How to Create and Foster Sustainable Smart Cities? Insights on Ethics, Trust, Privacy, Transparency, Incentives, and Success

Abstract

This paper describes the motivation, framework, concepts, method, implementation, and results of the HCII2023 Design Café, a dedicated participatory highly interactive design workshop held in Copenhagen in July 2023. Motivated by the ubiquitous challenges our world is facing, this initiative had the goal to explore six main issues from an interdisciplinary perspective with a focus on the UN SDG 11: “Sustainable Cities and Communities” and the HCI Grand Challenge 2 “Human-Environment Interactions.” The six issues were formulated as questions and presented to the participants working in small groups. (1) How to create inclusive and ethical smart cities? (2) How to establish trust between people and smart environments? (3) How to address privacy concerns in smart environments that adopt the “disappearing computer” paradigm? (4) How to promote explainability and transparency of policies and measures to citizens of smart cities or in smart environments in general? (5) How to design incentives and rewards for engagement and sustainable behavior in smart cities at a personal as well as collective/corporate level? (6) How to measure success and impact in sustainable smart city projects? The method and approach of the Design Café is a tailored composition of a guided, structured format combined with and inspired by processes of informal communication and exchange of knowledge and ideas. Aligned moderation, a minimum set of rules, a set of relevant topics and an interdisciplinary group of motivated participants working in rotating formations provides the structure for achieving results. The results show that the six issues are not independent of each other, but require a holistic view, considering the various dependencies as well as synergies when exploring solutions. Nevertheless, the importance of first establishing higher-level goals based on ethical and inclusive approaches fostering human dignity and human rights were key. Acceptance of overall goals, processes, rules, and regulations was considered as the fundamental pre-requisite for sustainable change towards a declared goal. Acceptance needs trust and privacy as well as explainability and transparency of policies and measures. The role of incentives and rewards for engagement and sustainable behavior was twofold. Offering incentives must include planning on how to measure their impact. Effective measurement of success and impact depends heavily on how the institutions address privacy. Explainability and transparency should become one of the ethical guidelines that steer and control concepts, decision making, and implementations of all activities. The scale of these societal challenges still needs to be recognized by those responsible. It is essential to educate decision makers in the psychological needs and effects of sense making, comprehensibility, and finally acceptance and well-being moving towards a humanity-centered design.

Keywords:

• Human-environment interaction
• smart environments
• HCI grand challenges
• participatory design
• human- and citizen-centered design
• humanity-centered design
• design café
• UN SDGs
• sustainability
• sustainable cities and communities
• sustainable behavior
• quintuple helix
• smart city
• humane city
• cooperative city
• disappearing computer
• artificial intelligence
• human-centered AI
• explainable AI
• privacy
• trust
• ethics
• inclusiveness
• explainability
• transparency
• observability
• incentives and rewards for sustainable behavior
• measuring success and impact
• information design, knowledge design

1. Introduction

The world is changing dramatically. We are facing multiple challenges, problems, and crises. Climate change with its increasingly dramatic implications in a wide range of areas is one of the major challenges to be addressed, e.g., with a strong awareness of the need to foster sustainability and act accordingly. In response, the United Nations (UN) declared the “2030 Agenda for Sustainable Development” with its 17 Sustainable Development Goals (SDGs, 2015) addressing the challenges in detail. Over the years, there are many hundreds of activities worldwide to address these challenges from multiple perspectives, too many to report here. For corresponding surveys see OECD (2017, 2019) and Global Survey (2020). Still, there is a universal lack of progress towards the goals of the UN SDGs as the recent report (SDGs-report, 2023) on the situation halfway to 2030 states clearly, asking for urgent action.

Cities are an important application domain and have a huge impact, because in 2017 already 55% of the world population lived in urban areas. Furthermore, it was projected already some time ago (UN, 2018) that by 2050 more than two-thirds will live in urban areas. Thus, we selected as a specific and relevant goal the UN SDG 11 “Sustainable Cities and Communities.” The 2023 report (SDGs-report, 2023) emphasizes this choice, because it states that this is the area with the least progress – less than 50%.

The role of Information and Communication Technology (ICT) is prominent in all venues of our lives. Our working and living spheres are more and more determined by technologies being ubiquitously integrated in our (smart) environments, e.g., via Internet of Things (IoT) infrastructures, increasingly based on Artificial Intelligence (AI) and Machine Learning (ML). Unfortunately, these developments are mainly technology-driven and do not take the needs and preferences of users, respectively citizens, and their rights sufficiently into account. Therefore, people are confronted with new challenges when interacting in and with technology-augmented environments in general, and urban environments, respectively smart cities, in particular.

These changes impact people and their behaviors and expectations, in cities, but also in general in our society. Thus, it is becoming increasingly urgent to actively shape our future and to find new ideas for innovative approaches to new solutions based on a humanity-centered design approach (Norman, 2023). Innovative design approaches for next generation cities are needed. But first, we must answer the fundamental question: what kind of cities do we want to live in?

The contribution of this paper is to report about a special activity, the “Design Café,” emphasizing a strong people-centered and participatory design approach involving an international group of people that came together in the context of the 25th International Conference on Human-Computer Interaction (HCII2023) in Copenhagen, Denmark. We describe the conceptual framework, report about the processes and the results and present our conclusions.

The general framework for our design considerations must take a global perspective as expressed in Figure 1 of the Quintuple Helix Model (Carayannis et al., 2012).

Figure 1. The five helices of the quintuple helix (modified illustration based on Carayannis et al., 2012).

Figure 2. Overview of the 17 SDGs of the United Nations (https://sdgs.un.org/goals).

This comprehensive model demonstrates the need to involve institutional and government bodies, academic and research spheres, industry and business sectors, the civil society with its citizens as well as the natural ecological environment in the innovation design process. The inclusion of the natural, resp. the socio-ecological environment implies considering issues of sustainability. In this context open innovation can act as an enhancer of sustainable innovation ecosystems. Costa and Matias (2020) describe the benefits as follows: “Creating an innovative ecosystem has a multi-layer effect: It contributes to regional digitalization, technological start-up emergence, open innovation promotion, and new policy enhancement retro-feeding the system. Public policy must create open innovation environments accordingly with the quintuple helix harmonizing the ecosystem to internalize emerging spillovers.”

The UNESCO (2019) Report on “Smart Cities – Shaping the Society of 2030” describes three Concepts of Urban Progress:

• Business City ⇨ Economy First

• Eco City ⇨ Ecology First

• Citizens City ⇨ Humans First

Our point of view is that any smart city should be a balanced combination of these three dimensions, but the mix is always dominated by a primary purpose. Our focus is to foster a vision which is based on a common purpose and values as well as design trade-offs of putting “Humans First” based on a Citizen-Centered Design.

2. Vision and goal of the HCII2023 Design Café

The 25th International Conference on Human-Computer-Interaction (HCII2023)¹ responded to the above challenges with a new format: the “Design Café.” The HCII2023 Design Café ² (from now on Design Café) was a new satellite event that allowed experts to actively participate in the discussion on selected most urgent topics as well as in the creation of new ideas for new solutions. The results are being published (for example with this paper) to make them available for a broader audience as inspiration and guidance for their work. Furthermore, every participant defined a personal “take-home idea” from the Design Café with the strong intention of realizing it in his/her individual business context.

The peculiarity of the Design Café format lies in the “power” which is given to the participants. While in standard workshop formats, participants remain more or less in a “listening mode” – apart from asking questions after a presentation – in the Design Café each participant takes an active role. It is the participant who actively selects his/her most relevant topic on which he or she wants to work according to personal or business interests. It is the participant who develops appropriate approaches for transfer into practice.

The setting of the Design Café provides an enabling space and inspiring framework for interdisciplinary exchange, open dialogue and discussion, professional knowledge transfer, meaningful high-quality collaboration, and creativity. It uses valuable and proven participation methods to engage stakeholders in the exchange of new ideas and approaches. They foster and stimulate open dialogue, constructive deliberations, and informal but meaningful collaboration. They empower creativity and inspiration in a casual atmosphere and promote innovative approaches for transforming ideas into practice. Thus, the Design Café is a moderated agile format, which facilitates participants to gain new perspectives, and create new ideas and approaches for meaningful relevant innovations. The approach is based on the experiences with previous editions of this format, e.g., in the context of addressing the challenges of designing “smart islands” as a lighthouse of research and innovation (Streitz & Riedmann-Streitz, 2022; Streitz et al., 2022).

With the Design Café in 2023, the HCII conference opened for the first time its doors for selected “outside” peer groups independent of conference participation, e.g., from the metropolitan area, where the conference takes place, in this case, Copenhagen, Denmark. Thus, topics of the HCII conference and appropriate approaches for transfer into practice could become part of the local economic and academic context. This paper describes the motivation and framework, the processes, and the results of the Design Café that was organized and conducted in Copenhagen in July 2023.

3. Big challenges that move us

The HCII Design Café was conceived and conducted as a platform for interdisciplinary discussion on and creation of ideas for major challenges. One of the big challenges that move us – and “us” here means the global world – is sustainability. Therefore, one main topic of this Design Café refers to the global “2030 Agenda for Sustainable Development” with its 17 Sustainable Development Goals (SDGs, 2015).

Other big challenges that move us were identified and published in the White Paper “Seven HCI Grand Challenges” (Stephanidis et al., 2019) as the result of a collective effort of an international group of 32 experts in the community of the HCII Conference series. It constitutes the second topic cluster of this Design Café.

Each HCII Design Café opens its doors to conference participants but also to peer groups independent of conference participation in the city and region where the conference takes place. As an esteeming reference to Copenhagen, the capital of Denmark, where the HCII conference took place in 2023, two specific topics out of the “2030 Agenda for Sustainable Development” and the “Seven HCI Grand Challenges” were selected for the HCII2023 Design Café.

The two specific topics refer to Copenhagen being considered as Europe’s largest Smart City Living Lab. Furthermore, the city of Copenhagen was officially designated as the UNESCO World Capital of Architecture for 2023 (Copenhagen, 2023) and reflected this in exhibitions and installations throughout the city.

Consequently, the UN SDG 11 “Sustainable Cities and Communities” was selected and since a city represents a living, leisure, and working environment for its citizens and visitors, the HCI Grand Challenge 2 “Human-Environment Interactions” was selected as the second topic to be discussed.

3.1. The UN sustainability development goals (SDGs)

3.1.1. Big picture

One of the crucial challenges for mankind is climate change. The concept of sustainability was identified and invented by Hans Carl von Carlowitz (1645 − 1714) in the early 18th century. The chief miner from Freiberg (Saxony, Germany) is considered to be the founder of sustainability (von Carlowitz, 1713; von Carlowitz & Hamberger, 2022). He recognized that important raw materials must be managed and used sustainably, and he was the first to take an intergenerational view on sustainability. The big issue of that time was the resource of wood, its use, and the correspondingly extensive and rapid deforestation. Carl von Carlowitz called for forest management, consistent reforestation and “sustained” use, which quickly became a technical term as sustainable forestry. He was the first to formulate one of the later principles of sustainability, that only as much wood should be felled as could grow back through planned reforestation, sowing and planting (von Carlowitz, 1713). This laid the foundation for one of the key principles of sustainability.

More than 300 years later, the environment has changed dramatically due to inventions such as new technologies, electricity, electronics, and Information Technology (IT), Virtual Reality (VR), and Artificial Intelligence (AI) as well as developments like digitalization, globalization, etc. However, regarding sustainability the world is threateningly lagging in implementation, although we developed the principles further and today also address the need for a circular economy (Ekins et al., 2019).

In 2015, the United Nations declared 17 Sustainability Development Goals (SDGs, 2015) as mandatory for all countries to be reached by 2030. “The SDGs build on decades of work by countries and the UN” and they are “an urgent call for action by all countries – developed and developing – in a global partnership.” For an overview of all goals see Figure 2. Since the UN SDGs were declared in 2015 with a time frame of 15 years for achieving them in 2030, there were only 7 years left (at the time of the HCII2023 Design Café).

