigrant participants shared some instances where they felt a physiological response of panic by the thought of some information about them being disclosed accidentally to people in their physical proximity, through their social media surveillance by the US embassy, people they work with, or someone monitoring them.Foreseeing potential threats from bureaucratic processes or people's reactions scared them viscerally.P10 shared the experiences of her friends who had developed obsessive compulsive disorder (OCD) 6 .
"There are people who just don't have any social media.Then I know some friends literally diagnosed with OCD and have issues related to privacy and data.They were scared of using their phone literally.I remember one of my friends saying, 'It's a fear of being watched all the time.' (P10, Citizen born) "

Mitigating Information Barriers
Although our participants found that most search mechanisms in ICTs were fawed in a variety of ways, many of which left them feeling excluded from the broader information culture, they also came up with some solutions to mitigate these barriers.
4.4.1 Cross-referencing to prevent inauthentic interpretations of Islamic teachings.In case of religious misinformation (reported in section 4.1.2),participants shared that they cross-referenced multiple sources to ensure they had the correct interpretation.Those who had a local circle relied on ofine sources like local scholars.
"It is important to be cautious.I have a lot of connections with scholars in the local community and I make sure to validate with them as well around whatever pieces of information I fnd.(P10, Citizen born) " Those who did not have a local circle relied on multiple online sources for cross-referencing.For example, P1 developed the habit of cross-referencing all religious information found online to match various sources and ensure that the interpretation is correct.Although she used cross-referencingas a method to authenticate in her home country too, she came across blogs that suggested that many Islamic websites in the West are fake.Therefore, she doubled the sources she uses to cross-reference information she fnds online in the US.
"Not that Pakistan or East wouldn't have weird internet sources, but there was also a lot of TV channels, a lot of books you could buy in the shops, a lot of other sources like Ulema [scholars], you could go to anything.(P1, Non-immigrant)" 4.4.2Avoiding search topics, rephrasing search terms, managing search history appearances .As a workaround to the concern of lack of tolerance in diferent states (reported in section 4.3.1),participants self-censor their expression of Muslim identity within their search activity.While traveling, participants assess the safety of searching for religious information, interpretation, or opinion depending on their location within the United States.They hold of searching for religious topics which they think could be traced back to their Muslim identities in conservative states.
"I think it was something semi religious not like fully religious...I know that all hotels and networks are public.I was searching and I thought hm... do I need to search it here.I am not sure how everybody would react to it. . .I can avoid the situation altogether, I can just go home and search. . . it was just my perception that I was out of Washington state (P1, Non-immigrant)" This behavior extends beyond traveling.Participants avoid searching for news related to terrorism incidents in their home countries or anywhere at all, including not clicking on such news links.
Multiple participants do not click on news or articles that contain the word 'terrorism' because they are concerned that anyone 'watching them' may get the wrong idea or because the algorithms used for surveillance may pick up the frequency of certain words that appear in the browsing history, but do not take the context of the search into account.P2 says: "When I was applying for my visa here, we have to hand in all of our email accounts, phone numbers, social media accounts, while flling out the visa application.I just have this fear that they do have those data points and can, at any point, get the history of those particular email accounts.I know, my intention is to only read the news, because I am concerned about my family's safety.But the algorithm that keeps track of my history may not know that.(P2, Non-immigrant)" Some participants also avoid searching for religious information on the Internet altogether because they do not want to appear overly religious.P2 refects: "If I follow a certain school of thought in Islam, and I want to know what a religious scholar in that particular school is saying about an issue that I'm facing, for example, depression, or stress.I would rather look up for information from doctors than what a religious scholar is saying out of the fear of coming of as an extremely religious person, even though I am religious, I don't talk about it with everyone, let alone looking up that information online.(P2, Non-immigrant) " As a workaround, P2 would then go about searching for the same information in person by reaching out to a scholar.They would do this when they visited the mosque for Friday prayers.
P15 shared that she avoids searching for certain Muslim advocacy groups, such as the Muslim Legal Fund of America (MLFA) website, because she is aware that these groups are being monitored.
"I also tend to avoid MLFA, which is Muslim legal fund of America.There's nothing wrong with the group.I just tend to avoid it because there's a lot of people on it suing the government.When I was in Turkey also, there were a lot of groups that you don't know where they're originally from.So I would avoid more of those things.So anything overseas I tend to avoid as well.(P15, Naturalized Citizen)" 4.4.3Intergenerational practices around safety.Muslim families understand that Muslims are likely to be monitored in the United States.This understanding is shared among family members regardless of how the family is spread out geographically between US and other countries.For participants born and raised in US, their parents told them to be careful with their online behavior since they were a child.Their parents used to discuss which topics to avoid discussing in school or avoid typing certain things altogether on the Internet.P5 refects, "It has been my routine since I was a child.My parents talked to me about it.Navigating the Internet carefully became a thing even when I was at school and there were no smartphones.(P5, Citizen born) " Participants P5, P14, and P15 shared that being a child of frst generation immigrant parents, their parents had instilled a fear within them to be careful while using technology, for P15 particularly while entering and leaving the country.As P15's parents are not technologically savvy, she also warns them about being careful in forwarding or sharing messages sent by aunties and uncles about various conspiracy theories.
"As a child with frst generation immigrant parents, they have a fear that they instill in you as a child.We came in July 2001.September 11 happened pretty soon after.I grew up in that fear-based [environment] when Muslims were being targeted a lot, and there was also a lot of discrimination, even towards kids...I think my parents are not very technologically fast.I have to warn them sometimes about forwarding Facebook messages to aunties and uncles.(P15, Naturalized citizen) " P15 shared that she deleted Whatsapp completely from her phone because it is owned by Facebook that is known to share data with third parties.Her extended family members, as noted by P15, in Pakistan, discuss political issues and share news articles on Whatsapp that insinuate the US involvement in Pakistani politics.She did not want to be associated with these discussions.
"The concern there was not because I was doing anything on it, but just because there's anything on there that can be used against me when I'm entering the country or leaving the country that they could cause me trouble.So that's sort of the concerns that are kind of always in the back of my mind.(P15, Naturalized citizen) " For participants who had relatives both in the US and in their home countries, a list of Dos and Donts for living in the US was suggested by their relatives and friends who had lived in the US longer than them.These included not trusting certain Muslim apps and not accessing certain news articles.
"My friends told me not to use Muslim Pro [app], because it keeps tabs on you.Don't share anything religious.[Avoid] anything that has the word terrorism or anything Islamic extremism in the content even on digital mediums.(P2, Non-immigrant)"

