数字媒体，文化与社会系列 (Digital Media, Culture and Society Series)

In January 2020, Alphabet Inc., Google’s parent company, became the third technology company in the world to reach a market capitalisation of 1 trillion US Dollars (Swartz 2020), only following Apple and Microsoft. However, unlike its two rivals, most of Google’s user-facing services — including the two most popular websites in the world, Google Search and YouTube (Alexa.com, n.d.) — are free. Instead of charging users to use its services, Google reached its trillion-dollar status through advertising — the company reported 134.84 billion US Dollars in global advertising revenue in 2019 (Statista.com, n.d.). Google is transparent about this fact, stating clearly on its policies website that “advertising keeps Google and many of the websites and services you use free of charge. We work hard to make sure that ads are safe, unobtrusive, and as relevant as possible” (Policies.google.com, n.d.).

2020年1月，谷歌的母公司Alphabet Inc.成为全球第三家市值达到1万亿美元的技术公司(Swartz 2020)，仅次于苹果和微软。 但是，与两个竞争对手不同，Google的大多数面向用户的服务都是免费的，其中包括世界上两个最受欢迎的网站Google搜索和YouTube(Alexa.com，nd)。 谷歌没有向用户收费以使用其服务，而是通过广告达到了数万亿美元的地位-该公司报告称，其2019年全球广告收入为1348.4亿美元(Statista.com，nd)。 Google对这一事实是透明的，在其政策网站上明确指出“通过广告，Google以及您使用的许多网站和服务都是免费的。 我们会努力确保广告安全，不干扰且尽可能相关”(Policies.google.com，nd)。

As an avid Google user, I created my first Google Mail account when I started university in 2010. For the past decade, Google has kept track of my online activities and built a comprehensive profile of me, which can then be aggregated with millions of its other users and sold to advertisers. In its attempt to be transparent, Google allows users to access their personal profile, which Google creates using “personal info you’ve added to your Google Account, data from advertisers that partner with Google, and Google’s estimation of your interests” (adsettings.google.com, n.d.) This “data body” (Kitchin 2017: 177) is a unique opportunity to investigate the prevalence of surveillance capitalism and the implications of using predictive analysis to increasingly personalise user experience.

作为Google的忠实用户，我于2010年上大学时创建了我的第一个Google Mail帐户。在过去的十年中，Google一直跟踪我的在线活动并建立了我的全面个人资料，然后可以将该人与其上百万其他用户并出售给广告客户。 为了透明起见，Google允许用户访问其个人资料，Google使用“您添加到您的Google帐户中的个人信息，与Google合作的广告商的数据以及Google对您的兴趣的估算”创建的个人资料(设置。 google.com，nd)这个“数据主体”(Kitchin 2017：177)是一个独特的机会，可以研究监视资本主义的盛行以及使用预测分析来日益个性化用户体验的含义。

Using key ideas from Fuchs (2012), Kitchin (2017), Van Dijck (2014) and Van Dijck et al. (2018), this essay will investigate how Google employs “the popularization of datafication as a neutral paradigm, carried by a belief in dataism and supported by institutional guardians of trust” (Van Dijck 2014: 206) to enable and naturalise its business practice and economic surveillance of its users (Fuchs 2012). Furthermore, this essay will critique the data body that Google has documented using my personal online activities over the past ten years to highlight the asymmetrical power relationship between Google and its users.

使用Fuchs(2012)，Kitchin(2017)，Van Dijck(2014)和Van Dijck等人的关键思想。 (2018年)，本文将研究Google如何利用“数据化的普及作为一种中立的范式，它由对数据主义的信念所载，并得到了信任的机构监护人的支持”(Van Dijck 2014：206)，以实现并使其业务实践和自然化。用户的经济监控(Fuchs 2012)。 此外，本文还将对Google在过去十年中使用我的个人在线活动记录的数据主体进行评论，以强调Google与用户之间的不对称权力关系。

Google提供免费服务，但将用户数据作为商品出售给广告商 (Google offers free service but sells user data as a commodity to advertisers)

Google’s insistence that the presence of advertising “keeps Google and many of the websites and services you use free of charge” (Policies.google.com, n.d.) is a prominent rhetoric employed by social media companies to justify the continuous surveillance of its users. In this sense, users’ personal information and metadata are used as a widely-accepted form of currency in exchange for service (Van Dijck 2014: 197–198). The normalisation of this barter relies on the popular assumptions that online activities “naturally [generate] vast quantities of meaningful data” (Couldry and Yu 2018: 4482) and that personal data “has no value unless used” (Couldry and Yu 2018: 4477) and as Van Dijck (2014) notes: “few people appear willing to pay for more privacy” (p. 200). From Google’s point of view, however, “the economic value of personal data is largely realized in the aggregate — patterns and future potential become visible when data points are linked together through data analytics and algorithmic processing” (Cinnamon 2017: 614).

