TY - JOUR
AU - Park, Young-Woo
AB - Abstract Adopting a re-materialization approach, we designed and implemented an everyday interactive artifact that enables an individual to monitor and establish the reconfirmation time of email data. This device represents a new means of handling and interacting with email. To investigate the value of the materialization of email data through a daily interactive object, Maili, we conducted a 1-month field study with five participants in their work environments. The results showed that applying physicality to email handling helped to increase accessibility to and interest in the data, as well as in the reconfirmation function of email. Results also indicated the value of combining non-digital (the tray) with digital functions. By presenting the process of using the dematerialized data, our findings offer new insights into how we can materialize digital information in everyday tangible artifacts. RESEARCH HIGHLIGHTS Design and implementation of a device called ‘Maili’ that can manage emails with physical interactions Understanding design process toward email data materialization Findings from a month-long in-field study of the device in an office environment Considerations for designing future everyday artifacts materializing digital data 1. INTRODUCTION Since the launch of the first Apple iPhone in 2007, multi-touch devices, including smartphones, tablets and others (Vitale et al., 2019), have dominated the way we interact with personal data (Van Campenhout et al., 2016). This change has been manifested in several types of intangible digital applications (Redström, 2001). These digital mediums have the advantage of accessibility across various environments and the accumulation of data in a device or cloud service for a long time (Odom et al., 2012). On the other hand, these same characteristics have resulted in issues such as continuous stacking of personal digitized data in storage (Borghouts et al., 2017) and causing difficulties in data management (Vertesi et al., 2016). These challenges have resulted in complications for the user in the use of digital data, mainly due to the formless and placeless character of digital content (Odom et al., 2013). This dematerialization has led to a loss of focus on the advantages of persistent and static characteristics that comprise physical world objects (Van Campenhout et al., 2013). One example is the reminder feature of Whittaker and Hirschberg’s (2001) work on paper archives. Their approach simply provides a clear physical cue of the archive’s availability, enhancing awareness for users in terms of their possession of required information. In this respect, researchers have started to explore the possibility of tangible and materialized data. Studies have been carried out in the human–computer interaction (HCI) community to convey personal digital information through a tangible approach: the utilization of random prints from the users’ digital photo collections (Odom et al., 2014), providing productive interaction for controlling music (Van Campenhout et al., 2013) and accessing users’ accumulated music data through physical interaction using metadata as a design material (Odom and Duel, 2018). The increasing proliferation of design research in the HCI community, aimed at reviving the physicality of digitized devices, has led to works exploring how the representation of digitally stored data could more naturally permeate into everyday artifacts. These existing studies provide examples of clarifying the roles of digital objects and adapting the devices to identify role changes, which then allows for separation of tasks and supporting user-centric activities (Dearman and Pierce, 2008). In line with this body of work, we identified a design space to investigate the materialization of email data. In this, we aimed to explore how a materialized approach might affect users’ engagement with email information in their everyday lives, focusing primarily on the office environment. To ground our thinking, we designed and developed an interactive device with digital email information through a re-materialization approach. The design case (illustrated in Fig. 1) was developed based on three guidelines from an existing dematerialized research framework (Van Campenhout et al., 2013). In using the design case, we attempted to understand the relationship between the physical product and the user from the perspective of digital function, interaction and product form. Using the Maili design case, through a concept-driven research approach (Stolterman and Wiberg, 2010), we conducted an exploratory investigation of product use. Our design was inspired by the understandings from dematerialization framework theory and and we explored what value the Maili design may provide to users by deploying it in a real use situation, in particular—an office environment. We selected this context because email usage in work environments is high (Bälter, 1998), with resulting user stress associated with email usage, management (Neustaedter et al., 2005; Grevet et al., 2014; Mark et al., 2016) and notifications (Pielot et al., 2014). FIGURE 1 Open in new tabDownload slide Design of Maili exploring email data materialization. FIGURE 1 Open in new tabDownload slide Design of Maili exploring email data materialization. Maili, as an everyday tangible artifact, enables users to check and set reconfirmation times for email. Time reconfirmation is identified in the current work as the process of assigning a time/date for an important email to be re-presented back to the user through a physical change in the Maili product. The process of time reconfirmation is as follows. First, the user sees the number of new emails and checks each email’s title and contents through Maili. Second, the device allows users to register reconfirmation time through the dial; for example, emails that users consider important can be registered in the Maili device by setting the date and time to reconfirm the email. When that time comes, a node on the front of the device pops out to inform the user that there is an email that requires reconfirmation. We have selected ‘email’ as the medium on which to impose physicality for several reasons. First, email as a platform for communication now plays an essential role in contemporary, everyday communication (Jovicic, 2000). Email is used as an essential tool for various purposes, from personal chatting and time administration (Mackay, 1988) to task management (Bellotti et al., 2003). Second, the number of email users and accounts is increasing. Email data now accounts for the most significant proportion of personal digital data for office workers (The Radicati Group, Inc., 2015), who can now check and respond to email messages anywhere and at any time (Bettina, 2018). Third, although there are significant benefits to the digitalization of communication, the increased frequency and the volume of email has resulted in issues around the management of email data (Whittaker et al., 2002). Stress related to email management is an example (Whittaker and Sidner, 1996); in particular, Mark et al. (2016) cited implications of new design interventions for email organization, which can support the receipt of timely information and adjust the tone of the received email. These issues pose new design challenges for the HCI community. We deployed Maili in the offices of five users over a month-long period to explore the impact of the Maili design case—as an example of email materialization with physical form and interaction—as means of improving the efficiency of email handling. We expected that Maili, which developed from our design process, could support the user’s email confirmation and may help the process of email checking and management from its design, function and interaction. Our research contributions can be described as follows: (i) the design of the tangible object, which enables email confirmation and reconfirmation through physical interaction; (ii) reflections from iterative design and prototyping process toward email data materialization; and (iii) results of in-field study regarding how materialization of email data through an everyday object affects user experience in work environment. 2. RELATED WORK 2.1. The Materialization of Digital Data Dematerialization happens when existing products lose their original form and are digitized (e.g. listening to music, paying with the touchscreen of a smartphone) (Belk, 2013). These features are available through personal devices and can bring the benefits of digital functionality, such as dynamic flexibility and information storage. Nevertheless, due to dematerialization, interactions between user and product are also restricted to narrow and screen-based touch interactions. This intangible interaction is in contrast to materialized product interactions through multi-mode sensations (i.e. visual, auditory and haptic) (Van Campenhout et al., 2019; Redström, 2001). 2.1.1. Frameworks for digital data physicalization and materialization The re-application of physicality to digital data has been studied from various aspects, including the form element of products (Vallgårda, 2014) and the quality of the prototype (Odom et al., 2016). For example, Vallgårda (2014) proposed that the form element of the interaction design combines three things: the physical form, including color, and size; the temporal form that displays the change of state of a device; and the interaction gestalt that indicates the movement according to the user’s object or environment. Additionally, Frens (2006) describes function, interaction and form as three elements of interactive products. At the same time, while traditional products are adequately linked with those elements, contemporary interactive products are becoming more diverse in their forms, while interaction is becoming standardized. More specifically, expressing data through physicalization have been explored to indicate its value for everyday user interactions. For example, Jansen et al. (2015) defined data physicalization as ‘a physical artifact whose geometry or material properties encode data’. The physical representation of such data and their interaction have various advantages. Physical representation can make interactions with data more accessible (Houben et al., 2016) and easier to grasp (Huron et al., 2017), help to stimulate perceptual exploration skills (Alexander et al., 2015) and promote a playful and engaging experience (Gallacher et al., 2015). Furthermore, the data materialization method pursued in this study aims to not only make data