Strong or thin digital democracy? The democratic implications of Taiwan's open government data policy in the 2010s

Abstract

What kind of “democracy” do new government-led digital initiatives facilitate? This paper discusses the issue by investigating the open government data policy in Taiwan in the 2010s, asking whether the policy encouraged “strong democracy.” Using interviews, written records, and an analysis of platform design, I argue that the implementation of Taiwan's open data policy has not institutionalized the engagement of civil society groups or ordinary citizens in government decision-making processes, which is at odds with the claims that open government data encourages “strong democracy.” Instead, open government data in Taiwan has facilitated monitorial democracy, which presupposes watchful but not active citizens, and neoliberal democracy, which presupposes profit-pursuing citizens. Both are more in line with “thin democracy,” which focuses more on individual rights and private interests than on participation and political community. The finding sheds light on why conservative governments around the world often embrace open government initiatives.

Keywords

Digital democracy open government data Taiwan monitorial democracy neoliberalism participatory democracy

Introduction

While the proliferation of data is often seen as fostering government surveillance of citizens (Brayne 2017; Parenti 2003; Tufekci 2014), recent trends in “open government” have promised citizens the right to engage in “sousveillance”—a “watchful vigilance from underneath” against government behaviors (Dennis 2008; Harvey et al. 2013). In the last fifteen years, one of the main developments in open government is “open government data” (OGD), a policy that requires governments to release data that can be freely used, modified, and shared by anyone for any purpose (Open Knowledge International, 2015). After the US President Barack Obama signed the Memorandum on Transparency and Open Government on his first day of office in 2009, open government data have been well institutionalized internationally and actively promoted by many governments in the world. In 2011, the global Open Government Partnership (OGP), led by the US and seven other countries, was created. Today, OGP already has more than 70 national members, and each member country publishes their Open Government National Action Plans every 2 years. Prominent international organizations, such as OECD, World Bank, and the UN, also joined ranks to promote open government data initiatives.

Through the free sharing of information between citizens and the public sector, open data promoters expect that it will aid the goal of “transparency, participation and collaboration” (White House, 2009), thereby having the possibility to facilitate “strong democracy” (Baack, 2015; Barber, 2004). The reality, however, is often not the ideal. Scholars have pointed out that OGD may result in “thinner” versions of democracy, such as “monitorial democracy,” which emphasizes transparency more than engagement (Ruijer et al., 2017), or neoliberal citizenship, which treats citizens more as consumers and entrepreneurs than civil society members (Birchall, 2015). Therefore, benign and empowering as it may seem, the implementation of OGD has complex implications for democracy (Worthy, 2015).

This paper aims to assess these contrasting views on the relationship between open government data and democracy by examining its policy implementation in Taiwan, including data selection and input, platform design, and data usage. Taiwan is a compelling case because it has been one of the most proactive countries in promoting open government data, ranking No. 1 in the Global Open Data Index in both 2015 and 2016–2017, ahead of the US, Great Britain, and other pioneering nations. This globally recognized performance makes Taiwan an exemplary case for studying the implications of OGD. If Taiwan's OGD implementation is considered one of the “best” by international standards, the effects on democracy there should represent OGD's potential to enhance democratic governance when they are adequately executed.

This paper focuses on the 2010s both because it was the period that experienced the most significant development of OGD policy in Taiwan and also because of the puzzling political contexts during this period. The Kuomintang (KMT) government started to introduce open government data policy in 2012, with the hope that it can boost economic growth and catch up with global trends of digital democracy (Chen, 2022). However, KMT's policymaking processes were widely characterized as exclusive and discouraging public engagement (Ho, 2014; Hsiao, 2019). Many large social protests emerged during this period to challenge KMT's undemocratic decision-making style, including the Sunflower Movement in 2014 (Fell, 2017). This phenomenon is not unique to Taiwan; many countries, such as the UK, Singapore, and Japan, have also proactively promoted open government data under conservative leadership (e.g. Stevens, 2019). Examining Taiwan's experience in the 2010s can thus help researchers understand the seemingly contradictory endorsement of open government data by conservative governments worldwide.

In the sections below, I first discuss the theoretical debates about the relationship between open data and democracy. I then describe my methods and the historical contexts of Taiwan's OGD policy. Next, I present the findings based on data I collected in 2019 and 2020, including interviews, documents, and observations of the open data platform. I find that that open government data in Taiwan promotes “thin” democracy that encourages citizens’ monitorial behavior and entrepreneurship, but not “strong” democracy that facilitates participation in the decision-making processes. Overall, this paper makes several critical contributions to the literature on digital democracy: it assesses the different views on the relationship between open government and democracy, provides a systematic framework to analyze data policies and applies it to a concrete empirical setting, and helps to understand why conservative governments might adopt open government policies.

Different views on the relationship between open data and democracy

Echoing the US White House's promise of OGD to promote transparency, participation, and collaboration, scholars point out the potential of OGD to facilitate democratic citizenship. OGD should create informed citizens, who are better equipped to actively engage in public discussion and affect policy outcomes (Baack, 2015; Janssen et al., 2012; Sivarajah et al., 2016; Yiu, 2012), and produce collaborative mechanisms between government and citizens (Baka, 2017; Bertot et al., 2014; Granickas, 2014; Susha et al., 2017). In other words, OGD paves the way for “strong democracy”—politics characterized by participatory decision-making processes and the creation of a political community (Baack, 2015; Barber, 2004). Although many of these scholars also mention barriers and challenges to achieve the ideal, they all recognize that OGD is essentially beneficial to democracy.

Other scholars, however, contend that open government data presupposes the ideal subject as “citizen–auditor–consumer–entrepreneur,” who takes over the responsibility of the state, thereby reducing the role of proper politics and participatory democracy (Birchall, 2015; Harvey et al. 2013). There are two major arguments in this line of critique: first, under the logic of neoliberalism, the government releases datasets not to enhance accountability and civic participation but to promote economic competitiveness and serve the interest of capitalists, who can profit from free datasets (Bates, 2012, 2014; Franceschetti, 2016); as a result, the democratic virtue of open government data is compromised in favor of economic productivity. Second, scholars also argue that the underlying notion of democracy behind open data—primarily focusing on transparency—reinforces a nonreciprocal concept of democracy that dismisses the engagement of citizens (Hansson et al., 2015; Moore, 2018). As Birchall (2015: 187) puts it, “it is made available rather than communicated; it pre-empts or intercepts communication.”

