Keywords

1 Introduction

In the context of the wicked issues facing education [1] and urban areas [2] in the 21st century, this work explores data challenges and opportunities for smart cities [3]. Schmitt [4] points to the importance of urban data emanating from “the city’s inhabitants and its infrastructure” in real time, “producing a constant flow” that may be understood “as the new building material of the future dynamic city.” Gordon and Mihailidis [5] claim that civic tech “typically refers to work within government” but that this space “has grown significantly” to include many businesses, groups and individuals. Where Townsend [6] noted that, “we experience the symbiosis of place and cyberspace everyday”, Konomi and Roussos [7] claim that, “we are now going beyond the last decade’s conception of smart cities” and moving “towards a deeper level of symbiosis among smart citizens, Internet of Things and ambient spaces.” As such, this work focuses on the problems associated with the sharing and using of data for learning and solution making in technology-pervasive urban environments. Moving beyond dichotomous dystopian and utopian views [6] of civic tech, smart cities, and urban data initiatives, this paper explores the potentials and opportunities identified by Hespanhol and Tomitsch [8] in relation to ‘responsive ambient’ interfaces and what Florida [9] refers to as the “legitimately exciting use of new data” in cities.

This work is significant in that it explores everyday understandings of data in smart cities from multiple perspectives through diverse voices; contributes to theorizing of the ambient data concept in the context of smart cities; and further develops the learning cities component of smart cities in relation to ambient data. The main objective of this paper is to explore existing and emergent understandings of data in contemporary urban environments, characterized by pervasive and aware technologies, in shedding light on complex challenges as opportunities for increased awareness, learning, openness, and meaningful engagement. A review of the research literature on smart cities and learning cities in relation to data, and more particularly real-time data, is provided in developing the theoretical perspective for this work and formulation of an ambient data framework. Using an exploratory case study approach, data collection techniques included a combination of interview and survey. In parallel with this study, qualitative data were also collected through group and individual discussions with a range of people across the city. Content analysis was used in the deductive and inductive analysis of data from the literature review and interviews/discussions, respectively. Additional details on methodology are provided in Sect. 3 of this paper.

What follows is the theoretical perspective for this paper including the research questions and propositions under exploration, the methodology, and a presentation and discussion of findings. The contributions of this work are identified along with the limitations and mitigations, followed by the conclusion.

2 Theoretical Perspective

This section provides a review of the literature for smart cities and learning cities in relation to data, data challenges, and data opportunities. With this background, formulation of an ambient data framework for learning cities and smart cities is advanced in support of an inquiry into the research questions and propositions for this paper.

2.1 Smart Cities and Learning Cities

Townsend [6] defines smart cities as “places where information technology is combined with infrastructure, architecture, everyday objects and even our bodies, to address social, economic, and environmental problems.” Referring to the smart city as a work in progress, and claiming that “there isn’t any single place we can go to see a smart city in its entirety” because building such a thing is “a long, messy, incremental process”, Townsend [6] poses what he considers to be the “more important and interesting question” of “what do you want a smart city to be?” Albino et al. [10] conducted a review of the literature and found that the smart city label “is a fuzzy concept” that “is used in ways that are not always consistent.” Incorporating the people component of smart cities enables an understanding of social infrastructure in cities and key drivers such as creativity, education, learning, and knowledge [10].

Nam and Pardo [11] identify a learning city in relation to the three main dimensions of a smart city (technology, people, and institutions) where “the critical factor in any successful city is its people and how they interact.” While “a smart city is also a learning city” according to Nam and Pardo [11], “learning cities are actively involved in building a skilled information economy workforce” and fostering “social learning for strengthening human infrastructure.” UNESCO describes a learning city as one that “effectively mobilizes its resources in every sector to promote inclusive learning from basic to higher education” [12] extending to workplaces, new technologies, and a culture of lifelong learning. In response to “why learning cities?” UNESCO maintains that “learning throughout life is becoming increasingly relevant in today’s fast-changing world where social, economic and political norms are constantly being redefined.” Indeed, UNESCO provides guidelines for building learning cities [13], arguing that, “a learning society must be built province by province, city by city, and community by community” [12]. Adams Becker et al. [1] note the “move away from traditional lecture-based programming” towards “more hands-on scenarios” where “university classrooms will start to resemble real-world work and social environments that facilitate organic interactions and cross-disciplinary problem solving.” Kearns [14] identifies three generations of learning cities, claiming that “Gen 1 Learning Cities were European in their orientation” and “Gen 2 reflected their East Asian environment” and speculates that “emerging Gen 3 Learning Cities well be fully international in drawing on ideas and experience from anywhere” in order to address “the big global issues confronting cities.”

