Designing and implementing a global multilingual real-time location-based game as a novel highly gamified spatial crowdsourcing platform Manuel F. Baera a Centre for eResearch, University of Auckland, New Zealand Abstract Spatial crowdsourcing and other participatory methods of high-quality spatial data generation are increasingly popular in academic research. However, participatory approaches face issues of user motivation and retention. Gamification has been identified as a valuable approach towards increasing user motivation, engagement and retention by offering an entertaining experience. This paper presents a case study of the literature-based development and implementation of Arcane Shift, a novel globally available location- based game as a highly gamified spatial crowdsourcing platform. In particular, this paper focuses on the first three activities (problem identification, problem definition and application development) of the Design Science Research Methodology and how the implementation was guided by the established Gamification Framework for Volunteered Geographic Information and the Mechanics, Dynamics and Aesthetics Framework. Internal testing revealed the game to be enjoyable, albeit in need of further content. Four concrete implementation recommendations and considerations (develop for testing, incorporate caching, build for longevity and spatial is special) were gleaned from the development process and are presented as a methodological contribution to the design and development of location-based games and spatial crowdsourcing platforms. Finally, a list of future research agendas is provided including urban data generation, educational content delivery and motivating physical activity. Keywords Design science research; mechanics, dynamics, aesthetics framework, gamification framework for volun- teered geographic information; spatial crowdsourcing; gamification; location-based game 1. Introduction initiatives (e.g. [13, 14, 15, 16]) and has shown to successfully motivate users to participate Participatory approaches to generating spa- [9, 17]. However, many gamification efforts in tial data for academic research have enjoyed a academic crowdsourcing have implemented a surge in popularity over the past two decades limited number of gamified elements such as [1, 2, 3, 4]. In particular, spatial crowdsourcing points, badges and leaderboards [18, 17] whilst approaches have shown immense potential in highly gamified applications such as location- generating rich high-quality spatial information based games remain rare. about our environments for comparatively low Meanwhile, commercial location-based games financial investments [5, 6, 7, 2]. However, a such as Ingress1 and Pokémon GO2 have moti- limiting factor of crowdsourcing approaches, vated millions of users [19, 20, 21] to contribute and participatory approaches in general, is data on unprecedented scales [22, 23]. Location- user motivation and retention [8, 9, 10, 11]. based games offer playful virtual interactions A maturing approach to increasing user en- according to the real-world coordinates of a gagement, retention and satisfaction in crowd- respective user [24, 21] and are predominantly sourcing tasks is gamifying the participatory played in urban areas [25, 24] (showing spa- process. Gamification, summarised as “hedonic tial biases towards wealthier neighbourhoods or entertainment-oriented technologies being re- [26, 19]). appropriated for productive use” [12, p. 191], Skillfully balancing entertainment and spa- has been applied to a number of crowdsourcing tial crowdsourcing tasks has seen success in a handful of academic studies [cf. 27, 13, 16], and 8th International GamiFIN Conference 2024 (GamiFIN seeing the great potential of using more ludic 2024), April 2-5, 2024, Ruka, Finland. applications such as location-based games to $ manuel.baer@auckland.ac.nz (M. F. Baer) motivate users to contribute data towards scien-  0000-0002-9474-3299 (M. F. Baer) © 2024 Copyright for this paper by its authors. Use permitted under Creative Commons Li- 1 cense Attribution 4.0 International (CC BY https://www.ingress.com/ (accessed: 19.11.2023) 4.0). 2 CEUR Workshop Proceedings https://pokemongolive.com/ (accessed: CEUR 19.11.2023) http://ceur-ws.org Workshop ISSN 1613-0073 Proceedings (CEUR-WS.org) CEUR ceur-ws.org Workshop ISSN 1613-0073 Proceedings 88 tific crowdsourcing efforts, this paper presents of the data used to understand the status quo the design, development and implementation of our environment are produced by sensors, of a globally available, modular, multilingual through computational algorithms or reported real-time location-based game as an underlying by specific demographic groups. Temperature crowdsourcing platform. Specifically, this pa- and pollution readings are captured by fixed per aims to make a theoretical contribution to measuring stations [38] or through remote sens- the design and implementation of entertaining ing [39], walkability analyses of urban and resi- spatial crowdsourcing platforms by presenting dential areas focus on tangible variables of the the case study of Arcane Shift, a location-based built environment disregarding individual per- game as a spatial crowdsourcing platform. This ceptions [40], and city planning is only now paper is particularly interested in how we can gravitating from top-down approaches towards design and implement a highly gamified spatial the inclusion of local citizens [41, 42]. crowdsourcing platform in the form of a glob- ally accessible location-based game by focusing 2.1. Spatial crowdsourcing and on the following research questions: participatory data collection • What are the key underlying game fea- Contrasting more automated spatial data col- ture requirements to develop a highly lection approaches are initiatives that include gamified spatial crowdsourcing platform non-expert users in spatial crowdsourcing tasks. and how do these map onto established Many approaches exist concerning the collec- implementation frameworks? tion of spatial information contributed by non- • How can underlying crowdsourcing func- experts including volunteered geographic infor- tionalities be complemented with enter- mation [5] where individuals voluntarily make taining game features offering global real- some form of spatial data available [1] and pub- time gameplay? lic participation geographic information systems • What are key considerations and recom- (PPGIS) [43] where individuals contribute data mendations for developing spatial crowd- through a participatory mapping effort. An sourcing platforms and location-based umbrella term for many participatory (spatial) games? data collection approaches is crowdsourcing In the following, I introduce relevant previous [1, 3, 4], “a type of participative online activ- work as well as the topical background before ity in which an individual, an institution, a diving into the specifics of the underlying sys- non-profit organization, or company proposes tems and how individual features were designed to a group of individuals of varying knowledge, and implemented. The final sections of the pa- heterogeneity, and number, via a flexible open per culminate in discussing the implemented call, the voluntary undertaking of a task“ [44, p. location-based game in light of the literature, 197]. This participatory approach can broadly presenting key lessons learnt during the design, be split into two categories: (1) active crowd- development and implementation processes as sourcing where data is actively contributed for a theoretical contribution to the design of gam- a specific task or question as part of a data ified spatial crowdsourcing platforms. Finally, collection effort (e.g. open street map (OSM)3 ), this paper presents further recommendations and (2) passive crowdsourcing where data is and propositions as well as future avenues of contributed for another intention than what it research. is ultimately used for (e.g. social media data) [1]. Spatial crowdsourcing initiatives have been 2. Previous work found to produce high-quality data at a frac- tion of the cost of more traditional expert- Multiple studies showcase the direct influence based approaches [1] and have been found to our surroundings have on our social, mental complement more technocratic and top-down and physical well-being [28, 29, 30]. Examples approaches to data collection (e.g. through range from street design increasing perceived sensory networks and remote sensing) by pro- safety [31, 32], over particular areas offering viding novel insights into more subtle dimen- tranquillity [33, 34] boosting mental well-being sions of human perception [cf. 45, 46, 33, 34]. [35, 36], to public transportation networks in- creasing physical activity [37] whilst reducing 3 https://www.openstreetmap.org/ (accessed: personal motorised vehicles. However, much 19.11.2023) 89 This can lead to a more holistic understand- ified urban data crowdsourcing effort and Fo- ing of emergent spatial issues such as urbani- toQuest Austria [13] as well as StarBorn [16], sation, sustainability and well-being, making entertaining approaches to collecting landcover spatial crowdsourcing an invaluable data gen- information. eration approach. Whilst successful large-scale Location-based games are entertaining ap- spatial crowdsourcing efforts in academia re- plications, commonly played on devices with main scarce, non-academic entities have identi- location tracking features such as a global po- fied the value of location-based user-generated sitioning system, where a player’s in-game po- content and boast several successful applica- sition reflects their real-world coordinates [24]. tions. Examples include the location-based Players navigate the virtual game world by games Ingress and Pokémon GO crowdsourcing changing their real-world position and in-game the collection of culturally significant objects interactions are dependent on the real-world through their Wayfarer system [23], Google position of a user and their proximity to virtual Maps collecting vast amounts of user-generated objects. As such, location-based games offer spatial information such as points of interest a particularly interesting platform to collect [47] and Waze focussing on mobility-related geospatial data from a heterogeneous group of data such as real-time traffic updates [17]. De- users. Pokémon GO and Ingress are highly spite having large repositories of spatial data, popular commercial