Put another way, the first line of critique suggests that OGD presupposes the neoliberal idea of citizenship, in which “all citizens … are encouraged to be enterprising and seize the opportunities of the global economy” (Kamat, 2004: 164). In the context of open data, neoliberal citizenship resembles what Davies (2010) identifies as the “individual choice/market participation” model. The second line of critique, on the other hand, points out the affinity between OGD and “monitorial citizenship,” in which citizens are “defensive rather than proactive,” and “they have more bits of information, but there is no assurance that they know at all what to do with what they know” (Schudson, 1998: 311). Both arguments are concerned about the potential that OGD may lead to “thin” democracy, in which “democratic values are prudential and thus provisional, optional, and conditional—means to exclusively individualistic and private ends,” and from which “no firm theory of citizenship, participation, public goods, or civic virtue can be expected to arise.” (Barber, 2004: 4)

The discussion above suggests that there are at least three models of democracy that may result from OGD policies: monitorial democracy, neoliberal democracy, and participatory democracy. First, in the framework of “monitorial democracy,” citizens are “alerted on a very wide variety of issues for a very wide variety of ends and may be mobilized around those issues in a large variety of ways.” Citizens don’t usually proactively seek out information, but they are attentive to what happens in society and are “poised for action if action is required” (Schudson, 1998: 310–311). Because citizens need basic information about a wide range of issues, the government's role is primarily to disclose information to them (Ruijer et al., 2017: 2). Even if the release of data itself does not necessarily entail monitoring behavior, information should be readily available such that citizens can learn about different issues when they choose to do so. However, since citizens are not considered active participants in policymaking, the government does not give civil society⁠ members much leverage in the decision-making process but sees them as watchful outsiders. While elected representatives and technocrats may consider public opinion, they ultimately retain authority over public policies. Thus, the emphasis of monitorial democracy lies on the ability to access information rather than on accountability and power-sharing.

Second, the “neoliberal democracy” model sees citizens as entrepreneurs and customers. The government provides data for entrepreneurs who can utilize datasets to make profits, and other citizens can consume the products and services that these entrepreneurs create. Compared to the impact of businesses, the role of civil society members is limited, for most are unable to create economic value.

Third, “participatory democracy” views citizens as active political actors who are able to engage in policy discussions and make policy decisions (Pateman, 1970, 2012); in other words, civil society groups are involved in what Fung (2006) calls “cogoverning partnership” with the government. In the era of digital media, this might also suggest that citizens and governments can collaborate to bring about common policies and projects (Noveck, 2009).

The first two models embody “thin democracy” because they emphasize individual rights and personal gains, whereas the last one expresses “strong democracy” because it emphasizes public goods, democratic participation, and the role of civil society. Of course, these three models do not exhaust all possibilities for democratic politics, but they are the most relevant to the proclaimed goals and actual consequences of open government data. Importantly, this paper focuses on the role of open government data in institutionalized politics, namely how citizens are included in governmental political processes, rather than how citizens use data to challenge and transform the existing system. Therefore, this paper does not discuss how open government data supports activism and protests outside of institutionalized politics or the self-organizing behavior of civil society (Meng et al., 2019).

This article is not the first to try to develop typologies about the relationship between open government data and democracy (for previous attempts, see Davies 2010; Ruijer et al., 2017; Sieber and Johnson, 2015). However, existing studies tended to treat citizens as a totality and did not distinguish between profit-making private companies and other citizens; in the present study, by establishing “neoliberal democracy” as a distinct typology, I highlight how private interests, rather than ordinary citizens or advocacy groups, can be the main beneficiaries of open government policies. Moreover, existing studies regarding OGD and democracy are mainly conceptual discussions, and most empirical studies have not gone into the details of policy implementation. By contrast, in this paper, I focus on the implementation of OGD in Taiwan to assess different theoretical views about the relationship between open data and democracy.

Methods

To explore the relationship between OGD and democracy in Taiwan, I collected data through three sources. First, I conducted interviews with 18 individuals familiar with the OGD policy in Taiwan from 2019 to 2020. The selection of interviewees was mainly based on the extent to which they were crucial to the policy process of open government data in Taiwan. I used two criteria to decide whether these actors were crucial: first, whether they were present in official documents or news reports regarding open government data; second, I also asked interviewees who they thought were important in the process. Based on these criteria, I interviewed 10 government officials, 6 tech community participants, and 2 researchers, which covered most crucial actors in Taiwan's open government data policy at least until 2016. Government officials include those from Board of Science and Technology (BOST), Industrial Development Bureau (IDB) of Ministry of Economic Affairs (MOEA), the Research, Development and Evaluation Commission (RDEC, later merged into the National Development Council or NDC), as well as top officials such as the former premier and ministers without portfolio who were responsible for the policy. In these interviews, I asked about the policy process of open data, the motivation for promoting open data, the experience in participating in the policy-making process, etc. These interviews took one to three hours and were all recorded and transcribed for analysis.

Second, I collect written regulations, official documents, reports, and news stories to supplement my analysis. For example, I used official reports on open government data conducted by the two leading agencies in the policy—Research, Development, and Evaluation Commission and Ministry of Economic Affairs—as well as reports generated by civil groups regarding the performance of open government in Taiwan. Finally, I also investigate the functioning of the open data platform as a material artifact to analyze the model of democracy inscribed in its platform design. In doing so, I follow the tradition of science and technology studies to treat material things seriously by examining their features and activities (Latour, 1987: Ch.2; see MacKenzie, 2019 as an example). The approach also echoes the “socio-materiality” theoretical framework, which emphasizes that “discourse necessarily entails materialization in some form” (Orlikowski and Scott, 2015: 12). In the next two sections, I will discuss my findings based on these data.

The historical background of Taiwan's open government data policy

Propelled by both the economic potentials of government data in the data economy, global normative pressures to promote better governance, as well as the demands from the civic tech community, the Taiwanese government initiated the open government data policy, under the rule of the leading party Kuomintang (KMT) (Chen, 2022). Building on the Freedom of Government Information Law passed in 2005, the Executive Yuan (the main executive branch of the government of Taiwan) passed the Strategies of Promoting Open Government Data in 2012, and in the following year, the first national open data platform of Taiwan was launched. The total number of open datasets had risen from 3377 in 2014 to 11,700 by the end of 2015 (Chuang, 2015), and Taiwan ended up ranked No.1 in the Open Knowledge Foundation's Global Open Data Index in two consecutive rankings. After the Democratic Progressive Party (DPP) took office in 2016, the new government largely continued existing efforts of open government data policy, as many bureaucrats in charge of the policy remained in office after the transition. During the first four years of the DPP government, there were little changes to the existing OGD policies.