3.1.2. UN SDG 11: Sustainable cities and communities

Motivated by the role of Copenhagen as described before, we selected the UN SDG 11 “Sustainable Cities and Communities” for the Design Café as one of the two topics. It states “Make cities and human settlements inclusive, safe, resilient and sustainable” as the UN request to all countries, cities, and communities on earth. It addresses amongst others the areas of

• public transport

• air pollution

• public spaces and streets

• human welfare (no poverty or slums).

The UN SDG 11 provides the framework that should be included in every Future City- and Corporate Strategy to achieve sustainability goals effectively. In addition, the following three dimensions are introduced (Streitz et al., 2022) emphasizing the hybrid character of many environments:

• urban and rural (areas)

• land and sea (earth and water)

• real and virtual (analogue/physical and online)

In our time, two pitfalls (among others) lurk in the transformation to a sustainable city or community:

• thinking from the perspective of what is technologically feasible.

• thinking in terms of the individual benefit for the players in industry, business, politics, or administration.

Both approaches counteract the holistic approach of this UN SDG 11. Thus, we focus in the Design Café on taking the perspective of humans (citizens, workers, visitors) addressing aspects as ethics and inclusion, trust, privacy, explainability and transparency, rewards for engagement, and finally measuring success and impact for continuous improvement.

3.2. The Seven HCI Grand Challenges

3.2.1. Overview

We adopted a second framework (besides the UN SDGs) for selecting the second main topic of the Design Café. It is based on the white paper on the “Seven HCI Grand Challenges” (Stephanidis et al., 2019) created as a collective effort of an international group of 32 experts in the community of the HCII Conference series. The goal was to identify and explore the grand challenges we are facing because of current and rapidly emerging technological developments towards interactive smart technologies (often called “intelligent”) in the context of increased and widened societal needs in all venues of our lives.

The Seven HCI Grand Challenges are: Human-Technology Symbiosis; Human- Environment Interactions; Ethics, Privacy and Security; Well-being, Health and Eudaimonia; Accessibility and Universal Access; Learning and Creativity; Social Organization and Democracy (Figure 3).

Figure 3. Overview of the Seven HCI grand challenges (Stephanidis et al., 2019, used with permission).

The current hype about AI and Machine Learning (ML), triggered by the fact that applications based on Generative AI, like ChatGPT, became available and accessible to a wide public, has additional new implications that need to be addressed. For example, the paper by Ozmen Garibay et al. (2023) is reflecting on “Six Human-Centered Artificial Intelligence Grand Challenges”. It advocates a “human-centered approach to AI that (1) is centered in human well-being, (2) is designed responsibly, (3) respects privacy, (4) follows human-centered design principles, (5) is subject to appropriate governance and oversight, and (6) interacts with individuals while respecting human’s cognitive capacities.” Additional material can be found in a recent Special Issue on AI in HCI (Antona et al., 2023) bringing together research findings and best practices from academia and industry, highlighting results of joined AI and HCI forces. A holistic perspective on human-centered design of AI was presented by Shneiderman (2022), arguing that Human-Centered AI will amplify human abilities, empower people, and ensure human control.

Our working and living environments are increasingly determined by technology being ubiquitously integrated and promoted with the claim that everything is “smart.”

This development can be attributed to what Streitz (2019) calls the “Smart-Everything” Paradigm, where every artifact and service must now be “smart”: smart phones, smart cars, smart homes, smart cities, etc. Since these developments are mainly technology-driven and do not take users’ needs, preferences, and their rights sufficiently into account, Streitz (2019) calls for “redefining the Smart-Everything paradigm by introducing appropriate design trade-offs” (e.g., empowerment of humans by keeping them in the loop and in control vs. automated/autonomous systems; user-defined and controlled privacy-by-design vs. importunate smartness). The goal is to move beyond “smart-only” cities towards humane, sociable, cooperative, and self-aware hybrid cities (Streitz, 2021b). Therefore, citizen-centered participatory design approaches are needed for creating and fostering viable sustainable cities and smart environments.

3.2.2. HCI grand challenge 2: Human-environment interactions

For the Design Café, we selected the HCI Grand Challenge 2 Human-Environment Interactions because it corresponds best to the UN SDG 11 Sustainable Cities and Communities selected above.

Cities represent urban environments. They are complex systems with a huge number of networked entities, communicating and interacting with each other. The complexity results from the fundamental diversity of these entities: living organisms like humans, animals, and plants; “bricks and mortar” constituting the physical environment like buildings, bridges, streets, etc.; active artificial artefacts with embedded IoT components. Furthermore, the real world is augmented by virtual counterparts or “digital shadows” of basically all organisms and artefacts (depending on the availability of appropriate models) constituting hybrid environments, e.g., digital twins. At another layer, we must also address networks of people with their social relationships in the real world as well as in the virtual world. Designing human-environment interactions, especially in urban environments raises new issues. Here are a few examples based on the “Seven HCI Grand Challenges” paper (Stephanidis et al., 2019):

• Interactions take place in the physical and digital continuum resulting in a wider design space for “hybrid” worlds/environments.

• New forms of interaction via smart wearable devices, such as glasses, watches, and bracelets.

• Implicit sensor-based interactions, e.g., gesture-based interaction with large interactive walls/displays, require new notions and concepts of affordances going beyond the traditional types of affordances used in HCI.

• Embedding hardware into the environment results in the “disappearance of the computer” as a “visible” distinctive device. It was initially propagated by Mark Weiser (1991) to achieve “calm technology” and adopted by the EU-funded “Disappearing Computer” proactive initiative (Streitz et al., 2007). This approach was meant to provide support to users in a smooth and unobtrusive way. At the same time, it causes privacy issues, because people are not aware of hidden devices and sensors being used for unwanted surveillance without their consent. This is a problem especially in public spaces.

• Interactions in public and transient spaces, where the ubiquity of technologies “blurs” the boundaries between private and public spaces and their corresponding constraints for interactions, e.g., available functionality, access rights, privacy.

• Interactions in virtual and augmented reality (VR, AR) require new forms of methods and affordances.

The Grand Challenge 2 also reflects the transition from traditional Human-Computer Interaction (HCI), as we experience it at the laptop computer or on the smartphone, to Human-Building Interaction (HBI) in homes, office buildings, schools, theatres/cinemas or even a sport stadium to Human-Environment Interaction (HEI) (Streitz, 2006). Special cases are Citizen-City Interaction (CCI) and Citizen- Environment Interaction (CEI) (Streitz, 2021a). Finally, we should be aware that there will be increasingly Human-AI Interaction in instrumented, smart environments (Challenge 6 in Ozmen Garibay et al., 2023).

4. How to make visions happen: The impact of participatory design

One decisive insight from transformation processes that the HCII Design Café took advantage of is that the initial approach of thinking determines the result. Transforming a city or a community or an organization is a process of change. Thus, it is of critical importance to follow the crucial factors of successful sustainable change. Unfortunately, most approaches are still top-down: Few people in charge decide on a plan, implement it and communicate it to their stakeholders. The result is very often alarming because of the low acceptance by stakeholders and – partly incomprehensible – errors in the implementation.

On the other hand, some municipalities have a long tradition of citizen participation – even beyond what is required by law. But they also face a variety of challenges. There is no comprehensive and convincing concept for bringing together the goals, expectations and resources of the city and the stakeholders in a structured way.

The successful development of a sustainable city (a community, an organization, or an island) and its lasting acceptance depend on the committed participation of all relevant stakeholder groups. This is of crucial importance, because change, the abandonment of the familiar and the accustomed and a multitude of vested interests are burdened with fears, anxieties, misunderstandings, and prejudices that massively hinder the path to success.

Broad acceptance, commitment, engagement, and ownership require participatory formats empowering and inviting active contributions, collaboration, co-creation and, if possible, co-decision. Only here, at this high level of participation, valuable knowledge is unlocked. There are many approaches to participatory design. For a recent overview, we refer to Bødker et al. (2022) as well as considering earlier work (Bødker & Kyng, 2018; Clarke, et al., 2018; Gooch, et al., 2018).

For the stakeholders of a city or an organization, there are considerable rational and emotional advantages in the possibility to actively participate – simply because participation

• is a basic principle of our democratic societies.

• allows orientation and control for every single stakeholder.

• fosters security and trust for every single stakeholder.

• supports belonging, which results in bonding, identification, self-esteem, and satisfaction which are part of the basic human needs.

• offers holism, i.e., meaning, being part of a larger whole.

General objectives in “stakeholder participation” are strengthening/increasing:

• Legitimacy [democratic function]

• Common Welfare and Identification through participation [democratic and social function]

• Civic resp. Employee Competence [democratic and societal function]

• Transparency [social and economic function]

• Acceptance [social and economic function]

• Quality (of results) and Satisfaction (of citizens resp. employees) [societal and economic function]

• Efficiency and Effectiveness [economic function]

Therefore, a special participatory design approach was developed for this Design Café that empowers participants by using the highest stakeholder participation level “collaboration and co-creation.” It follows the idea that a sustainable city and community must be humane, sociable, and cooperative (Streitz, 2021b, Streitz et al., 2022).

5. The approach of the HCII Design Café

5.1. Approach and method

The Design Café creates its special impact based on careful considerations in terms of overall guidelines and attention to detail. It is a composition of a guided format combined with and inspired by processes of informal communication and exchange of ideas. The setting facilitates the generation of new ideas and energy to pursue them. It fosters an interdisciplinary, open-minded, and curious exchange of people from all over the world who did not know each other beforehand and provides a fertile ground for new solutions. Appropriately aligned moderation, some minimal rules of the game and a set of selected topics provides the necessary structure for achieving results. The joy of exchange in groups and the enthusiasm to find new answers to urgent questions and/or problems inspires the participants. Accordingly, it is guided by principles that promote high motivation of the participants and high-quality results:

• Urgency/Relevance: The Design Café should address a relevant purpose. The topics show some overall urgency and have a personal significance for the participants (“what’s in for me”; new ideas for their own work, self-development, etc.).

• Voluntary: Participation is voluntary. The participant who attends is the right participant. The participants decide to which topics they want to contribute.

• Diversity: When addressing the participants (via the invitation to the Design Café), emphasis should be placed on interdisciplinarity, multiculturalism, multi-generations - as a foundation for diversity of perspectives and creativity.

• Positive Emotions: Providing a relaxed café atmosphere. Actively encouraging fun in participation and exchange – as a foundation for trust and creativity.

• Rules of the Game: Define and communicate the rules. They include: every idea is welcome. No idea is prematurely evaluated and criticized. All participants actively share their knowledge for creating a comprehensive pool of ideas. Be inspired by others and develop ideas further. Unplanned and unexpected content is welcome, as it can be useful for creativity and innovative ideas.

• Moderation: The moderator actively involves everybody, promotes mutual exchange and creativity, and structures the different phases of the format. Having a sound experience with moderation and agile participative formats is a prerequisite.

• Commitment: Clear statement in advance about what will be done with the results of the Design Café.

In relation to the number of participants, we recommend the following: A Design Café should provide at least 3 questions for discussion at the different bar tables. A maximum of 10 participants should take part at each bar table, because otherwise the groups will be too large to ensure an intensive exchange and the involvement of everybody. The format can be flexibly designed to scale via the number of tables (and corresponding table hosts), number of questions and number of participants.

5.2. Setting and process

The HCII2023 Design Café started with an introduction into the format, its goal, its mindset, its agenda as well as into the topics for discussion and creativity. This was accompanied by socializing among the participants in an informal café atmosphere with drinks and snacks. The opening speech on “Designing Human-Environment Interactions from the Perspective of the UN SDGs” provided information and inspiration for the participants. Prior to the Design Café, six issues were identified and elaborated by the organizers and framed as six questions (see below in Section 6) for discussion and creation during the Design Café.