location-based games [21] these are not easily accessible, nor are these shadowing other approaches in terms of user systems customisable to specific scientific data engagement, successfully motivating hundreds needs, and thus new academic initiatives are of millions of users to engage with the appli- called for. cation and contribute rich spatial information A shift towards more bottom-up approaches [22, 23]. This begs the question: can a location- to data collection in academic spatial research based game be implemented to crowdsource encourages a reformulation of traditional re- heterogeneous spatial information for research search agendas and has raised some underlying purposes? methodological questions about how we can ef- fectively include citizens in science and how we 2.3. Design science research and can motivate long-term engagement. However, questions of user motivation and user retention relevant frameworks arise [10, 11], especially when competing with Application development and implementation an onslaught of entertainment and otherwise processes concerning the construction of infor- available playful activities. mation systems as problem solutions to con- crete issues can be grounded in design science 2.2. Gamified crowdsourcing and research [48, 49, 50, 51] which "seeks to enhance technology and science knowledge bases via the location-based games creation of innovative artifacts that solve prob- A recent development sees the emergence of lems and improve the environment in which crowdsourcing efforts capitalising on entertain- they are instantiated" [51, p. 1], or more gener- ing features through gamification [9]. Gami- ally "the construction and evaluation of generic fication can be summarised as “a process of means–ends relations" [50, p. 470]. One widely enhancing services with (motivational) affor- used and well-established framework is the De- dances in order to invoke gameful experiences sign Science Research Methodology presented and further behavioral outcomes” [18, p. 3026]. by [49], dividing the overall goal of develop- Gamification shows great potential in generat- ing and evaluating an application aiming to ing crowdsourced urban data and complement- address a specific problem statement into six ing geospatial crowdsourcing more generally distinct activities (for a detailed discussion, see [9, 2]. In the context of crowdsourcing, gami- [49]): fication has successfully been incorporated in 1. Identifying concrete research problems a number of academic studies as a means of as well as justifying the importance of increasing user motivation (for an overview, the proposed solutions see [9]), albeit commonly implementing sim- 2. Defining specific objectives and require- ple competitive features such as points and ments of the envisioned solutions leaderboards. More gamified geospatial crowd- sourcing efforts include Urbanopoly [27], a gam- 3. Designing and implementing the applica- tion or artefact 90 4. Demonstrating the implemented applica- makes the Mechanics, Dynamics and Aesthet- tion’s or artefact’s ability to address the ics framework particularly interesting in terms identified problems of gamified application design is the emphasis 5. Evaluating the application’s or artefact’s on taking both the developers as well as the ability to address the identified problems user’s perspectives into consideration, allow- 6. Communicating the initial problem state- ing for a more holistic approach to application ments, the implemented application or design. artefact and the design process as well Drilling down to specific frameworks regard- as the evaluation to a heterogeneous au- ing the gamification of geospatial information dience collection, the Gamification Framework for Vol- unteered Geographic Information aims at bridg- Whilst offering a stable foundation to con- ing the divide between geospatial crowdsourc- duct research relating to application design ing applications and their gamified equivalent and offering clearly defined overarching activi- by mapping the terminology between the two ties, it makes sense to complement the Design domains [53]. When translating from a na- Science Research Methodology with additional tive volunteered geographic information ap- more specific frameworks when developing a plication to a gamified implementation, users specific application. To build a globally avail- become players, the map transforms into the able, modular, multilingual crowdsourcing plat- game board, geo-data is used to generate virtual form as a location-based game, further estab- goods and tasks translate to challenges. Addi- lished frameworks exist to ground specific de- tional gamification elements aim at increasing velopment decisions in theory and guide the collaboration and/or competition and include implementation. Two established frameworks leaderboards, social elements such as friends with increasing topic specificity include the Me- lists, votes as well as levels, points, badges and chanics, Dynamics, Aesthetics Framework [52] bonuses as typical rewards. concerning general gamified applications and This paper delves particularly into the first the Gamification Framework for Volunteered three activities of the design science research Geographic Information [53] which focuses on approach. After having identified the concrete volunteered geographic information, albeit ap- research problem of increasing user motivation plicable to geospatial crowdsourcing efforts in and retention in spatial crowdsourcing in the general. introduction and background sections (activity A framework seeing widespread adoption in 1), in the following I focus on defining specific both designing [15] as well as analysing [54] objectives and requirements of a solution (ac- games and gamified applications is the Me- tivity 2) and shed particular light on how the chanics, Dynamics and Aesthetics framework Mechanics, Dynamics and Aesthetics frame- [52] which segregates an application into three work as well as the Gamification Framework underlying categories. Firstly, the aesthetics for Volunteered Geographic Information were of an application is what the user directly ex- used to develop and implement the application periences and what invokes a user’s emotional Arcane Shift (activity 3). response towards specific parts of the appli- cation (e.g. the design of the user interface). An application’s aesthetics are of particular 3. Design and implementation importance and can further be subdivided into eight categories: sensation, fantasy, narrative, The location-based game named Arcane Shift challenge, fellowship, discovery, expression, and aims to combine various game elements into submission. Secondly, the dynamics represent an entertaining crowdsourcing experience. In the systems that influence or are influenced a first step, and using the two aforementioned by a player’s interaction with the application frameworks, a list of envisioned features was and which result in specific behavioural out- created to guide the development. Firstly, the comes (e.g. a button that opens a specific overarching theme of the application was set screen affording new interactions within the as a synthwave-inspired sci-fi experience where application). Lastly, the underlying algorithms users would captain a futuristic spacecraft. Set- and data structures are summarised within the ting the theme allowed for individual game mechanics of an application and build the back- features to be designed around said theme. bone of the gamified system (e.g. the database To allow for persistent player data as well as in which application data is stored). What to collect basic demographic information from 91 the participants, crucial data in active crowd- collected basic ores that were procedurally gen- sourcing platforms, a registration system and erated depending on the underlying real-world onboarding system were needed. Seeing the land cover of an area (Figure 1 G). Users could location-based nature of the application, an un- either combine specific combinations of ores derlying spatial system was needed to allow the into new materials increasing their value or sell generation and visualisation of location-based ores for gold rewards. Gold could then be used content, a common feature of location-based to construct buildings for their team (Figure games [24, 21, 23]. Tying into the spatial sys- 1 D) or to buy other ores. If a user’s ship had tem of the location-based game, the application enough energy (Figure 1 I ), enemy buildings required a number of spatial content systems (Figure 1 E) could be attacked (Figure 1 F ) (e.g. resource system, building system) to al- and destroyed freeing up an area for friendly low for the persistence and correct placement buildings. Further, users could accept and com- of in-game content such as items and struc- plete a variety of quests (individual crowdsourc- tures. To introduce a form of progression and ing tasks) from survey drones (Figure 1 B) or to reward the user for in-game activities, a their quest log (Figure 1 J ). Many in-game level system was required which is commonly activities such as completing quests, collecting found in gamified applications [53, 12]. Seeing resources, forging items and building friendly the application focuses on spatial crowdsourc- or destroying enemy buildings rewarded experi- ing, a questing system was needed allowing ence points (Figure 1 I ). Leaderboards showed the integration of a variety of crowdsourcing the top-performing users and an activity log tasks. Primarily, the application should allow (Figure 1 L) displayed a user’s past actions users to upload location-specific content such as well as the enemy users who attacked their as location-based natural language reports and buildings. By limiting a player’s view to a fixed ratings, commonly collected variables in spatial distance (approximately a radius of 1km) the crowdsourcing. To foster motivation and incen- game aimed to engage the users’ inquisitive tivise continued engagement with the platform, nature and hoped to motivate users to explore the application was in need of entertaining sys- their surroundings. tems offering a variety of in-game activities. In First user feedback revealed the core game line with the overarching theme of the applica- loop to be interesting, albeit lacking content to tion, a ship upgrade system, an attack system motivate long-term engagement. As such level- and a forge system were envisioned. Having dependent unlockable