However, the political engine of Taiwan's OGD policy in the 2010s—the KMT government—was widely considered by researchers as having conservative political agenda and lacking policy channel that civil society could access (Ho 2014). In fact, the largest social movement during this period, the Sunflower Movement in 2014, was to protest the KMT untransparent policy-making processes and “reject any agreement signed ‘in a black box.’” The conservative images of the Kuomintang government seem to be in direct contrast to the ideal of open government, which was supposed to enhance transparency, public participation, and collaboration. Yet, as Yu and Robinson (2012) point out, “open government” is ambiguous enough to allow for very different interpretations. While some initial promoters focus its potential to enhance political deliberation and civic participation, it can also be aligned with the political goals of “thin” democracy.

Indeed, my interview with one of the former prime ministers suggests that KMT elites had a particular vision about the relationship open data and democracy that focuses on transparency and access to information:

Democracy is being transparent, open to investigation, and keeping nothing as secrets, are these not just the requirements of democracy? Therefore, I think ordinary citizens can use open data to examine, let's say government service, is it that every part of it can be surveyed piece by piece? You don’t have to claim how uncorrupted or fair you are, open data can reveal whether you are really uncorrupted and fair.

However, this official was critical about giving decision-making power to people—the central tenet of participation democracy. This is what he said when he discussed the notion of “deliberation”:

So-called “deliberation” should not involve decision-making power. It is wrong that we made it associated with decision-making power. I think the “deliberation” in U.K is really just consultation, as it does not have decision-making power. Put simply, who gives you power? Where does your power come from? Deliberation can be deep and present both sides of opinions, but after that it only presents these opinions to the government. It is still the authority that should make decision, as they should take the responsibilities. If they think it makes sense, then they say “yes, and I’ll take responsibilities”; if not, they say “you’re wrong and I want to stick to my original plan,” then they’ll still have to take the responsibilities, and we’ll see how it plays out in the next election or the one after next.

The understanding of “deliberation” as mere consultation here is certainly different from most major theorists of deliberative democracy, which sees the influence on decision-making as important in the model of democracy (Dryzek, 2000). What is clear from these quotes is that for KMT elites, “democracy” is more about the goals of “thin” democracy like transparency and consultation than about “strong” democracy, namely participation from civil society in the decision-making process. However, for many social movement participants, a formally transparent policy process that includes open data and public hearings with no real influence on decision-making is still a “black-boxed” one. Therefore, underlying the ostensible disparity between seemingly democratic practices and an undemocratic government is the same “thin” notion of democracy, which does not challenge the political power of elites. Below, I will show that the implementation of open government data in the 2010s reflects such an understanding of democracy.

The implementation of OGD policies in Taiwan in the 2010s

Following the theories of “formatting” (Koopman, 2019) and bias production (Friedman and Nissenbaum, 1996), which posit that the politics of information can be analyzed from inputs, technical systems, to outputs, I will analyze the implementation of open government data policy through its three components: data selection and input, platform design, and data intermediaries. First, data selection and input is the process where the government decides which datasets to release on the open data platform. The process has important social consequences in terms of social inequality, as certain important datasets crucial for marginalized groups might not be available while other data might be published in ways that harm socially disadvantaged groups (D’Ignazio and Klein, 2020). The second process is platform design. Echoing Winner (1980) emphasis on choices of technological design, I will look into the features of the platform itself such as data formats, metadata (descriptions and categorizations of datasets), searching functions, and interactive design, such as places for citizens to request data.^[1] The third process is data intermediaries (Baack, 2015; Schrock and Schaffer, 2017) or infomediaries (Ulbadi, 2013), which refers to the process in which societal actors equipped with technical skills (such as media, developers, or NGOs) make use of and translate the open datasets on the platform, making them legible to the general public. The discussion of data intermediaries dovetails with the insight of science and technology scholars that the use of technology does not necessarily follow the design of it and might produce unexpected social consequences (Oodshoorn and Pinch, 2003). For example, widespread digital gaps might lead to the lack of projects that can benefit marginalized groups.

In each of these three processes of OGD policy, we can expect different outcomes according to different models of democracy (see Table 1). To repeat, monitorial democracy is about creating a rich informational environment for citizens to monitor various issues without their active involvement in policymaking processes. Therefore, in the data selection and input process, under monitorial democracy, the government would release datasets with limited civil society participation concerning what datasets they should release, as citizens are expected to know, monitor, and at most influence, but not participate in, decision making. In the platform design process, the government would ensure data formats and functions that are related to information display (e.g. data search and download) are well-equipped but does not guarantee the active interaction between the government and citizens. In the process of data intermediaries, the datasets would mainly be used by skillful citizens to promote better information presentation (such as data visualization), for the main objective of open data is to allow monitorial citizens to “scan the informational environment” more easily (Schudson, 1998: 311).

Table 1.

Three models of democracy and three processes of OGD implementation.

Models		Monitorial democracy	Neoliberal democracy	Participatory democracy
Subject citizen (vis-à-vis the government)		Watchdog/auditor	Entrepreneur/customer	Cogoverning partner in decision-making process
OGD Implementation	Data Selecting & Input	Government releases data with limited civil society participation	Government creates channels targeting businesses	Co-production of data between civil society and government
	Platform design	Information display	Information display about economic activities	Information display + Tools for collaboration
	Data Interme-diaries	Data used in better information presentation	Data used in profit-making activities	Data used in joint projects solving social problems

Second, the idea of neoliberal democracy centers around the pursuit of economic profits and entrepreneurship. Therefore, in the data selection and input process under the model of neoliberal democracy, businesses and startups would have special channels to request datasets, as citizens are expected to use these data to make profits. In the platform design process, the platform would ensure that data that are directly related to economic opportunities are of the best quality to be searched and used, as opposed to those that help citizens monitor government behaviors. In terms of data intermediaries, the datasets would be used by businesses to generate profits by creating applications or data services, and the government would support these business initiatives through some policy instruments (e.g. hackathons or subsidies).