Google坚持认为广告的存在“使Google以及您免费使用的许多网站和服务保持正常”(Policies.google.com，nd)是社交媒体公司用以证明对其用户进行持续监视的正当理由。 从这个意义上讲，用户的个人信息和元数据被用作一种广泛接受的货币形式，以换取服务(Van Dijck 2014：197–198)。 这种以物易物的标准化取决于普遍的假设，即在线活动“自然[生成]大量有意义的数据”(Couldry和Yu 2018：4482)，个人数据“除非使用，否则没有任何价值”(Couldry和Yu 2018：4477)。 )，正如Van Dijck(2014)指出的那样：“很少有人愿意为更多的隐私付费”(第200页)。 但是，从Google的角度来看，“个人数据的经济价值在很大程度上实现了聚合-当通过数据分析和算法处理将数据点链接在一起时，模式和未来潜力就变得可见”(Cinnamon 2017：614)。

Since Google’s primary source of revenue is advertising, Fuchs (2012) argues that the users who create content and produce data on free platforms are both “prosumers” — producers and consumers of online content and services — and “an audience commodity that is sold to advertisers” (Fuchs 2012: 43). Advertisers traditionally favour large audiences and efficient return on investment, so Google’s strategy to offer free service is only to acquire a large user base, along with their personal information and web activities, which makes Google’s advertising business more attractive to advertisers. Dataveillance, therefore, is an important factor in Google’s business model. According to Fuchs (2012), “Google engages in user surveillance for the end of capital accumulation. Google surveillance is, therefore, a form of economic surveillance” (p. 45). Google’s proprietary algorithms are also sold to advertisers as efficient ways to sort users into demographic and interest groups, which help advertisers reach their relevant audience and optimise their return on investment. This strategy is presented to Google users under a different name: “personalisation” — making ads more “useful” to users (Adsettings.google.com, n.d.). Calling this practice “personalisation” manipulates users’ perspective and shifts their understanding of predictive analysis to be primarily beneficial to individual users, while at the same time turning attention away from Google’s advertising business as if it is only a secondary goal. As we can see, Google’s mission statement excludes the advertising side altogether. Personalisation, however, has troubling implications, which I will revisit in a later section.

由于Google的主要收入来源是广告，Fuchs(2012)认为在免费平台上创建内容并产生数据的用户既是“生产者”(即在线内容和服务的生产者和消费者)，也是“向广告客户”(Fuchs 2012：43)。 传统上，广告商偏爱广大受众和有效的投资回报，因此Google提供免费服务的策略只是获得大量用户群以及他们的个人信息和网络活动，这使Google的广告业务对广告商更具吸引力。 因此，数据监视是Google商业模式的重要因素。 根据Fuchs(2012)的说法，“ Google进行用户监控以终止资本积累。 因此，Google监视是一种经济监视”(第45页)。 Google的专有算法还出售给广告商，作为将用户分类为人口统计和兴趣组的有效方法，这有助于广告商覆盖其相关受众并优化其投资回报率。 该策略以不同的名称呈现给Google用户：“个性化”-使广告对用户更“有用”(Adsettings.google.com，nd)。 将这种做法称为“个性化”会操纵用户的观点，并将他们对预测分析的理解转变为对单个用户的主要好处，与此同时，将注意力从Google的广告业务转移到另一个目标是。 如我们所见，Google的使命宣言完全排除了广告方面。 但是，个性化具有令人不安的含义，我将在后面的部分中再次进行讨论。

Google平台呈现透明且中立 (Google platforms are presented as transparent and neutral)