The civic technology landscape is described by the USDN [15] as “an interface for local government and community members to virtually interact” where the interaction is intended to solve something. For example, the use of “open data to improve communications or to spot trends in public transit use.” As distinct from the term smart cities, civic tech is said to be “the softer, community-facing side of technology” forming “a new normal for interaction between citizens and their governing bodies” while providing “a door to big data access and use through local crowdsourcing.”

Differences.

Citing a possible reason for the lack of a generally agreed upon definition for smart cities, Albino et al. [10] point to application of the term in “hard domains” including physical infrastructures for energy, transport and the like “where ICT play a decisive role” and in “soft domains” where “ICT are not usually decisive”, as in, “education, culture, policy innovations” to name a few.

Picking up on “so called soft factors”, Laitinen and Stenvall [3] refer to the progression of smart cities to smart learning environments, arguing that, “the learning dimension is becoming more central” requiring “more investments in training and in continuing education” in fostering the “learning and innovation capacity” of smart cities. With the purpose of creating “something new and unique” Laitinen and Stenvall [3] note that smart cities programs can be understood as a learning process involving universities “with many different partners/stakeholders.”

Visible perhaps in the smart cities and learning cities landscape is the deep learning concept which Goodfellow, Bengio, and Courville [16] explain “was known as cybernetics in the 1940–60s” undergoing a name change to “connectionism in the 1980s–1990s” with “a current resurgence under the name deep learning beginning in 2006.” Pointing to “the diminished role of neuroscience in deep learning today” Goodfellow et al. [16] note that “modern deep learning draws inspiration from many fields.” Goodfellow et al. [16] are careful to distinguish computational neuroscience as “the effort to understand how the brain works on an algorithmic level”, from deep learning, as the building “of computer systems that are able to successfully solve tasks requiring intelligence.” Deeper learning approaches in a higher education context are described by Adams Becker et al. [1], based on the work of the William and Flora Hewitt Foundation, as “content that engages students in critical thinking, problem-solving, collaboration, and self-directed learning.”

Similarities.

Albino et al. [10] highlight the types of emerging metrics for assessing smart city initiatives, citing the work of Zygiaris [17] who developed a measurement system based on six layers of a smart city. The six layers include: city layer (context); the green city layer (environmental sustainability); the interconnection layer (diffusion of green economies); the instrumentation layer (real-time smart meters and infrastructure sensors); the open integration layer (apps to communicate and share data, content, services, and information); the application layer (real-time responsive operation); and the innovation layer (foster new business opportunities). It is worth noting that school, as in education and learning, is located in the interconnection layer and data is situated in the instrumentation layer as well as the open integration layer.

Learning is occurring in a technology context where again, a layer approach is used in deep learning [16], described by Goodfellow et al. as “a type of machine learning” utilizing “a technique that enables computer systems to improve with experience and data.” Deep learning is advanced for use “in complicated real-world environments” through “representing the world as a nested hierarchy of concepts, with each concept defined in relation to simpler concepts” [16]. Adams Becker et al. [1] identify a range of wicked, as in highly complex, challenges for higher education including next generation learning management systems (LMS) as “more flexible spaces that support personalization” and natural user interfaces (NUIs). NUIs are “allowing humans to interact with machines similarly to how they interact with each other” incorporating “taps, swipes, touching, motions, body movement, and increasingly, natural language.”

2.2 Data in Smart Cities and Learning Cities.

In 2014 the World Council on City Data (WCCD) was launched along with ISO 37120 [18], an international standard for Sustainable development of communities: Indicators for city services and quality of life. The standard is based on the framework of the Global City Indicators Facility [19], consists of 100 indicators (46 of which are core) categorized into 17 themes, and is designed to measure the social, economic, and environmental performance of a city. Lea points to the complexity of the smart city standardization landscape [20], providing an overview and noting the work of The British Standards Institution (BSI) in development of data standards for interoperability and data sharing – PAS 182, Smart city concept model: Guide to establishing a model for data [21].

Van Zoonen [22] provides a preliminary overview of the city data landscape, shedding light on the diversity of data in smart cities in terms of size, regularity, purpose, complexity, ownership, and visibility, to name a few. Florida [9] points to the complexity of relations between data and cities, highlighting the importance of “nuanced human reasoning about cities” in the midst of “new data sources and analytic techniques.” Gil-Garcia, Zhang, and Puron-Cid [23] identify from the research literature “three approaches to the use of data in government” in smart cities as: (a) data that is “captured from both physical and virtual sensors” and is “related to the concept of Internet of Things” (IoT); (b) data emerging from an “enterprise computing and communication platform for interconnecting and integrating” content “from various public services”; and (c) leveraging of the first two types of data “for complex analytics, models, and simulation to make better operational decisions.” Laserson [24] describes deep learning in relation to smart sensors and the Internet of Things (IoT) enabling the transforming of building automation. Herrmann, Hildebrandt, Tielemans, and Diaz [25] demonstrate the purpose and value of interdisciplinary collaborations in the investigation of complex challenges associated with location-based data when explored from engineering, legal, and ethical perspectives.