Each issue was assigned to one of six bar tables distributed in a very spacious room. To foster a dynamic exchange, a special participatory setting (derived from the world café approach) was employed. Participants selected four issues out of six for their discussions and creative engagement. In a rotating process, each subgroup that worked at one of the six tables moved after a prescribed time to another table. This rotation happened four times. Thus, all six issues were discussed and elaborated four times from a different group of participants, building on what existed already at the table before.

Each table had an experienced researcher as a “table host.” The table hosts did not rotate but stayed at their table all the time. The role of the table host was to keep the discussion going, but not interfering with the content, and informing participants about the current state of the discussion on the issue at this table when a new subgroup arrived. The participants elaborated the issues in-depth by writing their ideas on a large paper tablecloth and on pin boards besides the tables, reflecting on them, and editing the notes if needed. A comprehensive focused picture emerged about key drivers and respective key activities for achieving the objectives stated in the six questions. After four rounds of rotation were completed, the table hosts presented the ideas and results on the specific issue/question at their table to all participants in a plenary session.

Finally, each participant used colored voting points to vote for the most important ideas out of all ideas that were created during the Design Café. The results are described for each issue/question below in Section 6.

To stimulate the transfer to practice in their everyday activities, all participants received a specially prepared card and were asked to write down a so called “take-home idea.” It should be their most important idea from this Design Café to take home for realization. They are invited to send a statement after 6 months on how they transferred their “take-home idea” into practice. The underlying rationale is that if one wants to have an impact in the real world, ideas and strategic questions must become actionable. We call this from “Think Tank” to “Do Tank.”

5.3. Perspectives for discussions and creation

The table hosts were provided with some guidance on how to stimulate creativity and to structure the discussion at the tables. Each issue and corresponding question could be discussed from different perspectives. The perspectives are based on the “Participatory Innovation Model” (Streitz et al., 2022). The six-dimensional model (Figure 4) facilitates a structured process, providing access to existing knowledge and information and allowing to exploit valuable synergies. It contains the essential “fields of action” in a city or community: Environment (energy and environment), People (access to information and education), Economy (industry, trade), Government (information, digitization, guidelines), Infrastructure (IoT, ICT, Security, Data Platforms), Living (Mobility, Health, Society, Culture). Its dimensions have an overlap with a taxonomy proposed by Giffinger et al. (2007) who had the objective to evaluate and rank cities. Among others, it differs with respect to positioning infrastructure and mobility.

Figure 4. The “participatory innovation model” (© MarkenFactory – used with permission).

6. Questions that move us

Based on the challenges and conceptual framework introduced in previous sections of this paper, we identified six issues of high relevance and formulated them as questions to be presented to the participants of the HCII2023 Design Café. In addition, a short abstract of the rationale for selecting it was provided for each issue. The six issues are:

Issue 1: Inclusive and ethical smart cities.

Question: How to create inclusive and ethical smart cities?

Abstract of issue:

This task is not only in terms of definitions a multidimensional challenge but also in terms of implementation and requires input from multiple disciplines and stakeholders. How much must be subject to regulations (e.g., GDPR, Ethics Guidelines for AI)?

How much can be achieved by appropriate voluntary behavior of platform providers and city authorities?

How much can be realized by creating a common purpose and culture?

Issue 2: Trust between people and smart environments.

Question: How to establish trust between people and smart environments?

Abstract of issue:

People want to trust smart environments in handling their data and making decisions that are appropriate, safe, and secure so that they can rely on the systems. This trust is being challenged when confronted with different/unpredictable reactions by smart environments in similar situations. (see also Guidelines on trustworthy AI).

Issue 3: Privacy in disappearing computer environments.

Question: How to address privacy concerns in smart environments that adopt the “disappearing computer” paradigm?

Abstract of issue:

Especially sensors and actuators, but also processing units, are hidden in the environment and thus invisible to people in the environment (e.g., a room, a building) or passing by (e.g., in a train station, airport or - in general - public spaces). This situation is not transparent and does not provide the necessary affordances for how to interact and what to expect.

Issue 4: Explainability and transparency of policies and measures.

Question: How to promote explainability and transparency of policies and measures to citizens of smart cities or in smart environments in general?

Abstract of issue:

Citizens will adopt and act according to policies and be in favor of measures if they understand the reason why and the underlying rationale. Like Explainable AI, it is a great challenge, but also opportunity. Citizens will engage once they see a mutual benefit.

Issue 5: Incentives and rewards for engagement and sustainable behavior.

Question: How to design incentives and rewards engagement and sustainable behavior in smart cities at a personal as well as collective/corporate level?

Abstract of issue:

There are in principle two options: influencing behavior by prescribed rules and legislation or triggering voluntary behavior by persuasive design and insight in the relevance of the desired behavior. Of course, it is never either or, but one must look at the design trade-offs between the options and meaningful combinations.

Issue 6: Measuring success and impact in sustainability projects.

Question: How to measure success and impact in sustainable smart city projects?

Abstract of issue:

Different organizations, institutions, stakeholders use different definitions of “sustainability” and accordingly different parameters and threshold values for defining their goals and how to measure them. This has implications for the goals and for when they are achieved. The challenge is to identify overarching success factors at a qualitative level.

The following Sections 6.1.–6.6. present first the context and theoretical background for each issue and then the results of the discussions at the six tables, each assigned to one issue. The pictures show the tablecloths where the participants placed their ideas using post-its or drawing on the paper directly. The tablecloths were initially placed horizontally on top of the round tables used for the discussions. The pictures here show them attached to the vertical flip charts used for the presentation of the results in the final plenary discussion.

6.1. Inclusive and ethical smart cities (Issue 1)

6.1.1. Context and theoretical background

Two overarching questions which arise when dealing with the development and deployment of smart cities are how to make smart cities inclusive, and how to make the technological infrastructure behave according to commonly accepted ethical rules. These two interrelated aspects constitute not only a multidimensional challenge, but also an objective which requires input from multiple disciplines and stakeholders.

Various definitions of inclusiveness are offered in the literature in relation to smart cities. According to de Oliveira Neto and Kofuji (2016), a smart city must reinforce the participation of everyone recognizing the diversity of citizens, struggle against the segregation of minorities, and try to eliminate as much as possible physical and digital barriers. Malhotra et al. (2021) claim that smart cities in order to be inclusive need to be accessible to all, adaptable and affordable, addressing the needs and aspirations of the elderly, the poor and disadvantaged and people with disabilities.

Fundamental aspects of inclusiveness in smart cities are the public availability of the collected data for everyone’s benefit (e.g., Lee et al., 2020; Mercille, 2021), as well as the collective participation and engagement of stakeholders and citizens throughout the design process (e.g., Laenens et al., 2019) as well as the life of the smart city (Paskaleva et al., 2015).

While a wider discussion of ethics is well beyond the scope of this paper, ethical issues in relation to Smart Cities can be roughly considered as mainly related to the collection, use and availability of data (Kitchin, 2016). Privacy, transparency, explainability, and lack of bias are often mentioned in the literature as the main ethical challenges in the context of smart cities (Chang, 2021; Ahmad et al., 2022; Ryan & Gregory, 2019). The latter is particularly important insofar the biased collection and use of data can influence decision-making in discriminatory ways (Bianchini & Avila, 2014). This is therefore a significant intersection between inclusiveness and ethics in Smart Cities. In order to address ethical issues, authors suggest the elaboration of rule frameworks based on a human-centered perspective and human values, as well as of evaluation frameworks to assess the degree to which a smart city respects the set rules (e.g., Bianchini & Avila, 2014). Another context is provided by the efforts of the IEEE Global Initiative on Ethics of Autonomous and Intelligent Systems (A/IS) and its recommendations for ethically aligned design (IEEE, 2021).

6.1.2. Results of the discussions

The discussion of the first group started with an analysis of the stakeholders of a smart city designed to be inclusive and ethical, including on the one hand typical decision makers such as the city government bodies and institutions, policy makers, professionals such as designers and IT developers, health providers, academic and research institutions, and on the other hand requirement holders such as the inhabitants at large, including vulnerable groups such as people with disabilities and older people, but also city visitors. See Figure 5 for an overview of the aspects discussed.

Figure 5. Tablecloth discussion items for issue 1 on flip chart for plenary presentation.

As mediators between decision makers and requirement holders, sociologists and activists were mentioned. Their role would be to raise awareness on human needs and requirements at a societal level, as well as enhance participation motivation for everyone in the city. Additionally, it was mentioned that philosophers have an important role to play in an inclusive and ethical smart city, since clear and operational definitions of the involved concepts are needed and methods for driving dialogue and agreement between different stakeholders are necessary.

The discussion then turned to investigate how the two main concepts of the question interrelate. The matrix above (Figure 6) was drawn and the four potential combinations emerging from it were analyzed.

Figure 6. Combination matrix of ethical and inclusive aspects (issue 1) (detail from Figure 5).

Apart the obvious cases of a smart city which is both inclusive and ethical, or both non inclusive and unethical, more interesting cases were the mixed ones. It was pointed out by the participants in the group that a smart city may be inclusive but unethical, because it violates other ethical principles not directly related to inclusion, such as, for example, privacy. Also, despite the initial denial, the group arrived at the conclusion that a smart city could potentially be non-inclusive but ethical, because ethic norms are culture-dependent, and some cultures could not value inclusiveness to the same extent as others³. This led to discussing the context- and culture-dependence of the values which should inform a smart city. Inclusion is also to be considered as dependent on various factors. To illustrate this issue, the example was mentioned of the lack of availability of women toilets in India and other countries, as witnessed by movies like “Toilet” and “Susu”⁴. This situation was considered as a case of gender discrimination which is culture-dependent and would have a significant impact on the realization of smart cities in India. It was suggested that the matrix could be used as an initial assessment tool for a smart city, followed by more specific instrument depending on the screening outcomes.

Overall, identifying needs and requirements was deemed as a basic challenge for the development of inclusive and ethical smart cities. A suitable process to achieve this objective should be transparent (making data public), continuous (as needs and requirements evolve over time) and involve multiple voices and perspectives, and especially those of people struggling with physical and digital accessibility issues.

Given the wide variety of needs that may emerge, the smart city should be flexible and provide adaptive support and options for persons with different requirements.

Appropriate technological tools may be designed to support related processes in the smart city. Suggested examples are a monitoring infrastructure for automatically capturing the evolving needs of the citizens and making the data publicly available, as well as AI-based dialogue support tools inspired to philosophical approaches such as the Socratic method (Harbers et al., 2019).

6.2. Trust between people and smart environments (Issue 2)

6.2.1. Context and theoretical background

Trust plays a crucial role in moderating the adoption of technology by end users, thus influencing its reception and eventual use. This is reflected in research on technology acceptance and the pertinent developed models. Trust is included as one of the most frequently used external variables that affect behavioral intention or use behavior and their moderating constructs, namely perceived ease of use, effort expectancy, social influence, and facilitating conditions (Williams et al., 2015). Simply put, when a person feels that they can trust a technology, they find it easier and more effortless to use, and as a result, they intend to use it. If the given technology actually proves to be trustworthy in practice, individuals are more likely to adopt it, and depending on various additional factors, use it on a regular basis.

Considering its importance and potential impact, trust has been recognized as an influential factor of technology acceptance across different contexts, such as e- commerce, trading, internet banking, assistive technology, and ubiquitous computing (Ntoa et al., 2017). Ubiquitous computing environments, proposed as environments wherein computing devices are interweaved and indistinguishable (Schmidt, 2010), can be seen as the precursor of Ambient Intelligence and smart environments that represent human-centered ubiquitous computing environments including a multitude of interconnected embedded systems, collecting data from sensors and devices and processing that data to aid the user in a pervasive, nonintrusive, and transparent manner (Ntoa et al., 2022). In Ubiquitous Computing trust is studied from two perspectives (Connelly, 2007), namely if the information collected about an individual is kept as confidential as possible, which is directly relevant to discussions in Question 1 of the Design Café, as well as in terms of trusting an application or system to behave as expected, considering its potential to adapt its behavior.