Google’s mission statement says that its purpose “is to organise the world’s information and make it universally accessible and useful” (About.Google, n.d.). Note that Google uses the word “organise” to assume the role of a transparent looking glass through which users can have access to useful information. This language is in line with what Van Dijck (2014) calls “the ideology of dataism” — “a widespread belief in the objective quantification and potential tracking of all kinds of human behaviour and sociality through online media technologies” (p. 199). The alleged neutrality of big data was first introduced by Mayer-Schönberger and Cukier (2013), and scholars like Stephens-Davidowitz (2018) have argued that data and metadata collected from online platforms like Google Search are objective and indisputable displays of human behaviours and psychology. This argument minimises the role of the platforms through which user data is collected as if they are only “neutral facilitators” (Van Dijck 2014: 199). However, as Van Dijck (2014) and Van Dijck et al. (2018) noted, the proprietary algorithms employed by the likes of Google are highly secretive, selective and manipulative — popular practices such as “data filtering and algorithmic manipulation for commercial or other reasons” are well-established (Van Dijck 2014: 200). The norms established by Google’s platforms also communicate to their users what it means to be popular or legitimate in the Google-verse and therefore, actively shape users’ behaviours on these platforms. For example, search engine optimisation (SEO), the practice of manipulating search engine algorithms to yield clicks, is now a required knowledge for every marketing professional. Furthermore, data is never “raw” or untouched — “data are always already prefigured through a platform’s gathering mechanisms” (Van Dijck et al. 2018: loc 823). In this sense, Google’s role is not just to “organise the world’s information”, but also to actively shape our understanding of “information” and “the world”.

Google的使命声明说，其目的“是为了组织世界各地的信息，并使其普遍可访问和有用”(About.Google，nd)。 请注意，Google使用“组织”一词来充当透明玻璃的角色，用户可以通过该玻璃访问有用的信息。 这种语言与Van Dijck(2014)所说的“数据主义意识形态”相符-“人们普遍相信通过在线媒体技术对各种人类行为和社会进行客观量化和潜在追踪”(第199页)。 所谓的大数据中立性是由Mayer-Schönberger和Cukier(2013)首次提出的，像Stephens-Davidowitz(2018)这样的学者认为，从Google搜索等在线平台收集的数据和元数据是人类行为的客观，无可争辩的展示。心理学。 该论点最大程度地减少了收集用户数据的平台的作用，就好像它们只是“中立的推动者”一样(Van Dijck 2014：199)。 但是，正如Van Dijck(2014)和Van Dijck等人(2014)。 (2018)指出，Google之类的公司采用的专有算法具有高度机密性，选择性和操纵性-诸如“出于商业或其他原因的数据过滤和算法操纵”之类的流行实践已得到广泛确立(Van Dijck 2014：200)。 Google平台建立的规范还向用户传达了在Google-verse中流行或合法的含义，因此可以积极塑造用户在这些平台上的行为。 例如，搜索引擎优化(SEO)是操纵搜索引擎算法以产生点击次数的一种做法，现在已成为每位营销专业人员必备的知识。 此外，数据永远不会“原始”或被篡改-“数据总是已经通过平台的收集机制预先制成了模型”(Van Dijck等人，2018：loc 823)。 从这个意义上讲，Google的作用不仅是“组织世界的信息”，而且是积极塑造我们对“信息”和“世界”的理解。

The notion of “data body” — how users are represented based on their online activities — is also important to question here. Using its automated surveillance system and predictive algorithm to sort users into different social and interest groups, Google keeps minutely detailed data profiles of its users, which users can access and modify as they wish (Adsettings.google.com). Within Google’s business models, these data bodies are not personal, but only meaningful in the aggregate sense. Surveillance, predictive analysis and automated profiling are useful to advertisers, as they “make human behaviours and preferences calculable, predictable, and profitable in aggregate” (Cohen 2017: 14). Thanks to the belief in dataism, advertisers believe that they have access to accurate representations of their audience: “the data body not only claims to have ontological privilege but actually has it. What your data body says about you is more real than what you say about yourself” (Kitchin 2017: 177–178). Of course, this is not the case, as we have already established that platforms and their algorithms are highly manipulative and misleading (Van Dijck et al. 201 8; Kitchin 2017).