content was introduced. unlockable and upgradable in-game items offers Thereafter, users could additionally spend gold a form of progression and positively affects a on unlocking various spacecraft component up- user’s motivation [53, 12]. Attackable in-game grades such as energy generators (e.g. increas- structures increase an application’s competitive ing energy regeneration rate), tractor beams elements, which have been found to positively (e.g. increasing number of simultaneous ore affect user engagement [17] and allowing users collections) and weapons (e.g. increasing dam- to forge new items offers a form of discovery age). Gold could also be spent on unlocking and a collaborative element, increasing the like- and constructing improved buildings for the lihood that the application is recommended user’s team such as forges with reduced forge [17]. Lastly, an alert system was deemed nec- times or markets with increasingly discounted essary to inform users of important in-game prices. An overview of the current core game activities. loop is presented in Figure 2. These envisioned features of the application In the following, I first present the infrastruc- were mapped to the broader categories of the ture with which the application was built before two aforementioned frameworks helping to struc- introducing core features as well as discussing ture the development and guide the implemen- the implemented systems. tation. Table 1 shows the implemented features of the location-based game and the correspond- 3.2. Infrastructure and core systems ing categories of the frameworks. The location-based game was built from scratch in the Unity4 game engine (version: 2022.2). 3.1. The core game loop Within the location-based game, users captain a futuristic spacecraft (Figure 1 H ) and initially 4 https://unity.com/ (accessed: 10.01.2024) 92 Table 1 Features of the implemented location-based game and how they map to the Gamification Framework for Volunteered Geographic Information and the Mechanics, Dynamics, Aesthetics Framework Game Feature Gamification Frame- Mechanics, Dynamics, Aesthetics Framework work for Volunteered Geographic Informa- tion registration user / player aesthetics: narrative, fantasy; dynamics: registra- and onboard- tion form, email confirmation, demographic infor- ing mation report, character selection, team selection; mechanics unity game code, firebase authentica- tion, firestore database spatial system map / gameboard aesthetics: discovery; dynamics: providing and translating coordinates, downloading, caching and displaying map tiles; mechanics: unity game code, MapBox tile server level system challenges, levels aesthetics: challenge dynamics: increasing levels unlocks various in-game items mechanics: unity game code, firestore database ship system virtual goods, chal- aesthetics: fantasy, narrative, challenge, discovery; lenges, levels dynamics: ship components can be unlocked and equipped changing the behaviour of a user’s ship; mechanics: unity game code, firestore database quest system tasks / challenges, geo- aesthetics: challenge, discovery, expression; dy- data namics: quests can be accepted and submitted, submitting quests rewards the user; mechanics: unity game code, firestore database resource sys- virtual goods, chal- aesthetics: fantasy, narrative, challenge, discovery; tem lenges dynamics: spawning resources depend on the real- world landcover in the spawn location, resources can be collected and forged into new resources, collecting resources rewards experience points, re- sources can be bought and sold; mechanics: unity game code, firestore database forge system virtual goods, chal- aesthetics: sensation, challenge, discovery; dy- lenges namics: forging allows new items to be discov- ered increasing their value, forging rewards experi- ence points; mechanics: unity game code, firestore database building system gameboard, social ele- aesthetics: fantasy, challenge, fellowship, expres- ment sion; dynamics: buildings can be built and inter- acted with by users on the same team. Enemy buildings can be attacked and destroyed; mechan- ics: unity game code, firestore database attack system gameboard, points, lev- aesthetics: sensation, fantasy, challenge, fellow- els, challenges ship, discovery; dynamics: in-game buildings of the enemy team can be attacked and the attack dam- age depends on a user’s ship’s weapon, destroying a building rewards experience points; mechanics: unity game code, firestore database activity alerts social element aesthetics: discovery, fellowship, challenge; dynam- ics: alerts of important in-game activities; mechan- ics: unity game code, firestore database Google Firebase5 and Firestore6 , massively scal- able NoSQL database solutions, were intro- 5 https://firebase.google.com/ (accessed: duced to provide authentication and storage for 10.01.2024) the application. Leveraging Firestore’s event 6 https://firebase.google.com/docs/firestore (ac- functionality I introduced real-time gameplay cessed: 10.01.2024) 93 Figure 2: Schematic representation of the current core game loop of the location-based game 3.2.1. Data downloading and caching In a first iteration, all game data was stored on the user’s device, however, to allow for rapid de- velopment and balancing during testing, game content must remain updatable allowing for small tweaks according to user feedback (e.g. weapon damage). Seeing the uneven cover- age of mobile broadband internet, especially in rural areas [55], and to limit needed mo- Figure 1: The main game screen. (A) settings; (B) bile data, specific game content is cached on location-based quest / spatial crowdsourcing task; (C) a user’s device. A caching pipeline was im- stylised base map; (D) friendly building; (E) enemy plemented where content is initially stored on building; (F) weapon charging and weapon projectile; a server and is only (re)downloaded from the (G) procedurally generated ores; (H) user’s spacecraft; (I) ship energy generator and user experience point server if the content is not found in the user’s indicator bars; (J) quest log; (K) ship menu; (L) in- persistent data cache or if the content has been game alerts updated on the server. The downloaded con- tent is then stored in a persistent cache on the user’s device as well as in memory, allowing fu- through subscribing to database document events ture access to up-to-date data without needing and acting accordingly. In other words, updat- to (re)download the content (cf. Figure 3). ing a document in the database triggers an update event, which in turn triggers all listen- 3.2.2. Application environments ing devices to call a respective method. This At the start of the development process, a sin- event-based approach allows for chosen in-game gle environment was used for development and actions to be propagated to all connected user testing. However, with growing complexity and devices, allowing for real-time gameplay. to allow for easy updates and continuous inte- gration [56, 57], the application was segregated For example: When a user attacks an enemy into three separate environments: a develop- building the persisting building data in Firestore ment environment for rapid prototyping, a stag- is updated, triggering all listening devices to call ing environment for beta-testing with test users the method to update the respective building’s and a production environment used for the live health. application. Each environment is accompanied by a decoupled database instance, allowing for 94 if the database document has been updated. This allows the textual content to be seamlessly translated and updated in the application, a key ingredient in contemporary application de- sign [60]. 3.3. Registration and onboarding To ensure an individualised experience, users must register and go through an onboarding process (Figure 4). After registering with email and password, users are required to confirm their email address to minimise misuse. When logging in for the first time, an onboarding pro- cess is started, where users are first presented with the free and prior informed consent of Figure 3: Schematic representation of the data the application’s intended use and how a user’s caching pipeline data will be used and processed. This allows users to make an informed decision on whether they would like to proceed, in line with con- large changes to be implemented and tested temporary data protection regulations when without affecting other environments. This ap- collecting user-generated data [61]. Users are proach follows general software development then introduced to the underlying narrative guidelines and allows for rapid prototyping, of the location-based game and are asked to continuous integration and stable production make a character, provide basic demographic environments. information and choose their preferred user- name and team. The onboarding process ties into the game narrative, which has been found 3.2.3. Internationalisation to increase user engagement and motivation Having multilingual content allows for a more [62]. Once a user has provided the required in- diverse audience and has been found to be an formation, they enter the location-based game important consideration in modern applications world. [58, 59]. Internationalisation of game content was thus implemented as an integral part of the 3.4. Spatial features application and allows the game to be played in multiple languages (at the time of writing, A defining feature of location-based games is this includes te reo Māori, English, French and the translation of a user’s real-world coordi- German). Chosen game content (e.g. textual nates into a (virtual) game world [24]. This content) was translated into the mentioned generally requires two broader systems: (1) languages and respective translations were ini- querying a user’s real-world coordinates as well tially stored using Unity’s Localization package. as (2) translating these to respective game- However, similar to the reasoning of moving world coordinates and constructing the game game content data to an external database to world accordingly. At first, the location-based allow for rapid prototyping and quick adjust- game was developed implementing the MapBox ments (Section 3.2), translation tables were Software Development Kit (SDK)7 for Unity. moved to the Firestore database. In the cur- However, the SDK was accompanied by three rent iteration, translated texts are referenced main limitations: (1) computational