Trust in smart and AI-enabled environments constitutes an area of scholarly discourse currently in the foreground, having an impact in the overall acceptance of AI technologies (Choung et al., 2023). In this regard, trust pertains to data collection, algorithm and model design based on the collected data, as well as to how the system behaves, but also on its understandability and explainability (Capel & Brereton, 2023). Shin (2021) explored through a survey with 350 participants the effect of explainability on user trust and attitudes toward AI, and in particular if perceived transparency, fairness, and accountability positively influence user trust in AI. Degen and Ntoa (2021), in the context of a co-creation workshop on Human-Centered AI, identified that trust was the most important aspect for participants and identified seven research questions related to AI and trust addressing explainability, trust measurement, and trust impact. Glikson and Woolley (2020), through literature review research, highlighted that trust is not only associated with the rejection of a technology (disuse), but also with misuse through inappropriate overreliance on technology and with abuse. Aiming to specify the qualities of an AI system in order to be trustworthy, the Independent High-Level Expert Group on Artificial Intelligence, set up by the European Commission, identified that trustworthy AI should be (1) lawful, respecting all applicable laws and regulations, (2) ethical, respecting ethical principles and values, and (3) robust, from a technical perspective but also taking into account its social environment (European Commission, 2019).

When it comes to smart cities, trust is closely related to security and privacy, revolving around the efficient and secure transmission of large volumes of data.

Challenges to trust in smart cities include GPS-based tracking data integrated with detailed personal information, communication between devices, the large amount of data possessed, as well as potential social inequalities and social bias (Ismagilova et al., 2022). Solutions proposed in the literature to enhance trust in the context of smart cities include blockchain solutions to ensure trusted transactions and better data control, federated identity to setup a circle of trust around an identity provider, identity management systems, as well as well-established techniques such as authentication, authorization, and access control (Hui et al. 2017; Ismagilova et al., 2022).

6.2.2. Results of the discussions

The importance of trust in intelligent environments was introduced to the Design Café participants, initially drawing a parallel with real-life and interpersonal relationships, in the context of which trust is not only gained between two parties but also maintained. Moving to trust between people and smart environments with a particular focus on citizens and smart cities, table discussions revolved around the following questions:

• (Q1) What does trust entail?

• (Q2) How can it be pursued?

• (Q3) What could and should be done, from a design and/or technical solutions perspective, to establish and maintain the trust of citizens in smart cities?

The table host passed these questions to the participants and asked them to freely share their thoughts, brainstorm, and draw or write down ideas, questions, and proposed solutions on the tablecloth paper. Participants’ contributions yielded results across all three explored questions organized on two large “tablecloth” papers (Figures 7(a,b)).

Figure 7a. Tablecloth discussion items for issue 2 on flip chart for plenary presentation.

Figure 7b. Tablecloth discussion items for issue 2 on flip chart for plenary presentation.

Regarding what trust entails, participants contributed short definitions as well as their perceptions of the components of trust in smart cities. Trust was defined as feeling that “the city is there for you” and feeling safe in the city, but also comfortable and free. Elaborations on what makes a city trustworthy revealed additional insight on the qualities that should be exhibited, namely accessibility and inclusivity – placing explicit emphasis on women and cultural diversity – responsibility, transparency, explainability, context awareness, and personalization. From the citizens’ point of view, it was suggested that they should have a good overview of the environment and understanding of the city rules, but also be in control of the interaction. It is evident that trust is not a monolithic concept but a multifaceted attribute (Holliday et al., 2016) that expands in complexity as the technological complexity increases and the scale of the application context broadens.

Having specified several aspects of trustworthy smart cities, the discussion deepened aiming to determine how trust can be pursued. Participants were urged to approach the problem at hand from multiple perspectives, exploring their experience as citizens, HCI and UX experts, or system designers, but also from a policy perspective. In this respect, the theories that “good design enhances trust” and that “successful UX plus successful decisions equals trust in the smart city” were postulated. Further drilling down to how good design and successful UX can be achieved, particular focus was placed on collective creativity, shared knowledge, and citizen participation in the design process, in accordance with the foundations of Human-Centered Design and Human- Centered AI (ISO, 2019; Margetis et al., 2021; Shneiderman, 2022).

From a design perspective the importance of ensuring that the environment makes optimal decisions was emphasized, as well as the need for capitalizing on successful solutions and employing in this respect recurring design patterns. Participants also identified that the design process should be driven by the motivation to meet the needs of all persons, while avoiding the pitfall of yielding infinite design solutions. This is aligned with the notions of Universal Access and Design for All, which advocate the need for proactively considering factors of diversity and accommodating them early in the design process, avoiding the need for a posteriori adaptation (Stephanidis, 2021).

From an operational perspective, it was stressed out that the design process should involve continuous learning and iteration to achieve a trustworthy smart city as identified in the first round of discussions, thus adopting the well-established concept of iterative design (Nielsen, 1993). Finally, from a policy perspective, the need for the establishment of a pertinent authority was pointed out. The role of such an authoritative body was shortly expanded indicating that it should oversee data management but also issue trustworthiness certificates, which could then be used by IT applications, services, and ecosystems in the smart city to certify that they adopt best practices or adhere to pertinent principles, guidelines, and even standards, if available.

Subsequently, discussions revolved around how the smart city should manifest its trustworthiness to citizens and what technological solutions could be developed in this regard. A considerable part of the discussion was devoted to privacy-related issues, with contributions highlighting that passive monitoring should cease, that the smart city, all stakeholders, and individual service providers should handle data responsibly, while special attention should be paid to data storage and data tracking issues. Citizens should be informed regarding which of their data are used and how and be provided with the possibility to opt out of their data being collected. Technological solutions suggested in this regard involved the visualization of citizens’ data paths in the city, which could be further strengthened with the use of Augmented Reality technologies, connecting the digital information with the built city environment. Considering trust as it was initially defined as “feeling safe” in the city, proposals on surveillance for high-risk areas to protect women and vulnerable persons were suggested. Further ideas were contributed in this regard with the aim to promote safety in a privacy-preserving way, suggesting that thermal cameras combined with other sensor data (e.g., measuring stress) could be used to identify persons at risk.

In summary, participants contributed ample perspectives and concerns and offered valuable suggestions. Trust was explored from various angles and several attributes were examined, however, in a nutshell, it could be deduced that a trustworthy smart city is one where its citizens feel safe, comfortable, and free, but also that they can count on it. In this regard, well-established HCI principles, methods, and approaches can be employed in the pursuit of designing and developing trustworthy smart cities. However, it is equally imperative to institute appropriate policy-making initiatives to prioritize, safeguard, and preserve a “trustworthiness by design” approach.

6.3. Privacy in disappearing computer environments (Issue 3)

6.3.1. Context and theoretical background

An important concern of the future disappearing computer environments (e.g., smart cities as envisioned in the context of UN SDG 11) is citizens’ safety and particularly their privacy safeguarding. In such environments, where each person constitutes a source of continuous data generation, aspects such as invasive data collection, lack of user awareness and protection, data ownership, and retention are of paramount importance to be considered and addressed.

Information privacy is defined by Westin (1967) as “the claim of individuals, groups or institutions to determine for themselves when, how, and to what extent information about them is communicated to others.” Although information privacy is an obvious right for individuals, in disappearing computer environments it is not self- evident. In such environments, persons are continuously being monitored by a plethora of sensors, such as cameras, microphones, etc. sharing their personal data, most of the time, unwittingly. Moreover, in disappearing computer environments the collected data can be transferred beyond physical boundaries (e.g., outside our home walls) to virtual services running on cloud data centers many miles away and probably in a different country. Könings et al. (2016), define another dimension of privacy that should be considered in pervasive computing environments, coined as territorial privacy, which should aim at controlling all the physical and virtual entities that are present in the user’s physical and virtual extended territory, excluding any unwanted entities from the private territory.

Although many steps have been taken lately by policymakers worldwide towards regulating to what extent digital or physical services can have access to human data and information (e.g., the General Data Protection Regulation - GDPR⁵), is not enough if it is not supported by user awareness and protection regarding their privacy and data control rights. In pervasive computer environments, Privacy Enhancing Technologies (PET) (e.g., Prasser & Kohlmayer, 2015; Adamakis et al., 2023; Pinkas et al., 2020; Bernabe et al., 2019) can constitute a key enabler for users’ privacy protection. A systematic analysis of PET is provided in Cha et al. (2019) aiming at identifying the current state of these technologies in various fields and exploring if they comply with the latest legal principles and privacy standards. As highlighted by this analysis, such technologies are not yet mature enough to address the principles summarized by well-established regulations and standards, such as GDPR and ISO 29100⁶, whereas more effort is needed for holistic privacy preservation. However, the utilization of emerging technologies such as blockchain and differential privacy can leverage PET’s effectiveness in reducing privacy threats. Despite the technological efforts to develop technologies that can protect users’ privacy, awareness still remains a challenge. A first step towards this direction is the privacy awareness cycle specified in Rizi and Seno (2022), identifying seven different phases in the private data lifecycle, and discussing actions that should be considered for user awareness towards mitigating the risk of unintentional provision of personal information.

Determining who owns the data collected in pervasive computing environments and how long it is retained is also critical. Without clear regulations and policies, data could be stored indefinitely or shared with third parties without individuals’ consent, leading to potential privacy violations and data misuse. This aspect is also known as data sovereignty. As elucidated in Hummel et al. (2021), data sovereignty refers to control, ownership, and claims regarding data, which affects individuals, societies, and nations. It deals with a complex interplay of values, particularly control and authority over data that intersects with considerations related to inclusive decision-making and fundamental rights. To enhance discussions related to digital data governance, a clearer distinction between the description of data sovereignty, the challenges it poses, and the strategies to address them is crucial.

6.3.2. Results of the discussions

The above-mentioned privacy aspects constituted the basis for setting up the narrative used for introducing Issue 3 to the Design Café participants. Specifically, all participants involved in the collaborative design for this issue were presented by the Table Host with the corresponding concerns as specified above, for being acquainted with the discussion topic and to start thinking of pertinent problems and solutions. After some initial general deliberations, the participants started exchanging some first thoughts and conveying their experience on the topic, while they were requested by the Table Host to start writing their thoughts on post-its available on the table. Initially, the process was a top-down approach that started with the specification of the problems corresponding to citizens’ privacy in disappearing computer environments.

As illustrated in Figure 8 (yellow post-its), the problems identified by the participants were classified into four main categories: (a) data invasion, (b) data sovereignty and data control, c) citizens’ awareness, and d) data minimization and retention. In more detail, participants expressed major concerns regarding data invasion problems in pervasive computing environments. They pointed out that these environments collect vast amounts of citizens’ data without their knowledge or consent. This behavior can lead to an ambient feeling of constant monitoring, which could ultimately result in a dystopian future (Orwell, 1949), where everyone and everything is surveilled by a “big brother”⁷ and social control becomes a norm. In terms of data sovereignty and personal data exploitation by 3rd party entities, the main participants’ concern was “how citizens’ data are sold,” determining the important aspect of who is the data owner, who is permitted to control it, and further process it. Regarding citizens’ awareness, most of the participants agreed that there are still a lot of steps to be taken towards making citizens aware of their data control rights, under which legal and ethical framework can be collected, and how their data can be used by other entities. A very coherent concern about the data collection and use was also identified as a risk, pinpointing the need for regulating a data minimization and retention policy.

Figure 8. Tablecloth discussion items for issue 3 on flip chart for plenary presentation.