Finally, participatory democracy emphasizes the participation of citizens in decision-making processes. Therefore, in the data selection and input process, the government would provide civil society with effective communication mechanisms about the types of datasets to release and even collaborate with civil society to produce some previously unavailable datasets. In the process of platform design, besides information display, the platform would also provide tools for interaction, participation, and collaboration, such as survey technology, discussion boards, wikis, crowdsourcing mechanisms, or social networking functions. And in the process of data intermediaries, datasets would mainly be used to assist the collaboration of governments and citizens in deliberative communication and joint projects to address public problems. For example, after the Aceh earthquake in Indonesia in 2016, government officials, local communities, and scientists came together to utilize the open data platform InaSAFE to cope with the disaster (World Bank, 2018). Similarly, a local government in the UK used open data in collaborative decision-making regarding adult learning fund allocation (Sivarajah et al., 2016). Below, I will use this framework to analyze the three aspects of OGD policy.

Data selection & input

From the beginning of Taiwan's OGD implementation, a government imperative to release as many datasets as possible was already salient. In 2012, when OGD was made official national policy, the government required that each ministry should release at least five open datasets in the first phase and 50 in the second; as a result, 1700 datasets were released by the end of 2013. From 2014 onwards, the total number of open datasets rose even more sharply, growing from 3377 in December 2014 to 12,200 in October 2015 (Research, Development and Evaluation Commission, 2015).

However, during this period, the decision-making process through which the government collected and released datasets was mainly internally driven within the government, thereby not fully considering the voice of citizens and civil society. The National Development Council (NDC), which oversaw intra-government OGD policy, urged ministries to release several open datasets as its crucial first step, regardless of whether these datasets were what citizens and civil society needed. As a former NDC official recalled: “Back then, I didn’t care so much about the importance of data, neither did I care about the value of data. We asked bureaus to release data first to test the waters, through which we wanted to assure them [that the policy was plausible].” A related story was from the Ministry of Education: when it was required to release five datasets, it provided the addresses of elementary schools, junior high schools, senior high schools, and universities, which were too basic to entail meaningful actions (Liberty Times, 2016). On the other hand, in 2014, a controversy over changes to the national curriculum occurred in Taiwan, and the Taiwan Association for Human Rights demanded that the government release the minutes of the meetings where decisions were made. The Ministry of Education rejected the request, and a judicial solution was then sought (Taiwan Association for Human Rights, 2016). This incident suggests that while the government proactively released datasets, it did not necessarily do so when citizens requested certain data.

That the government emphasized releasing data but not incorporating the voices of citizens into data selection and input processes reflected the idea of monitorial democracy. That is, the government only needs to provide a platform that provides as many datasets as possible, which ensures that citizens have access to “more bits of information” that enable them to “scan the informational environment” (Schudson, 1998: 310–311). On the other hand, participation from civil society did not fit the government's vision of monitorial citizens—the ones who often see, are willing to know, but seldom take actions, who talk about politics but don’t want to engage too deeply into it—and was thus not what the government should guarantee. In other words, by providing more information but not including civil society into the decision-making process, the government could create an environment in which “monitoring [was] abundant but accountability scarce” (Graves 2017: 1248).

After 2015, the government recognized this problem of insufficient participation and tried to solve it by establishing Open Data Consulting Groups, which were made up of officials and representatives from tech communities, scholars, and social groups, to encourage communication with citizens in the data selecting and input processes in each ministry. However, since meetings were held only every several months and the function of the consulting group was mainly consulting, group members could hardly influence what datasets the ministries decide to release. Jack, a prominent open data advocates in Taiwan, who was once highly engaged in these consulting groups, noted that the influence of the consulting group was “extremely small” because even though he could criticize the government and request something in the meeting, whether his opinions would be seriously considered “depended on the chairperson,” and often the meetings were just “everybody expressing their ideas in turn.” His experience suggests that citizens in these consulting groups were still treated as outsiders rather than active decision-making partners; consequently, even though the personal distance between citizens and officials might shorten, the burden of data selection and input remained predominantly (if not exclusively) on the shoulder of government officials. Therefore, communication intensive as it might seem, the consulting groups were far from achieving “participatory democracy.” After all, participatory democracy is essentially a way of collective decision making, not just consulting. If decisions depend mainly on the will of government officials, the “consulting” process can hardly be seen as a practice of participatory democracy. Moreover, genuine policy discussion, which is reflected in the idea of “deliberation,” should be “mutual communication that involves weighing and reflecting on preferences, values, and interests regarding matters of common concern” (Bächtiger et al., 2018: 2). However, if there are no procedures to ensure a genuine civil society-government discussion, and people are just expressing ideas without dialogue, the participatory effects of these meetings are even weaker.

The only non-governmental force that has substantial leverage on decisions regarding data release is business. In 2013, the Ministry of Economic Affairs (MOEA) created a government-business coalition, the Open Data Alliance, which held “open data interactional forums,” through which business leaders had direct channels to request the government to release certain datasets. The government was required to respond to these demands: MOEA would collate these requests, and the NDC would ask ministries to release those requested datasets. For example, pharmaceutical companies requested information regarding drug and health product inspections and the use of medical insurance, while shipping companies requested real-time data of vessels near Taiwan as well as the import and export of sea and air cargo (Ministry of Economic Affairs, 2015). As a result, 207 requested datasets in 2015 have been or will be opened (among 409 requests), and the figure was 222 (among 492 requests) in 2016 (Ministry of Economic Affairs, 2018)—notably, there are no corresponding official statistics for civil society or ordinary citizens. As a long-time civil society advocates for open data Richard observes, “because of the influence of the Ministry of Economic Affairs, Taiwan only has industrial policies but not public policies in this regard [open data].” The asymmetric channels between businesses and civil society in the data selecting and input process point to the underlying assumption of neoliberal citizenship—citizens are treated as entrepreneurs rather than civil society advocates.

To sum up, in the process of data selection and input, one can clearly see the expression of monitorial and neoliberal democracy; taken together, citizens are supposed not to actively engage in public affairs but are thought of as willing to take advantage of new economic opportunities. “Strong democracy” is still at best symbolic—while the government seems to have established communicative mechanisms, the function is far from the ideal of participatory democracy.

Platform Design

In this section, I will examine the features of Taiwan’s open data platform (data.gov.tw) to see how the platform design fits into the three models of democracy. The most basic function of an open data platform is information display. In this regard, scholars have pointed out the importance of data provision, data search and download, and metadata, among many others (Charalabidis et al., 2014). First, data provision is the most basic function of the open data platform, which requires a platform to display a great number of high-quality and reliable data from different thematic areas. Second, data search and download, embodied in functions like searching toolbars and proper data formatting, is crucial to information display because it lets users find and view the relevant information they need. In addition, metadata, which refers to a set of data that describes and gives information about the open datasets on the platform, is also important in information display, for it structures information and makes datasets more accessible, discoverable, and easier to interpret and link with other datasets (Zuiderwijk et al., 2012).