overhead as translation keys and translation data is han- due to the number of features, (2) complexity dled similarly to general game data in terms of extending the SDK and (3) discontinued de- of caching where data is only downloaded if velopment of the SDK leading to deprecated it is not found in the translation table cache third-party code. As such, two custom spatial on a user’s device, or if any updates to the systems were built to allow customisation and translation table are detected. In other words, integration in the game. translation tables for textual content are stored online and downloaded on demand and only 7 https://www.mapbox.com/unity (accessed: 05.09.2023) 95 Figure 4: Screenshots showing the onboarding in the application. (A) free and prior informed consent; (B) spacecraft selection and customisation; (C) personal information; (D) display name and referral code; (E) team selection In the implemented application, a location long-term in-game goals [53, 12]. The imple- provider was implemented which uses a respec- mented platform features experience points and tive device’s location features to periodically a levelling system, belonging to the most com- query the user’s current coordinates. A map- mon gamification elements [18]. Users initially ping engine was built to leverage the location start the game at level 1 and are rewarded with provider and convert between the retrieved real- experience points for several in-game activities world latitude and longitude coordinates and such as the completion of quests, destroying the game world coordinates. The mapping enemy buildings, building friendly buildings engine calculates the current map extent of and collecting or forging ores (cf. Figure 2). the player and identifies the needed map tiles. Once a level-specific threshold of experience The device’s cache is queried and existing map points is reached, the player’s level is incre- tiles within the user’s vicinity are retrieved mented. Initially, the application featured a and drawn to the screen as an underlying base linear progression system, where every level map in the game world. Missing map tiles are required the same amount of accumulated ex- downloaded from MapBox8 and cached on the perience points to advance. However, in a user’s device to reduce the bandwidth needed later iteration, this was changed to exponen- when a user revisits locations. When the user tial experience point thresholds to ensure long- changes their location in the real world their term challenges. With increasing player levels position within the game is updated to reflect new game content is unlocked, providing an the player’s new real-world position. This in additional incentive to accumulate experience return triggers the querying, downloading and points. displaying of new relevant tiles as well as dis- abling no longer needed tiles. Another key 3.6. Ship system implemented spatial feature of the mapping engine is the support for the global hexagonal Gaining increasing levels within the applica- indexing system H39 . This allows chosen con- tion unlocks further in-game content. In an ini- tent (e.g. quests, buildings, resources) to be tial version of the application, additional game placed in hexagonal tiles on a global grid. content was restricted to additional buildable structures. However, user feedback highlighted the lack of incentives and thus a ship com- 3.5. Level system ponent system was introduced, where users An important dimension in gamified applica- could upgrade individual components of their tions is progression and inducing a sense of spacecraft. Implemented ship components con- sisted of energy generators, tractor beams and 8 https://www.mapbox.com/ (accessed: weapon systems. The energy generator com- 04.01.2024) ponent provides a limited amount of energy 9 https://h3geo.org/ (accessed: 03.01.2024) that recharges over time and which limits the 96 allowed use of other components. Upgrades to and placed at specific locations to collect spe- the energy generator include a higher energy cific location-dependent information. In con- capacity and a faster recharge over time. The trast, the latter refers to crowdsourcing tasks tractor beam component uses energy to collect that are randomly generated throughout the resources. Upgrades include faster resource game world, independent of location. In a collection times and an increasing number of preliminary version of the application, each simultaneous collections. Finally, the weapon user could complete each quest once. However, system uses energy to fire projectiles at enemy many spatial crowdsourcing tasks are suited structures. Key upgrades include increasing to be repeated in different locations (e.g. per- the number of simultaneous projectiles and in- ceived scenicness of an area) or at varying times creasing each projectile’s damage. Users can (e.g. perceived safety during the day vs. dur- view key statistics as well as access their ship’s ing the night). Seeing the global nature of, cargo bay, ship components, a general market and the focus on, geospatial crowdsourcing, and the building tool through the ship menu quests were thus extended to be either local or (Figure 5). global and repeatable or non-repeatable. Table 2 shows an overview of the quest categories and examples of questions or tasks that could be incorporated. Quests are comprised of one or multiple quest steps, which in return are made up of one or multiple quest tasks (Figure 6). In addition, quest steps can be locked and only made avail- able to the user if other specific quest steps are completed first. This allows a sequence of quest steps to be combined into a journey, guiding the user through multiple locations in the real world. This in return opens the door to col- lecting specific spatially nuanced information or guiding users through educational journeys in a given landscape. Example: A quest could ask a user to go to the main train station and report on their ex- perience. This quest would be divided into two quest steps (1) navigate to the specific area of the train station and (2) report on their expe- rience, where (2) only becomes available after completing (1). Figure 5: A screenshot of the ship menu and the user’s key statistics as well as the available options 3.8. Resource system Collectable entities are common in location- based games and provide an incentive to nav- 3.7. Quest system igate the game world as well as a means of generating in-game value. Within Arcane Shift According to the Gamification Framework for users can collect a variety of basic ore crystals Volunteered Geographic Information, spatial which are procedurally generated depending data contribution tasks in crowdsourcing are on the respective real-world land cover of a mapped to challenges in gamified systems [53]. given area (Figure 1 G). Collected basic ore These are commonly implemented as quests, crystals are stored in the users inventory and short in-game tasks that afford progression in can be sold for in-game gold, a basic in-game the game, and commonly reward a player with currency. This currency can be used to buy virtual in-game items or points. The quests specific in-game ores as well as to unlock a va- implemented in Arcane Shift fall into the broad riety of in-game content (e.g. ship component categories of designed or generated. The former upgrades), supposing all level requirements are refers to quests that are purposefully designed 97 Table 2 Quest Categories and Types Category Type Description Examples of Underlying Tasks Designed Local non- Quests that are purposefully placed and that Complete the educational repeatable can be completed exactly once in each avail- coast walk and do the able location. quiz. Designed Local repeatable Quests that are purposefully placed and that Is any of the infrastruc- can be repeated multiple times in each avail- ture in this park in need able location, requiring a set amount of time of repair? to pass between submissions. Designed Global non- Quests that are purposefully placed and that Describe your favourite repeatable can be completed exactly once globally. place in the world. Designed Global repeatable Quests that are purposefully placed and that Tell us about your can be repeated multiple times globally, re- favourite new place you quiring a set amount of time to pass between have discovered this submissions. month. GeneratedLocal non- Quests that are randomly generated and What is your first impres- repeatable that can be completed exactly once in each sion of this area? available location. GeneratedLocal repeatable Quests that are randomly generated and How safe do you feel in that can be repeated multiple times in each this area? available location, requiring a set amount of time to pass between submissions. GeneratedGlobal non- Quests that are randomly generated and Do you prefer urban or ru- repeatable that can be completed exactly once globally. ral landscapes? These quests are useful to generate generic not necessarily spatial data. GeneratedGlobal repeatable Quests that are randomly generated and Upload an image of that can be repeated multiple times glob- your favourite place this ally, requiring a set amount of time to pass month. between submissions. met. gagement with an application as well as the likelihood of word-of-mouth recommendations 3.9. Forge system [65, 17]. Within Arcane Shift, the users must choose between one of three teams during the Exploration as a dimension of gamification onboarding process. The teams each boast ties into the Mechanics, Dynamics, Aesthet- an intriguing backstory and players within a ics Framework aesthetic of discovery and has team can interact with buildings built by other been found to increase user motivation [63, 64]. players within the same team. Buildings of Besides exploring the game world, another im- the same team allow for more in-game inter- plemented form of exploration revolves around actions (e.g. access to the forging mechanic the forging feature, where players can submit as shown in Figure 7) and as such, there is a four ore crystals and have the chance to dis- strong incentive for players to be in proximity cover new items (Figure 7). However, only to friendly buildings. In a first iteration, users specific combinations of ores will result in a could build buildings if they had sufficient gold successful forge rewarding a new item, meaning in their inventory and if there was an unoccu- players must explore the various combinations pied hexagonal H3 tile in the vicinity of the and potentially share their findings with other user. The application would display preview players. buildings and once a user chose the preferred building location by clicking on a preview, a 3.10. Building system loading screen would show whilst the build- ing was created and subsequently saved on the Another key dimension is collaboration, which server. However, constant loading screens were has been found to boost motivation in and en- perceived as disruptive to the game experience. 