“数据主体”的概念(即如何根据其在线活动来表示用户)在这里也很重要。 Google使用其自动监视系统和预测算法将用户分类为不同的社会和兴趣组，从而保留了用户的详细数据资料，用户可以根据需要访问和修改这些资料(Adsettings.google.com)。 在Google的业务模式中，这些数据主体不是个人的，而是仅在总体意义上有意义。 监视，预测分析和自动配置文件对广告客户非常有用，因为它们“使人类的行为和偏好可计算，可预测且可获利”(Cohen 2017：14)。 得益于对数据主义的信念，广告商相信他们可以访问其受众的准确代表：“数据主体不仅声称拥有本体特权，而且实际上拥有它。 您的数据主体关于您的言论比您对自己的言论更真实”(Kitchin，2017：177-178)。 当然，事实并非如此，因为我们已经确定平台及其算法具有高度操纵性和误导性(Van Dijck等人201 8; Kitchin 2017)。

Google担任信任的监护人 (Google assumes the role as guardian of trust)

In terms of privacy rights, Google repeatedly asserts that it is a trustworthy guardian of users’ personal data and gives users the ability to control their own privacy settings (Policies.google.com 2020). Van Dijck (2014) points out that in order for users to surrender their personal data, “trust in the (institutional) agents that collect, interpret, and share (meta)data” (p. 199) is a crucial factor. Faith in Google’s ability to protect users’ privacy and shield user data from exploitation one of the most important factors that maintain the integrity of Google’s business model. However, according to Van Dijck (2014), trust is an “embattled notion in an ecosystem of connectivity where all online platforms are inevitably interconnected, both on the level of infrastructure as on the level of operational logic” (p. 204). The credibility of the platform is carefully constructed to help the platform acquire users’ trust and at the same time, manipulate their online experiences and commodify their data. For example, Google’s policies page states that they do not sell user data to third parties, and that “we keep your personal information protected and private” (Safety.google. N.d.). Google even goes one step further in allowing its users to control their privacy settings by adjusting what information that advertisers can see about them. However, turning off ad personalisation does not necessarily mean that Google will stop surveilling you or that you will stop seeing advertisements altogether, control is only an illusion in this sense.

在隐私权方面，Google反复宣称它是用户个人数据的值得信赖的监护人，并使用户能够控制自己的隐私设置(Policies.google.com 2020)。 Van Dijck(2014)指出，为了让用户交出其个人数据，“对收集，解释和共享(元)数据的(机构)代理人的信任”(第199页)是至关重要的因素。 对Google保护用户隐私和保护用户数据免遭利用的能力的信任，这是保持Google业务模型完整性的最重要因素之一。 然而，根据范·迪克(Van Dijck，2014)的说法，信任是“在连接生态系统中难以言喻的概念，在该生态系统中，所有在线平台都不可避免地相互连接，无论是在基础架构层面还是在运营逻辑层面”(第204页)。 该平台的信誉经过精心构建，以帮助该平台赢得用户的信任，同时操纵他们的在线体验并使其数据商品化。 例如，Google的政策页面指出，他们不将用户数据出售给第三方，并且“我们将您的个人信息保护为私人”(Safety.google。Nd)。 Google甚至更进一步，允许其用户通过调整广告商可以看到的有关他们的信息来控制其隐私设置。 但是，关闭广告个性化设置并不一定意味着Google会停止对您进行监视或完全停止查看广告，在这种意义上，控制权只是一种幻想。

A rhetorical analysis of Google’s privacy policy by Lindh and Nolin (2016) also found that Google’s presumed trustworthiness is “built around a crucial distinction between data and information” (p. 651), although these concepts are never officially clarified. While “personal data” or “user data” is not defined in Google’s current Privacy Policy (policies.google.com, n.d.), Google states that it collects information from users’ activities on its platforms, including (but not limited to) “things that you search for, videos that you watch, ads that you view or click, your location, websites that you visit, apps, browsers and devices that you use to access Google services” (safety.google, n.d.). Personal information is defined as “information that you create or provide to us”, including (but not limited to) “your name, birthday and gender; your password and phone number, emails that you write and receive on Gmail, photos and videos that you save, Docs, Sheets and Slides that you create on Drive, comments that you make on YouTube, contacts that you add, calendar events” (safety.google, n.d.). Lindh and Nolin (2016) also observe that when Google refers to “your data” or “user data” in its policies, it is usually communicated as a necessary feature that ultimately benefits the users and that it is not exploited (p. 651). This strategy allows Google to simultaneously presents itself as a public service and at the same time, exploit user data to serve its advertising business.