DISCUSSION: TOWARD A CONCEPTUAL FRAMEWORK OF DIGITAL BORDERS
In this paper, we show that Muslims, a racialized and marginalized group in the US, often experience ICTs as 'digital borders' (Section 5.1) thus extending the place where physical bordering takes place (e.g.customs and border patrol at the airport) to online environments.In this section, we map our fndings to our theoretical grounding in border studies.In doing so, we ofer a few key characteristics to identify digital manifestations of border governance.This juxtaposition is particularly realized in the web-based information-seeking experiences of Muslims in the US (Section 5.1.1).Our fndings further show that borders afect mobility not only at physical borders but also across online spaces (Section 5.1.2).
In this analysis, we also show that our participants often internalized border imperialism, self-policing their search activities (Section 5.1.3).Despite the similarities among the participants, we highlight the sources of heterogeneity in how our participants experience digital borders (Section 5.1.4).We show that digital borders have epistemic implications (Section 5.2) as they limit the participation of Muslims in epistemic circles due to self-censorship in information disclosures for information seeking (section 5.2.1) and restricting themselves in their information seeking and search activities (section 5.2.2).

Experiencing ICTs as extensions of physical borders
Typically border studies use the term digital borders to refer to the technologies of surveillant assemblages used to enforce, enact, and materialize borders, such as the use of databases, facial recognition, AI, biometric systems, etc. [78] on borderlands, a practice known as 'securitization' [83].We argue that the experiences of borders are no longer limited to the physical or material geographies of land, but follows marginalized, racialized, and criminalized groups to the cyberspace as well.
It is important to note that our site of analysis is the technology practice of information seeking using ICTs.Technology practice is defned by the triad of artifacts, practices, and beliefs [10,33], and the technology practice of information seeking involves a variety of artifacts including websites, applications, search engines, and social networks.It is also worth noting that search engines are a gateway to the Internet, and consequently to online information.The frequency with which they appear in our interviews is indicative of our participants experiencing them as such.
Our interrogation of the technology practice of informationseeking leads us to uncover how search technologies on various platforms including search engines like Google Search pose as gatekeepers of cyberspace.The experiences many of our participants had at or around physical borders make them aware of their own rights to privacy in the face of ongoing border surveillance and control.Borders are then reproduced online whenever a visa application requests information on online activities from applicants.Similarly, criminalizing certain search topics by certain religious groups is reproducing borders online by nudging these groups to alter their search and browsing behaviors or face disproportionate consequences.