98 Figure 6: A screenshot of an example location-based Figure 7: A screenshot of the forging mechanic, only quest or spatial crowdsourcing task with two depen- accessible through friendly buildings dent quest steps view of a specific building. This was achieved As such, loading screens were masked with a by: (1) starting a weapon charging animation new building warp mechanic, where the appli- to mask client-server communication lag, (2) cation would show a warping animation instead updating the persistent building data in the of a loading screen (Figure 8). In addition, fur- database, (3) firing a document update event ther unlockable building tiers were added as informing all connected devices of a change and additional game content. (4) updating a respective building’s health on all connected devices. 3.11. Attack system Competition, if provided in adequate amounts, 4. Internal user testing has been found to increase user motivation and engagement in games and gamified applications A testing pipeline was set up on the Google [9, 17]. As such, providing the right amount of Playstore to perform internal user testing. This competitive elements is a valuable tool within allowed invited test participants to download gamification efforts. Within the location-based the application through the Google Playstore game, besides incorporating a variety of leader- and update the application as soon as new ver- boards, competition is fostered by the limited sions of the application became available, a building space in the virtual world and enemy crucial feature to allow for iterative develop- buildings being destroyable to make room for ment. First internal pilot tests were conducted buildings for one’s own team. If a user’s ship’s with a small group of three users to test the sta- generator has enough energy, a user can click on bility of the main game systems and to identify an enemy structure to fire the ship’s weapons at points of improvement. All users were between the enemy structure, inflicting weapon-specific the ages of 32 - 45 years old and covered all damage to the structure. Once a structure’s levels of expertise with location-based games health depletes, it is destroyed to make room for including one user with no experience, one ca- buildings for one’s own team, leading into the sual user and one expert user. The users were dimension of competition. Building updates given access to the application and received must be propagated to all other players with an overview of the location-based game, the 99 volved around user convenience and interaction features such as auto-login or revising the touch detection and tap interactions within the ap- plication. The location-based game will be trialled with a larger test user group before being opened up for public beta testing. 5. Discussion and outlook The developed and implemented location-based game is a novel highly gamified crowdsourcing and citizen science platform, upon which var- ious interesting systems can be built. This paper presents the implementation and the literature-based reasoning behind various de- sign decisions as well as gives an overview of the current state of the application and fu- ture potential. In the following, I first discuss gamifying spatial crowdsourcing in general be- fore presenting key recommendations for build- ing highly gamified spatial crowdsourcing plat- forms and location-based games, gleaned from Figure 8: A screenshot of the building warp mechanic lessons learnt during the development of Ar- which masks loading screens with a warping animation cane Shift. Finally, I lay out specific future whilst data is being persisted to the server agendas for the platform and summarise the findings in the conclusions. various features of the application as well as de- 5.1. Gamifying crowdsourcing tailed instructions of the testing process. The internal test users were asked to pay particular through a location-based game attention to the core systems (e.g. questing, Common approaches implementing the design collecting, building, attacking) and were asked science research method identify a specific topi- to report if they encountered any errors or un- cal research question or organisational problem, expected behaviour. In addition, the users were which is subsequently distilled into application asked for their opinion of the first test version features to be implemented [48, 49, 50, 51]. The and potential improvements that would make approach presented here attempts to address the game more enjoyable. Feedback was col- the more general problem of how users can be lected through an online form, as voluntarily motivated to participate in spatial crowdsourc- submitted additional written feedback and as ing tasks through a gamified application, in one-on-one discussions. return allowing a plethora of domain-specific The users generally enjoyed the location- research questions to be asked. Specifically, based game and reported it being engaging Arcane Shift was built to allow for a variety and mostly stable (n = 3). The users reported of user contributions, allowing for most crowd- liking (n = 1) or loving (n = 2) the overall aes- sourcing tasks to be translated into location- thetics of the game and it being a little (n = 1), based game quests, offering a novel approach very (n = 1) or extremely (n = 1) motivating in to crowdsourcing and citizen science. its current state. Chosen features of the game The implemented location-based game fur- such as upgrading ship components, gaining ex- ther transcends traditional crowdsourcing plat- perience points and levelling up were reported forms in regards to the special focus on en- to be enjoyed by the test users (n = 3). How- tertaining and motivational elements, allow- ever, many features such as collecting resources, ing for high-quality data generation whilst of- constructing buildings and buying items from fering entertaining virtual experiences. This the in-game market were reported to be only contrasts with many more traditional crowd- moderately enjoyed due to the limited amount sourcing approaches, where participants are of in-game content. Further key feedback re- 100 motivated through monetary rewards, scien- provements or important considerations in the tific or other forms of acknowledgements or are design and implementation process. encouraged to participate as part of a curricu- lum [1, 9, 10, 2]. Scientific inquiries leveraging 5.2.1. Develop for testing, test for crowdsourced spatial data are plentiful with development examples ranging from characterising specific landscapes [66], over detecting popular view- Rapid prototyping and user testing are the ports [67] and analysing urban mobility [68], to backbones of creating new information systems mapping a city’s smellscape [45]. These gener- and have proven to be invaluable in the develop- ally use the rich spatial information contained ment of Arcane Shift. Developing applications within geolocated social media data as well as can be seen as an iterative circular process start- other forms of (online) user-generated content ing with implementing features, subsequently and are a welcomed source of easily accessi- testing features and collecting feedback and fi- ble data from a heterogeneous crowd. How- nally refactoring or implementing new features ever, such datasets are generally noisy with according to the collected feedback, effectively many irrelevant data points. The implemented starting a new iteration loop. This approach location-based game aims at minimising irrele- however requires two key implementation con- vant contributions through an active approach siderations from the start of a project. to collecting data. Being able to strategically Firstly, a given application should be divided place crowdsourcing tasks (quests) at specific into at least three distinct environments: (1) locations and the ability to create tailored tasks a development environment in which develop- to collect any information needed makes the ment takes place, (2) a staging environment implemented application a novel highly gami- with which internal or open user testing is con- fied crowdsourcing platform, transcending more ducted and (3) a production environment en- passive approaches to generating crowdsourced compassing the live application. These envi- data. ronments can further be extended according Combining the ease of adding new crowd- to the needs of the project (e.g. additional sourcing tasks either globally or in specific internal, closed and open beta testing envi- locations with the general goal of developing ronments). Using modern versioning software extendable core systems makes this platform such as GIT10 , the code base for each envi- especially interesting for longitudinal studies. ronment can be tracked and changes can be As such, this surpasses general academic crowd- merged as required. Finally, each environment sourcing efforts which are typically developed should incorporate a decoupled database in- for a specific project after which they tend stance so changes to one environment’s back- to disappear [69], potentially due to the dif- end database does not influence the other en- ficulty of updating the underlying systems to vironments. Seperating environments and us- accommodate new research endeavours. The ing version control is common practice in soft- location-based game presented in this paper ware development [71, 72] and despite the ad- takes a different approach: the core systems ditional overhead of setting up versioning and are built to function independently of overar- distinct environments with decoupled database ching projects and research questions. instances, this has proven invaluable to allow for simultaneous testing of multiple versions as well as continued development. 