A Toolkit and Lessons for Designing Smart Things with Children for Outdoor Environments* Rosella Gennaria, Maristella Materab, Alessandra Melonioa, Mehdi Rizvib, and Eftychia Roumeliotia a Free University of Bozen-Bolzano, Piazza Domenicani 3, 39100 Bolzano, Italy b Politecnico di Milano, via Ponzio 34/5, 20133 Milan, Italy Abstract Smart things are present in children’s life, from smart watches to smart toys. Designing smart things with children can help them understand the inner working of smart things and reflect on the usage of technology. This paper stems from action research around smart-thing design with children, conducted across three years, in diverse countries and with different children, and mainly for outdoors’ environments. The analysis of data gathered across the research enabled authors of this paper to distil practical guidelines for smart-thing design toolkits for children, which helped children design for a given environment and reflect across design. They might help other researchers organise smart-thing design with children, encompassing different design stages, and make them reflect in design, in relation to the chosen environment. Keywords Children, design, smart thing, nature, outdoors, action research 1. Introduction The role of children in design has been explored extensively in the Child-Computer Interaction and related communities. In the participatory design tradition, researchers or practitioners team up with children so as to support the ideation of novel design solutions and, lately, to reflect while ideating so as to foster critical skills in children’s approach to technology. In the making tradition, children are invited to program and prototype solutions, and tinker with different sorts of material, in an informal learning context. In both traditions, children’s benefits should be considered in the design process. However, in order to ensure that children’s benefits are considered in the entire design process, from ideation to programming and prototyping, design needs to be framed within an encompassing research approach, which plans for and assesses such benefits. Authors of this manuscript chose to frame design with children with action-research [1]. Specifically, this paper stems from the analysis of an action research experience, longer than three years, with 70 children, 41% females and 59% males, aged 8–16 years old, and across different towns in Italy and Greece, e.g., [3,4,6]. See the table below. Along years, children were challenged to design smart things, mainly for outdoor-nature parks, e.g., a smart tree which invites park visitors to listen to its story. The SNaP design toolkit was purposefully developed and evolved, so as to support children’s gameful design of smart things, according to the chosen environment. Notice that, whereas playful design aims at affording so-called “paidic qualities”, characteristic for unstructured play, gameful design aims at embedding “ludic qualities or gamefulness (the experiential qualities characteristic for gameplay)” in design [2]. Proceedings of the NatureHCI 2021 workshop, co-located with the CHItaly 2021 conference, July 12, 2021, Bolzano, Italy. EMAIL: gennari@inf.unibz.it (A. 1); maristella.matera@polimi.it (A. 2); alessandra.melonio@unibz.it (A. 3); syedmehdi.rizvi@polimi.it (A. 4); eftychia.roumelioti@stud-inf.unibz.it (A. 5) ORCID: 0000-0003-0063-0996 (A. 1); 0000-0003-0552-8624 (A. 2); 0000-0001-6655-1946 (A. 3); 0000-0001-8386-5779 (A. 4); 0000- 0003-3293-4521 (A. 5) © 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Wor Pr ks hop oceedi ngs ht I tp: // ceur - SSN1613- ws .or 0073 g CEUR Workshop Proceedings (CEUR-WS.org) Actions were the smart-thing design workshops, in which children used the toolkit for designing and reflecting in design. In other words, in line with the perspective adopted by the workshop but with children as main actors, actions aimed at making children explore the possible usages of technology, reflect on technology and critically develop technology for nature environments. Data were uniformly gathered and analysed, leading to the reflections which are the focus of this paper. The reflections are shaped as lessons concerning gameful smart-thing design for children, which we hope can spark interest in the community of researchers willing to engage in design of technology- enhanced nature environments. When How Where Participants Summer and Autumn In presence Athens, Milan, Bolzano 12 (8F, 4M) 2018 Summer 2019 In presence Bolzano 27 (10F, 17M) Autumn 2019 In presence Bolzano 4 (1F, 3M) Summer 2020 At a distance, in Ioannina, Milan, Bolzano 7 (3F, 4M) presence Winter 2020 At a distance Salerno 20 (7F, 13M) Total n. of participant children: 70 (29F, 41M) FabLearn Europe / MakeEd 2021, June 02–03, 2021, Virtual Conference First Author, et al. 157 The initial reference 2. Design framework for structuring the design process in workshops with children was the one by Smith Model 158 et al. [2015a]. This was adapted to the requirements for smart thing design with children across the COVID-19 pandemic 159 A reference and actions, modeltangibles as interactive for structuring evolved the design in the workprocess by Rizviinetworkshops with children is by Smith al. [17, 18, 36]. 