Then, for each of the defined problems, the participants were asked to elaborate on principles, approaches, or solutions that could be applied in the context of pervasive computing environments to tackle and eventually overcome the defined problems, if possible (Figure 8, green post-its). To that end, several propositions were placed on the table, many of them with a big potential for complementarity. For example, regarding data invasion, an easy-to-implement solution would be the notification of citizens about their data collection when they reside in public places or interact with online services. Regarding data sovereignty and control, several solutions were proposed, spanning from specifying strict regulations about data processing from third parties to making citizens able to have full control of their data by defining which can be open to third entities and which cannot be accessed at all. In terms of citizens’ awareness, three were the main pillars suggested for addressing the pertinent problems: a) awareness campaigns, b) integration of privacy awareness in education, and c) specification of legislation focused on protecting the digital rights of users. Regarding data minimization and retainment, the participants indicated that technology could help towards this dimension, e.g., using anonymizing approaches before the data is stored and then processed, as well as, raising the awareness of people via social events and citizen meet-ups.

After discussing potential approaches to tackle the problems in each of the specified domains, participants began to consider practical applications and services to implement these solutions. Many of the proposed applications are characterized by their simplicity and can serve as a solid foundation for immediate action by numerous cities. In addition to offering valuable recommendations for expanding current regulations and legislation to address the identified problems, which is already a priority for policymakers worldwide, including the United Nations, as well as on a regional level such as the EU and the USA, many other solutions were proposed. For example, related to controlling the way that personal data can be used by third persons or legal entities, a personal digital platform could be provided through which each citizen can get detailed information regarding how their data are used and specify who can use them, by adjusting the sharing level of their data. Such mechanisms, already exist in several social media applications, so from a technological perspective such a platform can easily be devised, at a local or regional level at least, acting as a meta-service that aggregates users’ privacy adjustments in one place. Finally, another, important aspect that was pointed out by the participants, was the necessity for interventions in public education towards a diversity of actions, including the integration of privacy as a topic in school curricula, or the use of social media for reaching out to young persons and educating them.

6.4. Explainability and transparency of policies and measures (Issue 4)

6.4.1. Context and theoretical background

When trying to answer the trigger question of Issue 4 “How to promote explainability and transparency of policies and measures to citizens of smart cities or in general smart environments?,” one should first clarify what is to be understood by these terms in the current context. While there have always been requests for explainability and transparency, especially in political decision-making contexts, the issue became more urgent with the rise of decisions made by algorithms using machine learning based on

models that are not transparent and, in many cases, misleading or completely wrong (Streitz 2019) and even creating AI-based disinformation campaigns (Zhou et al., 2023). Due to the increase of critical comments (e.g., bias of test data, black box models), Explainable AI (Kandul et al, 2023; Saranya & Subhashini, 2023; Samek et al, 2019) is becoming a prominent and much needed area of research. It has been identified by the

U.S. government as a key tool for developing trust and transparency in AI systems (CRS Report, 2021). The OECD formulated their Principle 1.3 “Transparency and Explainability” as a major request for their set of AI principles (OECD-AI, 2019). As a result, many players in the field do not tire of stressing that their systems are meeting high standards of transparency, which are difficult to verify.

The concept of smart cities and smart environments assumes that large amounts of data can be and in fact are collected, analyzed, and exploited for a wide range of applications. These data are, on the one hand, used by city administrations and municipal authorities for planning and decision-making purposes as well as controlling their infrastructures, e.g., via traffic control centers. In many cases, there is an overlap with surveillance methods motivated by security aspects, but they also have an impact on privacy issues. On the other hand, data will be – hopefully more and more – made available to citizens or used to inform citizens. An important dimension is to describe and explain policies and measures that govern planning and decisions by municipal authorities. Citizens will adopt and act according to policies and be in favor of measures if they understand the reason why and the underlying rationale. Like Explainable AI (Ozmen Garibay et al., 2023; Wachter et al., 2017), it is a great challenge, because one is usually dealing with very complex systems, but also an opportunity. Citizens will engage once they see a mutual benefit and a win-win situation. Moving beyond only informing citizens is to provide - even raw - data for further processing by citizens in their activities like neighborhood initiatives, improving local infrastructures, etc. This is often part of a more general Open Data Model approach as, e.g., in Amsterdam and other European cities (Open Data, 2022). They provide a basis for co-creation and co- design in city planning and other joint collaborative city-citizens activities as proposed in the Citizen ⬄ Cooperative City Contract (Streitz, 2021b).

Objectives of transparent descriptions and explanations include the following: to show policies and regulations; to clarify who oversees data collection, decision making, and distribution of measurements made regarding the policies; to inform how citizens can participate in evaluating and acting upon this information, and to track the record of success in this overall effort. When emphasizing “explainability and transparency,” it is not only important that it happens at all but how it is implemented. Thus, the design and communication of information plays an important role. It goes without saying that advanced techniques are required to reduce misinformation and disinformation to a minimum, especially in a time of increasing fake news shared in social media. The information provided must be trusted, because trust is a highly valuable and needed ingredient for all relationships between citizens and city administration. At the same time, the request for appropriate explainability and transparency is a measure of controlling political and economic decision makers, hopefully also preventing mismanagement and corruption.

It would be beyond the scope of this section, to go into detail what good and appropriate information design for citizens is all about. There are many books and papers about it (Marcus 2015; Rendgen & Wiedemann 2019, 2020; Tufte, 2001). In the USA, and elsewhere, many organizations and publications have attempted to provide official information. For example, the US Census Bureau and other government agencies published atlases and other documents with their latest data.

Books have appeared aiming to help citizens understand how to interpret and use these data (Bouk, 2022). Explaining which factors influence relevant parameters, e.g., energy consumption of a city, can also be used to persuade people to change their behavior. Aaron Marcus and Associates (Marcus, 2015) developed ten award-winning mobile applications (e.g., the Green Machine) using persuasion theory to change people’s behavior regarding ten different areas, e.g., nutrition, energy conservation, wealth management, driving, education, innovation, marriage, and happiness.

A particular prominent example of providing transparency in smart cities is the mandatory register for sensors in Amsterdam. In 2021, the municipality issued a regulation (Cities Today, 2021) that all private companies, research institutions and government organizations are now obliged to report their sensors deployed in public spaces. The information is being displayed via an online map (Figure 9) to give residents more insight into how, where, and what data are collected from sources such as cameras, air quality and traffic sensors, Wi-Fi counters, and smart billboards. The map shows the type of sensor, the owner, and whether personal data is processed (Amsterdam Sensorenregister, 2023). There are plans to extend these registers to a National Sensor Registry – SensRNet (van Andel et al., 2021). In addition, Amsterdam – together with Helsinki – were the first municipalities to launch artificial intelligence (AI) registers detailing the AI systems in use, including information on datasets, data processing, and whether the tools have human oversight (Cities Today, 2021; Amsterdam Algorithm Register, 2023). These are examples where measures for introducing transparency also provide contributions to the issue of privacy. Citizens are informed and can avoid areas if they do not want to be monitored, e.g., by cameras in a particular public space.

Figure 9. On-line map of Amsterdam mandatory Register for sensors (screenshot from website https://sensorenregister.amsterdam.nl/).

6.4.2. Results of the discussions

The findings of behavioral science show us again and again in the most diverse contexts, that people will adopt and act according to policies and be in favor of measures if they understand the reason why and the underlying rationale. This applies to employees in companies as well as to residents of a city. Like Explainable AI, it is a great challenge, but also an opportunity. Citizens will engage once they see a mutual benefit. For this reason, the participants of the Design Café examined the question of how to promote explainability and transparency of policies and measures to citizens of smart cities or in smart environments in general?

A key outcome of the discussions was that the scale of these challenges has yet to be recognized by those responsible. Even if artificial intelligence helps to organize all data, it must first be defined which data a citizen needs to see, which data should be shared in the public community, and how privacy can nevertheless be protected. From a UX perspective, the interaction between citizens and data must not only be facilitated, but made simple, so that every citizen can easily access this data, but also be able to understand it, interact with it and get meaningful insights, regardless of their background or expertise (Vitsaxaki et al., 2023). This is a responsibility especially for experts on design, user experience, and usability. It faces the major task of developing conventions and standards for information presentation and use.

Another key outcome of the discussion was that the processes should not be developed and implemented top-down. A participatory involvement of all relevant stakeholders is essential for successful implementation and corresponding acceptance by citizens. As instruments, it was suggested, for example, that townhall meetings should be used to discuss together with interested stakeholders in the city what information is needed and what conditions for data protection, openness, and diversity need to be considered.

The experts agreed that every municipality and its administration with great responsibility must pay attention to appropriate transparency and explainability. It must strike a balance between the amount of information (avoiding information overload) and the protection of privacy (collect only the data that is necessary and useful in each case) and emphasize the value of private data. Too much information can lead to mistrust, ignorance, or confusion, too little privacy protection can lead to mistrust and lack of acceptance. City governments need appropriate methodological competence for these tasks.

The experts came up with concept-ideas on how to promote explainability and transparency of policies and measures to citizens of smart cities or in smart environments in general (Figure 10).

Figure 10. Tablecloth discussion items for issue 4 on flip chart for plenary presentation the ideas and recommendations by the participants can be organized in six clusters.

6.4.2.1. Education and involvement

The development of Educational Programs at universities, design schools and other educational institutions - adapted to the respective target group - is essential to convey appropriate information. Where such courses already exist, they would need to be significantly expanded in terms of scope, variety, and topics. Creating Exhibitions on a regular schedule that highlight important information play an important role, too. These exhibitions could be physical, could be accessed online, or encountered via augmented reality at numerous locations. Design documents can be presented as Cartoons or Graphic Novels employing non-textual means to convey key information. Examples of recent documents include the graphic novel versions of Sapiens in two volumes by Harari (2020, 2021). Furthermore, Information Museums might show historical trends for various classes of information. These could be virtual museums that citizens could visit as well as physical museums.

6.4.2.2. Transparency and understanding

Digital Shadows are very relevant for transparency and understanding. They are design representations of data collected and disseminated as abstractions, which can represent alternative formulations or historical versions. Additionally, it is of importance to create a Sign System – the experts recommended, e.g., Emoji Stickers etc. – that help all viewers to understand key information more quickly. In general, the experts emphasized designing simple, clear, consistent explanations based on a Set of Conventions (customized for specific demographics, languages, education, etc.) that everyone can become familiar with and use. In other words, standardized data display conventions would be as well-known and understood as the Fahrenheit/Celsius thermometer scales.

6.4.2.3. Information and usage

The experts focused on applications which are highly beneficial for citizens. Examples are Design collections of Biometric Data that inform the population of trends, threats, or successes. These data could be available online, through augmented reality viewing or physical displays. Furthermore, building a set of standardized Charts, Maps, and Diagrams could be considered that enable people to become more familiar with the organization and classes of data and their typical patterns. Strategies should be developed that take advantage of Social Media (as TikTok, etc.) to communicate policies and measures, successes and failures that encourage discussion, reward participation/viewing, and diminish circulation of misinformation and dis-information. Moreover, appropriate Avatars could be designed to represent classes of information, information resources, or known clusters/patterns of information. Large-scale Ambient Displays could be considered to convey information about policies and measurements. For example, lights on the San Francisco Bay Bridge have been used as an artwork to present large-scale light displays. The Eiffel Tower in Paris has been used as a social-/political sign using different-colored lights. Advertisement Billboards can be designed so that they announce new data patterns and gather citizens awareness and attention to these displays. In effect, the Data Information Network might resemble the old television networks with competing offerings, and multiple channels of information. In this context, it is important to study and optimize the best Combination of text, graphics, and audio to convey information for specific topics to specific groups. An early example of combining Augmented Reality (AR) with traditional games resulting in pervasive gaming applications was provided by Magerkurth et al. (2003). Also, one can explore how Bluetooth Beacons and AR signs embed large-scale displays on buildings and in environments providing key information indicators appropriately updated.