5.2. Recommendations and lessons A second important consideration is mak- learnt ing content easily updatable and propagating During the implementation of Arcane Shift the updates to test users. This is especially im- application has gone through several versions portant for data where minor changes might and substantial amounts of code were refac- become necessary, for example for balancing tored with each new iteration. This dynamic (e.g. weapon damage, energy use, recharge and evolving nature of the application is in rates). Storing such content online and hav- line with modern agile software engineering ap- ing the application download updated values proaches [70, 71] and allows for flexibility in when the application starts allows developers design and development. In the following, I to change values as needed without needing to touch upon four key areas of location-based distribute a new application build to the test game design that have proven to be key im- 10 www.git-scm.com (accessed: 03.01.2024) 101 users. However, downloading content on the coordinates. This becomes particularly impor- fly has the caveat of needing an internet con- tant when incorporating an increasing amount nection, a particularly important consideration of third-party software with unclear develop- when designing mobile applications. ment and maintenance roadmaps. Adhering to established principles where sensible ensures 5.2.2. Incorporate caching and on-demand code is reusable and remains maintainable. Us- content retrieval ing interfaces and favouring composition over inheritance also allows for the easy addition of Moving content to online storage and download- new items such as different ores and a variety of ing needed data when starting an application new buildings or new ship component upgrades. allows for rapid value updates, however, load- This effectively allows for large updates of the ing content from external sources can result in location-based game with new content, keeping notable amounts of mobile data usage. Dur- players engaged and motivated (with infrequent ing the development of Arcane Shift it became large updates reported as preferable over fre- evident that the majority of content remained quent small updates [76]), and as such arguably unchanged between application usages. This increasing the likelihood of players contributing highlighted the importance of caching content to the various crowdsourcing tasks. on the device to minimise mobile data usage. Caching was introduced where content was only 5.2.4. Spatial is special downloaded from an external source if the on- line content was updated. This allowed the A final key area worth considering is the spe- seamless updating of existing content as well as cial nature of developing a globally available the addition of new content, whilst minimising spatial application. Firstly, visualising a digi- needed bandwidth and mobile data usage. This tal representation of a user’s surroundings as is in line with modern online application design the game world requires careful consideration best practices, advocating for fast content load- of the level of abstraction of the spatial fea- ing and easy navigation [73, 74]. Carefully con- tures. Adding too many details such as points sidering which content is immutable and which of interest or cluttered map labels can result content requires dynamic updating is thus a in cognitive overload [77] whilst reducing the key consideration when building location-based available screen real estate for virtual game games or spatial crowdsourcing applications. objects. Reducing map complexity could also be in part responsible for the lack of altitude 5.2.3. Build for longevity information in many location-based games [21]. However, not adding enough spatial informa- The location-based game was built keeping fu- tion can confuse the user as to their in-game ture development in mind and thus, particular location and the translation between real-world focus was set on building expandable core sys- and game-world coordinates. As such, striking tems and interchangeable modular components a balance between incorporating a minimalist with accessible interfaces. An important consid- base map with enough spatial information to eration when building an application is follow- ground the user in the game world is needed. ing established software development principles. Secondly, seeing the potentially global scale of Modern well-known software design principles location-based games, I recommend pursuing advocate single responsibility classes (such as a threefold content creation strategy: (1) de- advocated by the SOLID principle [75]) and signed content, (2) user-generated content and urge not to repeat code. Developing an applica- (3) procedurally generated content. Developing tion in line with these principles ensures a mod- the application to accommodate these three ular system with accessible interfaces making distinct content generation strategies will allow code reusable and testable. An added benefit for specific areas of the game to be purpose- is the ease of updating individual components fully designed especially in highly frequented of the application. For example, seeing the areas inviting the creation of custom content. location provider implemented in Arcane Shift In addition, having user-generated content al- exposes the current coordinates of the player lows for a dynamic and changing game world through a public interface that other classes can and procedurally generated content offers glob- consume, the location provider can be updated ally available content. Finally, of particular or completely changed at any point as long as it importance when designing applications geared exposes the same public interface that provides towards crowdsourcing spatial information is 102 considering the spatial and temporal availabil- vate physical activity), and location-based games ity and repeatability of crowdsourcing tasks. in general, have been found to increase physi- During the development of Arcane Shift, four cal activity and in return affect users’ overall overarching categories of tasks emerged: spa- health and well-being [86, 87]. This offers a tially and temporally repeatable, spatially but third and important line of research for the not temporally repeatable, temporally but not implemented application: questioning the abil- spatially repeatable and neither spatially nor ity of the location-based game to motivate in- temporally repeatable. This allows for the col- creased physical activity. Motivating users for lection of detailed information about specific physical activity in urban environments has be- areas over time as well as increasing spatial come an important dimension of urbanisation coverage. research and entertaining motivational incen- tives such as location-based games show great 5.3. Future agendas potential. Within this paper, I have presented the design decisions and the reasoning behind the individ- 6. Limitations ual implementation steps in creating Arcane Shift, a global gamified crowdsourcing platform. The implemented location-based game is ac- In the following, I divulge three key lines of companied by various limitations, key of which research building on top of the implemented are listed here. platform. These aim to showcase the potential Firstly, a large active player base and broad of the application for sparking future research spatial coverage of game content are needed to in a wide variety of domains. make a virtual world feel alive and motivate Crowdsourcing and participatory approaches user engagement [88]. This limitation can be have been identified as valuable complementary somewhat mitigated by procedurally generated approaches in policy and decision-making con- content and non-player characters or large-scale texts [78, 79, 80]. A first scientific avenue to in-game events, conjuring the illusion of an explore is thus the potential of using the imple- active and dynamic game world. mented location-based game to collect policy A second key limitation is the requirement and decision-making relevant spatial datasets. of an adequate internet connection. This ex- Seeing the particular need for urban data to cludes many rural areas which have generally pave the way for future urban living [81, 82] been underrepresented in location-based games in combination with the high spatial coverage [19], potentially reinforcing the digital divide of mobile broadband internet in urban areas, [5]. However, an active internet connection the implemented location-based game offers is required to allow for real-time updates as a particularly interesting platform to collect needed for the game experience. As such, an in- a variety of urban information from the local evitable tradeoff between real-time multiplayer citizens themselves. features and asynchronous offline gaming must Secondly, location-based games have shown be carefully considered. One possible approach great potential in delivering certain educational to address this limitation is to implement spe- content [83, 84]. The possibility of implement- cific offline features accessible anywhere (e.g. ing chained quest steps as journeys opens the accepting a quest) which are not dependent door to investigating the platform’s ability to on real-time updates from other players or the deliver location-based educational content, which game world. may be of particular interest to educational One final major limitation of the implemented professionals seeking new ways to engage with location-based game is the dependence on com- their students in a time of increased blended mercial products. The implemented platform and gamified learning [85]. The platform of- was built from scratch using state-of-the-art fers the ability to create tailored educational technology and commercial products (e.g. Unity, journeys, guiding learners through specific real- MapBox, FireStore) were used or integrated as world locations. Thus, testing educational vari- a means of minimising overall platform main- ables such as knowledge acquisition and learner tenance. Access to these products may change motivation could become an interesting future in the future, raising an important issue of investigation. longevity. I try to address this issue by im- Lastly, exergames (applications that moti- plementing many of the core systems to be modular and adapted if need be. For exam- 103 ple, the developed mapping system displaying References the base map can easily be modified to dis- play map tiles of other providers such as Open [1] L. See, P. Mooney, G. Foody, L. Bastin, Street Map. A. Comber, J. Estima, S. Fritz, N. Kerle, B. Jiang, M. Laakso, H.