Lindh和Nolin(2016)对Google隐私政策的言辞分析还发现，尽管这些概念从未得到正式澄清，但谷歌的推定可信度“建立在数据和信息之间的关键区别之上”(第651页)。 虽然Google当前的隐私权政策(policies.google.com，nd)中未定义“个人数据”或“用户数据”，但Google声明其从其平台上的用户活动中收集信息，包括(但不限于)“搜索的内容，观看的视频，观看或单击的广告，您的位置，访问的网站，用于访问Google服务的应用，浏览器和设备”(safety.google，nd)。 个人信息被定义为“您创建或提供给我们的信息”，包括(但不限于)“您的姓名，生日和性别； 您的密码和电话号码，在Gmail上收发的电子邮件，您保存的照片和视频，在云端硬盘上创建的文档，表格和幻灯片，在YouTube上所做的评论，添加的联系人，日历事件”(安全.google)。 Lindh和Nolin(2016)还观察到，当Google在其政策中提及“您的数据”或“用户数据”时，通常会将其作为最终使用户受益的必要功能进行传播，并且不会被利用(第651页) 。 通过这种策略，Google可以同时将自己展示为一项公共服务，同时可以利用用户数据来为其广告业务提供服务。

Google作为经济监视 (Google as economic surveillance)

The three assumptions discussed above — that data is an accepted form of barter for free service, that data and platforms that collect them are objective, that platforms are institutions of trust — pave the way for dataveillance to be accepted “as a ‘normal’ form of social monitoring” (Van Dijck 2014: 2016). As Fuchs (2012) has argued, Google’s surveillance of its users is a form of automated economic surveillance, which is “networked, spreads and tries to reach all over the world wide web” (p. 45). An empirical study by Gomer et al. (2013) found that online tracking technology, most prominently cookies and third-party widgets (for example, “Like” or “Share” buttons hosted by news websites), are ubiquitous tools that enable automated, widespread tracking of users by only a few tech giants: “after visiting just 30 search results, the probability of getting cookies from all top 10 third party domains is 99.5%” (Gomer et al. 2013: 553).

上面讨论的三个假设-数据是接受免费服务的易货贸易形式，数据和收集数据的平台是客观的，平台是信任的机构-为数据监视被接受为“正常”形式铺平了道路社会监控”(范迪克2014：2016)。 正如Fuchs(2012)所言，谷歌对其用户的监视是一种自动经济监视的形式，它是“网络化，传播并试图遍及全球的互联网”(第45页)。 Gomer等人的实证研究。 (2013)发现在线跟踪技术，最主要的是cookie和第三方窗口小部件(例如新闻网站托管的“赞”或“共享”按钮)，是无处不在的工具，可以自动，广泛地跟踪用户科技巨头：“仅访问30个搜索结果，从所有排名前10位的第三方域中获取Cookie的可能性为99.5％”(Gomer等，2013：553)。

Google’s ubiquitous dataveillance practice is in line with the three characteristics of automated surveillance, which Andrejevic (2019) observes to be “‘operational’ in the sense that it privileges intervention over the symbolic power of the monitoring apparatus”, “‘environmental’ in its mode of governance and ‘frameless’ in scope” (p. 12). Thanks to the perceived objectivity of big data and technology, the end goal of automated surveillance, Andrejevic (2019) argues, is the “automation of judgement” (p. 12): pre-empting risks, identifying underlying patterns to predict future behaviours and attitude (Andrejevic 2019, Andrejevic and Gates 2014, Van Dijck 2014, Lyon 2012). In automated surveillance systems, the goals of data collection only come after the fact — hidden patterns and correlations reveal themselves in unexpected ways, which is why the scope of dataveillance is frameless and ever-expanding (Andrejevic and Gates 2014: 187). Andrejevic (2019) also observes that when algorithms fail to preempt risks or predict behaviours, the justification is always “attributed to incomplete or inaccurate information: the system just needs to know everyone better” (p. 11).