6.4.2.4. Relevance and acceptance

Iconic Figures and Cuteness were identified as strong levers to increase relevance and acceptance on the part of the users: The participants talked about Iconic Figures like Smokey the Bear, which was used for many decades to promote care for US forestry lands. They proposed also real or fictional avatars who take advantage of their popularity promoting the use of official descriptions and explanations and persuading the public to engage more significantly with these data. Examples like the iconic figures of Taylor Swift, Lady Gaga, Margot Robbie (Barbie), Ryan Gosling (Ken), Hello Kitty, or the rabbit Miffy could present the latest and greatest and most significant information patterns for appropriate age, language, culture, and education groups. Accompanying, a strategy could be developed to use cuteness to make information more explainable and attractive to certain parts and peer groups of society. Perhaps branded characters could be developed to convey specific topics to specific audiences. Generally, the experts emphasize the approach to develop a strategy and plan for how to involve artists and designers in thinking out and assessing all information policies and their communication. Perhaps ongoing panels or reviews could be available to disseminate their thinking.

6.4.2.5. Privacy

The participants agreed on the importance of Free Areas for Privacy, i.e., areas of geographic space or conceptual spaces without data collection, in other words: Safe Cities or Safe Environments. These suggestions correspond with proposals like “cold zones” with “disposable identities” (van Kranenburg et al., 2020) and “ambient privacy” (Streitz, 2021b). Data Privacy is a most important value and asset. Therefore, appropriate levels of privacy for data collection and data distribution should be designed. Furthermore, appropriate techniques should be developed for gaining the Trust and Consent for data collection from all citizens for all circumstances.

6.4.2.6. Data collection and classification

How data can be obtained and classified is anything but insignificant. The experts created the following ideas: (1) Levels of Information collection per building: Consider how individual buildings might cluster the information that they collect, accounting for social classes, income/education levels, etc. (2) Reporting: Consider how and when a regular report from appropriate authorities might present information about what data has been collected, used, and disseminated. (3) Consider how and whether collecting information in parks, public spaces, etc. (4) Consider how to define Default Settings for all classes of information, how data might be collected, who could access the data, and how it might be displayed. This might require numerous Town Hall gatherings and discussions to evaluate policies. (5) Developing strategies, policies, and tactics in advance for When and How data are collected and disseminated. (6) Hidden Worlds: Consider how and whether to design “invisible” information displays for almost every place and time that could be accessed through appropriate codes on mobile devices.

This might be designed as games with incentives, like Pokémon Go, which became a world-wide phenomenon in the mid-2010s.

6.5. Incentives and rewards for engagement and sustainable behavior (Issue 5)

6.5.1. Context and theoretical background

Climate change is accelerating, creating a pressing need for sustainable practices. Environmental, social, and economic challenges demand innovative solutions that encourage individuals and organizations to engage in sustainable behaviors. To address this, urban planners and governments have developed incentives and rewards for encouraging people to adopt sustainable behavior. A recent review of 159 articles (Tura & Ojanen, 2022) on the interlinked concepts of smart city and sustainability-oriented innovation summarized growing trends on smart city development. The authors found that human-centric approaches, including incentives and rewards that encourage citizen engagement are the most successful. For example, recycling incentives provide better sustainable outcomes for waste management within smart cities. The authors also concluded more work was needed in this area.

Smart cities leverage cutting edge technology and data-driven solutions to improve the quality of life for their citizens (Sujata et al., 2016), while practicing sustainability. In this context, incentives and rewards for engagement and sustainable behavior provide a foundation for creating the desired outcomes. Smart cities can encourage citizens to adopt eco-friendly practices such as recycling, energy conservation and public transportation usage. These incentives, ranging from discounts on public transport fares to rewards for recycling, create a real benefit for citizens, motivating them to actively participate in sustainable initiatives.

Smart cities utilize data analytics to identify specific behaviors that need encouragement and then tailor incentives accordingly. By harnessing the power of data-driven incentives, smart cities as a technological and conceptual framework can catalyze change (Papangelis, et al., 2023). They can not only foster a culture of sustainability but also create a more engagement and environmentally conscious community, leaning to a greener, healthier, and more harmonious urban environment.

The concept of creating incentives and rewards to achieve desirable outcomes draws upon a rich theoretical background, encompassing both behavioral economics (Kahneman, 2003) and behavioral psychology (Zachrisson & Boks, 2012). Behavioral psychology offers valuable insights for designing strategies that can stimulate desired behavioral patterns while also mitigating undesirable ones. Through the application of behavioral psychology principles to innovative incentive programs, individuals and organizations can be motivated to embrace sustainable practices, thereby contributing to a more sustainable and responsible world.

To motivate people to change or adopt behaviors, incentives are known to work. People can be motivated to change behavior when provided with financial rewards and gaming applications work to motivate users. In one case, a game called “Payback” was used to motivate high school students to make smart personal financial choices. In one experiment, high school students were taken through the game while connected to galvanic skin response and eye tracking to gather biometrics to understand whether behavioral finance could change behavior. The findings showed that gaming was successful in providing incentives and motivations to change behavior (Rosenzweig, 2020). This behavior modification can be generalized to influence people to adopt any behavior, including sustainable practices.

Social norms can be incentivized to get people to cooperate and engage in environmentally responsible actions for the greater good. Downs (1957) expounded on this in his book “An Economic Theory of Democracy.” In the age of social media, social norms become important influencers of people’s behaviors. This can be seen in looking at the wide-ranging power of norms to influence energy use, although people may not be aware they are being influenced (McDonald & Crandall, 2015).

6.5.2. Results of the discussions

The discussion started with establishing a foundation and theoretical basis for further brainstorming and problem solving, by reviewing theories for motivation and successful incentive programs and this created the scope for further debate. Among the initial theories discussed a few were repeated in the brainstorming, such as gamification, behavioral finance (Miltenberger, 2008) and the pressure of social norms. Game Theory is quite effective in influencing people to change their behavior. Therefore, game theory principles come into play when designing incentives for sustainable behavior, particularly in communal or organizational settings. By framing sustainability as a collective game where everyone benefits from participation, individuals, social norms heavily influence human behavior. When individuals perceive that a behavior is widely accepted or encouraged within their social or professional circles, they are more likely to adopt it. Incentive programs can leverage this by creating social rewards or recognition for sustainable practices, such as giving public recognition to eco-conscious organizations or individuals.

The scope laid out the use cases for incentives and the first one discussed was bottle recycling. The bottle recycling programs charge consumers a deposit for every bottle purchase, and the deposit is returned when the bottle is returned. The idea of recycling trash has been a program in one way or another since the 19^th century in the United Sates when peddlers travelled the countryside to sell manufactured good, also purchased recyclable materials from households they visited (Time Magazine, 2016). Recycling has expanded over the years with various incentives which have resulted in more sustainable behaviors. A study focusing on virtual reward tokens created to encourage recycling among families, using incentives and rewards to improve recycling behavior, producing a webapp prototype to register the recycled plastic. By the end of the 6- week pilot project, 1053 families were registered on the app, resulting 10% of that target population. Gamification played a key role in engaging citizens and shows that people can be influenced into creating more sustainable practices by means of innovative incentive programs (Gibovic & Bikfalvi, 2021).

The entry point for the discussion was how to design incentives and rewards at both a personal and corporate/collective level. Once the foundation and framing for the use case and problem we were solving, the group used a brainstorm method to generate idea (Paulus & Kenworthy, 2019). The brainstorm portion of the session produced many examples of incentives and motivators for the use cases of recycling and transportation for smart cities. As common in brainstorm, the first pass was focused on getting ideas on paper, no order or taxonomy was applied at this stage. There were many ideas, some complementary, some duplicates of the same idea but perhaps from a different point of view.

The second iteration of brainstorming was to first editing out duplicate ideas, and once that was done sorting the remaining ideas into a taxonomy. Figure 11 shows the overview of the generated ideas. The taxonomy was a draft of a categorization scheme, still in draft form and from the creative point of view. (Jablin, 1981). Once the final taxonomy had been created, the ideas were prioritized using dot voting (Dalton, 2019).

Figure 11. Tablecloth discussion items for issue 5 on flip chart for plenary presentation the discussion concluded with an exploration of final outcomes and notes.

Individuals might be motivated by different things and at different times then groups and organizations. Organizations such as companies, NGO and municipalities could be motivated because of collective issues such as improving transit to get people to work and school, or recycling because it impacts the municipal systems. Behavioral psychology in the theories mentioned above have shown positive outcomes in getting people to change their behavior through small changes maintained over long periods of time. A good example would be an incentive for recycling bottles through the bottle deposit system whereby consumers pay a little more to use a bottle and will get a few cents back when they recycle the bottle. That has proven to produce good incomes in the United States, whereby people whose budget might be tighter are motivated to get the rebate, while the others a reminded of the importance of recycling because of the program. Some specific motivations that could be for both individuals and organizations are

• Rewards for recycling.

• Rewards for taking public transportation.

• Rewards for making homes more sustainable.

• Increase identity and positive encouragement through T-shirts and other messaging.

• Apply social pressure.

6.6. Measuring success and impact in sustainability projects (Issue 6)

6.6.1. Context and theoretical background, related work

Smart sustainable city projects are springing up all over the world. They use different smart technologies, rely on their respective economic geography, and achieve different focus goals. Consequently, a multitude of evaluation systems has been developed, composed of different parameters, calculation methods, thresholds, and weighting systems. Choosing an appropriate evaluation method to measure its success and impact is the first challenge that decision-makers will face before implementing a smart city project.

The complexity of smart cities makes it difficult to assess them simply.

Although the abundance of available urban data enhances the feasibility of implementing an intricate evaluation system, it also leads to the risk of decision-makers becoming overwhelmed by many indicators, being unable to assess the project comprehensively (Shamsuzzoha et al., 2021). Moreover, different smart city projects employ diverse evaluation systems. Each of them has its own strengths and weaknesses. The lack of uniform standards further complicates the comparison between projects.

To address the above challenges, many organizations and researchers have been working towards standardizing evaluation systems for smart sustainability cities, including the International Organization for Standardization (ISO), the International Telecommunication Union (ITU), and alliances such as the European Committee for Standardization (CEN) (Huovila et al., 2019). ISO alone has published more than 4,000 standards that can be used to assess the success of sustainability goal (Kristiningrum & Kusumo, 2021). Among them, SNI ISO 37122:2019 has been seen as the most widely adopted standard due to its structured approach, ability to measure progress over time, and potential for benchmarking against other cities (Lacson et al., 2023). In-depth discussions have been conducted on these indicators’ differences and applicability, and have provided usage recommendations for different scenarios (Huovila et al., 2019). Based on these standards, more operational evaluation systems have been proposed including the overall performance measurement framework developed by the European Union’s CITYkeys project (Bosch et al., 2017), the Uniform Smart City Evaluation (USE) framework (Kourtzanidis et al., 2021), and evaluation systems that can be directly quantified based on theoretical models (Abu-Rayash & Dincer, 2021).

Those evaluation systems serve primarily two groups of stakeholders (Bosch et al., 2017). One group comprises the managers of smart city projects. The evaluation system assists them monitoring project progress, establishing a common language among departments, enhancing governance transparency, and serving as decision support (Huovila et al., 2019). The other group consists of decision-makers within the government. The evaluation system helps them understand the status of projects and optimize future project decisions. It also contributes to the government’s city auditing, accountability, and sustainable governance, as well as residents’ welfare and energy consumption (Alsaid & Ambilichu, 2023). A robust evaluation system can also be used for computational decision support in digital twins, offering computational decision support for sustainable development planning (Corrado et al., 2022).

When the design of smart city evaluation systems is primarily intended to serve managers and decision-makers, it raises the question of how to prove its value for both the city and its citizens (Caird & Hallett, 2019). Faced with numerous measurement systems and indicators, decision-makers must consider not only the current economic status and level of data intelligence in the city but also the future urban vision and strategic development. Evaluation process is not just a post-project outcome assessment but also helps cities identify gaps and emerging innovation opportunities in their development. They need both quantifiable indicators that are easy to apply, and comprehensive indicators to reflect the multifaceted nature of smart cities and the complexity of urban systems.