physical but also focus on a single function in a digital application as an independent everyday artifact that provides separation of critical tasks affecting management of personal data (e.g. emails). In this respect, studies on the interaction of dematerialized digital data have emerged, based on the tangible user interface (TUI) framework that utilizes digital functions alongside tangible interactions (Ishii and Ullmer, 1997). Those studies aim to achieve the advantages of both physical interaction (Frens, 2006) and direct manipulations of the digital (Spindler et al., 2010), thereby adding value at the border between the digital and tangible (Van Campenhout et al., 2013). As users live and act in the physical world (van Dijk et al., 2013; Hornecker, 2011), how digital information is linked to a physical direction is positioned as a more intuitive way of interacting with digital products (Ishii and Ullmer, 1997). In order to merge the flexibility of digital data with the richness of tangibility, research on data materialization has proposed three design requirements (Van Campenhout et al., 2013, 2016, 2019): 1) design seeks a balanced harmony of form, function and action based on the rich interaction framework (Frens, 2006); 2) design focuses on a single function-specific operation that is linked to the following interaction; 3) design considers a collection of dematerialized information as an essential part of its physical structure. The framework emphasizes the merits of digital functions while simultaneously presenting benefits related to interactions through the physical form of a device. A few related studies were identified that explore the opportunities for materialized devices through an iterative design process using the research through design approach (Gaver, 2012; Frens, 2006; Zimmerman et al., 2007) based on the three design requirements. One recent study developed prototypes to provide material components to a dematerialized audio system. This applied digital functionality into diverse physical territories called modules (Van Campenhout et al., 2013). In another case, a physical payment device called experimental payment terminal (EPT) was developed. A user study was conducted to identify the value from a laboratory investigation with comparing this device, EPT, with an existing commercialized payment terminal (Van Campenhout et al., 2019). Based on this existing literature, we found a design space to develop an independent device while pursuing a combination of digital functions, physical interactions and form. To that end, we have further explored research aimed at developing everyday design objects, utilizing various characteristics of digital information. These are illustrated in the following section. 2.1.2. Representation of digital data in everyday artifacts HCI and design communities have long been exploring how to develop digital information to independent everyday artifacts (Gaver and Höök, 2017; Nam and Kim, 2011; Gaver, 2012), applying physicality with embedded interaction (Redström, 2001; Vallgårda, 2014) and utilizing the characteristics of digital information. For instance, by utilizing personal data accumulated slowly in a user’s life to physical devices (i.e. digital photographs, Chen et al., 2019; Odom et al., 2014; music, Odom et al., 2019), research has suggested a feasible way to match users with their data. Besides, studies have explored practical applications aimed at enhancing the convenience of accessing digital information by transmitting traffic data as well as the user’s schedule data (Kim et al., 2018), printing information on the paper beyond presenting schedules on display (Jang et al., 2019) and storing family photo data through tangible printing (Bennett et al., 2012). Also, by utilizing public geographic data, a research product conveyed local event information around the home by LED and movement of acrylic columns (Ylirisku et al., 2016). Other research has explored the application of hedonic value with tangible interaction (Alonso and Keyson, 2005), bringing the advantage of pleasure in controlling data physically beyond a practical purpose (Hassenzahl, 2018). By combining the playful elements through tangible interaction, other research was conducted aimed at understanding how people can participate in surveys in public places through a tangible approach (Golsteijn et al., 2015). Most of these existing studies have been conducted in the laboratory (Bennett et al., 2012; Alonso and Keyson, 2005) or field studies through installations in public (Quintal et al., 2018) or private spaces, ranging from 1 week (Jang et al., 2019) to several months (Odom et al., 2014). These example works appear to offer new directions in data control, transmission and storage to increase functional usability and enjoyment of use through everyday physical devices. 2.2. Tangible and intangible interaction with email Since email is used as an essential means of communication in modernity (Jovicic, 2000), earlier studies of email were conducted to improve our understanding of email usage. For example, research was conducted to measure frequencies of email inbox use over time through a large-scale quantitative study (Alrashed et al., 2018). To manage an email, search and threading methods may increase the efficiency of use to a greater degree than the conventional approach of folder creation (Whittaker et al., 2011). Other studies explored how individual businesspeople utilize email in the working context (Bälter, 1998; Quaresma et al., 2013). Building upon these works, further studies have identified challenges in email usage, such as email triage (Neustaedter et al., 2005) and email overload (Whittaker and Sidner, 1996; Grevet et al., 2014), related to an increasing number and frequency of emails and the challenge in their management. For example, research illustrated how a smartphone’s email or message alarm led to negative emotional response (Pielot et al., 2014). One study proposed a method to improve work management efficiency by re-establishing and showing work-oriented information based on email information (Bellotti et al., 2003). Furthermore, several methods have been proposed to deal with the above issues. For instance, one study combined mobile text messages to alert users of critical emails (Rector and Hailpern, 2014). Alternatively, another focused on the meeting schedule in users’ emails, developing a system that proactively displays schedules (Zhao et al., 2018). Other studies have explored utilizing email data with physical interaction through printed emails for people who cannot access computer email properly (Cannata, 2006) or comparison of the user’s response when checking email through printed text compared to the computer (Hill and Monk, 2000). Another proposed a voice interface with a traditional telephone to assist the elderly when sending emails through voice recording (Brewer et al., 2016). Regarding tangible interaction with voice messages, one of the earliest studies expressed voicemail information in a token format. To achieve the expression of information related to voice messaging, a rolling ball was used as an indicator of a voicemail received. In a rich interaction approach, the ball could be re-located back into a hole when the user wanted to hear the voicemail (Bishop, 1992). These studies appeared to illustrate how tangibility in email function supported improved usage through a playful interactive device and further showed potential in reducing stress (Gaunt et al., 2013). Through this literature review, we identified the scope, importance and challenges of contemporary email use, which were then considered in the design and development of our design case. Collectively, the strands of research have contributed to understanding how tangible forms can be designed based on dematerialized information, and how ongoing HCI research has begun to deal with the application of digital data in tangible, everyday objects. While this research is promising, less is understood about how to physically convey email and email-related functions that utilize email data as part of a design method and any resulting benefit of a materialized approach for email users. There is growing interest in representing digital information with tangible artifacts. Thus, we see opportunities to investigate how dematerialized email data can be designed with physical form and interaction and what kinds of value can be delivered with an everyday object to users. 3. MAILI DESIGN AND IMPLEMENTATION Based on the concept of dematerialization (Van Campenhout et al., 2013, 2016, 2019), we first explored various ideas to show email while considering the specific target context as the work environment. During the process of exploration, we identified users’ stress associated with email management and many inbox visits (Whittaker and Sidner, 1996; Grevet et al., 2014). For this issue, we targeted the possibility of sorting important emails (Neustaedter et al., 2005) and considered a design to register and inform users of important emails through a tangible product. Within the existing function of reconfirmation of emails, we noticed that users had five actions when visiting an inbox: (i) open, (ii) delete, (iii) reply, (iv) organize and (v) download attachment (Alrashed et al., 2018). From this, we explored features that may help users open and organize emails. As a result, we devised three ways to deliver emails to users (Fig. 2a): (i) print important emails, allowing users to carry relevant emails or attach them to walls to be visible within their working environment; (ii) delivering email tokens for notification, that acted as triggers to remember the information contained within relevant emails, with users able to check emails by putting the token back into the device; and, lastly, (iii) protruding nodes from the front or side of the device to notify the arrival of relevant emails. FIGURE 2 Open in new tabDownload slide Design process of Maili. FIGURE 2 Open in new tabDownload slide Design process of Maili. Of those design ideas, we selected the protruding node type because we considered this approach would bring an exciting notification with a change of product appearance, while at the same time providing an intuitive trigger. Existing research on dematerialization suggested mode-relevant action possibilities (Van Campenhout et al., 2013) and the opportunity for inducing user interaction by changing the product appearance (Rasmussen et al., 2012). This means of expressing data through actuation is summarized by Jansen et al. (2015) as the notion of actuated physicalization. We applied this minimally to an everyday object that shows physical states based on the email data. In addition, we combined the protrusion of nodes with sensing for synthetic interaction and designed node transformation (as indicator of email data) as an angle change, rather than a vertical height change to provide the physical affordance of pushing the node back to its original state. While this existing research is focused on shape change following the users’ input, our design (i.e. Maili) utilized this approach to notify users and induce interaction with emails. We also considered it could be a pertinent way to express digital information (e.g. important emails) physically (Jansen et al., 2015). Based on this, the product design was progressed with the consideration of internal structure design, usability and aesthetics for an office environment. The process started from sketching to modeling, along with diversifying colors and shapes (Fig. 2b). From this, we developed our first design into a working prototype (Fig. 2c). Detailed considerations in the design process are included in the following sections. 