One can easily see Taiwan's open data platform perform well in data provision and data search and download. In terms of data provision, the number of datasets in the platform has exceeded 40,000, and free and up-to-date data are publicly available in a wide range of topics such as government budget, national statistics, procurement, air quality, election results, land ownership, etc., as shown in OKFN's Global Open Data Index, in which Taiwan is ranked No. 1.^[2] For data search and download, the platform provides the basic searching toolbar and download functions, and more than 64% of datasets are released in machine-readable formats (Radio Taiwan International, 2019). More importantly, Taiwan's government has established a detailed regulation on metadata,^[3] ensuring that all datasets on the platform are categorized and indexed properly and discoverable by citizens. These features have certainly provided monitorial citizens with abundant information for “scanning the informational environment.”

However, despite generally good information display, the quality and usability of data on the platform are not the same across datasets of different topics. As Taiwan Open Government Report (a civil society-led assessment of open government in Taiwan) points out, the datasets about economic activities are more complete and of high quality, whereas datasets about law and organization, government accountability, and government operations scored lower in the report's assessment (Lee and Tseng, 2017). This disparity reflects the underlying neoliberal idea of democracy behind the OGD implementation in Taiwan: the government is more likely to encourage entrepreneurship and economic activities rather than strengthen democratic participation.

On the other hand, in terms of “strong democracy,” the performance of the current platform is unsatisfying. The main interactional channel on the platform is called “I Want More Data,” through which citizens can request datasets that they want the government to release. After receiving the request, related government officials should reply in the “response” column and explain whether they will open the dataset. Nevertheless, a civil society-led project called “Gov Data Opener” has analyzed government responses to these requests and found that the government promised to release less than 20% of all requested datasets (Gov Data Opener, 2019). Chang, the project leader of Gov Data Opener, summarizes three major problems of government-citizen interaction on the platform: many government agencies simply didn’t respond to citizens; among those who did respond and reject the requests, the reasons provided were often unreasonable or ambiguous; and even if the government agreed to release the data, they often didn’t specify the timing of release or provide updates to the citizens requesting the dataset (Chang, 2020).

This lack of effective responses suggests that the design of the interactional channel on the platform is unfavorable to deeper participation—we hardly see policy discussions on the platform, not to mention attempts for collaboration to overcome the obstacles to data releasing. More specifically, merely designing a “response” column is prone to one-time “request-response” interaction without following accountability, and this is why Gov Data Opener asked the government to re-design the platform to incorporate modules that facilitate unambiguous responses – for example, a checkbox of reasons for not providing data, and a system to track government promises. In addition, various tools for live video streaming, ideation and brainstorming, collaborative document writing, etc. (as listed in Participate DB: http://www.participatedb.com) would also facilitate deeper citizen engagement.

In sum, the design of Taiwan's open data platform allows the platform to display information well—and especially well in economy-related data—but fails to facilitate deliberation and collaboration. As a result, the platform design is still primarily reflecting the idea of monitorial and neoliberal democracy: the government gives citizens high-quality information, encourages them to pursue profits, but does not treat them as decision-making partners.

It should be noted that these features of Taiwan's open data platform in the 2010s were not intrinsic characteristics of the technologies themselves but rather the result of intentional design processes. In the Written Report on Research and Analysis on Value-added Applications of Open Government Data, a research report commissioned by the government and conducted by a private enterprise in 2012 to assist with the implementation of open government data policy, the report recommends that the most essential function of the platform is that it should be “easy to search and use.” It provided seven recommendations: adopting an open platform architecture, following relevant Internet standards, focusing on data platform interoperability, clearly labeling license terms and usage agreements, ensuring ease of use and stability of data access, providing APIs and mass access to data, and including a user feedback mechanism. Most of these recommendations aim to ensure the quality and accessibility of data, aligning with the concept of monitorial democracy that emphasizes information display and provision. The “user feedback mechanism” mentioned does not refer to substantive participation and government-citizen interaction but rather focuses on “the correction of errors in a single piece of data or the optimization of fields/formats of the whole data set (e.g. adjusting an EXCEL file from a two-star file to a three-star CSV file)” (Research, Development and Evaluation Commission, 2013: 62–63).

Moreover, as suggested by the term “value-added” in the title of the report, the emphasis of the report is on the goal of economic growth behind the open data platform. In the introduction of the report, the author argues that “access to government information not only enables officials and government organizations to be monitored, but also generates economic benefits, whether in the form of direct savings in government expenditure, improved service delivery, more efficient procurement, or indirect economic and other benefits to the community.” Given the emphasis on economic benefits, it is not surprising that data related to economic activities are of better quality on the platform.

Data intermediaries

Finally, I analyzed data intermediaries—the process through which people with technical and communication skills use datasets to produce relevant information for general audiences. I visited the “Applications” area on the open data platform in December 2019, which displayed 81 applications of open data by citizens, to see how these datasets are used. Projects aligned with the ideal of “monitorial democracy” should present information in an easily understandable way to assist citizens in monitoring the government without engaging in policy decisions, such as through data visualization and information searching. Conversely, projects aligned with the ideal of “participatory democracy” should utilize open datasets to transform existing policymaking processes, such as through data-based policy deliberation or collaborative problem-solving.⁴

I find that among the 81 projects displayed on the platform, 23 are data visualization projects, the topics of which include air quality, natural disasters, legislator performance, epidemics, wi-fi locations, etc.; 49 are information searching projects for public transport, weather, animal adoptions, restaurants, prices of agriculture products, tourist attractions, and other themes; and 9 are related to personalized services, such as calorie control, tools for electronic receipts, health management, etc. Many of these projects, especially data visualization and information searching projects, are useful for “environmental surveillance”—through which watchful, monitorial citizens can “keep an eye on the scene” (Schudson, 1998: 311)—but it is also astonishing that given the abundant data, none of these projects create channels for deliberative discussion or collaborative problem-solving. For example, many projects are related to air quality monitoring, suggesting that environmental pollution is a big concern for many citizens; nevertheless, these projects only visualize, enable searches, and notify, but do not encourage evidence-based collective problem-solving. The phenomenon clearly illustrates the underlying assumption of “monitorial democracy” behind most open data usage in Taiwan.