UNESCO provides a framework of the key features of learning cities along with a more detailed listing of key features and measurement indicators [26]. Data sources are identified as a combination of official city data, expert reviews, and statistical data in survey/review results.

As noted by Schmitt [4], while data generation and collection have always occurred in cities, “new today is the ability of any person using computational devices to generate large amounts of data” that is “real time data.” As such, “the entire city becomes an information organism” the visualization of which “creates new knowledge about the city and is able to make the invisible visible.” It is the “constant flow of data” produced by “the inhabitants of the city” that gives way to the notion of ambient data. From a legal perspective [27], EDRM defined ambient data as “data stored in non-traditional computer storage areas and formats, such as Windows swap files, unallocated space and file slack.” For the purposes of this paper, the ambient data concept is expanded for use in the context of smart cities and learning cities. For example, Schmitt [4] calls for the need to “derive a connection between data, activities, and locations in a meaningful sense by establishing relations between data” in forging “the new building material” of the dynamic city.

2.3 Data Challenges for Smart Cities and Learning Cities

Lea [28] argues that smart city issues “are not strictly” about “technologies,” suggesting that, “the problems are more around exposing data, sharing data and using the data” in the two areas of “infrastructure data” and “citizen data.” Cohen et al. [29] explore emerging tensions among innovators and entrepreneurs engaging with local governments and citizens including those related to open and generative data. Fruehe [30] points to the importance of whether data is “accurate, secure, and actionable.” In addition to concerns with trust around the sharing of data, Gurevich, Hudis, and Wing [31] describe the inverse privacy problem as – “the inaccessibility to you of your personal information” in a “safe scenario” such as “your favorite supermarket.” Cinnamon and Schuurman [32] highlight the data divide in the context of spatial turns focusing on volunteered geographic information (VGI) enabled through the “techno-social revolution in geospatial technology and data.”

Kurose and Marzullo [33] note that “the rapid increase in the quantity of personal information being collected and retained” together “with the increased ability to analyze and combine it with other information” gives rise to “concerns about privacy.” Kurose and Marzullo [33] highlight the importance of “investment in research and development in privacy-enhancing technologies” (PETs) and “encouraging cross-cutting research” of a multi-disciplinary nature involving “not only computer science and mathematics, but also social science, communications, and legal disciplines.” The goal of the strategy according to Kurose and Marzullo [33] is to “produce knowledge and technology that will enable individuals, commercial entities, and the Federal Government to benefit from technological advancement and data use while proactively identifying and mitigating privacy risks” and aiming to “inspire parallel efforts in the private sector.”

From a legal perspective, Solove [34] claimed the privacy concept to be in disarray, such that “nobody can articulate what it means.” Van Zoonen [22] refers to the privacy paradox involving “people’s clearly expressed concerns about their privacy” juxtaposed with “a simultaneous lack of appropriate secure behavior.” Acknowledging this “disorder in the field”, Van Zoonen [22] identifies three consistent and persistent “factors influencing people’s concerns about privacy” as “the type of data, the purpose of data collection and usage, and the organization or persons collecting and using the data.” Van Zoonen develops a smart city privacy challenges framework as a sensitizing instrument in a 2 × 2 frame of data as personal or impersonal in relation to the purposes of service or surveillance.

Concerned with “autonomic smart environments” that “take an unprecedented number of decisions both for the private and the public good”, Hildebrandt and Koops [35] point to the novel concept of ambient law to address the challenges of “errors, loss of autonomy and privacy, unfair discrimination and stigmatisation, and an absence of due process.” More recently, Hildebrandt [36] argues that “technological infrastructures matter” and “require our attention” at a time now when “data-driven agency builds on an entirely different grammar.” According to Hildebrandt [36], the era of data-driven agency is resulting in a shift away from “the linearity and sequential processing demands of written text” using instead “the building blocks” of “information and behavior.”

Williamson [37] refers to the neuro-turn for smart cities where, “transformed by actors such as IBM into brain/code/spaces”, education is being driven by learning algorithms such that “nonconscious computing brains are embedded in the functioning of the environment and intended to weave into the cognitive experience of citizens.”