Google无处不在的数据监控实践与自动监控的三个特征相吻合，Andrejevic(2019)认为这是“可操作的”，因为它在监视工具的象征性权力上享有干预的特权，“在其监视工具的象征性特权中”治理模式和范围内的“无框架””(第12页)。 安德列耶维奇(Andrejevic，2019)认为，由于大数据和技术的客观性，自动化监控的最终目标是``判断的自动化''(第12页)：先发制人的风险，识别潜在的模式以预测未来的行为和态度(Andrejevic 2019，Andrejevic and Gates 2014，Van Dijck 2014，Lyon 2012)。 在自动监控系统中，数据收集的目标仅是事实-隐藏的模式和相关性以意想不到的方式展现出来，这就是为什么数据监视的范围是无框且不断扩大的(Andrejevic and Gates 2014：187)。 Andrejevic(2019)还观察到，当算法无法抢占风险或无法预测行为时，其理由总是“归因于不完整或不准确的信息：系统只需要更好地了解每个人”(第11页)。

我的Google数据主体 (My Google data body)

Sources used to generate my data body (Adsettings.Google.com, n.d.):

用于生成我的数据主体的来源(Adsettings.Google.com，nd)：

• You added this to your Google Account. Updates you make across Google services.
  您已将此添加到您的Google帐户。 您在Google服务中所做的更新。
• This advertiser shows you ads based on: Your visit to the advertiser’s website/app.
  该广告客户向您显示广告的依据是：您对广告客户的网站/应用的访问。
• Google estimates this demographic because your signed in activity on Google services (such as Search or YouTube) is similar to people who’ve told Google they’re in this category.
  Google估算了此受众特征，因为您在Google服务(例如Search或YouTube)上的登录活动类似于告诉Google他们属于此类别的人。
• Google estimates this interest, based on your activity on Google services (such as Search or YouTube) while you were signed in.
  Google会根据您登录后在Google服务(例如Search或YouTube)上的活动来估算这种兴趣。
• Google estimates this interest, based on your activity on non-Google websites and apps while you were signed in.
  Google会根据您登录后在非Google网站和应用上的活动来估算这种兴趣。

Among the five sources that Google uses to generate my data body, only two result from my direct disclosure of personal information or direct visit to the advertiser’s website. The three remaining categories are estimated using Google’s algorithm to determine my demographic information and interests. This Google data body, which Cheney-Lippold (2011) calls “a ‘new algorithmic identity’, an identity formation that works through mathematical algorithms to infer categories of identity on otherwise anonymous beings” (p. 165), displays the prominent logic of economic dataveillance: the use of predictive profiling to sort users into smaller categories in order to optimise economic value, the opaque system employed to surveil and sort users, and the imbalanced power relationship between users and the dataveillance system.

在Google用于生成我的数据主体的五个来源中，只有两个是我直接披露个人信息或直接访问广告客户网站的结果。 使用Google的算法估算剩下的三个类别，以确定我的人口统计信息和兴趣。 这个Google数据主体被Cheney-Lippold(2011)称为“一种“新算法身份”，一种通过数学算法推断出其他匿名生物上身份类别的身份形成”(第165页)，显示了经济数据监视：使用预测分析将用户分类为较小的类别以优化经济价值，使用不透明的系统对用户进行监视和分类，以及用户与数据监视系统之间的不平衡权力关系。

预测分析的使用 (The use of predictive profiling)

My data body is an example of Cheney-Lippold’s (2011) argument that “our identifications are largely made for us” (p. 165) in this online environment. From my online activities, Google sorts me into high specific demographic and interest groups (which are not necessarily correct). From Google’s and advertisers’ point of view, however, I am not an individual but a member of these specific demographic and interest groups, which made it easier to manipulate my online experiences. It is worth noting, however, that these are not “social” categories, per se, but categories whose definitions are “wholly-embedded within the logic of consumption, where categorical behaviours are statistically defined through a cybernetics of purchasing and research that marketers have deemed valuable for identification and categorization” (Cheney-Lippold 2011: 170–171). Being female, single, 25–34 years old, in this sense, are assigned with definitions that Google and its advertisers deem most profitable. According to Cheney-Lippold (2011), the logic of predictive profiling actively excludes us from choosing our own identities online, as well as what those identities mean.