On the other hand, official documents and interviews suggest that open data usage by businesses is significantly encouraged by the government. The Ministry of Economic Affairs (MOEA) initiated plans to promote economic usage of open data, the instruments of which include: (1) competitions, workshops, and hackathons, in which startups and techies use the open datasets provided by the government to produce new ideas and applications; and (2) subsidies based on the proposals that businesses submit (Ministry of Economic Affairs, 2015, 2018). A former official in the government, Mike, observes that “the major target of the government is businesses […] in other words, to promote new industries and new sources of profits.” Similarly, in a report published by the Ministry of Economic Affairs to summarize the implementation results of OGD, it boasted that “through subsidies, we encouraged 15 businesses to utilize Open Data to innovate service models, assisting enterprises in enhancing decision-making quality and strengthening competitiveness, thereby driving private investment and generating derivative revenue of 120 million NT dollars” (Ministry of Economic Affairs, 2018: 239). The tendency to focus on profitability again suggests that the idea of neoliberal citizenship is salient in the process of data intermediaries.

Up to 2018, no such institutionalized mechanisms were established to promote usage from civil society groups, not to mention efforts to increase the collaboration between citizens and governments on open data usage. However, the Taiwanese government started to hold “presidential hackathons” from 2018, which provided citizens opportunities to propose projects based on open datasets and collaborate with the government to solve issues like disasters, youth employment, and social security. The government promised that the five best proposals selected by the evaluation committee, which consisted of both government officials and civil society actors, will be prioritized to be implemented. Although the new initiative did suggest a turn to more participatory democracy, there are still important limitations. First, because the hackathon allowed government agencies to participate, many proposals were led by the government and lacked the participation from citizens. For example, if one looks into the top five final proposals that were selected by the committee in 2018, four of them were proposals by government or quasi-government agencies. If only one project that has genuine citizen participation is selected each year, it is hard to argue that there is a robust relationship between open data and participatory democracy. Moreover, even though the government promises to implement the proposals, there is no legally binding power that can make sure the government implement them as proposed by citizens. The decision-making power, in the end, still lies in the hands of government officials.

To sum up, the fact that most projects that used open data were used for information display, and that the government did not encourage participatory and collaborative data usage, suggest that the components of participatory democracy remain largely absent in the process of data intermediaries. Therefore, in terms of data intermediaries, we can see a similar pattern between open government data and democracy: the applications of data are more emblematic of the notion of monitorial and neoliberal democracy. There has been attempts to advance participatory democracy after 2018, yet the effects have been fairly limited as of 2019.

Conclusion

In this article, I argue that the implementation of the open data policy in Taiwan in the 2010s has not fully incorporated civil society groups and ordinary citizens in government decision-making processes, which is at odds with the claims that open government data can achieve “strong democracy.” Instead, open government data in Taiwan have facilitated monitorial democracy, which presupposes watchful but not active citizens, and neoliberal citizenship, which presupposes profit-pursuing citizens. Both are more in line with “thin democracy,” which focuses on individual rights and private interests rather than on participation and political community. This general argument is consistent throughout the three processes of the operation of the open data platform: in the process of data selecting and input, the government emphasizes data releasing per se but not civil society engagement, and only businesses have real leverage on the decision-making process; in the process of platform design, the open data platform has good information display functionality (especially in data related to economic activities) but lacks effective interactional mechanisms; and in the process of data intermediaries, data are more likely to be used in information transformation and profit-making activities than in participatory and collaborative problem-solving. In conclusion, instead of facilitating “strong democracy,” the OGD implementation in Taiwan has created an open government with limited civil society participation and collaboration, a far cry from fulfilling the original promise of OGD policy.

The analysis in this paper contributes to the literature on digital democracy and open government in several ways. First, by establishing the typology of “neoliberal democracy,” I emphasize that “citizens” are not homogenous; rather, the “citizens” that are empowered by open government data in Taiwan are predominantly businesses and private interests, rather than civil society groups or ordinary Taiwanese, highlighting the asymmetrical power dynamics among different stakeholders. Second, while previous studies on open government and democracy are mostly conceptual discussions, I provided a systematic empirical framework to analyze data policies, which encompasses data selection and input, platform design, and data intermediaries, that can be applied elsewhere.

Third, this paper contributes to the literature on digital democracy by assessing different views on the relationship between open government data policy and democracy. In contrast to the optimism of many open data practitioners who believe that new technologies releasing government data can fundamentally enhance democratic participation, my findings suggest that even in a “successful” case like Taiwan, the implementation of open government data largely aligns with Birchall (2015: 191) argument that “[the] dual function of open data—to answer the demands of democratic accountability and economic growth—configures the imaginary identity of the ideal data subject as a citizen–auditor–consumer–entrepreneur.” Put another way, the ideals of monitorial democracy and neoliberal citizenship are intertwined, which, taken together, transfer the burden of government bureaucracies to the market and depoliticize the public sphere into an arena of transparency without participation and effective response.

Finally, the close coupling between open government data and thin democracy helps explain why conservative governments can easily accept and even promote open data. For example, other studies show that countries like the UK emphasized transparency and the use of digital technology more than collaboration and participation when they promoted open government (De Blasio and Selva, 2016). Similarly, semi-authoritarian countries like Singapore promote open government data in pursuit of economic development (Stevens, 2019). More broadly, these findings highlight the limitations of transparency and data availability in an age of widespread digital tools. While openness and transparency can serve basic democratic functions and promote entrepreneurship, they do not necessarily transform existing power relations and decision-making processes.

Having said this, one can still see a silver lining amid some of the empirical realities presented in my findings. Taiwan's government has indeed established some participatory mechanisms which, if run properly, could be transformed into effective channels for civil society to engage in decision-making processes. For example, the Open Data Consulting Groups could be run like a participatory forum, where procedures ensure that the meetings can effectively change the patterns of data selection and input. The presidential hackathons could be expanded to include more citizen participation and targeted problem-solving, a direction the government appears to be moving towards in the 2020s (i-media, 2023). The government could also link the open data platform with some highly praised platforms in Taiwan that facilitate open consultation with citizens (such as vTaiwan) or allow citizens to raise new ideas and discuss policies (such as join.gov.tw), thereby enabling citizens to use data to engage in public decision-making processes. Of course, whether these reforms are possible depends on the intentions of political elites, the power balance between different actors, and the awareness of citizens about the significance of data. After all, it is not open data per se that leads to limited participation, but the social contexts that shape its patterns of implementation. With a society and political leadership that genuinely emphasize civic participation, open data can still facilitate “strong” democracy.

Footnotes

Acknowledgements

I thank Bart Bonikowski, Kuo-Ming Lin, and Martin Eiermann for their invaluable guidance and feedback. I also appreciate the feedback I received during the American Sociological Association Annual conference in 2022.