2.4 Data Opportunities for Learning Cities and Smart Cities

Goldsmith and Crawford [38] articulate the notion of the responsive city, highlighting community engagement and “data-smart governance” with a shift “from a compliance model to a problem-solving one.” Russon Gilman [39] looks “beyond open data” in promoting partnerships and providing “access to the tools and computing power” in order to “engage people who are typically disconnected from ‘data’ to create their own framework to analyze and understand data” in support of finding “creative uses for the data.” Townsend [40] challenges engineers to design for urban inclusion and the work of Janet Echelman [41] who “explores the cutting edge of sculpture, public art, and urban transformation” provides vibrant examples of what is possible through her collaborations with “aeronautical and mechanical engineers, architects, lighting designers, landscape architects, and fabricators.” Echelman “creates experiential sculpture at the scale of buildings that transform with wind and light” advancing the idea of working with data to “use it to do serious things in a playful manner” [42]. For example, Echelman describes a request from the Biennial of the Americas in Denver to “represent the 35 nations of the western hemisphere and their interconnectedness in a sculpture.” Echelman noticed the “earthquake in Chile” and how the ensuing “tsunami that rippled across the entire Pacific Ocean shifted the earth’s tectonic plates, sped up the planet’s rotation and shortened the length of the day.” Based on Tsunami data shared by NOAA (National Oceanic and Atmospheric Administration), Echelman translated the data into an artistic rendering titled 1.26 that “refers to the number of microseconds that the earth’s day was shortened.” This visualization of data provides an artistic rendering of the interconnectedness of 35 nations that now forms “part of the fabric of the city” of Denver.

Recognizing the need to modernize law in the face of pervasive, aware, and smart technologies, Hildebrandt [36] identifies the opportunity to rethink and redesign the “architecture of the Rule of Law” through the “framing of law as information.” As such, Hildebrandt [36] calls for “collaboration between lawyers and computer scientists” that is “in-depth as well as hands-on” in order to “redesign the upcoming data-driven architectures to accommodate human action, to safeguard the fundamental uncertainty and indeterminacy it assumes” while protecting “the pinch of freedom and autonomy that defines us.”

Data figures strongly in four of the 14 dimensions of smartness in government identified by Gil-Garcia, Zhang, and Puron-Cid (e.g., integration, evidence-based decision making, equality, openness) [23]. Collaboration figures in six of the dimensions (e.g., integration, evidence-based decision making, effectiveness, citizen engagement, openness, resiliency, technology savviness) while education figures strongly in six (e.g., integration, creativity, equality, entrepreneurialism, citizen engagement, openness). It is worth noting that information figures strongly in nine dimensions (e.g. integration, evidence-based decision making, citizen centricity, effectiveness, equality, citizen engagement, openness, resiliency, technology savviness) while knowledge figures in five (e.g., integration, creativity, entrepreneurialism, citizen engagement, technology savviness). And finally, relationships figure strongly in two of the dimensions (e.g., citizen engagement and technology savviness). Gil-Garcia et al. [23] cite the work of Gil-Garcia, Pardo, and Burke [43] where “government inter-organizational information integration” is advanced as including “components in a continuum from social to technical aspects” as – “trusted social networks, shared information, integrated data, and interoperable technical infrastructure.” Lämmerhirt, Jameson, and Prasetyo [44] advance a framework for collaboration in “making citizen generated data work.”

Lee et al. [45] identify the value of smart city data hubs for interoperability in that “federation of smart city functionality can focus on hub integration, rather than the integration of individual city sub-systems” enabling developers to “more easily create reusable applications that work in multiple cities.” Lea et al. [45] point to the importance of this type of open innovation platform for “the variety and quality of data streams generated by city infrastructure and citizens.” Fan, Chen, Ziong, and Chen [46] articulate the Internet of Data (IoD) concept, noting that “data standing alone has little or no meaning” and that “when they interoperate” it is “the information of the relations between data” that “become more significant and useful.” In the context of the architecture of collaboration for a smart city, Snow, Håkonsson and Obel [47] define infrastructures as “systems that connect actors” allowing them “to connect with one another” and “access the same information, knowledge, and other resources.” Russon Gilman [39] argues that “interactive and fun” initiatives “help build civic involvement with government, encourage experimentation, and collaboration within governments.” Further, Russon Gilman [39] notes that, “educating more people about the innovative work of government can build stronger, more resilient systems while also fostering more trust in governance institutions.” In an interview, O’Malley describes the work of the MetroLab, highlighting the importance of “using data in city government” for application to human services problems based on “new ways of governing and getting things done” that are “very much aided by technology” [48].

Gurevich et al. [31] argue that “shared data decays into inversely private.” The “sharing back” of such data from, say Fitbit, can have benefit for people in terms of learning about their exercise patterns for example. As such, Gurevich et al. [31] argue that inversely private data will be enormously influential in evolving social norms around data sharing and use.

Data innovation is advanced in a development context as “the use of new or non-traditional data sources and methods to gain a more nuanced understanding of challenges” [49]. Further, the combining of data sources is used “to reframe issues and shed new light on seemingly intractable problems” possibly drawing on an array of digital spaces such as “social media, web content, transaction data, GPS devices” in seeking “more complete, timely, and/or granular information” [49].