我的数据主体是Cheney-Lippold(2011)论证的一个例子，该论证是在这种在线环境中“我们的识别码主要是为我们制作的”(第165页)。 Google通过我的在线活动将我归类为特定的特定人群和兴趣组(不一定正确)。 但是，从Google和广告商的角度来看，我不是个人，而是这些特定受众和兴趣群体的成员，这使我的在线操作更加轻松。 但是，值得注意的是，这些本质上不是“社会”类别，而是其定义“完全嵌入消费逻辑中”的类别，其中分类行为是通过营销商拥有的购买和研究控制论从统计学上定义的被认为对识别和分类有价值”(Cheney-Lippold 2011：170-171)。 在这种意义上，单身，年龄在25-34岁之间的女性被赋予Google及其广告客户最赚钱的定义。 根据Cheney-Lippold(2011)的研究，预测分析的逻辑积极地使我们无法在线选择自己的身份以及这些身份的含义。

Cinnamon (2017) and Yeung (2018) also warn us of the danger of sociocultural discrimination through predictive profiling using opaque, proprietary algorithms. Consumer exploitation practices like “differential pricing by retailers, predatory lending to vulnerable groups, racial profiling, and higher life insurance rates for those suspected of having a disease” (Cinnamon 2017: 616) are increasingly difficult to detect and challenge because each individual is served by a discriminatory digital shopfront disguised under the flag of “personalisation” (Yeung 2018). Thus, these practices continue to produce and reinforce social inequalities without being challenged.

Cinnamon(2017)和Yeung(2018)还通过使用不透明的专有算法进行预测分析来警告我们，社会文化歧视的危险。 诸如“零售商的差别定价，对弱势群体的掠夺性贷款，种族歧视以及对那些被怀疑患有某种疾病的人的更高的人寿保险费率”之类的消费者剥削做法(Cinnamon 2017：616)越来越难以发现和挑战，因为每个人都被服务由伪装成“个性化”标志的歧视性数字商店(Yeung，2018年)。 因此，这些做法在没有受到挑战的情况下继续产生并加剧了社会不平等。

Part of the reason that this discriminatory business practice goes largely unchecked is explored by Ruckenstein and Granroth (2020), whose research finds that “personalisation” is considered a natural part of the online experience — users only question it when personalisation is felt to be “intrusive, scary, or creepy” (p. 15). They found that when the works of surveillance and predictive analysis are incorporated seamlessly into the user experience, it can “remain imperceptible to those whose data traces are being used” (p. 14) and even create “pleasurable moments of being ‘seen’ [by the algorithm]” (p. 21).

Ruckenstein和Granroth(2020)探索了这种歧视性商业行为基本不受抑制的部分原因，他们的研究发现“个性化”被认为是在线体验的自然组成部分–用户仅在认为个性化是“个性化”时才提出质疑。侵入，吓人或令人毛骨悚然”(第15页)。 他们发现，将监视和预测分析工作无缝地整合到用户体验中时，“对于那些正在使用其数据跟踪的人来说，它仍然是不可察觉的”(第14页)，甚至创造了“可观的时刻”。通过算法]”(第21页)。

用于监视和分类用户的不透明系统 (The opaque system employed to surveil and sort users)

As displayed in my Google data profile, Google does explain the source of its categorisation but remains purposely vague in its language — the source of data or how Google determines my demographic information and interests are never clearly explained. Although users can choose to turn off categories they deem incorrect or unhelpful, Google does not allow users to challenge how its algorithm came to this algorithmic categorisation, or how its advertisers may use this categorisation to manipulate their online experiences. In other words, although Google appears to be transparent by disclosing this data profile, its system remains opaque, and therefore, highly difficult to critique. This opacity, according to Yeung (2018), is grounded in their complexity, their proprietary nature, and “the way in which they operate automatically, seamlessly, and unobtrusively integrate into users’ daily routines” (p. 264).

正如我的Google数据资料中显示的那样，Google确实解释了其分类来源，但故意用其语言含糊-数据源或Google如何确定我的人口统计信息和兴趣从未得到清楚的解释。 尽管用户可以选择关闭他们认为不正确或无用的类别，但Google不允许用户质疑其算法如何进行此算法分类，或者其广告客户如何使用此分类来操纵他们的在线体验。 换句话说，尽管通过公开此数据配置文件Google看起来是透明的，但其系统仍然不透明，因此很难进行批判。 根据Yeung(2018)的说法，这种不透明性的基础是其复杂性，专有性以及“它们自动，无缝且无干扰地集成到用户的日常工作中的方式”(第264页)。