Declaration of conflicting interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: I would like to thank the Department of Sociology at New York University for funding the publication of this article.

ORCID iD

Terrence Ting-Yen Chen

Notes

References

Baack

(2015) Datafication and empowerment: How the open data movement re-articulates notions of democracy, participation, and journalism. Big Data & Society 2(2): 1–11. doi:https://doi.org/10.1177/2053951715594634

Bächtiger

Dryzek

Mansbridge

, et al. (2018) Deliberative democracy: An Introduction. In: The Oxford Handbook of Deliberative Democracy. Oxford, UK: Oxford University Press, pp.1–35. Available at: http://oxfordhandbooks.com/view/10.1093/oxfordhb/9780198747369.001.0001/oxfordhb-9780198747369-e-50 (accessed 14 December 2019).

Baka

(2017) Co-creating an open platform at the local governance level: How openness is enacted in Zambia. Government Information Quarterly 34(1): 140–152.

Barber

(2004) Strong Democracy: Participatory Politics for a New Age. Berkeley: University of California Press.

Bates

(2012) This is what modern deregulation looks like”: Co-optation and contestation in the shaping of the UK’s open government data initiative. The Journal of Community Informatics 8(2): 1–20. doi:https://doi.org/10.15353/joci.v8i2.3038

Bates

(2014) The strategic importance of information policy for the contemporary neoliberal state: The case of open government data in the United Kingdom. Government Information Quarterly 31(3): 388–395.

Bertot

Gorham

Jaeger

, et al. (2014) Big data, open government and e-government: Issues, policies and recommendations. Information Polity 19(1,2): 5–16.

Birchall

(2015) Data.gov-in-a-box’: Delimiting transparency. European Journal of Social Theory 18(2): 185–202.

Brayne

(2017) Big data surveillance: The case of policing. American Sociological Review 82(5): 977–1008.

10.

Chang

(2020) 踩到政府資訊不公開的雷區?看公⺠的進擊之路! In: OCF Lab. Available at: https://lab.ocf.tw/2020/03/12/opendatagovinfo/ (accessed 2 June 2024).

11.

Charalabidis

Loukis

Alexopoulos

(2014) Evaluating second generation open government data infrastructures using value models. In: 2014 47th Hawaii International Conference on System Sciences, January 2014, pp.2114–2126. Waikoloa, HI: IEEE. Available at: http://ieeexplore.ieee.org/document/6758865/ (accessed 14 December 2019).

12.

Chen

TT-Y

(2022) The developmental state and its discontent: The evolution of the open government data policy in Taiwan. Third World Quarterly 43(5): 1056–1073.

13.

Chuang

Y-J

(2015) Open government data strategy and outlook in Taiwan. Archives Quarterly 14(4): 22–31.

14.

Davies

(2010) Open data, democracy and public sector reform: A look at open government data use from data.gov.uk. Available at: http://www.opendataimpacts.net/report/.

15.

De Blasio

Selva

(2016) Why choose open government? Motivations for the adoption of open government policies in four European countries. Policy & Internet 8(3): 225–247.

16.

Dennis

(2008) Viewpoint: Keeping a close watch – the rise of self-surveillance and the threat of digital exposure. The Sociological Review 56(3): 347–357.

17.

D’Ignazio

Klein

(2020) Data Feminism. Cambridge, MA: MIT Press.

18.

Dryzek

(2000) Deliberative Democracy and beyond: Liberals, Critics, Contestations. Oxford political theory. Oxford: Oxford University Press.

19.

Fell

(2017) Taiwan’s Social Movements under Ma Ying-Jeou: From the Wild Strawberries to the Sunflowers. (1st ed.). London, UK: Routledge. Available at: https://www.taylorfrancis.com/books/9781317198550 (accessed 2 June 2024).

20.

Franceschetti

(2016) The Open Government Data Policy as a Strategic Use of Information to Entrench Neoliberlaism? The Case of Italy. Partecipazione e conflitto. Available at: http://siba-ese.unisalento.it/index.php/paco/article/view/16315/14066 (accessed 18 November 2019).

21.

Friedman

Nissenbaum

(1996) Bias in computer systems. ACM Transactions on Information Systems (TOIS) 14(3): 330–347.

22.

Fung

(2006) Varieties of participation in complex governance. Public Administration Review 66(s1): 66–75.

23.

Gov Data Opener (2019) 資料申請小幫手. Available at: https://hackmd.io/PWJKFlU8TqiaUoUHVL2-Yw?view/.

24.

Granickas

(2014) Building Community around Open Government Data. European Public Sector Information Platform Topic Report 2014/09.

25.

Graves

(2017) The monitorial citizen in the “democratic recession”. Journalism Studies 18(10): 1239–1250.

26.

Hansson

Belkacem

Ekenberg

(2015) Open government and democracy: A research review. Social Science Computer Review 33(5): 540–555.

27.

Harvey

Reeves

Ruppert

(2013) Anticipating failure: Transparency devices and their effects. Journal of Cultural Economy 6(3): 294–312.

28.

M-S

(2014) The resurgence of social movements under the ma ying-jeou government: A political opportunity structure perspective. In: Cabestan

J-P

deLisle

(eds) Political Changes in Taiwan Under Ma Ying-Jeou: Partisan Conflict, Policy Choices, External Constraints and Security Challenges. London and New York: Routledge, pp.118–137.

29.

Hsiao

H-HM

(2019) The tripartite links of middle class, civil society and democracy in Taiwan 1980–2016. In: Hsiao

H-HM

(eds) Middle Class, Civil Society and Democracy in Asia. Routledge contemporary Asia series 65. Abingdon, Oxon; New York, NY: Routledge, 23–28.

30.

i-media (2023) 2023總統盃黑客松聚焦零碳轉型號召黑客攜手解題朝幸福家園邁進. 25 April. Available at: https://tw.yahoo.com/style/news%2F2023#x7E3D;統盃黑客松聚焦零碳轉型-號召黑客攜手解題朝幸福家園邁進-071400431.html.

31.

Janssen

Charalabidis

Zuiderwijk

(2012) Benefits, adoption barriers and myths of open data and open government. Information Systems Management 29(4): 258–268.

32.

Kamat

(2004) The privatization of public interest: Theorizing NGO discourse in a neoliberal era. Review of International Political Economy 11(1): 155–176.

33.

Koopman

(2019) How We Became Our Data: A Genealogy of the Informational Person. Chicago: The University of Chicago Press.

34.