2.5 Research Questions, Propositions and Ambient Data Framework

The literature review and theoretical perspective provided in this paper enables formulation of an ambient data conceptual framework for learning cities and smart cities (Fig. 1). The framework is operationalized for use in this paper in responding to the research questions under exploration in this study. As depicted in Fig. 1, this work is concerned with urban data generated through city infrastructures and human infrastructures. Challenges are explored in relation to the constructs of awareness, learning, openness, and engagement. And it is these challenges that simultaneously provide opportunities for: exposing, sharing, and using data; data privacy, trust, and control; and data purpose, value, innovation, and other emergent elements.

Fig. 1.
figure 1

Ambient data framework for learning cities and smart cities

Full size image

The research questions for this paper focus on the key constructs of awareness, learning, openness, and engagement, as follows.

In contemporary urban environments:

  • Q1: How is awareness aided by data?

  • Q2: How does learning benefit from data?

  • Q3: What are the challenges of openness for data in relation to civic tech in the public realm?

  • Q4: What is the opportunity for engagement with data as a building material?

The research questions are formulated as propositions for exploration as follows:

In contemporary urban environments:

  • P1: Ambient data forms part of the critical infrastructure by contributing to greater awareness for action, creativity, and interactions

  • P2: Enabled by ambient data, learning becomes more adaptive, contextual, situational, and emergent

  • P3: Openness is supported by ambient data, driving new conceptualizations of sharing, privacy, and other complex challenges

  • P4: Opportunities for more meaningful engagement arise shedding light on the purpose, value, and innovative aspects of ambient data as a connective and connecting element of the urban social fabric.

3 Methodology

The research design for this study employed an emergent and exploratory case study approach, said to be particularly appropriate for the exploration of contemporary phenomena in a real-world context [50].

A website was used to describe the study and enable people to participate. Demographic data were gathered at sign up including age range, city, and self-identification in one or more category types (e.g., city official, student, instructor, local business, community member, or visitor). Participants were invited to complete an online survey with the option to discuss smart cities and their experience of the city. More specifically, questions invited discussion of learning cities and smart cities and provided the opportunity to think about the city as an emergent environment for leveraging data while exploring challenges, opportunities, practices, and approaches.

Interview data served as the main source of qualitative evidence for this study along with responses to open-ended survey questions. In parallel with this study, group and individual discussions were also conducted with a wide range of people in small to medium to large cities in Canada (e.g., from Langford to Victoria to Vancouver to Toronto).

Content analysis was used as an analytic technique for qualitative data involving inductive analysis to identify emerging terms on the one hand, and deductive analysis of terms from the literature review on the other. Qualitative data gathered from discussions in parallel with this study supported further analysis, comparison, and triangulation of evidence.

Diverse voices emerged from this work across multiple cities and countries from individuals involved in business (e.g., architectural design, ecology, energy, sustainability, technology, tourism); government (e.g., city councilors, policymakers, IT staff); educators (secondary and postsecondary, researchers/lab directors); community members (IT professionals, urban placemakers and designers, engagement leaders, policy influencers); students (postsecondary – engineering/design/computing/education/media).

Overall, data were analyzed for an n = 51 spanning the age ranges of people in their 20 s to their 70 s with 35% females and 65% males. Response to the study was received mostly from Canadian cities (e.g. Langford to Victoria to Ottawa to St. John’s) and also extending to cities in Europe (e.g. Jyvaskyla to Valetta). This study spans a 1.5-year timeframe from mid 2015 through 2016.

4 Findings

Findings are presented in response to each of the four propositions for this study in relation to the key constructs of awareness, learning, openness, and engagement.

Awareness.

A community member in Greater Victoria identified the desire to share information with city officials about bike path areas requiring improvement for increased safety stating that “I would love to have an app that” enables me to “tell the city that” a particular street or road “needs a bike lane.” City IT staff noted that “historically we published datasets that have very little value” adding that now there is a focus on “normalizing the dataset” and “making sure the data has some value and structure.” Citing concerns such as privacy regulations, City IT noted the tendency toward “being very conservative when it comes to data collection” in relation to “the purpose, and how we track, how we manage” the data. A student in Valetta suggested that “as we work out how to sort out the data that is constantly being made” and “built, essentially in social media” spaces, “we will know more” about “what to do with it.” Regarding technology in smart cities, an educator in St. John’s observed that, “we’re not smart on how we use it” calling for “improved communication about the transportation system” and “anywhere Internet” as “particularly useful.” A student in Valetta commented that “an absolute goal of smart cites” should be “total, free wifi” in urban areas. An urban environmental designer in Victoria commented that “I guess its kind of a tradeoff of awareness of what’s around, what I can perceive without the technology to what is perceived through the technology.” A community engagement leader speculated on whether there are “roles for technology” in the transition of downtown parking, as in, “an app that could help me find parking or the closest open stall” and “maybe even reserve it by paying.” An innovation designer for education and cities in Vancouver observed that “young people are not really interested in the archive, the data management” and “they’re not interested in holding on to things in some ways” thus, contributing possibly to the notion of ambient data. An IT leader in higher education in Greater Victoria added that for youth “their legacy is change, their life’s work is not materials, its social, social capital” aided by technology. Yet, referring to the physical coming together of people in the city, the innovation designer suggested that, “the meeting becomes the technology that changes everything.”