-Y. Liu, G. Milčin- ski, M. Nikšič, M. Painho, A. Pődör, 7. Conclusions A.-M. Olteanu-Raimond, M. Rutzinger, Crowdsourcing, citizen science or volun- Within this paper, I have presented the devel- teered geographic information? the cur- opment and implementation of a novel spatial rent state of crowdsourced geographic in- crowdsourcing platform in the form of a glob- formation, ISPRS International Journal ally available real-time location-based game. of Geo-Information 5 (2016) 55. doi:10. The implemented application was not devel- 3390/ijgi5050055. oped to address a specific research question [2] M. Bubalo, B. T. van Zanten, P. H. Ver- in favour of serving as an underlying gamified burg, Crowdsourcing geo-information platform to host a variety of spatial crowdsourc- on landscape perceptions and prefer- ing initiatives globally. The design and imple- ences: A review, Landsc. Urban mentation of the application were grounded Plan. 184 (2019) 101–111. doi:10.1016/ in the Design Science Research Methodology j.landurbplan.2019.01.001. and the application was built using two well- [3] Y. Tong, Z. Zhou, Y. Zeng, L. Chen, established frameworks. The location-based C. Shahabi, Spatial crowdsourcing: a game itself revolves around collecting and forg- survey, VLDB J. 29 (2020) 217–250. ing resources, upgrading one’s spacecraft, build- doi:10.1007/s00778-019-00568-7. ing structures, attacking enemy buildings and [4] R. Lenart-Gansiniec, W. Czakon, completing location-based quests which are Sułkowski, J. Pocek, Understand- crowdsourcing tasks. I further presented a ing crowdsourcing in science, Rev. list of key design an implementation recom- Manag. Sci. 17 (2023) 2797–2830. mendations, distilled from the lessons learnt doi:10.1007/s11846-022-00602-z. during the implementation of the application, [5] M. F. Goodchild, Citizens as sensors: to contribute to the theoretical discussions on the world of volunteered geography, Geo- implementing gamified spatial crowdsourcing Journal 69 (2007) 211–221. doi:10.1007/ platforms or location-based games more gener- s10708-007-9111-y. ally. Finally, by presenting three distinct future [6] L. See, A. Comber, C. Salk, S. Fritz, research agendas, I hope to have highlighted M. van der Velde, C. Perger, C. Schill, the intended impact and present real-world sci- I. McCallum, F. Kraxner, M. Obersteiner, entific domains that would benefit from the Comparing the quality of crowdsourced implemented platform. Arcane Shift hopes to data contributed by expert and non- pave the way towards an era of entertaining experts, PLoS One 8 (2013) 1–11. doi:10. engagement in science and societally relevant 1371/journal.pone.0069958. data collection. [7] A. Ghermandi, M. Sinclair, Passive crowd- sourcing of social media in environmen- Acknowledgments tal research: A systematic map, Glob. Environ. Change 55 (2019) 36–47. doi:10. I would like to extend my gratitude to the in- 1016/j.gloenvcha.2019.02.003. ternal test users for their amazing feedback. I [8] A. Baruch, A. May, D. Yu, The moti- would also like to thank the anonymous review- vations, enablers and barriers for volun- ers for their detailed feedback, considerably tary participation in an online crowdsourc- improving the quality of this publication. This ing platform, Comput. Human Behav. 64 project was funded by the Swiss National Sci- (2016) 923–931. doi:10.1016/j.chb.2016. ence Foundation (SNSF) under grant number 07.039. P500PT_214436. [9] B. Morschheuser, J. Hamari, J. Koivisto, A. Maedche, Gamified crowdsourcing: Conceptualization, literature review, and future agenda, International Journal of Human Computer Studies 106 (2017) 26– 104 43. doi:10.1016/j.ijhcs.2017.04.005. on System Sciences, 2014, pp. 3025–3034. [10] L. F. S. Pinto, C. D. dos Santos, Motiva- doi:10.1109/HICSS.2014.377. tions of crowdsourcing contributors, Inno- [19] A. Colley, J. Thebault-Spieker, A. Y. Lin, vation & Management Review 15 (2018) D. Degraen, B. Fischman, J. Häkkilä, 58–72. doi:10.1108/inmr-02-2018-004. K. Kuehl, V. Nisi, N. J. Nunes, N. Wenig, [11] F. R. Vianna, J. Peinado, A. R. Graeml, D. Wenig, B. Hecht, J. Schöning, The Crowdsourcing platforms: Objective, ac- geography of pokémon GO: Beneficial and tivities and motivation, 25th Americas problematic effects on places and move- Conference on Information Systems, AM- ment, arXiv (2017) 1179–1192. CIS 2019 (2019). [20] S. Laato, A. K. M. N. Islam, T. H. Laine, [12] J. Koivisto, J. Hamari, The rise of mo- Did location-based games motivate players tivational information systems: A review to socialize during COVID-19?, Telemat. of gamification research, Int. J. Inf. Man- Inform. 54 (2020) 101458. doi:10.1016/j. age. 45 (2019) 191–210. doi:10.1016/j. tele.2020.101458. ijinfomgt.2018.10.013. [21] M. Baer, T. Tregel, S. Laato, H. Söbke, [13] J. C. L. Bayas, L. See, S. Fritz, T. Sturn, Virtually (re)constructed reality: the rep- C. Perger, M. Dürauer, M. Karner, resentation of physical space in commer- I. Moorthy, D. Schepaschenko, D. Do- cial location-based games, in: Proceed- mian, I. McCallum, Crowdsourcing in- ings of the 25th International Academic situ data on land cover and land use using Mindtrek Conference, ACM, New York, gamification and mobile technology, Re- NY, USA, 2022. doi:10.1145/3569219. mote Sensing 8 (2016) 1–18. doi:10.3390/ 3569339. rs8110905. [22] S. Laato, S. M. Hyrynsalmi, M. Palo- [14] S. Hantke, T. Appel, B. Schuller, The heimo, Online multiplayer games for inclusion of gamification solutions to en- crowdsourcing the development of digi- hance user enjoyment on crowdsourcing tal assets: The case of ingress, in: Lec- platforms, 2018 1st Asian Conference on ture Notes in Business Information Pro- Affective Computing and Intelligent In- cessing, Lecture notes in business infor- teraction, ACII Asia 2018 (2018). doi:10. mation processing, Springer International 1109/ACIIAsia.2018.8470330. Publishing, Cham, 2019, pp. 387–401. [15] T. Ivanjko, Crowdsourcing image descrip- doi:10.1007/978-3-030-33742-1\_31. tions using gamification: A comparison [23] S. Laato, T. Tregel, Into the unown: between game-generated labels and pro- Improving location-based gamified crowd- fessional descriptors, 2019 42nd Inter- sourcing solutions for geo data gather- national Convention on Information and ing, Entertain. Comput. 46 (2023) 100575. Communication Technology, Electronics doi:10.1016/j.entcom.2023.100575. and Microelectronics, MIPRO 2019 - Pro- [24] D. Leorke, A definition and brief his- ceedings (2019) 537–541. doi:10.23919/ tory of location-based games (2001–08), MIPRO.2019.8756841. in: Location-Based Gaming: Play in [16] M. Baer, F. M. Wartmann, R. S. Purves, Public Space, Springer Singapore, Sin- StarBorn: Towards making in-situ land gapore, 2019, pp. 17–44. doi:10.1007/ cover data generation fun with a loca- 978-981-13-0683-9\_2. tion based game, Trans. GIS (2019) 1–21. [25] De Souza E Silva, L. Hjorth, Playful urban doi:10.1111/tgis.12543. spaces: Historical approach mobile games, [17] B. Morschheuser, J. Hamari, A. Maed- Simulation & Gaming 40 (2009) 602–625. che, Cooperation or competition – when [26] L. Juhász, H. H. Hochmair, Where do people contribute more? a field exper- to catch ‘em all?–a geographic analy- iment on gamification of crowdsourcing, sis of pokémon go locations, Geo Spat. International Journal of Human Computer Inf. Sci. 20 (2017) 241–251. doi:10.1080/ Studies 127 (2019) 7–24. doi:10.1016/j. 10095020.2017.1368200. ijhcs.2018.10.001. [27] I. Celino, D. Cerizza, S. Contessa, [18] J. Hamari, J. Koivisto, H. Sarsa, Does M. Corubolo, D. Dellaglio, E. D. Valle, gamification work? – a literature review S. Fumeo, Urbanopoly - a social of empirical studies on gamification, in: and location-based game with a pur- 2014 47th Hawaii International Conference pose to crowdsource your urban data, 105 in: Proceedings - 2012 ASE/IEEE acoustics.asn.au, 2010, pp. 23–27. International Conference on Privacy, [36] G. Waters, B. Warren, E. Ratcliffe, Security, Risk and Trust and 2012 J. Godefroy, Tranquil city: identifying ASE/IEEE International Conference on opportunities for urban tranquillity to pro- Social Computing, SocialCom/PASSAT mote healthy lifestyles, Cities & Health 2012, 2012, pp. 910–913. doi:10.1109/ 5 (2021) 138–144. doi:10.1080/23748834. SocialCom-PASSAT.2012.138. 2019.1617918. [28] A. Abraham, K. Sommerhalder, T. Abel, [37] C. Rissel, N. Curac, M. Greenaway, Landscape and well-being: A scoping A. Bauman, Physical activity associated study on the health-promoting impact with public transport use–a review and of outdoor environments, Int. J. Pub- modelling of potential benefits, Int. J. En- lic Health 55 (2010) 59–69. doi:10.1007/ viron. Res. Public Health 9 (2012) 2454– s00038-009-0069-z. 2478. doi:10.3390/ijerph9072454. [29] C. W. Thompson, Linking landscape and [38] I. Batur, S. A. Markolf, M. V. Chester, health: The recurring theme, Landsc. A. Middel, D. Hondula, J. Vanos, Street- Urban Plan. 99 (2011) 187–195. doi:10. level heat and air pollution exposure in- 1016/j.landurbplan.2010.10.006. formed by mobile sensing, Transp. Res. [30] L. Menatti, A. C. Da Rocha, Landscape D Transp. Environ. 113 (2022) 103535. and health: Connecting psychology, aes- doi:10.1016/j.trd.2022.103535. thetics, and philosophy through the con- [39] K. Fuladlu, H. Altan, Examining land cept of affordance, Front. Psychol. 7 (2016) surface temperature and relations with 1–17. doi:10.3389/fpsyg.2016.00571. the major air pollutants: A remote sensing [31] C. Llinares, J. L. Higuera-Trujillo, A. Mon- research in case of tehran, Urban Clim. tañana, N. Castilla, Improving the pedes- 39 (2021) 100958. doi:10.1016/j.uclim. trian’s perceptions of safety on street 2021.100958. crossings. psychological and neurophysi- [40] H. Wang, Y. Yang, Neighbourhood walk- ological effects of traffic lanes, artificial ability: A review and bibliometric analy- lighting, and vegetation, Int. J. Envi- sis, Cities 93 (2019) 43–61. doi:10.1016/ ron. Res. Public Health 17 (2020) 8576. j.cities.2019.04.015. doi:10.3390/ijerph17228576. [41] C. Legacy, Investigating the knowledge [32] E. Zeng, Y. Dong, L. Yan, A. Lin, Per- interface between stakeholder engagement ceived safety in the neighborhood: Ex- and plan-making, Environ. Plan. A 42 ploring the role of built environment, so- (2010) 2705–2720. doi:10.1068/a43164. cial factors, physical activity and multiple [42] R. G. Hollands, Critical interventions into pathways of influence, Buildings 13 (2022) the corporate smart city, Camb. J. Reg. 2. doi:10.3390/buildings13010002. Econ. Soc. 8 (2015) 61–77. doi:10.1093/ [33] F. M. Wartmann, K. F. Tieskens, B. T. cjres/rsu011. van Zanten, P. H. Verburg, Explor- [43] G. Brown, M. Kyttä, Key issues and ing tranquillity experienced in landscapes research priorities for public participa- based on social media, Appl. Ge- tion GIS (PPGIS): A synthesis based ogr. 113 (2019) 102112. doi:10.1016/j. on empirical research, Appl. Geogr. 