160 et al. [7]. This was adapted to the requirements for smart-thing design with children along three 161 162 163 164 165 166 167 168 169 years of action research. Figure 1. The design model Fig. 1. The smart thing design framework 170 171 In the resulting model, smart-thing design starts with the exploration of smart things, so as to 172 In thefamiliarise resulting framework, smart children with thething design design starts with language the exploration and what of smart smart things things, are made so as todevices of—input familiarise children 173 with thewith design language certain and whatoutput properties, smart devices things are made with of—input other devices properties, with and certain things of properties, the chosenoutput naturedevices with 174 environment, e.g., trees other properties, and things of anorenvironment. benches. It steps through It steps thetheir through ideation of smart ideation, when things, when di�erent different ideas are brainstormed 175 ideas are brainstormed over, and conceptualisation, when children converge on an idea and 176 over, and conceptualisation, when children converge on an idea. It moves then children into the programming and play it out. It moves then children into the programming and prototyping of their ideas of smart 177 prototyping of for things theirthe ideas of smart chosen things. AllAll environment. stages stagesareare intertwined intertwinedwith multiple with re�ection multiple stimuli, reflection from peers and stimuli, 178 experts alike. from See peersFigure 1. All stages and experts alike.areSee made “tangible” Figure 1. All and connected stages are madeby “tangible” means of the toolkit, and explained connected by next. 179 means of the toolkit, explained next. 180 3.2 A Toolkit for Smart Thing Design 181 3. The SNaP Design Toolkit 182 The framework itself is centred around a gami�ed toolkit for designing smart things with children, namely SNaP, 183 which evolved with toolkits Generative the framework are veryalong oftenaction-research used in designcycles. Figure processes to 2facilitate recaps all thethe main actions ideation and the main stage of 184 components of the toolkits smart-thing design,which were whereas used in the actions. programming and prototyping toolkits are employed in the related 185 stages of smart-thing design. These tools can serve as a common “design language” for 186 designers, researchers and users in the design process. Game cards, in particular, have been 187 188 189 190 191 and expanded on them through a physical version of SNaP, and companion programming technology. These children were able to take up others’ ideas and programmed them to completion. ⌅ Hybrid design should promote collaborations (e.g., asynchronous), taking care of ethical considerations, and include children across frontiers. used to engage non-experts in a wide range of design processes, acting as sources of inspiration 5 GUIDELINES FOR SMART-THING DESIGN TOOLKITS and a tangible, play design material, e.g., [3,4]. 5.1 Toolkits Game withcards can be of Components used to motivate Smart Things participants towards the design goal and help them understand its context. Examples are scenario and mission cards of Tiles [5]. Although not In case of toolkits for generic end users, technology tends to be represented in an abstract manner on cards. Examples necessarily in the form of game cards, motivation material is frequently used in workshops are Tiles cards, withwhich represent children, e.g.,what the people work can by do with et Smith input al.,and output e.g., theydevices of smart things, used briefing such as statements to human start actions (e.g.,immersing touch) and receiving childrenfeedback (e.g., sound) in a scenario [7]. [31]. SNaP Similarly, the IoT Design also employed Deckbriefing similar includes statements input and output for cards oriented to a perception and action mechanism, for helping non-experts design the behavior of smart devices for immersing children in a context, as well as motivation cards in the form of mission cards [10]. giving the design goals; by playing these cards, each player chooses her or his own The IoT Service Kit contains cards for sensors and interactions to represent functionalities of input and output devices mission for the smart things under design. Cards for smart things also tend to have so-called technology in an abstract way [6]. The Lighting User Experience (LUX) cards contain input cards to describe the data source for decks for input (e.g., buttons) and output devices (e.g., LED matrix), which are related to the design ofphysical smart lighting devices.solutions Cards[9]. Generic need input also to and output represent the cards, thingssuch as made to be the above ones, smart, have and thusthethey potential are advantage ofhighly being usable for ideating di�erent smart things. However, they can be di�cult to context dependent. In the work reported in this paper, they were related to park elements, match with technology for programming such smart things,or as benches and engage trees. non-experts Examples in this. of such decks of cards are in Figure 2, left side. Cards for Cards areinpart children, of a board particular, need togame, be havephysical or digital, more concrete which guides representations thanchildren cards for(1) to explore adults cards if their aim is to and hence components of smart things, (2) to brainstorm and ideate with cards, engage children in the programming and prototyping parts. For example, Know-Cards include input and output cards (3) and finally to conceptualise ideas of smart things. The game board embeds different reflection lenses that that represent speci�c electronic components for understanding and designing IoT devices [8]. There are also cards that children explore while designing, e.g., “does your idea make sense for the mission of making represent inputs andinteract people outputs forwithspeci�c natureprogramming elements?”