3.1. First design trial We placed a dial controller at the front where it was expected to be the most comfortable placement for two interactions, rotating and pressing to set the reconfirmation time of emails that users consider to be important. The dial was designed to be slightly inward from the front side of the product, preventing the finger from slipping out when rotating. Also, we attempted to hide the display when the device is not in use, allowing users to push the top side of the display bezel to check the contents of emails. The aim was to provide an impression of not bringing an additional display in the office environment (Fig. 3a). FIGURE 3 Open in new tabDownload slide Interaction in the first prototype. FIGURE 3 Open in new tabDownload slide Interaction in the first prototype. We fabricated the display bezel and dial using steel and black finishing, which provided a different feel from the body of the device. This made it easier to distinguish these two parts as interaction points. The display bezel was designed to simulate a handle form, protruding slightly upwards to allow the user to press and eject the display conveniently (Fig. 2c). We made the color of the nodes to be distinguished with the body of the device to make users to easily notice when the node protrudes based on the settled reconfirmation time of the emails. 3.2. Preliminary evaluation After the initial design and implementation, we conducted an ideation focus group with five additional researchers in design or engineering, to consider overall form, usability, function and roles in the working environment in order to make modifications of the design. The issues were analyzed in terms of use situation and theory according to the existing literature (Stolterman and Wiberg, 2010). The reasons to conduct this preliminary evaluation were as follows: (i) to test its usability in advance with people who were among the first to use Maili, (ii) to determine whether the Maili’s email reconfirmation and notification functions could be appropriately delivered through nodes and (iii) to identify any problems that may occur in advance in a limited environment before user study. The use of researchers for this preliminary phase was chosen as a more pragmatic means to quickly trial the initial Maili design with more experienced designers. In this way, we hoped to obtain more focused, expert opinion on both the functionality of the design and how it may be improved. 3.2.1. Perspective from use situation It was found the vertical display angle made it difficult for users to see the email content. Besides, some researchers suggested that the way of pushing the display bezel to see email content could be cumbersome to check emails as it required additional interaction. To address this, we applied a static and smaller (1.5 inch) monochrome OLED display. We revised the angle of the display to be laid down for users to view email contents while sitting at a desk easily. Also, from the initial design trial, there were opinions that the dial parallel to the front side of the product is difficult to rotate (Figs 2c and 3b) and did not correctly reflect the user’s ordinal routine of controlling digital information. To address this challenge, we analyzed the usage pattern in the existing time-setting application and physical rotating device. Thus, it was modified to the vertically rotatable design. 3.2.2. Perspective from theory From existing dematerialization studies (Van Campenhout et al., 2013, 2016, 2019), we identified products composed of modules, with each module designed to have a specific function. In this respect, the first design trial was not well modularized between dial and nodes components. For this, we designed a separate location for checking email and added placing nodes for the physical interaction (Fig. 4). FIGURE 4 Open in new tabDownload slide Three divided modules of Maili. FIGURE 4 Open in new tabDownload slide Three divided modules of Maili. 3.3. Final design of Maili 3.3.1. Design to fit the office environment After going through the initial prototype development (Fig. 2c), we explored how the design might better fit into an office environment. As a result, a tray-type feature (Fig. 2e) was incorporated into the design (Sheldon, 2017; Studio PESI, 2017), considering the tray could act as storage for stationery (David Allen Company, 2016) along with providing a reason for the user to situate Maili on the desks. In particular, to improve the efficiency of work, the tray needed to be located within reach of the user (David Allen Company, 2016). It was expected that the position of the device would naturally be closer to the user’s hand. For this, we went through four different tray designs (Fig. 2d), exploring the height and width of the main body, including nodes, dials and displays. Through this approach to design, we attempted to increase the value of the digital product by combining non-digital with digital function to set the reconfirmation time of the important emails. Additionally, 15 different shade samples were prepared to determine the color. Throughout this design process, we attempted to improve the overall aesthetics and accessibility of Maili, believing this approach may improve usability (Frens, 2006). 3.3.2. Three tangible modules The dematerialized data were assembled into tangible entities; therefore, digital information needed to be considered an essential element in the physical structure of the device (Van Campenhout et al., 2016). As such, we designed Maili with two tangible digital modules for digital function and a third module for an analog function (Fig. 4). The first digital module was designed to interact with users through appearance changing, notifying the registered email to the user by changing of node angle (Fig. 4, Module 1). A second digital module consisted of a display and a rotating dial for checking email content (Fig. 4, Module 2). Providing digital functions in each visible component within Maili’s design, digital information was shown as it flowed from one module to the next. Lastly, the analog function, a physical tray, was situated as a third module (Fig. 3, Module 3). 3.3.3. Protruding nodes from flat surface and rotating interaction Inspired by the form of a mouse wheel used when scrolling email, we implemented a rotating interaction for inbox scrolling (Fig. 5a). In this way, we aimed to simulate similarity existing interaction. For example, when looking at the time-setting interaction in an existing smartphone alarm application, the interaction that rotates the visual dial-up and down is mainly adapted. For this reason, we decided that vertical interactions would be appropriate for physically representing the scrolling and time setting of email and naturally encouraging users to turn and press the dial (Kim and Eune, 2015). FIGURE 5 Open in new tabDownload slide Interaction of Maili. FIGURE 5 Open in new tabDownload slide Interaction of Maili. Maili always displays information instantly. As a result of this, it was expected to shorten the process of turning on a smartphone or a computer’s internet browser, making it directly recognizable. Maili presents the title of each email one-by-one to prevent the user from passing over an important email. Through a rotating interaction, users can register time to reconfirm emails deemed necessary. When the registered time is reached, the product’s appearance changes through the protruding nodes from the flat surface to indicate an important email. From the fixed exterior form, physical changes of the artifact, such as a protruding node in the Maili, serve to attract the user’s attention and induce new action (Frens, 2006). We also wished to enhance usability by indicating notifications through an appearance change as nodes protrude from the flat surface, leading to the interaction, in which participants naturally push nodes back to the original position (Fig. 5b). The movement of the node in this study is applied as a function of actuation with (i) the characteristic of change in spatial positions of objects (Poupyrev et al., 2007) by deforming the angle of a node and (ii) the characteristic of change in force applied to the user (Poupyrev et al., 2007) by holding the position of the node while the node protrudes (Fig. 5b). In other words, the user is required to touch the node, feel the material and provide more force than a certain pressure on the node to return it to its original position. These kinds of design interactions are referred to as familiarity and usability in rich interaction (Frens, 2006). This concept suggests that the general user is familiar with action-based interaction in a physical environment. Therefore, physical interaction does not have to be generalized; instead, it can be identified according to the situation. In this study, the design provides a reconfirmation of email notifications through changing a physical state by a protruding node. It is expected to provide a more familiar notification by pressing the node back to its original state. On the functional dimension, the node interaction was also expected to remind users of the multimodal sensation of touching. Additionally, the protruding node helps to reduce the additional work such as revisiting the email inbox. This function is expected to decrease email overload (Whittaker and Sidner, 1996), miss important mail (Jovicic, 2000; Rector and Hailpern, 2014) or abandon the organization (Bentley et al., 2017) through tangible interaction. In the previous dematerialization study, Lukas et al. recommended developing a product that focuses on the main task through a combination of physical and digital aspects (Van Campenhout et al., 2013). For this, the primary function of Maili is to check email through physical interaction with the tangible product. Additionally, in other research based on the dematerialized information, the authors showed how the combination of sub-dematerialized function was related to the primary function, thereby improving usability (Kim et al., 2018). From this, as a sub-function, we added to register time to reconfirm important emails. We expected to increase the functional value of Maili by complementing the auxiliary function with the primary function. 