In this context, our discussion aims not to design a complete and efficient smart city evaluation system, but to explore dimensions that can make recommendations for the design, implementation of evaluation systems, with a focus on human involvement.

6.6.2. Results of the discussion

In the discussion – Figure 12 shows the ideas and relationships created for issue 6 - the first focus was on those obvious, straightforward, and quantifiable indicators in evaluation systems. Simple quantifiable indicators reduce the complexity of data acquisition, making them more intuitive and easier for decision-makers and urban residents to understand. Group members discussed dimensions of indicators similar to previous research, including inputs, outputs, processes, outcomes, and impacts of projects (Bosch et al., 2017).

Figure 12. Tablecloth discussion items for issue 6 on flip chart for plenary presentation.

An example that was brought up in the discussion is as follows: the project’s outcome may be measured by the energy required to reduce outdoor temperatures by 2 °C during the day in response to the urban heat island effect. Under this assumption, compared to the construction of artificial mechanical systems that extract excess heat from pavements and lower their surface temperature (Vasilakopoulou et al., 2014), the Singapore smart city project demonstrates superior performance. It uses an intelligent environmental decision support system to simulate and optimize the city’s green structures, reducing urban temperature with minimal additional energy consumption (Lim et al., 2021). Similarly, when comprehensive assessment frameworks are too broad, standard setters can focus more on key performance indicator in specific domains (Caird & Hallett, 2019). For instance, sustainability performance of projects can be directly measured through the use of clean energy (Chen & Dagestani, 2023).

When considering evaluation indicator for project impact, the members of the discussion group place greater emphasis on people in the city, including the groups involved in the project and the groups affected by the project. The more people involved in the implementation of the project, the greater the impact of the project. A higher level of citizen participation consistently leads to better sustainable smart city outcomes (Alamoudi et al., 2023). Similarly, the more residents trust the project and approve of the project results, or the faster they trust, the greater the influence of the project.

At the same time, the well-being and behavioral changes of residents can more directly reflect the impact of smart sustainable city projects on people. Returning to the example of the urban heat island effect, when the urban heat island effect is mitigated, residents are less dependent on artificial cooling devices such as air conditioning during hot summer days, and the energy costs for cooling their homes are also reduced. The performance of a smart city project in improving the urban heat island effect can be measured by the indicator of residents’ summer cooling electricity expenses. The concept of citizen well-being we discussed here is broad, encompassing not only physical aspects that can be quantified, such as average lifespan but also psychological, emotional, and economic aspects. For example, measuring the frequency of happy and depressive emotions to assess people’s happiness index (Wang & Zhou, 2023).

The measurement of changes in human behavior and how projects promote those changes are important too. For example, one can measure residents’ mobility (Fernández-Aguilar et al., 2023) as an indicator of changes in residents’ behavior, reflecting the impact of various public transportation development initiatives in smart city projects. As for how projects change people, an important case discussed is how smart sustainable city projects reward people for pro- sustainable behaviors. The timeliness, effectiveness, ease of promotion, and engagement of the reward methods all ultimately impact the success of the project. A noteworthy case is the plastic recycling incentive project conducted in Pla de l’Estany, Catalonia, Spain (Gibovic & Bikfalvi, 2021). The project uses tokens, lotteries, and prizes to gamify incentives for residents participating in plastic recycling. It also uses a web application for direct communication with citizens, collecting and registering feedback after material recycling.

Diversified, effective, and innovative incentive programs change people’s recycling habits more effectively. Therefore, assessing how smart city projects change the sustainable behavior of residents is also an important part of the evaluation system. A good smart city plan should not only contain measures to encourage sustainable behavior of residents, but also form a sustainable cycle chain. For example, a group member suggested that there could be tax incentives for products made from recycled materials to reduce costs and prices. In this way, residents can not only buy environmentally friendly products, but also get benefits.

After analyzing numerous smart city cases, one can observe that successful projects not only change residents’ individual sustainable behaviors but also influence the behavior and structure of organizations, particularly government management departments. The development of smart cities has led to increased inter-departmental integration within government institutions. It contributes to the overall enhancement of urban planning and management, helps break down departmental silos, and enables governments to respond more dynamically and reflectively to various urban challenges, resulting in significant cross-departmental achievements in sustainable development (Haarstad & Wathne, 2019). Therefore, smart city projects should be able to drive cross-sectoral or even cross-city, cross-nation collaboration across all dimensions. Changes in organizational behavior due to smart city projects are equally important to consider in the design of evaluation systems.

Localization is another key focus during our discussion. Still using urban temperature adjustment as an example, when Singapore, located in a tropical region, needs to take action to reduce urban temperatures, Nordic cities are more concerned about how to maintain indoor temperatures with low energy consumption (Taveres-Cachat, 2019). This enlightens us that successful measurement methods for smart sustainable city projects should take local factors into account. Differences in geographical location (Perboli & Rosano, 2020), disparities in economic development levels (Lacson et al., 2023; Wang et al., 2021), as well as social and cultural differences (Šulyová & Vodák, 2020), can all have an impact on the effectiveness of sustainable plans. Correspondingly, the standards for evaluation systems should also be flexible and adaptable.

However, even with significant differences among cities on a global scale, our group still think that more cities should actively participate in the standardization of evaluation systems. Collaboration among cities, and even government departments, plays an irreplaceable role in promoting scientific development of smart city project evaluation systems. Cross-city and cross-country cooperation in sustainable projects maximizes resource efficiency and achieves the maximum impact of smart city projects. In the current context, laws and regulations remain the most effective way to change people’s sustainable behavior (Alremeithi et al., 2023).

7. The big picture: Dependencies and synergies

Our planet, our economy, our society are undergoing a serious process of change. These changes impact people, behaviors, and expectations. One of these crucial challenges is climate change which is not controllable without the consistent alignment of all thinking and action towards sustainability. As already stated in 1972 by the Club of Rome in its report “The Limits to Growth” (Meadows et al., 1972) and proven by behavioral sciences, real change to “a better world” starts at the individual and community level and the corresponding individual commitment and behavior.

Consequently, the first HCII Design Café in 2023 addressed the Sustainable Development Goals (SDGs, 2015) of the United Nations which are called “a shared blueprint for peace and prosperity for people and the planet, now and into the future.” Living in the Urban Age, we selected the UN SDG 11 on “cities and human settlements” as a focus. In addition to the sustainability target, challenges arise in the current and emerging landscape of rapid technological evolution, especially in the interaction of humans with technology embedded in the environment. In our context the environment is a city or human settlement. Thus, we selected challenge 2 “Human-Environment Interactions,” one of the “Seven HCI Grand Challenges” (Stephanidis et al., 2019), as a complementary perspective. As a result, the six issues of the HCII2023 Design Café (Section 6) were discussed under the umbrella of sustainability and human-environment interactions in cities and human settlements.

A key factor in reducing the complexity of very comprehensive and complex issues is to identify and explain the overall goals, dependencies and synergies, prerequisites and/or requirements, conclusions, and consequences. An additional welcome effect is that it fosters effectiveness and impact in implementation. One must always be aware that improvement is an ongoing process and never completed. In this respect, our conclusions refer to the insights of change and continuous improvement processes in complex environments.

The extent of complexity shows up again in the opportunities for synergies. Since all issues are interrelated in an intricate way, there are many dependencies and interdependencies as described in the following and shown in Figure 13 above.

Figure 13. The six guiding issues of the HCII2023 Design Café, their interactions, dependencies, and synergies.

7.1. Overall goal

The overall goal of conceiving, designing, and implementing sustainable human settlements with human-environment interactions appropriate for fostering human dignity and human rights is to ensure Inclusive and Ethical Cities and Environments. In terms of inclusivity, emphasis was placed by the participants on user communities at risk of exclusion, such as minorities and women. In this regard, the active participation of all target user groups in the design process was strongly advocated.

The challenge with data-based automated technologies and processes is that they are not objective or fair per se, although they are often advertised as being objective and error-free. Humans are increasingly removed from being in charge (thus, losing control), because they are – at an ostensible level of the discussion – considered to be the cause of errors, although the opposite is true in many cases. Results of automated algorithmic decision making are based on the data available and on probabilities. They are very much dependent on the quality of test data sets and accordingly very vulnerable. In any case, there are no ethical considerations.

Furthermore, although there were no explicit discussions on ethics as a discipline, several of the points raised pertain to attributes of an ethical city, such as ensuring the freedom of its citizens, being responsible, and transparent. Naturally, this goal is not independent, but has implications and needs fundamental pre-requisites. It is strongly related with the other issues discussed at the HCII2023 Design Café.

7.2. Fundamental pre-requisites and requirements

Acceptance of overall targets, activities, processes, rules, and regulations is the fundamental pre-requisite for sustainable change towards a declared goal. Acceptance is the foundation of peoples’ engagement and ownership. Acceptance needs Trust between People and Smart Environments, Privacy in Disappearing Computing Environments as well as Explainability and Transparency of Policies and Measures, which are three of the six issues.

Trust between People and Smart Environments is considered as a critical factor in user acceptance of technologies, described in Section 6.2.1. This was also reflected in the discussion results of the Design Café. But trust does not stand alone. It was found to be related to most other issues, e.g., privacy, inclusive and ethical cities, explainability and transparency, incentives, and rewards. Trust was identified as a notable pre-requisite for an inclusive and ethical city. The participants mentioned explicitly the issue of trust, especially in relation to data availability, transparency, and lack of bias, as trust was considered a fundamental component for a fruitful dialogue between citizens and decision-makers in smart cities, leading to a human-centered smart city where citizens enjoy being engaged and providing their contribution.

Overall, based on the discussion, one can posit that when citizens are in control of their data, they are more inclined to develop a relationship of trust with the smart city. Otherwise, concerns about how their data is handled can undermine their engagement with the smart city, especially considering the vast potential for mass data collection, surveillance and monitoring enabled by the technological infrastructure.

Trust includes safety of private information and data in ubiquitous computing environments. Data driven incentives allow to be focused on specific areas that might need improvement and thus provide a rich data set to define which incentives would motivate people (Whittaker et al., 2021).

Privacy in Disappearing Computing Environments was determined by the Design Café participants as an essential prerequisite for acceptance and emerged as a prime issue in relation to ethical rules for the smart city. One reason is that privacy built upon ensuring fundamental human rights allows everyone, regardless of background or situation, to an improved well-being. To that end, if privacy is ensured, it can prevent discrimination, an essential factor for inclusivity, facilitating citizens to access and benefit from smart city services without fear of unnecessary intrusion or bias. The discussion showed that ensuring privacy is essential for citizens’ trust in smart cities, a finding which is aligned with recent studies in the field (Sah & Jun, 2023). Citizens are more likely to engage with their smart city if they feel confident that their data are handled under a responsible and transparent manner in line with the best of their interest. Another approach for assuring privacy was the proposal of providing “free areas for privacy,” i.e., defined “safe” geographic areas or conceptual spaces without data collection. These suggestions correspond with proposals of providing “cold zones” combined with “disposable identities” (van Kranenburg et al., 2020) and the notion of “ambient privacy” (Streitz, 2021b).

Citizens’ privacy determines how measurement of success and impact in sustainability projects can be performed as it requires the collection and analysis of data to evaluate the efficiency of different initiatives. To this extent, it is important to keep the balance between acquiring the necessary data and ensuring individual privacy.

Privacy-preserving methods can uphold the protection of personal data while also enabling meaningful impact assessment.

Explainability and Transparency of decisions, policies, and measures was considered as a foundation for several of the other issues discussed at the HCII2023 Design Café. It plays a role regarding the inclusivity of citizens, the respect of appropriate privacy concerns and creating or diminishing public trust in the data. Explainability and transparency are essential when determining what measurements and associated data collections will be accepted by citizens. They also facilitate the choice of incentives citizens might trigger to provide data and to participate in experiments with new services.