Enhanced Elasticity.
As suggested by border study scholars, "seeing" like a border involves considering the places where borders are enacted and by whom (discussed in section 5.1.3).Extending Walia's argument on elasticity of borders [83, p.78], we show that digital borders are elastic and follow our participants around wherever they are connected to the Internet and seeking information.We argue that digital borders are temporally elastic as the political climate also afects how they are experienced and expected by the participants (section 4.3.2);spatially elastic as they can convert or converge into the physical border when travelers devices are inspected at the airports (section 4.2.1);geographically elastic as they are coincidental with participants' frst interaction with the US visa and immigration processes in their home countries (section 4.4.2) but also because the global hegemony of surveillance capitalism allows visa approval or disapproval to be linked to search histories and social media activities.Once non-immigrants reach the US, they are aware of the possibility of surveillance and constant reevaluation for future visa renewals based on their digital footprint (section 4.4.2).
Our participants' ongoing sensenmaking of digital borders or the aspect of their activities being monitored is linked to their experiences with and information about the US visa application process, interviews at the embassies and at border, etc.These experiences directly contribute to their sensemaking about digital borders and are also evident of the enhanced elasticity of borders beyond physical and extending to the digital cyberspace.Hence, our participants' conception of what might be monitored, and trouble-inducing differs based on their experiences and knowledge.Sensemaking for our participants involves the constant conceptualization of the watcher and hence are experienced diferently by participants as they travel or work from home and ofces and use personal or organizationissued devices for searching information (section 4.3.1).
5.1.2Mobility and access.As discussed in section 2.3, borders have the ability to grant and restrict movement.We demonstrate that digital borders also control mobility not only in the physical realm but also across cyberspace by rendering some topics (section 4.2.6), applications, websites, communities and search behaviors potentially risking profling and surveillance, prosecution, and deportation for Muslims in the US (section 4.2).Customs and Border Patrol accessing the devices of international travelers' and going through their data and social media are a breach of privacy and an added form of surveillance on the physical border (section 4.2.1).Targeting of specifc Muslim people traveling to and from other Muslim majority countries indicates preemptive criminalizing and controlling their movement into and out of the US.It also indicates the confuence of physical borders with digital borders in cyberspace and the use of physical border crossings to breach into the private cyberspace of individuals in search of evidence.
Another way in which web-based information search technologies manifest as digital borders is the strengthening of orientalist views of Islam through location based search results (section 4.1).Said pointed to orientalist leaning permeating through majority of the Western mass media and academic literature.Our fndings (section 4.1.3)show that orientalist leanings of Western literature and information sources have permeated to search results as well providing a very distorted view of Islam and Muslims (section 4.1.2).Digital borders also manifest as the diference in access to knowledge bases across physical borders (section 4.1).In addition to diferences in location-based search results, our participants also experience a change in information ecologies between their home countries and the US which requires them to change their information practices.They have to rely more on online sources but at the same time they are concerned about how their search topics and browsing activities are categorized while running into the problems of misinformation and inauthenticity of sources.On the one hand, the change in information ecology increases the need for purposive information seeking and the dependence on ICTs to seek information for religious practice.On the other hand, searching for religious information online makes their Muslim-ness more visible and a potential target for racialization.