Learning.

An urban environmental designer commented that, “one thing I’m learning is” the importance of “using data from technologies to improve aspects of the city” on the one hand, and “the human component” on the other “in order to enable all those connections to happen.” An IT professional in a higher education institution (HEI) in Vancouver pointed to the importance of data sharing to verify the anonymized presence of people in buildings on campus for security, safety, and emergency purposes. Another IT professional in an HEI described smart cities and learning cities activities in terms of collaboration between the university and the local city indicating that, “we’ve been working on whole systems on the campus looking at everything from environmentally friendly all the way out to how we integrate new building into the infrastructure” and “working with” the city “on tying things in” since “the city has been working on a fiber project across all the city.” Regarding emergent opportunities for learning from urban data flows, city IT staff commented that, “we are a little immature with respect to that level of data mining.” Indeed, it was noted that opportunities to share data exist “even inside the processes of City Hall” whereby “we’re slowly getting those connections so that engineering and planning” can work together. Regarding data analytics, learning challenges and opportunities were identified by City IT staff in that, “we’re starting to look at the tools to help us mine the data that we already have an interest in.” City IT staff added that, “going outside of our little silos” and “getting that semi-structured, that unstructured data and trying to then look at it through a different lens is alien to us because we’re not data scientists” acknowledging that “we’re very much immature in that overall data sense.” As such, City IT staff commented that, “the idea of data between two different datasets, we’re just starting to look at the tools that would give us the visualizations of that.” An educator in St. John’s commented that, “I’m always looking for the more visual stuff” as in, how cities present “the visual sense of the city.” An educator in Jyvaskyla pointed to the potential of a mobile application for civic purposes “that people could use to capture ideas, to capture evidence of problems on the move, capture evidence of potential solutions and evidence of impact.” Noticing the use of infrared camera technology by downtown businesses to gather data on pedestrian volume and movement in certain areas, a community engagement leader in Victoria stated “I’d love to do some of that kind of stuff with some of our public spaces, like plazas” in order “to have some of that anonymous kind of count” data. Discussing apps for civic engagement such as PlaceSpeak, a community engagement leader highlighted such platforms as a way to “encourage people to do some observations of space and people and post stuff” in a space “that’s interactive” as a kind of activity to “look at any given space and analyze what’s working here and what isn’t and what could improve it.”

Openness.

On a community level in Victoria the issue was identified of “not being able to find out really what’s going on in the city.” City IT staff commented that, “fundamentally there is a desire to be very, very open with public data” while adhering to “privacy regulations” and being cognizant of “the public’s preferences.” An educator in St. John’s suggested the need to “provide user friendly and not too dense information on what’s happening at a particular point” in the city “so that you can access information and know what’s going on in real time almost.” An educator in Jyvaskyla suggested that civic apps “could be used from ideation to real life experimentations” establishing “proof of concept” and the “next stages of open innovation” and “capturing the evidence of the impact.” Referring to a mobile app for urban spaces, the educator pointed to the capability of being “able to open this kind of feedback” potential to anyone in the city as a way “to transform contributions both in terms of unique ideas and patterns into the design of some urban space or buildings” as in “city smart infrastructure” emphasizing “energy and resources” enabling city officials “to listen and understand what people want in the form of evidence.” A community member in Toronto noted that, “the number one requested dataset from the city by the public is zoning and development data,” adding that this “is not currently available as open data to cities around Canada at least.” An IT professional at a higher education institution in Vancouver raised the issue of learning to work across systems and infrastructures of data involving buildings and room bookings, for example, and “how we exchange that data in a way that is safe” to generate “real occupancy information” in conjunction with “devices of people to verify that someone is actually in the room.” A student in Valetta speculated on ways to address the wicked challenges of personal data privacy in order to “devise ways whereby you can create some kind of security to anonymize data and thereby make the data itself open and shared in some way that would still provide some kind of smart delivery too, so you could actually make use of the data” which could be “learning data, information data about people, things, events, places.”

Engagement.