Citizens are more likely to engage with their smart city if they feel confident that their data are handled under a responsible and transparent manner in line with the best of their interest. A trustworthy smart city should also be founded on the principle of explainability and transparency. When people understand how their data is being used and protected, it increases their trust in the smart environment and enables them to make informed decisions. Transparency helps to identify potential privacy risks, while explainability of the decisions taken by the smart city assists individuals in understanding their rationale and deciding whether to accept or reject them.

7.3. Additional dependencies and synergies

There are two directions of the role of Incentives and Rewards for Engagement and Sustainable Behavior. On the one hand, city administration or service providers can provide incentives and rewards to citizens for getting engaged and be willing to provide their data. On the other hand, citizens can also provide awards and ratings about the city’s behavior and policies. A strong dependency and synergy were noted between trust and incentives and rewards, centering around citizens rewarding the city. More specifically, citizens can incentivize and reward their city for its trustworthy behavior, by providing ratings and potentially by renewing their contract with the city, thus being in the loop and in control.

For incentives to work, they must be connected to the people they are aiming to motivate. Accordingly, incentives should be based on a deep understanding of people’s attitudes and needs as well as the fundamental principles and insights of sustainable motivation. Thus, the incentives must be inclusive and ethical, and they must be built on trust between the citizens and the smart city.

Incentives and rewards should be devised to respect privacy relying on the anonymized data of the citizens and supporting their opt-in participation. To this extent, sustainable positive behaviors are encouraged while individuals’ personal privacy is kept protected. Vice versa, incentive or rewarding solutions offered by a smart environment can foster the citizens’ awareness about taking care of safeguarding their personal data and its use by third parties.

Furthermore, data-driven incentives provide a solid foundation to apply data driven metrics to test whether the incentive was motivating the target population. The need for appropriate assessment frameworks was mentioned by the participants of the Design Café, as adequate instruments should be available to evaluate and monitor the level of inclusiveness of the city and the achieved compliance with ethical rules.

Measuring Success and Impact in Sustainability Projects encompasses a holistic consideration of various facets of their inputs, outputs, processes, outcomes, and impacts. As an important component of smart city development, privacy protection, trust building, and incentivizing sustainable behavior were recurrent themes within this discussion at the Design Café. It was considered imperative that these facets be integrated into the evaluation system as key indicators. Explainability and transparency of smart cities also necessitate measurement. On the other hand, during the design process of the assessment system, careful consideration must be given to its own interpretability and transparency.

Successful and effective measuring of success and impact depends heavily on how the institutions address the issue of privacy of the citizens involved. The collection and analysis of data is required for measurements evaluating the efficiency of different initiatives. To this extent, it is important to keep the balance between acquiring the necessary data and ensuring individual privacy. Privacy-preserving methods can uphold the protection of personal data while also enabling meaningful impact assessment.

Finally, there is a relevant synergy between the issues of measurement of impact and the provision of incentives and rewards. Since creating and offering incentives and rewards without planning on how to measure their impact does not make much sense, there is a need for a coordinated approach.

8. Evaluation and conclusions

The HCII 2023 conference responded to an environment which is changing dramatically by introducing a new format: the HCII Design Café. Based on the insights about the key factors of change processes, the Design Café confirmed its value as a proven participatory method for engaging relevant stakeholders in the creation and exchange of new ideas and approaches. The HCII conference opened its doors for the first time for selected “outside” peer groups independent of conference participation, e.g., from the metropolitan area where the HCII takes place, this year in Copenhagen. Thus, topics of HCII and appropriate approaches for transfer into practice became part of the local economic and academic context.

After the discussion at the tables and the presentation and discussion of the results in a plenary session, a follow-up activity was introduced. Based on the approach, to give power to the participants, they could actively select their most relevant issue (4 out of 6) on which they want to work according to their personal interest. This process resulted in valuable ideas and findings for the further development of our cities and environments. Moreover, each participant was asked to develop and write down his/her most important idea on an especially prepared “my take-home-idea” card to be taken to wherever he or she lives and works. Thus, there is a great potential impact beyond the HCII2023 Design Café event into daily life and work. Finally, the feedback by the participants showed that they were very pleased and satisfied with the format, very motivated by the issues and questions and are looking forward to a continuation. Thus, we can consider the Design Café as an engagement with a sustainable future.

The development of ethical and inclusive sustainable cities and settlements combined with human-centered, trustworthy, and engaging human-environment interactions is one of the most important endeavors in our times. This process must not be subordinated to individual political and economic interests or to what is technologically feasible. Furthermore, the process must not be organized in a top-down fashion but facilitated by instruments of participatory design in a balanced and iterative combination of bottom-up and top-down at the respective participatory levels. Ethical values, participation, understanding, and identification are essential for sustainable engagement and commitment and to make human settlements livable and prosperous in the long term. Acceptance and success are in strong synergy with each other. Change without success loses acceptance. Acceptance without success leads to doubt and uncertainty.

There is no success without measurement. However, the measurement and success criteria must also be well thought out. This implies, for example, to overcome the underrepresentation of evaluating inclusivity and ethical considerations within city projects in contemporary assessments. Cities’ performance in inclusivity and ethical aspects is hard to quantify using simplistic quantitative indicators. Factors such as morality, ethics, feasibility, social impact, environmental sustainability, and data security need to be taken into further account in future standard assessment systems. Engaging directly with city residents, monitoring sustainability behavior, ensuring ease of use, and providing real-time feedback are crucial. A key aspect of measuring success is that those who contribute to the development of a smart city through their activities and behavior can also perceive and recognize the kind of progress made. This is crucial because many developments imply a long-term process during which commitment can easily wane if success is not noticeable. The principle of ambitious small steps, which should only be as big as success can be achieved and recognized serves as one guideline. If this principle is disregarded, frustration will soon arise, and commitment will wane significantly. It is equally important to communicate the purpose of the joint efforts again and again when measuring and presenting the results. This provides an intrinsic incentive and motivation not to give up “on the last few meters.”

One of the key outcomes of the discussions on explainability and transparency was that the scale of these challenges still needs to be recognized by those responsible. We observe this also in the current debate on the proposed limits, guidelines, and regulations of artificial intelligence, where Explainable AI is an important ingredient of the Human-Centered AI approach. It is essential to educate decision makers in the psychological needs and effects of sense making, comprehensibility, and finally acceptance and well-being. Explainability and transparency should become one of the ethical guidelines that steer and control concepts, decision making, and implementations of all activities, in the broadest sense of the word, in the context of human settlements and the corresponding human-environment interactions, bearing the “invisible technology” always in mind.

The results of our HCII2023 Design Café are in line with results from related participatory foresight research. For example, Ystgaard and de Moor (2023) outline how the shift towards future human-centered intelligent environments and current and future technical visions can contribute to genuinely humane, fair, and equal real-world outcomes. The specific insights from our Design Café on ethics, trust, privacy, transparency, incentives, and success contribute to a more comprehensive perspective towards societal challenges and the need for a humanity-centered design approach.

According to Norman (2023), it represents the ultimate challenge for designers to help people improve their lives, because “humanity-centered” expands the view to the societal level of world populations. This corresponds highly to the call for action expressed by the United Nations when declaring the 17 Sustainability Development Goals (SDGs, 2015) as mandatory for all countries to be reached by 2030, which was the starting point of our motivation (described in the introduction) to organize and implement the HCII2023 Design Café in Copenhagen.

Acknowledgements

We would like to thank all participants of the HCII2023 Design Café in Copenhagen for being highly motivated and engaged in their collaborative efforts of discussing the six issues and creating ideas and solutions on how to address the pressing problems. A special thank you goes to Constantine Stephanidis, General Chair of the HCI International 2023 conference for his strong support since the early phase, when Christine Riedmann-Streitz proposed the concept of this new event in a Program Board meeting of the DUXU conference, all the way to implementing it in a joint effort at the conference premises in Copenhagen, Denmark, in July 2023.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Christine Riedmann-Streitz

Christine Riedmann-Streitz, Managing Director of MarkenFactory and Lecturer at Goethe-University, Frankfurt, Germany. Expert in brand and identity, innovation and change, leadership and culture. Facilitator, consultant, trainer, empowering people in different organizations. Focus on digital and cultural transformation, new work, stakeholder participation, and resilience. Program board member of international conferences.

Norbert Streitz

Norbert Streitz, Scientific Director of Smart Future Initiative, Germany. Prior positions: Deputy Director of Fraunhofer Institute IPSI; Assistant Professor at Technical University Aachen (RWTH); Postdoc University of California, Berkeley, visiting scholar: Xerox PARC and Intelligent Systems Lab, Tsukuba Science City, Japan. Co-Chair DAPI conference. Elected Member ACM CHI Academy.

Margherita Antona

Margherita Antona is collaborating researcher at the HCI Laboratory of ICS-FORTH. Her research interests include design for all, adaptive and intelligent interfaces, Ambient Intelligence, HRI and Human-Centred AI. She is Co-Chair of the UAHCI Conference, coeditor the UAIS Journal, and member of the Editorial Board of the IJHCI Journal.

Aaron Marcus

Aaron Marcus, Principal, AM+A, Berkeley; Editor-in-Chief Emeritus, User Experience; Editor, Information Design; member, ICOGRADA Hall of Fame (2000); Fellow, AIGA (2007); member, CHI Academy (2009); published 50 books, 300+ articles; designs mobile HCIs, information visualization, persuasion apps, cross-cultural communication. Past Faculty, Design Institute, IIT/Chicago and Design/Innovation College, Tongji University/Shanghai.

George Margetis

George Margetis is a Postdoctoral Researcher with the HCI Laboratory of ICS-FORTH. His research interests include intelligent environments, Human-Centered AI, tangible and embodied interaction, XR. He has been a scientific and technical responsible for numerous European R&D projects in the above areas. He is Co-Chair of the MOBILE Conference.

Stavroula Ntoa

Stavroula Ntoa is a Postdoctoral Researcher with the HCI Laboratory of ICS-FORTH. Her research interests include UX research in intelligent environments, Human-Centered AI, digital accessibility and universal access. She is Co-Chair of the AI-HCI Conference and member of Editorial Board of the UAIS, IJHCI, and Frontiers in Virtual Reality journals.

Pei-Luen Patrick Rau

Pei-Luen Patrick Rau, Professor at Department of Industrial Engineering at Tsinghua University, Beijing, China. Vice Dean of Tsinghua Global Innovation Exchange (GIX) Institute. Research areas: Human Factors Engineering, Human Computer Interaction, User Insights and Experience Innovation, Cross-Cultural Research, Design for Well-being, Service Design and Evaluation.

Elizabeth Rosenzweig

Elizabeth Rosenzweig is an author, HCI researcher, product designer, Adjunct Faculty member at Brandeis University and Principal at Bubble Mountain Consulting. Elizabeth holds 4 patents in intelligent user interface design and is Founder and Director of World Usability Day, organizing global conferences in over 80 countries since 2005.

Notes

1 https://2023.hci.international/

2 https://2023.hci.international/DesignCafe.html

3 https://www.hindustantimes.com/india-news/woman-battlesopen-defecation-in-hindi-movie-but-reality-in-india-is-different/story-is1RK2yTt6HtvByr2xm0tN.html

4 https://timesofindia.indiatimes.com/entertainment/bengali/movies/news/parno-mittra-women-need-clean-public-toiletssusu-is-the-first-indian-film-to-address-the-issue/articleshow/90393851.cms

5 https://gdpr-info.eu/

6 https://www.iso.org/standard/85938.html

7 See for example the book by George Orwell (1949). Nineteen Eighty-four. Secker & Warburg. (https://en.wikipedia.org/wiki/Nineteen_Eighty-Four).