4. INTERACTION WITH MAILI Figure 6 shows the function of checking email and registration of a reconfirmation time, and Figure 7 demonstrates the screen design of Maili during the whole operating process. Figure 8 indicates how the user can reconfirm registered email with a pushing node. FIGURE 6 Open in new tabDownload slide Checking the content of email and registering reconfirmation time for email. FIGURE 6 Open in new tabDownload slide Checking the content of email and registering reconfirmation time for email. FIGURE 7 Open in new tabDownload slide Screen design of Maili: (a) loading email, (b) new emails or no new email, (c) inbox with mail title and sender, (d) body of the email, (e) time set-up for reconfirmation and (f) showing the title and body of the registered email. FIGURE 7 Open in new tabDownload slide Screen design of Maili: (a) loading email, (b) new emails or no new email, (c) inbox with mail title and sender, (d) body of the email, (e) time set-up for reconfirmation and (f) showing the title and body of the registered email. FIGURE 8 Open in new tabDownload slide Check registered email through pushing nodes. FIGURE 8 Open in new tabDownload slide Check registered email through pushing nodes. Email read mode: when a new email arrives, the phrase ‘New Mail’ and the email’s identification number is presented on the display (Fig. 7b). If there is no new mail, ‘No New Mail’ appears on display (Fig. 7b). Every 30 minutes, a new email from Gmail is read. This time duration was selected based on our research team’s several trials with different time variations, to make Maili less disruptive in showing the new emails in real time or every 2–3 minutes. Moreover, we expected that urgent email confirmation could be an issue for the users if the duration was longer. In Fig. 6a, users can scroll and see the content of the title and sender in the inbox screen by rotating the dial. Detailed design of inbox with title and sender is like Fig. 7c. Maili presents the contents of a specific email by pushing the dial (Fig. 6b). The body of the email is shown like Fig. 7d, and the content is scrolled up automatically without the need to rotate the dial. Maili only receives the title and contents of emails delivered to users’ Gmail accounts. The current means of checking email through Gmail’s function (e.g. mark as an important email, snoozing or starring) are not represented in the Maili device. This was because we wanted to focus on the major aspect of ‘receive new email, setting reconfirmation time and notification through node pop-up’. Register reconfirmation time: in Fig. 6c, for important emails, the user presses the dial to open the email reconfirmation time set-up window (Fig. 7e). Users can set up the day and hour to reconfirm. The user can set the day first, then push the dial one more to move on to the hour setting (Fig. 7e). By turning the dial-up and down, it is possible to adjust the date and time. If the user appoints the reconfirmation time, the display returns to the inbox screen (Fig. 7c). Registered email notification and confirmation: Maili delivers registered emails to the user by protruding nodes, and the user can check these emails by pushing the node back. If there is an email registered to reconfirm at a specific time, nodes will protrude corresponding to the number of emails received (Fig. 8a). The user can press the node back (Fig. 8b) into its original position for checking the registered email, and the contents of the registered email will be revealed on display (Fig. 8c). The user can re-register the email reconfirmation time or delete a registered email by pushing the dial (Fig. 8d). There are three nodes in Maili. When there are more than three registered emails, the user pushes the node to check the mail, and the nodes protrude again, depending on the number of remained registered mails. 5. IMPLEMENTATION While the external design was a consideration, the top priority was to develop the software to take the user’s email data and register reconfirmation times according to the user’s input. Python code runs in the Raspberry Pi2, and physical elements, including servo motors and sensors, are operated by Arduino. These two computing elements are connected via USB and employ serial communication. In this relationship, the Raspberry Pi2 focuses only on managing emails, and the Arduino only operates the sensors and actuators, thereby reducing the functional burden. Using the Python script in the Raspberry Pi2, it retrieves 100 recent emails with the Gmail API. These email groups are converted to a CSV file format. This device is an independent device that merely retrieves emails from the user’s account without affecting a user’s existing Gmail account. When a new email arrives, the Python script sends a signal to the Arduino to indicate how many new emails have come in. The OLED display (Fig. 9) displays that information. Because the OLED display does not support any different characters besides English, all of the email content that will be presented on display is converted into image files. When the user registers the reconfirmation time for email, the content and time information of the emails are made into a list in Python. Information about reconfirmation is stored as a text file so that the data can be maintained even if the power is off because of a malfunction. When the reconfirmation time is reached, Maili activates the servo motors, following the number of registered reconfirmation emails. The movement of the node is realized by using the servo motor (Fig. 9). The servo motor was applied because it can actualize push and pull functions with simple rotating. For bringing constant torque to the node, a magnet is attached to the bottom of the node to fix the angle of the node when it protrudes forward. Infrared sensors are installed on the back of the nodes, and it can recognize when the nodes come closer. The rotary encoder is applied to perceive the user’s spinning input, and the micro push button is right behind the rotating button to observe push interaction. FIGURE 9 Open in new tabDownload slide A detailed structure of Maili. FIGURE 9 Open in new tabDownload slide A detailed structure of Maili. 6. FIELD STUDY We conducted a series of field studies (Brown et al., 2011; Rogers and Marshall., 2017) with Maili to explore the experiential value of the functional and emotional meaning of the design, as well as the interaction and practical aspects obtained from the materialization of email data through a physical device. The basic objective was to see how participants responded, changed and harmonized Maili in their day-to-day routine and within their normal work context. We aimed to understand (i) how the materialized email applies to an everyday tangible device and affects attitudes toward email usage and (ii) how Maili influences controlling and checking email through physical interaction when utilizing this device in the office environment. We hope that the evidence above may provide some insight regarding how the Maili intervention influenced the product experience in terms of email management and how our rematerialized approach implicated the experience of utilizing digital data. 6.1. Participants A total of five subjects participated in a study using Maili for 1 month (30 days). We aimed to explore how the Maili design might influence user habits from tangible email usage over 1 month. Selection criteria for participants aimed to identify subjects working within an office, who regularly communicate (e.g. more than a few times a day) through email with colleagues, supervisors and other companies. Specifically, we considered the following two additional points other than the amount of their email usage. First, we checked the available space on their desk. Three participants’ desks did not have much room (P2, P4, P5), and others had much free space (P1, P3). Here, the free space was determined based on whether there was enough space to install Maili, or whether it was necessary to re-organize the desk, workspace or move existing items to install the device. At the same time, there was a concern that the product size could be too large. However, when we looked at the usage patterns of three participants who had complex desk circumstances in the field study, we could see that they placed the email-related things (e.g. post-its and business cards) and got used to the tray part of Maili. Second, we asked how applicants managed their email and could hear that one did not use a specific technique to manage emails (P1), two placed tags on their emails (P2, P4) and two organized their emails by frequent deletion of unnecessary emails (P3, P5). Maili used Gmail API only; therefore, we gathered participants who used Gmail primarily. If users utilized other email accounts at the same time, we asked them to forward mail from these other accounts to Gmail during the test period. Two participants (P2, P5) needed to forward emails; thus, we received permission from a person in charge of the participant’s company or organization. The device was installed on the participant’s desktop only. Also, employees belonging to the same company were not required to avoid influencing each other’s operation and feedback. Participants were recruited through online advertisements. A total of five subjects participated in the study. P1 (female, mid-20s): an employee at an enterprise for product development and trading. P1 mostly contacts outside manufacturers about content related to transaction orders and the company CEO about the progress of the project through email. P2 (male, mid-20s): works as a product design intern in a start-up. P2 communicates with the CEO, the supervisor in the company for delivering the product development process and also with other organizations related to the project funding issue by email. P3 (male, late-20s): is the CEO of a start-up. P3 communicates with start-up employees and start-up funding staff by email. P3 usually sends and receives emails about applicable funding projects and the situation of the current funding with funding staff and shares product development information with team members. P4 (male, mid-30s): is a senior researcher in a national technology promotion agency. P4 communicates with the research team and other institutions through email. P5 (male, late-20s): works as a researcher in a design lab of a national university. He communicates about the progress of project and paper writing through email with researchers in other countries. 