A community member in Victoria referred to “a lot of advantages to having” smartphone data “to enhance the urban experience and to be able to get from point a to be faster.” City IT staff cited the example of an eTownHall, enabled through the use of technology, observing that, “bringing questions in and sharing answers through Twitter, Facebook” provided “a dataset of documented engagement.” As such, City IT staff noted that “using different tools to interrogate and present data becomes that much more important and it contributes to how we successfully engage with the citizenry.” Commenting on city dashboards, a student in Valetta stated that “you can take this kind of data and rather than just present numbers you can make beautiful artistic visualizations” adding that “this is how to get people on board.” An educator in St. John’s commented that “much more collaboration is necessary” in how technology is used. An educator in Jyvaskyla pointed to a mobile app in terms of data that is “evidence-based so the decision-makers can show at any meeting” that “people said this and that is why we decided to do it.” Hinting at unspoken elements in support of engagement, the educator commented that “if we forget about the technology for a moment its almost” as if “we assume there is this wireless connectivity between us happening.” A community member in Toronto pointed to the importance of “marketing and communications” when releasing a dataset, identifying what could potentially be built with the data, adding the need to point out that “the government just doesn’t have the means to make it happen and we’re hoping the people will.” Regarding zoning and development data, the community member stated, “imagine if there was an app that told you all the notifications on this corner” and because “people would be affected, they would pay attention and they would suddenly be engaged” since “their environment is being changed” and “right now, nobody knows what’s happening.” Regarding urban parking, a community engagement leader wondered whether technology could help with getting more clarity on whether there is an oversupply or undersupply of parking in terms of “people’s perception” versus actual data.

In summary, Table 1 provides an overview of findings for this exploration of ambient data in relation to the four constructs of awareness, learning, openness, and engagement. Associated challenges and opportunities are identified in the context of learning cities and smart cities.

Table 1. Overview of findings.
Full size table

A range of ambient data types for learning cities and smart cities are highlighted from city datasets to civic apps. Challenges presented by ambient data are identified in relation to analysis, collaboration, people and technology interactions, privacy, safety, and visualization. Opportunities presented by ambient data are also identified, including but not limited to: analysis, real-time sharing, purposes/users, value, and the potential for listening and action by decision-makers.

While this overview (Table 1) is not intended to be exhaustive, it does provide an early glimpse of the emerging ambient data landscape for small to medium to large-sized learning cities and smart cities in Canada and extending to Europe. Further, Table 1 provides an indication of how all four propositions under exploration were supported in this study.

Regarding proposition 1, ambient data is seen to form part of the critical infrastructure by contributing to greater awareness for action, creativity, and interactions in relation to bike paths, city datasets, social media, and parking apps.

For proposition 2, enabled by ambient data, learning becomes more adaptive, contextual, situational, and emergent in relation to urban spaces, university/city collaborations, and urban data analysis including emerging data sources and tools.

For proposition 3, openness is supported by ambient data, driving new conceptualizations of sharing, privacy, and other complex challenges in relation to real-time sharing and analysis and the development and use of civic apps.

For proposition 4, opportunities for more meaningful engagement arise shedding light on the purpose, value, and innovative aspects of ambient data as a connective and connecting element of the urban social fabric in relation to everyday urban experiences, documented engagement using social media and other aware technologies during eTownHalls, and the visualization of city information.

5 Discussion

The exploration of ambient data in relation to the four constructs of awareness, learning, openness, and engagement is depicted in terms of opportunities for learning cities and smart cities in Fig. 2. This glimpse of the emerging fabric of ambient data in smart cities and learning cities provides a rendering of the four constructs as overlapping and interweaving, yielding a continuously unfolding interplay of elements characterized as adaptive, contextual, connective, and interactive that enable action, sharing, creativity, purpose, value, and other emergent potentials. This dynamic urban fabric of data is giving way to opportunities for a rethinking of the information landscape, in support of innovative initiatives such as inverse privacy while contributing to a discourse, practice, and research space for ambient privacy.

Fig. 2.
figure 2

Ambient data opportunities for learning cities and smart cities

Full size image

Individuals in both Canadian and European cities identified the need to do more with the data being generated in urban contexts. Further, concern with data delivery was highlighted in terms of “mechanisms for delivery” in Canada and as “smart delivery” in Europe. Encapsulated in the words of a student in Europe – “the more that technical infrastructure can be made to constantly reciprocate the data flows that are happening between people, formal and informal, the better” – is support for real-time data opportunities and possibly emergent understandings of the ambient data concept.

In a higher education context, IT professionals identified multi-purpose opportunities for leveraging and exposing “some of the information in a way that is safe” to “make it available for other things.” The need for data sharing opportunities was identified on the community level in the words “nobody ever really knows what’s happening” in the city and reinforced by individuals in other cities. Of particular note is the opportunity identified by city IT staff in gaining new literacies associated with data sharing, analysis, and use – improving skills within and across organizations to advance beyond current levels of maturity.

Questions about dataset veracity were identified on the one hand by city IT staff in terms of completeness of datasets and on the other by a student with respect to social media data where people are said to often mix fictional elements into their profiles whether for fun, privacy, or other reasons. Urban collaborations and civic tech platforms such as PlaceSpeak provide an example of the potential for building trusted spaces for learning cities and smarter cities based on emerging understandings of ambient data and other evolving collaborative opportunities.