Internalized governmentality.
Borders are constituted by a range of bureaucratic processes.In case of digital borders, surveillance capitalism and privacy preferences sufer from information asymmetries where users are always less aware than the corporations about the data being collected and its use.Users rely on their interactions with the state bureaucracy to create their impression of the state's view of them.Johnson et al. [40] suggest that to understand borders, researchers adopt the perspective of 'seeing like a border' which acknowledges that borderwork is not only done by the state but by citizens and non-citizens too.We argue that some of the border work is being undertaken by our participants themselves by internalizing the state's view of acceptable behavior of a 'good' Muslim as a condition of residing in US.Borders can be as much perceived as they are tangible; "borders exist in our mind by virtue of the fear we have of the unknown of the 'there' and which, in turn, causes us to stay on our side of the border in the 'here'" [51].We add the perspective of how politically marginalized and racialized groups might perceive digital borders and the fear of the consequences of defying the rules of digital borders leads to change in information practices.These modifed information practices involve engaging in sensemaking to visualize their datafed and surveilled selves.The former involved thinking of the data points the US government might have on them (section 4.3).Sensemaking involves thinking about visa application experiences, news articles about targeted surveillance and deportation of Muslims, and community experiences of surveillance and their meaning for online surveillance of Muslims.Thus by policing themselves, our participants internalize borders and perform borderwork themselves by creating boundaries as to which information and information sites (websites, apps, online communities) they can or cannot access.Even as nobody holds down our participants or asks them to act in certain ways, they are engaging in self-policing watching or rephrasing their keywords for web searches, holding back from clicking URLs, avoiding online community groups, calibrating their search activity based on location, educating their friends and family and getting educated in the Dos and Donts of information seeking while living in the US.Digital borders are diferent from physical manifestations of borders in that they are more internalized to self than externalized.Our participants have internalized the notion of a border that they or their parents once crossed in their lifetime.
Singh [74] argued that high-resolution citizens fnd it easier to align their data with their way of life but at the cost of loss of privacy and surveillance.We argue that when the intent behind creating high-resolution residents is in fact othering of certain groups, it can create misalignment with their information needs.As we can see Muslims are made high resolution and highly visible by putting them in immigration databases, highlighting them in transit authority systems for inspection (4.2.1), and targeted surveillance of their communities, community leaders/members and through techonologies beyond physical borders (4.2.2).Such experiences with bureaucratic processes of the government make them aware of the State seeing them as a constant 'other' and a threat to security, which afects not only how others see them but also how they see themselves.These experiences of high resolution follow or strengthen their fears of targeted surveillance when seeking information online while at the same time being unaware of what might actually trigger a system to hyper-focus on them as an individual.Muslim residents then work to resist this hypervisibility through various ways in attempts to lower their resolution that range from non-use of certain apps or social media, avoiding certain websites, online community groups and news articles, selfcensoring completely to calibrating to place, time, device or websites.While Singh argued for data driven marginalization driven by low resolution, we argue that Muslims are facing marginalization through the creation of high-resolution residents.They are visualizing hyper-visibility through high resolution and self enforcing limits on the informational topics they explore, the sources they use or do not use, and their input to systems which lead to information retrieval (section 4.4.2).

Heterogenous experience of marginalization.
In our study, we see that even though the majority of our participants share similarities of South Asian origin, race and religion, their experiences of digital borders vary.Muslims in the US belong to diferent ethnicities, genders, races, socioeconomic groups, citizenship and immigration statuses, and countries of origin.Experience of digital borders may vary.Even within our small subgroup, digital borders are experienced diferently by our participants depending on their age, gender, nationality, educational background, technological literacy, number of years they have lived in the US, geographic exposure, and location within the US.Most of our participants are highly educated and multiple had advanced degrees in information technology disciplines providing them with a higher level of understanding of the workings of search engines, algorithms, surveillance, networks and use of technologies for social justice (section 4.3).Participants on non-immigrant visas were more careful and concerned about their online activity than citizens.Similarly, younger men were more cautious than women or older men.Participants of Pakistani origin were aware of the reputation of Pakistan and expected relatively higher scrutiny of any activity that might draw suspicion (section 4.2.5).We, therefore, do not claim generalization even within the subgroup of our participants.Heterogeneity within our dataset indicates that experiences of digital borders can be widely heterogeneous even when people do not have wildly diferent backgrounds and should be understood not based on people's backgrounds with respect to just nationality but also other sociopolitical markers that shape individuals' experiences within the nation-state.

Epistemic implications of digital borders
In this section we discuss the epistemic implications of digital borders for Muslims in the US.We discuss the resultant chilling efects of surveillance in the form of self-censoring and restrictive information seeking both of which lead to epistemic injustice for Muslims in the US by limiting their participation in the knowledge circles [46].Muslims' as knowers are marginally socially situated in the US making them vulnerable to capitalistic and government surveillance through search systems, creating a heightened sense of the resulting efects [77].On one hand, their interdependence on the knowledge accessible through these engines, created by the dominant groups, makes this knowledge less suited for their information needs (section 4.1.2,4.1.3),on the other hand, the self-censorship and limitations to freely accessing information without any fear of criminalization (section 4.2, 4.4.2,4.4.3) as chilling efects of surveillance prevent them from efectively participating in these knowledge circles both as knowers (section 4.2.6) and contributors [77] further strengthening the willful hermeneutical ignorance defned as "instances where marginally situated knowers actively resist epistemic domination through interaction with other resistant knowers, while dominantly situated knowers nonetheless continue to misunderstand and misinterpret the world"[57, p.2].