6.2. Data collection In order to collect qualitative data during the user survey, 25–35-minute interviews were conducted once a week with each participant over 4 weeks during the user study period. A total of 553 minutes of interview data was gathered. We also gathered usage log data to see how many emails were registered to reconfirm through the dial of the Maili. More specifically, we collected the number of days and hours changed when the users were setting up the reconfirmation time. Log data were automatically saved to the Raspberry Pie when the user registered the reconfirmation time. The interview was conducted in Korean in a semi-structured manner. The overall questions were followed as illustrated in Table 1. However, queries were immediately made regarding additional questions or changes in attitudes compared to the previous week’s interview. We recorded the entire interview; its contents were promptly transcribed to confirm any missing parts. Following this, we reviewed the contents and added additional questions that we wished to pose in the next week’s interview. TABLE 1. Interview contents following the date. Interview week . Interview contents . Week 1 The perception from checking email through a tangible device with rotating interaction The perception from checking registered email with angle changing node and pushing interaction Functional and emotional feedback from utilizing tray combined with an email device Criteria for choosing email for reconfirmation through Maili The difference in the email checking experience while using Maili compared to other digital devices: computer or smartphone Week 2 Change in perception related to utilizing Maili compared with Week 1 Week 3 Change in reaction related to email checking experience and using the reconfirmation function of Maili compared with Week 2 Week 4 The overall reaction to the design and interaction of Maili Overall opinions on the functional aspect of Maili and change of interest toward the device during 4 weeks Interview week . Interview contents . Week 1 The perception from checking email through a tangible device with rotating interaction The perception from checking registered email with angle changing node and pushing interaction Functional and emotional feedback from utilizing tray combined with an email device Criteria for choosing email for reconfirmation through Maili The difference in the email checking experience while using Maili compared to other digital devices: computer or smartphone Week 2 Change in perception related to utilizing Maili compared with Week 1 Week 3 Change in reaction related to email checking experience and using the reconfirmation function of Maili compared with Week 2 Week 4 The overall reaction to the design and interaction of Maili Overall opinions on the functional aspect of Maili and change of interest toward the device during 4 weeks Open in new tab TABLE 1. Interview contents following the date. Interview week . Interview contents . Week 1 The perception from checking email through a tangible device with rotating interaction The perception from checking registered email with angle changing node and pushing interaction Functional and emotional feedback from utilizing tray combined with an email device Criteria for choosing email for reconfirmation through Maili The difference in the email checking experience while using Maili compared to other digital devices: computer or smartphone Week 2 Change in perception related to utilizing Maili compared with Week 1 Week 3 Change in reaction related to email checking experience and using the reconfirmation function of Maili compared with Week 2 Week 4 The overall reaction to the design and interaction of Maili Overall opinions on the functional aspect of Maili and change of interest toward the device during 4 weeks Interview week . Interview contents . Week 1 The perception from checking email through a tangible device with rotating interaction The perception from checking registered email with angle changing node and pushing interaction Functional and emotional feedback from utilizing tray combined with an email device Criteria for choosing email for reconfirmation through Maili The difference in the email checking experience while using Maili compared to other digital devices: computer or smartphone Week 2 Change in perception related to utilizing Maili compared with Week 1 Week 3 Change in reaction related to email checking experience and using the reconfirmation function of Maili compared with Week 2 Week 4 The overall reaction to the design and interaction of Maili Overall opinions on the functional aspect of Maili and change of interest toward the device during 4 weeks Open in new tab Table 1 summarizes the questions that we raised during the 4-week interview. Through these questions, we expected to derive a general user’s opinion regarding the function, design and tangible interaction provided from Maili along with the changed pattern of their weekly uses. 6.3. Data analysis Interview data were analyzed using thematic analysis to determine the results of the study (Maguire and Delahunt, 2017; Braun and Clarke, 2006). Researchers created groups by continuously and repeatedly reading raw interview transcripts. Relevant or interesting quotations were marked with memos on a first review of the data. Besides the users’ participation, responses were grouped into themes since they emerged consistently during the interview analysis (Maguire and Delahunt, 2017). Throughout the interviews, we identified constant, opposing opinions from each participant with an across-case approach (Ayres et al., 2003). Through this, we sought to identify the same or different users’ responses following different email usage patterns. All resulting topics were organized after identifying that three out of five participants’ responses were similar or opposing on the same topic. The interview data were grouped as follows: (i) usage patterns of the existing email system, (ii) overall Maili usage patterns and opinions, (iii) Maili as an independent single-purpose device, (iv) reconfirmation function, (v) product design factors, (vi) interactions with the product (tray, dial and nodes) and (vii) the value of tray. From this, we chose five main topics related to the research questions as follows: (i) insights regarding the single-purpose tangible device to represent digital information, (ii) interest and usability of physical interaction to control email with tangible interaction, (iii) users’ opinions on the overall design of Maili within the context of office environment, (iv) utilization of reconfirmation function and feelings toward protruding node notification and (v) perceptions toward the function or design during their 30 days of use. 7. FINDINGS 7.1. Interacting with email desk object with non-digital and digital roles In a previous study regarding dematerialization, separating digital functions by physical modules (Van Campenhout et al., 2016) indicated the flow of digital information according to user manipulation. In this study, by combining two types of tangible parts that include a non-digital (the tray) and a digital module (email management) into a desktop artifact named Maili, we explored opportunities to increase overall usability by providing the value of physical richness and encouraging the function of the digital module. 7.1.1. Maili as an independent device on the desk Maili became a reminder of the importance of email by performing as an independent artifact on user desk environments. All participants showed a similar reaction to this. As stated by P4, ‘I considered that email was one of the important means of communication. Therefore, I felt better when utilizing Maili. It is a valuable addition to the office, and it can be operated usefully. I think it is worth the space it takes on the desk.’ After the fourth week, as P4 got used to Maili and learned how to utilize Maili to manage email. In the last interview, P4 remarked, ‘At first, I did not do like this. However, I found I could see only emails in this device, and I did not need to open the web site. I just rotated the dial to check it. It was fresh and good for confirming email with other actions without mouse clicking.’ P4’s reference to seeing only emails in a single device indicated the singular function of Maili, as a standalone design for managing email data, appeared valuable to the user. The physical interaction of the dial (to check emails) also appeared to provide novelty against standard, windows-based, email handing (i.e. open the web site). In addition, P4 mentioned the positive aspects of controlling a physical product with behaviors different from existing smartphone uses and said that it felt like using an analog product on a desk in this digital age. Primarily, this advantage came from their usage patterns while inducing natural management of their email by showing information that a new email had arrived through the display. Through this, it unconsciously encouraged email confirmation within the user’s sight. P2 said, ‘I think it has become standard for me to read the email by seeing new email messages on display. In the past, I used to check the inbox when I was bored during working. However, I think it made a system to check email after detecting email has come or not for certain on display.’ This also helped users to focus on their work while reducing unnecessary time checking emails. All participant opinions were similar. For example, P3 pointed out, ‘I feel like saving time. I’m not going to keep pressing f5 (refresh) to check the inbox. I’m focused on working because I know that new email information is coming (through Maili) at a specific time.’ As mentioned in a previous study (Mark et al., 2016), email use patterns, especially the duration of using emails, are significantly associated with stress. In this regard, Maili supported users in spending reduced time in checking their emails because email checking was limited to only when a new email was displayed by the device; this also seemed to be derived from the experience of getting emails every 30 minutes and not receiving them in real time. We could see Maili’s mode of email display and infrequent updates helped to avoid unconscious revisits to the email inbox with the aid of the precise planning of emails’ reconfirmation time. Moreover, the duration of displaying new emails in Maili can be determined more accurately by identifying users’ breakpoints and showing the emails at those points rather than with immediate delivery (Iqbal and Bailey, 2008). This mode may further explain the users’ swifter reactions to email notification from Maili. 7.1.2. Non-digital tangible form affects the digital function We identified the possibility of combining the non-digital tangible module part, a tray, with digital function modules. This consolidation supported increasing the usability of the device by adding simple storage with the primary digital function. This insight was evidenced by the possibilities to support reminders for work-related emails through items placed on the tray. P3 noted, ‘In the existing tray I previously had, I just put something on it and brought it if something was needed. However, this tray (in Maili) is connected to the email function. Therefore, I kept looking at it due to checking email. I put a business card or post-it on the tray (Figure 10, left a). For example, I was supposed to send an email to the person who gives me the name card. However, after the meeting, I forgot to do that. I came to my desk and did something else. (In Maili) I spontaneously looked at the email on the display and tray in order and brought that business card onto the tray. It reminded me to send an email to the person.’ FIGURE 10 Open in new tabDownload slide Maili in situ: (a) P1’s Maili in her office desk, (b) P2’s Maili placed on his design work table, (c) P3’s Maili in his start-up office, (d) P4’s Maili on his personal desk in a large office shared by 30 people and (e) P5’desk place on his office desk. FIGURE 10 Open in new tabDownload slide Maili in situ: (a) P1’s Maili in her office desk, (b) P2’s Maili placed on his design work table, (c) P3’s Maili in his start-up office, (d) P4’s Maili on his personal desk in a large office shared by 30 people and (e) P5’desk place on his office desk. Participants expressed the opinion that Maili’s design and function have the strength of being able to check email with the help of the non-digital tray. We could see the possibilities that the materiality of the tray provided benefit to email function by deploying the Maili design in the working context. Participants remarked that the activity of placing things into the tray also functioned as a work-related reminder (i.e. sticky notes flagging important tasks derived from email content). 7.2. Interaction with node: reminding by intentional unaligned form expression From notifying reconfirmation emails through changing the arranged node to an unaligned status by protruding nodes from a flat surface, Maili offered the benefit of communicating information in a noticeable manner, inducing natural user interaction. Four participants positively reacted to the physical angle change notifications compared to the digital alarm. P2 mentioned, ‘If the node is protruding while I am working, it attracts my eyes. It was an interesting element in that it showed that there are reserved emails.’ Previous studies have found that dynamic data can be expressed using form transformation (Coelho and Zigelbaum, 2011), showing possibilities to deliver meaning through shape changes. Similarly, the current study identified the benefit of adaptive shape change as opportunity to express the dynamic aspect of digital data. Regarding the pushing node interaction, all participants mentioned that the changing shape encouraged them to press the node back into place. This insight could also be divided into two types of findings related to the emotional perception toward the node shape change. P1 and P3 expressed the impression of pushing the node in order to restore Maili to its original shape. P2, P4, and P5 responded positively to completing or starting work by pressing one node. Specifically, the node for protruding and pushing interaction successfully stimulated a positive experience because it was natural to respond to and promoted use. More specifically, P4 remarked, ‘It (the node) gave me a novel feeling by moving mechanically, and it is fun to push the node because it provided the perception of accomplishing something by pressing it back. The notifications in the ordinal digital system finished after time passed. This node system is good for promoting the action to push the node to the original position. It brings me a positive feeling to complete the job.’ From here, pushing the node back to the original position could bring perception of finishing a specific job clearly by making the overall shape complete again. Besides, this node interaction of Maili promoted engagement with the device by providing a different type of tangible interaction (e.g. mouse-clicking or typing), which was not common in phones. P5 mentioned, ‘Touching the phone did not have a movement expression after I pressed. It was fun to see the motion from Maili that nodes coming out, hearing the sound of node protruding and to feel the texture while pressing. These things induced me to do new interesting actions.’ We found that Maili’s interaction allowed natural user engagement, while making users look at the protruding node and conveying a positive feeling through the push motion. However, one participant expressed concern that continuously inducing these physical interactions could cause fatigue for some users. As P1 remarked, ‘I’ve been too busy these days. In that situation, it feels like I can get a little tired of having to take action.’ In this regard, we could also see the need for careful consideration in designing the pop-up timing, frequency and intervals between nodes confirmation, when users engage in physical interaction with Maili. In particular, P1’s comment indicated consideration must also be given to how the appropriateness of node pop-ups is also context and activity sensitive. For a busier user, use-context, too much materialized feedback appears inappropriate. 7.2.1. Awakening memories and reducing checking steps by reconfirming important email The email reconfirmation function was used primarily for large-scale events the participants were required to prepare. In other words, the user registered the time of their plan and utilized this function similar to a calendar application. Five participants used the time-setting function through the dial for a total of 86 emails. Among them, 51 emails were delayed with days for the reconfirmation. Thirty-five emails were delayed by the only hours change without delaying the date. In the way of using the reconfirmation function, different patterns appeared according to user habits. P1 said that she set the alarm time of a day to finish the task in the mail (Week 3). Reflecting this, P1 showed a pattern (76%, 13 emails from a total of 17) of utilizing hour change alone rather than date change, compared to other users (Table 2). In the last week, P1 registered similar content of emails to be checked at once, regardless of reconfirmation time. Following these patterns, P2 and P4 primarily used date changes. Among them, P3 showed two different types of usage patterns. In weeks 1 and 2, P3 received notification 2 days before the schedule in the email. In Week 3, P3 utilized the reconfirmation function to complete the work in the email by the time he registered. Following this, P3 tended to use similar rates of date and hour within the day change during the experiment. P2, P3 and P5 checked the contents of their email once more before the due date when there was a task to be done. P4 appeared to register time to reconfirm whether P4 handled the business in the email properly or not. TABLE 2. Ways of setting the duration of reconfirmation times for each participant during 1 month; shows if the participant used only hour change for same day email reconfirmation (e.g. recheck 2 hours later) or used date + hour change to reconfirm an email a few days later (e.g. recheck 3 days + 6 hours later). . P1 . P2 . P3 . P4 . P5 . No. of only hour change 13 4 14 2 2 No. of date + hour change 4 12 17 12 6 . P1 . P2 . P3 . P4 . P5 . No. of only hour change 13 4 14 2 2 No. of date + hour change 4 12 17 12 6 Open in new tab TABLE 2. Ways of setting the duration of reconfirmation times for each participant during 1 month; shows if the participant used only hour change for same day email reconfirmation (e.g. recheck 2 hours later) or used date + hour change to reconfirm an email a few days later (e.g. recheck 3 days + 6 hours later). . P1 . P2 . P3 . P4 . P5 . No. of only hour change 13 4 14 2 2 No. of date + hour change 4 12 17 12 6 . P1 . P2 . P3 . P4 . P5 . No. of only hour change 13 4 14 2 2 No. of date + hour change 4 12 17 12 6 Open in new tab In sum, we could categorize usage patterns of email reconfirmation function, and those were divided into four types: (i) setting up an alarm time to complete a task (P1, P3), (ii) checking similar emails at once (P1), (iii) receiving notification before the due date mentioned in the email (P2, P3, P5) and (iv) receiving feedback (through shape change of the node) once more for checking duty in the email whether the work has been appropriately handled (P4). Looking at the number of days and hours users delayed in the date change (Table 3), most participants utilized the feature to reconfirm the email within 2 days (87%, 75 emails among 86). During Week 3, P5 remarked, ‘It was relaxing for me to check email one more time as soon as possible because that email contains things I need to do or confirm.’ As such, users showed a pattern of reconfirming the email within a short period. TABLE 3. Number of reconfirmation time set email about how many days and hours were delayed. Changed days . 1 hour ~ 2 days 0 hour . 3 ~ 7 days . Over 7 days . No. of Mails 75 7 4 Changed days . 1 hour ~ 2 days 0 hour . 3 ~ 7 days . Over 7 days . No. of Mails 75 7 4 Open in new tab TABLE 3. Number of reconfirmation time set email about how many days and hours were delayed. Changed days . 1 hour ~ 2 days 0 hour . 3 ~ 7 days . Over 7 days . No. of Mails 75 7 4 Changed days . 1 hour ~ 2 days 0 hour . 