City IT staff highlighted the importance of evolving approaches to sharing whether infrastructure, data centers, datasets, and the like as opportunities, emphasizing that “trying to get” existing mindsets “to change is the biggest challenge.” Engagement of people meaningfully in the city and with technology-based initiatives is seen to be key in coming to contemporary understandings, interpretations, and directions for deeper learning as it relates to education on the one hand and people and technology interactions on the other. The visualization of data for more meaningful engagement was identified with the suggestion that data be combined and rendered in more artistic ways.

Speculating on the innovative potential of data collection, sharing, and analytics, a city IT professional was particularly fascinated to observe the accidental, serendipitous, and unintended uses that occur when new mechanisms for sharing data are made available. In this context of emergent purposes and value for data collection, sharing, and use, the challenges posed open opportunities for increased data literacies in relation to privacy, security, analysis, and visualization. In response to privacy requirements to fully know in advance what the purpose, use, and value of data collection will be, it is precisely the accidental, serendipitous, and unintended opportunities that reflect the “human action” identified by Hildebrandt [36] and “the fundamental uncertainty and indeterminacy it assumes.”

It is worth noting that the civic tech engagement platform, PlaceSpeak, is architected to incorporate Privacy by Design (PbD) principles [51]. However, the embedding of privacy into applications and systems may prove to be too brittle, in some cases, a criticism leveled by Townsend [6] about systems that are not highly adaptive, responsive, or resilient. Going forward, and in parallel with PbD, perhaps opportunities exist for more adaptive and flexible privacy architectures conceptualized as ambient privacy [52]. The privacy requirement that data purpose be knowable and provided up front prior to the granting of use may be a sound principle in traditional information environments but less tenable in the emergent, uncertain environments of 21st century cities. As such, more flexible, adaptive, and dynamic privacy principles are required, perhaps in the form of new conceptualizations of privacy as ambient privacy [52], inverse privacy [31], and the like. As such, this work points to the potential for an extending of Sassen’s [53] urbanizing concept and the need for smart city technologies to be able to “work within a particular urban context”, to McKenna’s articulation of urbanizing the ambient [54], to the urbanizing of ambient data for smart cities and learning cities.

In summary, multiple learning opportunities emerged in this work from the perspective of those in higher education, city governments, and urban communities contributing to evolving understandings of learning cities and smart cities.

6 Contributions

This work is significant in several ways in that it: (a) explores the ambient data concept in the context of smart cities; (b) theorizes and operationalizes a conceptual framework for ambient data as dynamic, real time, and adaptive enabled by aware, pervasive, and other emerging technologies and human factors; (c) extends the discourse space for ambient data to civic technologies, the public realm, and other emergent socio-urban intersections; and (d) relates deep learning conceptualizations to learning cities and smart cities involving the interactive people and technologies dynamic.

7 Challenges and Mitigations

A key limitation of this study was sample size, mitigated by the diverse voices and rich content shared from multiple, small to medium to large sized cities in several countries. Challenges related to geographic location, size, and urban characteristics are mitigated by the potential to extend this work to other cities, including megacities and regions exceeding 10 million people. The challenge of studying emergent, invisible, and ambient data is mitigated by opportunities to creatively discover and collaboratively explore the opportunities and potentials for the making of spaces, infrastructures, and environments to advance understandings through real-world research and practice.

8 Conclusion

This work explores and theorizes ambient data in urban environments in the context of 21st century smart cities. Through the interactive dynamic of people – technologies – cities, the constructs of awareness, learning, openness, and engagement are explored in relation to ambient data and associated challenges.

Key contributions of this paper include: (a) conceptualization of ambient data in contemporary urban environments; (b) further development of the research and practice literature for smart cities, learning cities, and ambient data; (c) theorizing and operationalization of the ambient data framework for learning cities and smart cities; and (d) the opening of a discourse space for evolving understandings of data as dynamic, emergent, and continuous in the context of smart cities and learning cities. Of particular value in this work is the emergence of the diverse voices of people across the city from business to city officials to postsecondary educators and students to community members and information technology staff (e.g., city and higher education institutions). Also of note is the literature review that introduces a range of challenges and opportunities for data, and more particularly ambient data, in contemporary society. What can be learned in the process of smart city discussion and development is highlighted along with emerging understandings of the deep learning concept in relation to people on the one hand and to technologies on the other, in the context of learning cities and smart cities.

A key take away from this work is the importance of the ambient data concept, opening the way for new discourse spaces and explorations in both research and practice for learning cities and smart cities. It is hoped that the notion of ambient data will contribute to opportunities for innovating data related challenges from sharing to privacy to collaboration going forward.

This work will be of interest to urban innovators, educators, learners, city officials, community members, and anyone concerned with evolving approaches to civic tech and ambient data for learning cities and smarter cities.