Google’s insistence in its neutrality, transparency and trustworthiness while continuing to exploit user data and experiences highlight the imbalanced power relationship between Google (the platform that collects and effectively owns the data), its advertisers (those who have access to and can analyse the data), and its users (those who give up their data but have little power over how this data is collected or used). As Lyon (2014), Ruckenstein and Granroth (2020) and Silverman (2017) agree: “individuals have been made vastly more transparent, while authorities and corporations have become more opaque” (Silverman 2017: 149) and people “have few options not to participate as data-generating subjects” (Ruckenstein and Granroth 2020: 13). Although privacy policies are in place, Lindh and Nolin (2016) point out that they are not crafted in meaningful ways to users: companies only have them to prevent liabilities, and users rarely read or understand terms of use statements, which are often written in lengthy legalese, before accepting them. Although Google does enable its users to access and modify their data profiles as well as their privacy preference, it is not practised in any meaningful way: users’ options to resist tracking, avoid intrusive advertising, or gain insight in Google’s ecosystem of dataveillance, are still limited.

Google坚持其中立性，透明性和可信赖性，同时继续利用用户数据和体验，突显了Google(收集并有效拥有数据的平台)，广告客户(有权访问并可以分析数据的人)之间的不平衡力量关系。 ，以及它的用户(放弃数据但对如何收集或使用此数据几乎没有权力的用户)。 正如里昂(2014)，鲁肯斯坦和格兰罗斯(2020)以及西尔弗曼(2017)所同意的那样：“人们变得更加透明，而当局和公司变得更加不透明”(Silverman 2017：149)，人们“没有其他选择参与作为数据生成主体”(Ruckenstein和Granroth 2020：13)。 尽管已经制定了隐私政策，但Lindh和Nolin(2016)指出，隐私政策并不是以有意义的方式为用户制定的：公司只有它们可以防止负债，用户很少阅读或理解使用条款声明，这些声明通常以冗长的法文，然后再接受。 尽管Google确实允许其用户访问和修改其数据配置文件以及其隐私首选项，但并未以任何有意义的方式进行实践：用户可以选择拒绝跟踪，避免进行侵入性广告宣传或了解Google的数据监控生态系统仍然有限。

结论 (Conclusion)

Google continuously presents itself as a public servant whose ultimate goal is to benefit its users, whose advertising business only exists to keep its service free, who takes prides on its transparency and who empowers its users to control their privacy and personalisation options. This essay has critiqued the flawed assumptions of datafication that has allowed Google and similar platforms to thrive with this business model and naturalise their economic dataveillance, namely, the acceptance of personal data as a form of barter to use free service, the belief in the objectivity of big data and technology, the faith in Google as a guardian of trust, and the naturalisation of dataveillance. Furthermore, an analysis of my personal data profile also indicates that Google’s promises to privacy and transparency are merely rhetorical device employed to instil trust in its authority, these privacy and personalisation options are not actually meaningful or useful to users. The fantasy of convenience offered by personalisation also hides a sinister purpose of dataveillance, one that is deeply grounded in social sorting and discriminatory practice but goes largely undetected and unchecked because of their secretive nature. As Google continues to expand its services and advertising business, new definitions of digital privacy rights and systems that hold tech giants like Google accountable for their practice are required to ensure that Google actually works to benefit its users, just like it has promised.

Google不断地将自己定位为公务员，其最终目标是使用户受益，其广告业务仅是为了保持其服务免费，以其透明性为荣，并赋予其用户控制其隐私和个性化选项的权利。 本文批评了错误的数据假设假设，这些假设使Google和类似平台能够在这种商业模式下蓬勃发展，并使他们的经济数据监控自然化，即接受个人数据作为使用免费服务的易货贸易形式，对客观性的信念大数据和技术，对Google信任的监护人的信念以及数据监控的自然化。 此外，对我的个人数据资料的分析还表明，谷歌对隐私和透明性的承诺仅仅是用来灌输对其权威的信任的修辞手段，这些隐私和个性化选项实际上对用户没有任何意义或有用。 个性化提供的便利幻想也掩盖了数据监视的险恶目的，该目的深深扎根于社交分类和歧视性实践中，但由于其秘密性而在很大程度上未被发现和检查。 随着Google继续扩展其服务和广告业务，需要新的数字隐私权和系统的定义，要求像Google这样的技术巨头对其行为负责，以确保Google确实如其承诺那样，使用户受益。

This essay was written in May 2020, as a part of my assignments for my MA in Digital Media, Culture and Society at the University of Brighton, United Kingdom.

这篇文章写于2020年5月，是我在英国布莱顿大学获得数字媒体，文化和社会专业硕士学位的一部分。