3 ~ 7 days . Over 7 days . No. of Mails 75 7 4 Open in new tab In Table 4, delayed hours only within the day (without changing the day), the majority of emails (83%, 29/35) were delayed in less than 3 hours. TABLE 4. Number of reconfirmation time set email about how many hours were delayed within the day. Changed hours (within the day) . 1 h . 2 h . 3 h . 4 h~ . No. of mails 12 12 5 6 Changed hours (within the day) . 1 h . 2 h . 3 h . 4 h~ . No. of mails 12 12 5 6 Open in new tab TABLE 4. Number of reconfirmation time set email about how many hours were delayed within the day. Changed hours (within the day) . 1 h . 2 h . 3 h . 4 h~ . No. of mails 12 12 5 6 Changed hours (within the day) . 1 h . 2 h . 3 h . 4 h~ . No. of mails 12 12 5 6 Open in new tab In particular, when the hour was set within the day, as shown in Table 4, it was mainly used to register remained hours to finish the work described in the email at a specific time. Participants tried to finish that work before protruding the node. P3 noted, ‘I made myself a promise to finish work in the email up to this hour by registering reconfirmation time.’ P3 also mentioned the merits of Maili’s reconfirmation function compared to the existing digital calendar function, ‘When I registered an alarm in the existing calendar applications, I wrote only a simple information (e.g. ‘making a presentation file’). However, Maili made me think about the schedule in the email as I turned the dial to register reconfirmation time.’ It was found that the process of registering email by turning the dial helps users to remember the task in the email. Three participants mentioned similar answers to this benefit of thinking about email content through dial registering. Registering the reconfirmation time in such a short time provided the user with two advantages: (i) a reminder of essential emails and (ii) a shortening of the email reconfirmation process with no need to revisit the inbox. First, the interaction features of Maili have helped to remind all the users of important events. P3 mentioned that, ‘In the reconfirmation of email, it shows information I did not catch before. Even though I used to watch a smartphone or computer to check schedules, sometimes, I might have missed something. In the morning today, I forgot to prepare a presentation file that I had to make yesterday. Maili notified me in the morning (through protruding node). I was seriously shocked. Maili sent only one or two important registered emails which I couldn’t miss. I survived thanks to Maili.’ This email reconfirmation feature helped to identify the value of using it as a reminder of forgettable tasks or a tool that sends notifications to start a specific task when the reconfirmation function was activated. Cooperatively, Maili provided possibilities to support the user email management processes through relationships between appearance, interaction and functionality by the reconfirmation of important email by physically protruding nodes. As P2 mentioned, ‘It is satisfactory for the nodes to come out and push them back in. There are certain inputs and outputs, and not all ordinal electronics in the office react physically. Because of this, I focused on email notification once more. It was fun in that sense. I felt like working, processing, and finishing one work by pressing one node. It played a role as a reminder function. It was helpful because I did not have to go to the inbox and check multiple times.’ Second, in the existing literature, employees who used email in their work repeatedly accessed the inbox to obtain information about meetings (Zhao et al., 2018). Participants in our research showed similar patterns. P2 and P4, who used tags on their essential emails, reacted positively to shortening the process so that they did not have to repeatedly enter the mailbox. P4 stated, ‘The equipment gave a notification (through protruding node); therefore, I did not have to go looking again to the inbox. It saved me time in checking important emails. I used to access the inbox. Now I can check one exact email I needed.’ As indicated in the studies by Mark et al. (2012, 2016), reducing the duration of interaction with emails allows users to focus on their work. This reduction also appears to provide opportunity to lower stress levels as a necessity to continually revisit inboxes is reduced. Other studies have also stated that people tend to get lost and distracted because they read and respond to emails several times a day (Hanrahan and Pérez-Quiñones, 2015), which then requires task switching to emails (Borghouts et al., 2018). In line with those studies, Maili’s interaction helped to reduce visiting times and frequencies of visits to users’ email box. In line with previous work, we expect Maili’s physicalization approach appears to provide opportunity for improving work efficiency and reduced stress. 8. DISCUSSION AND IMPLICATIONS The materialization of email data, providing physical interaction to its handling, brings new possibilities as it provides improved access to information and expressive communication with users. Although Maili was another device added to people’s use of various, existing digital products, we were able to identify that the separation of a task in email management (e.g. setting reconfirmation time) through an independent daily object helped people to effectively control the way of using email and the timing of mail revisits. In this regard, Brudy et al. (2019) suggested that more research is needed to understand ecologies of multi-device systems through in-the-wild deployment studies and that it is necessary to rethink the boundaries, purposes and scope of devices within complex ecosystem of cross-device computing. During our field study in users’ offices, it was interesting to confirm that participants interact with and discover value in Maili, which was dedicated solely to email, rather than using their phones, devices or desktop computer. This also supports the work of Dearman and Pierce (2008), who highlight the importance of devices more explicitly supporting separation of work and personal roles. At this point, our design approach to assign a specific role to an additional device showed potential in reducing the frequency of revisiting one’s email inbox and the amount of time spent on emails, which may further decrease stress levels and support users in focusing on their primary work. Our findings resonate with the primary outcome of rematerializing the dematerialized (Van Campenhout et al., 2013, 2016, 2019). We offer new insights into how this design philosophy could apply to other domains, specifically email communication. We investigated opportunities for broadening this concept by applying materialized email data to an everyday tangible device to explore its effect on attitudes toward the email data, concerning its functionality and the emotional response of users. In the next section, we present research and design considerations for the HCI community. Notably, we suggest considerations for future research and product development in designing digital functions, interactions and forms that leverage the re-materialization of digital data. 8.1. Designing materialized interaction 8.1.1. Dividing input following functions Based on existing research (Van Campenhout et al., 2019), we have found from our user study that physical interaction can attract interest by providing a novel form of control. We divided the function of Maili into two: inbox control and registered mail reconfirmation. By applying the interaction considering the characteristics of the distinct digital functions, we attempted providing physical interaction which followed digital functions. Previous research was also able to identify the possibility of controlling complex Internet of Things functions by modularizing interactions according to increasing digital functions (Frens et al., 2018). Rather than controlling multiple digital functions with a single physical controller, in designing physical products with digital information, our results indicate the possibility of inducing the user’s interest while providing a functional advantage through the division of each disparate physical interaction. For example, in evolving Maili design, we might consider dividing digital function between inbox control, setting up a time for reconfirmation and checking registered emails. Moreover, we could also implement physical interactions to more clearly represent each digital function. 8.1.2. Context and moment of the materialized interaction In previous research of applying physicality to dematerialized information (Van Campenhout et al., 2019), the correct contexts in which a device for materialized interaction can be situated were explored. For a physical money payment terminal, its proper context was suggested as a stylish and expensive shop where shopping experience matters. From this point of view, we could identify the possibility of increasing the experiential value of materialized interaction with the products by considering the place, environment or even a specific moment in the context. This may positively highlight the tangible interaction and being in control by which the user can touch, control and actively see the flow of information. Through our design and deployment of Maili, physical interaction, especially the node input and output interaction, brought positive, pleasurable feelings from being in control and showing the flow of information by pushing the node back to see the email contents through the display. Based on this, in designing tangible data artifacts, limiting the coverage of data to represent into the artifact only to vital (e.g. VIP emails or essential date/event notification emails) as a way to further enhance the value of its interaction. The combination of physical interactions with careful consideration of the encountering moment with users’ data might promote the user’s experience to be more meaningful and engaging. Besides, we could see the potential context of materialized interaction from participants’ reactions to using Maili. Be it in a comfortable space, free of mental pressure, affording enough time to enjoy the experience of physical interaction. In other words, the environment in which the user experiences materialized interactive products will be most valuable, for example, a music player designed on the concept of re-materialization, with a movable physical module that displays the cover of the music album and its song lyrics (Van Campenhout et al., 2013). 