46 apgeog.2019.102112. (2014) 122–136. doi:10.1016/j.apgeog. [34] N. Fagerholm, B. Martín-López, M. Tor- 2013.11.004. ralba, E. Oteros-Rozas, A. M. Lechner, [44] E. Estellés-Arolas, F. González-Ladrón- C. Bieling, A. Stahl Olafsson, C. Al- De-Guevara, Towards an integrated bert, C. M. Raymond, M. Garcia-Martin, crowdsourcing definition, J. Inf. Sci. N. Gulsrud, T. Plieninger, Perceived con- Eng. 38 (2012) 189–200. doi:10.1177/ tributions of multifunctional landscapes to 0165551512437638. human well-being: Evidence from 13 eu- [45] D. Quercia, R. Schifanella, L. M. Aiello, ropean sites, People and Nature 2 (2020) K. McLean, Smelly maps: The digital life 217–234. doi:10.1002/pan3.10067. of urban smellscapes, Association for the [35] G. Watts, R. Pheasant, K. Horoshenkov, Advancement of Artificial Intelligence 12 Tranquil spaces in a metropolitan area, (2015). arXiv:1505.06851. in: Proceedings of the 20th International [46] L. M. Aiello, R. Schifanella, D. Quercia, Congress on Acoustics, Sydney, Australia, F. Aletta, Chatty maps : constructing 106 sound maps of urban areas from social quality in urban, rural, and tribal con- media data, R. Soc. open sci. 3 (2016). texts, in: 2021 International Confer- [47] G. Lamprianidis, D. Skoutas, G. Pap- ence on Computer Communications and atheodorou, D. Pfoser, Extraction, in- Networks (ICCCN), IEEE, 2021. doi:10. tegration and analysis of crowdsourced 1109/icccn52240.2021.9522152. points of interest from multiple web [56] P. M. Duvall, S. Matyas, A. Glover, Con- sources, in: Proceedings of the 3rd tinuous Integration: Improving Software ACM SIGSPATIAL International Work- Quality and Reducing Risk, Pearson Edu- shop on Crowdsourced and Volunteered cation, Philadelphia, PA, 2007. Geographic Information, ACM, New York, [57] M. Shahin, M. Ali Babar, L. Zhu, Contin- NY, USA, 2014. doi:10.1145/2676440. uous integration, delivery and deployment: 2676445. A systematic review on approaches, tools, [48] S. T. March, G. F. Smith, Design challenges and practices, IEEE Access and natural science research on infor- 5 (2017) 3909–3943. doi:10.1109/access. mation technology, Decis. Support 2017.2685629. Syst. 15 (1995) 251–266. doi:10.1016/ [58] A. Henry, L2 motivation and multilingual 0167-9236(94)00041-2. identities, Mod. Lang. J. 101 (2017) 548– [49] K. Peffers, T. Tuunanen, M. A. Rothen- 565. doi:10.1111/modl.12412. berger, S. Chatterjee, A design sci- [59] A. Cardinal, L. Gonzales, E. J. Rose, ence research methodology for infor- Language as participation: Multilingual mation systems research, J. Manag. user experience design, in: Proceedings Inf. Syst. 24 (2007) 45–77. doi:10.2753/ of the 38th ACM International Confer- mis0742-1222240302. ence on Design of Communication, ACM, [50] R. Winter, Design science research in eu- New York, NY, USA, 2020. doi:10.1145/ rope, Eur. J. Inf. Syst. 17 (2008) 470–475. 3380851.3416763. doi:10.1057/ejis.2008.44. [60] D. M. Da Rosa, M. L. L. Souto, M. S. [51] J. vom Brocke, A. Hevner, A. Maedche, Silveira, Designing interfaces for multi- Introduction to design science research, lingual users: a pattern language, in: in: Progress in IS, Springer International Proceedings of the 27th European Con- Publishing, Cham, 2020, pp. 1–13. doi:10. ference on Pattern Languages of Pro- 1007/978-3-030-46781-4\_1. grams, ACM, New York, NY, USA, 2022. [52] R. Hunicke, M. Leblanc, R. Zubek, MDA: doi:10.1145/3551902.3551964. A formal approach to game design and [61] F. Kreuter, G.-C. Haas, F. Keusch, game research, in: Proceedings of the S. Bähr, M. Trappmann, Collecting sur- Challenges in Games AI Workshop, AAAI vey and smartphone sensor data with Press, Menlo Park, CA, 2004. an app: Opportunities and challenges [53] R. Martella, C. Kray, E. Clementini, A around privacy and informed consent, Soc. gamification framework for volunteered Sci. Comput. Rev. 38 (2020) 533–549. geographic information, in: F. Bacao, doi:10.1177/0894439318816389. M. Y. Santos, M. Painho (Eds.), AGILE [62] S. Nicholson, A RECIPE for meaning- 2015: Geographic Information Science as ful gamification, in: Gamification in Ed- an Enabler of Smarter Cities and Com- ucation and Business, Springer Interna- munities, volume 217, Springer Interna- tional Publishing, Cham, 2015, pp. 1–20. tional Publishing, Cham, 2015, pp. 73–89. doi:10.1007/978-3-319-10208-5\_1. doi:10.1007/978-3-319-16787-9\_5. [63] H. Wang, C.-T. Sun, Game reward sys- [54] G. P. Kusuma, E. K. Wigati, Y. Utomo, tems : Gaming experiences and social L. K. Putera Suryapranata, Analysis meanings, in: Proceedings of DiGRA 2011 of gamification models in education us- Conference: Think Design Play., 2011, pp. ing MDA framework, Procedia Comput. 1–15. Sci. 135 (2018) 385–392. doi:10.1016/j. [64] J. Brich, Motivational game design fac- procs.2018.08.187. tors in player-game adaptivity, in: CHI [55] V. Adarsh, M. Nekrasov, U. Paul, PLAY ’17 Extended Abstracts: Extended T. Mangla, A. Gupta, M. Vigil-Hayes, Abstracts Publication of the Annual Sym- E. Zegura, E. Belding, Coverage is not posium on Computer-Human Interaction binary: Quantifying mobile broadband in Play, 2017, pp. 683–686. doi:10.1145/ 107 3130859.3133222. 6 (2014) 1–9. doi:10.5815/ijmecs.2014. [65] S. Fritz, L. See, M. Brovelli, Motivating 06.01. and sustaining participation in VGI, in: [74] Y. Wang, Y. Huang, J. Li, J. Zhang, S. Fritz, L. See, G. Foody, P. Mooney, A.- The effect of mobile applications’ initial M. Olteanu-Raimond, C. C. Fonte, V. An- loading pages on users’ mental state and toniou (Eds.), Mapping and the Citizen behavior, Displays 68 (2021) 102007. Sensor, Ubiquity Press, 2017, pp. 93–118. doi:10.1016/j.displa.2021.102007. [66] O. Koblet, R. S. Purves, From online [75] M. E. Rana, E. Khonica, Impact of design texts to landscape character assessment: principles and patterns on software flexi- Collecting and analysing first-person land- bility: An experimental evaluation using scape perception computationally, Landsc. flexible point (FXP), J. Comput. Sci. 17 Urban Plan. 197 (2020) 103757. doi:10. (2021) 624–638. doi:10.3844/jcssp.2021. 1016/j.landurbplan.2020.103757. 624.638. [67] M. C. Hartmann, O. Koblet, M. F. Baer, [76] X. Zhong, J. Xu, Measuring the effect R. S. Purves, Automated motif identifica- of game updates on player engagement: tion: Analysing flickr images to identify A cue from DOTA2, Entertain. Comput. popular viewpoints in europe’s protected 43 (2022) 100506. doi:10.1016/j.entcom. areas, Journal of Outdoor Recreation and 2022.100506. Tourism 37 (2022) 100479. doi:10.1016/ [77] C. Guo, S. Wei, Y. Chen, Cogni- j.jort.2021.100479. tive processing of information visualiza- [68] G. Marzano, J. Lizut, L. O. Siguen- tion, Encyclopedia of Computer Graph- cia, Crowdsourcing solutions for support- ics and Games (2018) 1–12. doi:10.1007/ ing urban mobility, Procedia Comput. 978-3-319-08234-9\_95-1. Sci. 149 (2019) 542–547. doi:10.1016/j. [78] C. Certomà, F. Corsini, F. Rizzi, Crowd- procs.2019.01.174. sourcing urban sustainability. data, people [69] D. Nevo, J. Kotlarsky, Crowdsourcing as and technologies in participatory gover- a strategic IS sourcing phenomenon: Criti- nance, Futures 74 (2015) 93–106. doi:10. cal review and insights for future research, 1016/j.futures.2014.11.006. J. Strat. Inf. Syst. 29 (2020) 101593. [79] A. Taeihagh, Crowdsourcing: a doi:10.1016/j.jsis.2020.101593. new tool for policy-making?, Policy [70] P. Abrahamsson, N. Oza, M. T. Sipo- Sci. 50 (2017) 629–647. doi:10.1007/ nen, Agile software development methods: s11077-017-9303-3. A comparative Review1, in: Agile Soft- [80] T. Aitamurto, K. Chen, The value ware Development, Springer Berlin Heidel- of crowdsourcing in public policymak- berg, Berlin, Heidelberg, 2010, pp. 31–59. ing: epistemic, democratic and economic doi:10.1007/978-3-642-12575-1\_3. value, Theory Pract. Legis. 5 (2017) 55–72. [71] L. Williams, G. Brown, A. Meltzer, N. Na- doi:10.1080/20508840.2017.1282665. gappan, Scrum + engineering practices: [81] S. Barns, Smart cities and urban data Experiences of three microsoft teams, In- platforms: Designing interfaces for smart ternational Symposium on Empirical Soft- governance, City Cult. Soc. 12 (2018) 5– ware Engineering and Measurement (2011) 12. doi:10.1016/j.ccs.2017.09.006. 463–471. doi:10.1109/esem.2011.65. [82] H. Niu, E. A. Silva, Crowdsourced [72] D. Jones, A. Nassehi, C. Snider, J. Gop- data mining for urban activity: Re- sill, P. Rosso, R. Real, M. Goudswaard, view of data sources, applications, and B. Hicks, Towards integrated version con- methods, J. Urban Plan. Dev. 146 trol of virtual and physical artefacts in (2020) 04020007. doi:10.1061/(asce)up. new product development: Inspirations 1943-5444.0000566. from software engineering and the digi- [83] S. Schaal, Location-based games for tal twin paradigm, Procedia CIRP 100 geography and environmental educa- (2021) 283–288. doi:10.1016/j.procir. tion, in: Geography Education in the 2021.05.121. Digital World, 1st edition ed., Rout- [73] H. K. Flora, X. Wang, S. V.Chande, An ledge, 2020, pp. 168–178. doi:10.4324/ investigation into mobile application de- 9780429274909-15. velopment processes: Challenges and best [84] J. Sheahan, H. Davies, L. Hjorth, The art practices, Int. J. Mod. Educ. Comput. Sci. of tacit learning in serious location-based 108 games, Front. Educ. 6 (2021) 686633. doi:10.3389/feduc.2021.686633. [85] S. Hrastinski, What do we mean by blended learning?, TechTrends 63 (2019) 564–569. doi:10.1007/s11528-019-00375-5. [86] C. Mihale-Wilson, P. Felka, O. Hinz, M. Spann, The influence of location- based games on traditional entertain- ment products, Decis. Support Syst. 149 (2021) 113604. doi:10.1016/j.dss.2021. 113604. [87] J. C. Davis, L. G. Killen, J. M. Green, H. S. Waldman, L. G. Renfroe, Exergaming for physical activity: A systematic review, J Am. Coll. Health (2022) 1–9. doi:10.1080/ 07448481.2022.2103377. [88] P. Fränti, N. Fazal, Design principles for content creation in location-based games, ACM Trans. Multimed. Comput. Com- mun. Appl. (2023). doi:10.1145/3583689. 109