.platforms, See Figure such2 as forMaker andlevel the first Scratch of cards [38, 40]. the board However, game, for these cards do not support exploring children thing cards in thetechnology and ideation part, and and cards they for are starting strictly related to a very the ideation of speci�c technology. smart things. Figure 2: Examples of cards (left), and the first level of the ideation tool of SNaP, which contextualises Fig. 5. Smart this thingdesign stageand components in aboard nature forenvironment familiar the environment relatedto to participating children a nature outdoors (right) park, part of the SNaP toolkit used in Summer 2019 Besides tools for ideating, the SNaP toolkit also supports the transition of ideas into interactive Therefore,prototypes of smart design toolkits things.children for engaging In the research across an reported in this entire design paper, process the come should digital version with of SNaP technology cards enables children to automatically generate a basic program to start from in the block-based or elements that are easy for children to match with technology for children, and yet su�ciently general to apply to Makecode programming environment (available at makecode.microbit.org/). See Figure 3. The 9 program is generated from the input and output cards which are part of children’s conceptualisation of a smart thing. Then children can continue exploring how to make their ideas evolve through programming and prototyping, besides reflecting in such stages, as well, on the technology under design. FabLearn Europe / MakeEd 2021, June 02–03, 2021, Virtual Conference First Author, et al. 261 programming environment of children’s choice, e.g., with visual blocks of Makecode. The following guideline is thus 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 Figure 3: The latest version of the SNaP toolkit, digital, which automatically generates programs for 277 children’s ideas in Fig.the Makecode design 3. An encompassing block-based programming process, centred around SNaP environment: see 278 279 snap.inf.unibz.it/play.php 280 derived, whereas guidelines for each single stage are discussed separately below. 281 282 4. ⌅Lessons Design withand Conclusions children should encompass diverse design stages, centred around a design toolkit: (1) exploration 283 and familiarisation with smart thing design; (2) ideation and conceptualisation of smart thing ideas; (3) 284 285 Along three years programming of research, and prototyping children of smart used the SNaP toolkit. Across all years, data were thing ideas. 286 uniformly gathered and classified, e.g., in relation to children's usages of SNaP. Data were 287 4.1.2lately processed Exploration with a contentThe and familiarisation. analysis and reflected design process suggestedover so as researchers, by several to extract guidelines foretsmart- such as Iversen al. [2017], 288 thing toolkits for children. The main lessons are extracted and elaborated in the remaining does not always involve a structured exploration stage. It is important, however, that the process is structured part so thatofchildren this paper. 289 become familiar with the components of the things under design before they move on to the next 290 291 design stages [49]. In smart thing design with SNaP, the exploration and familiarisation stage was structured by 292 4.1. Gameful Design the SNaP toolkit, as follows. The exploration stage employed SNaP cards for input and output devices, and it asked 293 children to play and match them with devices for programming and prototyping, e.g., via quizzes. Last but not least, 294 Games or game elements help motivate and guide children in the design of their smart things. the exploration stage o�ered children many sca�olding examples, to freely tinker with. For instance, in Summer 295 In the research reported in this paper, design was made gameful with the SNaP toolkit, offering 2020, children had SNaP boards with ideas of smart things and matching programs to tinker with in the Makecode 296 children a fun and not-intimidating way to explore the design process. Children, being in 297 programming environment general familiar with[30], e.g., they playing could board changeeasily games, the threshold graspedfor how the temperature input device to use SNaP’s boardsandand makeits the 298 smart thing react and mechanics, whenthey the environment was hot were motivated to instead try to of cold.the winning condition even in cases of no reach 299 300 former experience, e.g., to make the town park more attractive for their peers. ⌅ The exploration stage should make children familiar with smart things (i.e., what they are composed of) 301 and the design toolkit. Many sca�olding examples need to be o�ered, which children can tinker with. 302 The design process should be structured as a game or with game elements so as to motivate 303 and guide children in design. 4.1.3 Ideation and conceptualisation. During ideation and conceptualisation of smart thing ideas, divergent and 304 convergent thinking enable children to open up their design process, consider new perspectives and subsequently 305 4.2. Story-line discard aspects during their attempt to reach a design solution [23]. Ideation and conceptualisation should guide children 306 307 accordingly, and tangibly so. For instance, in the research reported in this paper, that was supported by the SNaP toolkit, 308 In games, story-lines or narratives, when present, help motivate players towards the game goal, and especially its game boards for �rstly ideating as many ideas as possible (divergent thinking) and then re�ecting 309 and immerse them into the game. In the same vein, briefings were used in design with children 310 on an to idea to conceptualise communicate (convergent the design goal,thinking). e.g., [7].Tangible outcomes were In SNaP–framed children’s design, boards,goal the design one per waschild, sharedwhich 311 conceptualised an idea ofofa smart via the story-line SNaP,thing to carry whereas on in thecards mission last design stage.turned the goal into objectives for of SNaP 312 smart things. For instance, in Summer 2019, the story-line started as follows: “we need to help 6 the Mayor of our town to design a new nature park with smart things for your peers” [4]. A mission card for a smart thing by a child was thus “make people interact in the park”. The story-line or narrative should motivate children and make the design goal tangible (e.g., through missions for players to accomplish). 