Latour

(1987) Science in Action: How to Follow Scientists and Engineers through Society. Nachdr. Cambridge, Mass: Harvard Univ. Press.

35.

Lee

Tseng

(2017) Taiwan Open Government Report, 2014–2016. Open Culture Foundation. Available at: https://opengovreport.ocf.tw/assets/pdf/report-en.pdf.

36.

Liberty Times (2016) 張善政批教育部：敷衍、無心做事. 15 May. Available at: https://news.ltn.com.tw/news/focus/paper/989801.

37.

MacKenzie

(2019) How algorithms interact: Goffman’s ‘interaction order’ in automated trading. Theory, Culture & Society 36(2): 39–59.

38.

Meng

DiSalvo

Tsui

, et al. (2019) The social impact of open government data in Hong Kong: Umbrella movement protests and adversarial politics. The Information Society 35(4): 216–228.

39.

Ministry of Economic Affairs (2015) Open Data Application and Promotion Project. NSC102-3114-Y-510–015, Report of National Science Council Funded Projects.

40.

Ministry of Economic Affairs (2018) Report of Data Driven Service Industry Application and Development Project (Year of 2017) .

41.

Moore

(2018) Towards a sociology of institutional transparency: Openness, deception and the problem of public trust. Sociology 52(2): 416–430.

42.

Noveck

(2009) Wiki Government: How Technology Can Make Government Better, Democracy Stronger, and Citizens More Powerful. Washington, DC: Brookings Inst. Press.

43.

Open Knowledge International (2015) The Open Definition. Available at: http://opendefinition.org.

44.

Orlikowski

Scott

(2015) The algorithm and the crowd. MIS Quarterly 39(1): 201–216.

45.

Oudshoorn

Pinch

(2003) Introduction: How users and non-users matter. In: Oudshoorn N and Pinch T (eds) How Users Matter: The Co-Construction of Users and Technology. The MIT Press, pp.1–26.

46.

Parenti

(2003) The Soft Cage: Surveillance in America, from Slavery to the War on Terror. New York, NY: BasicBooks.

47.

Pateman

(1970) Participation and Democratic Theory. Cambridge: Cambridge University Press.

48.

Pateman

(2012) Participatory democracy revisited. Perspectives on Politics 10(1): 7–19.

49.

Radio Taiwan International (2019) 促資料開放法制化國發會：草案預計明年底出爐. 17 October. Available at: https://www.rti.org.tw/news/view/id/2038230.

50.

Research, Development, and Evaluation Commission (2013) 政府資料開放加值應用研究分析書面報告. Available at: https://file.data.gov.tw/content/about/#x653F;府資料開放加值應用研究分析書面報告.pdf.

51.

Research, Development, and Evaluation Commission (2015) The Concrete Effects of Open Government Data [政府資料開放(OPEN DATA) 具體成效]. Available at: https://www.ey.gov.tw/File/866046C230EA406C?A=C.

52.

Ruijer

Grimmelikhuijsen

Meijer

(2017) Open data for democracy: Developing a theoretical framework for open data use. Government Information Quarterly 34(1): 45–52.

53.

Sandoval-Almazan

Gil-Garcia

(2012) Are government internet portals evolving towards more interaction, participation, and collaboration? Revisiting the rhetoric of e-government among municipalities. Government Information Quarterly 29: S72–S81.

54.

Schrock

Shaffer

(2017) Data ideologies of an interested public: A study of grassroots open government data intermediaries. Big Data & Society 4(1): 1–10. doi:10.1177/2053951717690750

55.

Schudson

(1998) The Good Citizen: A History of American Civic Life. New York: Martin Kessler Books.

56.

Sieber

Johnson

(2015) Civic open data at a crossroads: Dominant models and current challenges. Government Information Quarterly 32(3): 308–315.

57.

Sivarajah

Weerakkody

Waller

, et al. (2016) The role of e-participation and open data in evidence-based policy decision making in local government. Journal of Organizational Computing and Electronic Commerce 26(1–2): 64–79.

58.

Stevens

(2019) Open data, closed government: Unpacking data.gov.sg. First Monday. Epub ahead of print 1 April 2019. DOI: 10.5210/fm.v24i4.9851.

59.

Susha

Janssen

Verhulst

(2017) Data Collaboratives as a New Frontier of Cross-Sector Partnerships in the Age of Open Data: Taxonomy Development. In: Hawaii International Conference on System Sciences, 2017. Available at: http://hdl.handle.net/10125/41481 (accessed 12 November 2019).

60.

Taiwan Association of Human Rights (2016) 為德不卒的資訊公開:台權會對課綱微調案政府資訊公開行政訴訟判決的公開聲明. Available at: https://www.tahr.org.tw/news/1706#:∼:text=#x53F0;權會認為，依照,後積極提供相關檔案%E3%80%82 (accessed 2 June 2024).

61.

Tufekci

(2014) Engineering the public: Big data, surveillance and computational politics. First Monday 19(7).

62.

Ubaldi

(2013) Open Government Data: Towards Empirical Analysis of Open Government Data Initiatives. OECD Working Papers on Public Governance 22, 27 May. OECD Publishing. Available at: https://www.oecd-ilibrary.org/governance/open-government-data_5k46bj4f03s7-en (accessed 12 November 2019).

63.

White House (2009) Transparency and Open Government: Memorandum for the Heads of Executive Departments and Agencies. Available at: https://obamawhitehouse.archives.gov/the-press-office/transparency-and-open-government (accessed 27 August 2019).

64.

Winner L (1980) Do artifacts have politics? Daedalus 109(1): 121–136.

65.

World Bank (2018) In Thessaloniki, Citizen Engagement and Open Data Play Key Role Protecting Against Natural Disasters. Available at: https://www.worldbank.org/en/news/feature/2018/06/20/in-thessaloniki-citizen-engagement-and-open-data-play-key-role-protecting-against-natural-disasters.

66.

Worthy

(2015) The impact of open data in the UK: Complex, unpredictable and political. Public Administration 93(3): 788–805.

67.

Yiu

(2012) Fulfilling the promise of open public data in the UK. Policy Exchange Research Note. Available at: http://www.policyexchange.org.uk/publications/category/item/a-right-to-data-fulfilling-the-promise-of-open-public-data-in-the-uk.

68.

Robinson

(2012) The new ambiguity of ‘open government’. UCLA Law Review Discourse 59: 178–208.

69.

Zuiderwijk

Jeffery

Janssen

(2012) The potential of metadata for linked open data and its value for users and publishers. JeDEM - eJournal of eDemocracy and Open Government 4(2): 222–244.