5.2.1
Rethinking self-censorship in information seeking.As discussed in section 2.2, the behavior of citizens is shaped through technologies of government where overt disciplinary power creates fear of negative consequences and covert control mechanisms encourage desired behaviors.In the same vein, self-censorship can also be a result of a censorship regime or exercised by the authors by their own volition.The driving force can originate either from the author himself or from the external regime [36] depending on who the censor and censee are and their interaction.When selfcensorship is in reaction to a public censorship regime, it is an instrument of censorship and referred to as public self-censorship.Whereas private self-censorship is an act of suppression by the censee themselves in the absence or irrelevance of a public censor, making censor and censee the same [17].We argue that when specifc groups are disproportionately targeted, surveilled, racialized and criminalized generally with a potential negative consequences for online browsing and search activities, they internalize self-censorship in information seeking (section 4.2.6).It might appear private in the absence of an overarching public censorship but experienced diferently by diferent groups through mechanisms of control described above forcing them to think like a nationstate, self-imposing rules of good, acceptable behavior (4.2.5, 4.2.6, 4.3.1, 4.4.2, 4.4.3).Public self-censorship in information seeking is, therefore, not a generalized phenomenon experienced by all US citizens but a form of internalized oppression and marginalization experienced by Muslims.Because of the elastic nature of digital borders and their presence across devices, networks, time, space and geography (section 5.1.1),as soon as Muslims engage in information seeking, they start imagining the watcher (government and its agencies, IT network administrator, employer, tech companies) (section 4.3), their datafed selves and managing their presentation to the watcher (section 4.3.1,4.4.2).

Restrictive information-seeking as an information access issue.
The problem of access to digital services for individuals, referred to as the 'digital divide' [6], is thought of as lack of physical access to digital devices, internet connectivity, etc. and commonly discussed in the development context.However, fnding meaning in technology by an individual also requires access to be able to consume, learn, critique, and make [34,50] technologies.We have shown that access can also be limited among highly educated users, due to internalized governmentality of marginalized and racialized citizens (section 4.3).Their access is not limited by lack of afordability, physical access, cognitive access, but by their social and racial location.The US is promoted as a free speech country, an information economy, promising the American dream and a nation of immigrants, with millions immigrating here out of necessity or aspirations in mind.However, Muslims moving to or already residing in the US are at a disadvantage when it comes to seeking information.Unlike traditional notions of development, increased physical access to information does not translate into social access, hindered by digital borders that manifest online for Muslims.

CONCLUSION
Muslims have been otherized and racialized in the United States for a long time now, a history that predates 9/11 [65].Through the exercise of the state's disciplinary power, Muslims have been disproportionately targeted by creating laws such as the Patriot Act.Surveillance exceptionalism already expanded the surveillant assemblages to intrusive lengths.
The asymmetries in power, knowledge, and malleability of privacy are growing between consumer and corporations in favor of the latter.For Muslims in the US, internalizing the state and the West's view of themselves infuence their use and perceptions of information communication technologies (ICTs).In particular, this is enacted by self-censorship in online spaces and search engines, leading to epistemic injustice where Muslims are not able to freely contribute to knowledge spaces or even access them for information retrieval.
Driven by neoliberal and racial logics, borders afect diferent groups diferently [83].In our paper, we demonstrate how Muslims, a racialized and marginalized group in the US, experience search engines as 'digital borders.'The connection between borders and online search afects mobility not only at physical borders, but also across online spaces.We urge social computing and HCI researchers to pay special attention to the ways in which ICTs perpetuate border imperialism, further isolating and otherizing the already racialized.

4. 1 . 3
Perceptions around search engine optimization (SEO) and its role in exacerbating Islamophobia.Some participants such as P1, P2, P5, P10, with academic and professional backgrounds in computing or adjacent felds, relied on their technical knowledge of search engines to hypothesize ways in which the problem of misinformation and Islamophobia could be exacerbated by the way search engine optimization (SEO) and location-based search results work.P10 is of the opinion that location-based search results can skew the perception of Islam in a negative manner by showing more Islamophobic content in the US.Based on her own experience of the high degree of variation in the results shown by the Google search for girls' education in Pakistan, P10 observed "orientalist leanings" in the search results, with Western authors dominating the results in the US and painting a failing picture of Pakistan."Ihad limited results, they [Pakistani friends searching from Pakistan] had more results.And in my limited results, most of the papers were written by people in Western academic settings and not necessarily people from Pakistani academics.(P10, Citizen born)"4.2Surveillance of Muslim lives and its consequences on information practicesAlmost all of our participants (both citizens and non-immigrants) believed that surveillance is part of their lives in the US; something they have accepted as reality.P5 (citizen born) specifcally shared that they expect the US to have a 'keen eye on Muslims, ' and further adds: "I think I have zero [privacy].I am sure the government is not interested in me as an individual [...] I go on about my daily life.But if they wanted anything they could have it from the press of a button.(P5, Citizen born)"

Table 1 :
Demographics of Participants, Muslims living in the US (Total : 15)