4.3. Mechanics and Aesthetics In SNaP-framed design, the mechanics and aesthetics of the toolkit helped children navigate through the design process and contextualise design in the chosen environment, e.g., the Talvera park of Bolzano represented in Figure 2, right side. The closed rule-bound nature of games stimulates an awareness of structure and function, and it can transform spontaneous decisions into more formal understanding. Children, when playing for the second time with SNaP, were in fact able to remember the game rules and levels which helped them proceed with design with no help from adults, indicating their understanding of the process. The game mechanics and aesthetics should be designed so as to offer a familiar rule-bound structure, which smoothly guides children along design and helps them become aware of its stages in relation to the given environment. 4.4. Player Roles In SNaP-framed design, roles for players were partly bound by the game mechanics, which helped make children own their ideas, clarify adults' roles in design, and reflect with others with specific roles. Without clearly specified roles, adults can greatly influence children's design activities even without meaning it. The mechanics of the toolkits for supporting design could thus be used to specify or negotiate the roles of all participants so as to embed them clearly for all into the design process, as in the case of SNaP-framed design. Adults' roles, in particular, should be defined according to children's varying requirements and benefits. For instance, in in-presence workshops with SNaP, adults' role was part of the game rules and guided seemingly by these, so that scaffolding was gradually decreasing as per children's learning of design, one of the expected benefits of their participation in design. Their learning was assessed by triangulating and processing different data, e.g., learning questionnaire data and the evolution of children’s smart things over time [4]. Player roles should be defined in relation to design roles, so as to clarify responsibilities and tasks in design, e.g., playing the expert of a design heuristics. In particular, the role of adults should be adapted to the requirements of children and their expected benefits in design, e.g., engagement and learning in smart-thing design. 4.5. Reflections for and by Children Toolkits should have lenses that help children critically reflect on specific aspects of the things under design, in relation to the environment and technology-related risks. For instance, SNaP has different reflection lenses, e.g., related to the consistency of the smart thing under design for the chosen mission or safety-related risks for the environment or people for which/whom the smart thing is designed. Lenses should come with questions and probes that help children reflect critically and elaborate on their solutions. Moreover, reflection lenses should adaptable and embed children’s own reflections in design. Tangible and adequate reflection lenses, with questions and probes, should be embedded in playful toolkits for children, in relation to “things” of the environment and risks technology can bring therein, besides able to capture children’s own reflections in design. Acknowledgements Authors of this paper acknowledges the contribution of all participant children across three years of work, besides their parents or teachers. Financial support was granted through the SNaP, GEKI and EMPATHY projects. References [1] Adelman, C. (1993). Kurt Lewin and the Origins of Action Research. Educational Action Research 1, 1 (1993), 7–24. https:// doi/abs/10.1080/0965079930010102 [2] Deterding, S. (2015). The lens of intrinsic skill atoms: A method for gameful design, Human- Computer Interaction 30 294–335. [3] Gennari, R., Matera, M., Melonio, A., and Roumelioti, E. (2019). SNaP 2: The Evolution of a Board Game for Smart Nature Environments. In Extended Abstracts of the Annual Symposium on Computer-Human Interaction in Play Companion Extended Abstracts (CHI PLAY '19 Extended Abstracts). Association for Computing Machinery, New York, NY, USA, 405–411. https://doi.org/10.1145/3341215.3356281 [4] Gennari, R., Matera, M., Melonio, A., Rizvi, M., and Roumelioti, E. (2020). Reflection and Awareness in the Design Process Children Ideating, Programming and Prototyping Smart Objects. Multimedia Tools and Applications (2020). https://doi.org/10.1007/s11042- 020- 09927- x [5] Mora, S., Gianni, F., and Divitini, M. (2017). Tiles: A card-based ideation toolkit for the internet of things. In Proceedings of the 2017 Conference on Designing Interactive Systems, DIS ’17 (pp. 587–598). New York, NY, USA: ACM. [6] Roumelioti, E., Gennari, R., Matera, M., Melonio, A., and Rizvi, M. (2020). Towards Making Children Independent in Design. In Companion Publication of the 2020 ACM Designing Interactive Systems Conference (DIS' 20 Companion). Association for Computing Machinery, New York, NY, USA, 227–232. https://doi.org/10.1145/3393914.3395849 [7] Smith, R.C., Iversen, O.S., and Hjorth, M. (2015). Design Thinking for Digital Fabrication in Education. Int. J. Child-Comp. Interact. 5, C, 20–28. https://doi.org/10.1016/j.ijcci.2015.10.002