8.2. Enriching re-materialization: analog function to digital product Our approach to assigning a physical interaction to each aspect of the digital function and combining those with an analog part (the tray), opened a space for participants to relate analog storing function with the email handling digital functionality of Maili. Previous studies of dematerialized devices have indicated that all tangible elements should be given a digital function (Van Campenhout et al., 2013). However, these studies further mention that there is no fixed point as to where and how the combination of digital and tangible should occur; instead, it depends on the designer’s skill (Van Campenhout et al., 2016). In other words, existing work has argued for the importance of the competence of designers and researchers who utilize the theory. In Maili, an independent tangible element without digital information (the tray) was installed, which users could interact and utilize in any direction they wanted. The tray also demonstrated the value of blending the device into a working environment by incorporating digital functional modules for checking email with the tray’s analog function, keeping physical items (post-its or business cards) (Fig. 10). Even though many objects are now digitized (Belk, 2013), some analog artifacts are still maintained in their physicality to provide functional through tangibility. For example, a paper-calendar utilized within the office for personal and public use, offering value by users not needing more than a glance to check the schedule and to annotate directly onto the calendar as required (Tungare et al.,2008). Depending on the user’s pattern of use, products without digital technology that are still necessary for our lives remain on our desks. Identifying the hidden value of maintaining tangible areas and extending the functionality of the product through the combination of the digital and the non-digital are expected to increase the likelihood of designing new digital artifacts for the contemporary context and also leveraging the user’s embodied experience of a material world. 8.3. Synchronization of users’ action with the digital archive Maili, as an independent object and the user’s manipulation of the device, did not affect the user’s inbox data. Also, existing research has applied digital data, such as music (Odom and Duel, 2018), calendars (Lee et al., 2017) and news (Gaver et al., 2010) into tangible devices. Deleting elements or adding from the materialized artifact did not affect the digital archives. Based on the user’s existing use pattern when engaging with email, it was found that there are many actions involved in deleting an email (Alrashed et al., 2018). People usually delete old or spam email to prevent email overload (Murillo et al., 2018). In other words, email deletion itself helps to manage email. The participants of our user study also revealed patterns of deletion for organizing email. However, they had to remove email in the computer or smartphone application during the user study because Maili did not support a delete function. This throws up the possibility of considering direct manipulation of email data itself, including deletion through a physical device, beyond bringing the existing email data to the tangible device. It could be a solution to enhance the usability of an independent email device. From a broader perspective, it may suggest a potential design space that actively involves information manipulation through a tangible device and physical interaction. 9. LIMITATION First, our design, Maili, did not provide users with the ability to customize the tray to fit better with their desk space and the specific order of positioning the stationary equipment. The study focused on how the combination of digital (email data materialization) and analog parts affects the users’ experience handling email in their workspaces. Thus, we deployed a single type of design during the field study and found that participants wanted to modularize or personalize Maili to fit their office. As Maili only offered a tray with one wide independent space section, participants expressed the need to change the size and shape of the tray in various ways. We found the need for users to customize Maili’s design matched the same way they used to store things in the office environment. Second, Maili utilizes a small OLED display, which is perceived as not giving additional display for the office. This may show a need for the trade-off between the materialization of digital data and the addition of another digital display in the work environment for email content delivery. Some participants suggested adding a way to completely hide the display when not in use or allowing the user to quickly turn the display on and off. This requirement is related to the accessibility of the Maili, which displays the number of new emails continuously. Therefore, if the display is covered, it is necessary to define the role of the display more clearly, such as showing minimal information for key tangible interactions with the device. This relates to the design of not including all the input functions into one dial, which reduces the digital feeling but separates the dial interaction into day and hour settings. Also, according to the study, most users reconfirm email within a short period. Consequently, the device may focus on providing interactions that establish short reconfirmation time quickly. From a different point of view, previous research noted that money payment prototype based on dematerialization studies (Van Campenhout et al., 2019) was suitable for luxury retailers such as wine shops. This shows that even if the required interaction time is long, the device can provide positive value to the user through its fruitful interaction. In this study, all participants were in an open office environment with other coworkers, and users frequently stopped to quickly check email while they were busy concentrating on their work. This suggests that if users are in a solitary environment or a position to manage their email in a more relaxed manner, the experience of a new everyday artifact providing digital data materialization can be enriching. For example, additional research in the private spaces, such as craft workrooms and independent professional offices, might be a way to look deeply into how the materialization of email data handling may affect their the way of email management. 10. CONCLUSION We designed Maili to investigate how dematerialized email data can be applied to a tangible object. Maili’s design focused on blending into the office environment through tangible tray-type design by offering physical interaction with rotation and nodes protruding from the flat surface to naturally induce users’ manipulation and an additional reconfirmation function to offer easy email management. The 1-month in-field study with five office workers raised the possibility of a materialized device in their office environment to manipulate digital data (i.e. email). We identified two specific points: (i) Maili’s contextual value as an independent device for merging digital function modules with an analog physical tray. (ii) It demonstrated the functional advantage in checking new emails instantly, reminding and shortening the process by bringing email checking and additional management operation. This brought interest and convenience to the user through the physical interaction for controlling email data. Based on the findings, we propose new implications for designing tangible devices from the dematerialized information, email. Our work highlights the need for future research to explore the possibility of dividing input following digital functions and how the moment of interaction and the environment in which the device is situated can affect the experience of using a materialized artifact. We also suggest new design opportunities by synchronizing the user’s actions on the data artifact with the actual digital archive and integrating the digital and the non-digital (analog) parts. Funding National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. NRF-2020R1F1A1054047); Korea Institute for Advancement of Technology (KIAT) grant funded by the Korea Government (MOTIE) (P0012725, HRD Program for Industrial Innovation). REFERENCES Alexander , J. , Jansen , Y., Hornbæk , K., Kildal , J. and Karnik , A. ( 2015 ) Exploring the challenges of making data physical . In Proc. 33rd annual ACM conf. extended abstracts on human factors in computing systems , pp. 2417 – 2420 . Alonso , M. B. and Keyson , D. V. ( 2005 ) MusicCube: making digital music tangible . In CHI'05 extended abstracts on human factors in computing systems , pp. 1176 – 1179 . Alrashed , T. , Awadallah , A. H. and Dumais , S. ( 2018 ) The lifetime of email messages: a large-scale analysis of email revisitation . In Proc. 2018 conf. human information interaction & retrieval , pp. 120 – 129 . Ayres , L. , Kavanaugh , K. and Knafl , K. A. ( 2003 ) Within-case and across-case approaches to qualitative data analysis . Qual. Health Res. , 13 , 871 – 883 . Google Scholar Crossref Search ADS PubMed WorldCat Bälter , O. ( 1998 ) Electronic mail in a working context . Doctoral Dissertation, KTH . Bellotti , V. , Ducheneaut , N., Howard , M. and Smith , I. ( 2003 ) Taking email to task: the design and evaluation of a task management centered email tool . In Proc. SIGCHI conf. human factors in computing systems , pp. 345 – 352 . Belk , R. W. ( 2013 ) Extended self in a digital world . J. Consum. Res. , 40 , 477 – 500 . Google Scholar Crossref Search ADS WorldCat Bentley , F. , Daskalova , N. and Andalibi , N. ( 2017 ) “If a person is emailing you, it just doesn't make sense” exploring changing consumer behaviors in email . In Proc. 2017 CHI conf. human factors in computing systems , pp. 85 – 95 . Bennett , P. , Fraser , M. and Balaam , M. ( 2012 ) ChronoTape: tangible timelines for family history . In Proc. sixth int. conf. tangible, embedded and embodied interaction , pp. 49 – 56 . Bettina , S. ( 2018 ) Email client market share trends for the first half of 2018. LITMUS . https://litmus.com/blog/email-client-market-share-trends-first-half-of-2018 (accessed June 25, 2019) . Bishop , D. ( 1992 ) Durrell bishop marble answer machine . http://vimeo.com/19930744 (accessed September 15, 2020) . Borghouts , J. , Brumby , D. and Cox , A. ( 2017 ) Batching, error checking and data collecting: Understanding data entry in a financial office . In Proc. 15th European conf. computer-supported cooperative work-exploratory papers . European Society for Socially Embedded Technologies . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Borghouts , J. W. , Brumby , D. P. and Cox , A. L. ( 2018 ) Looking up information in email: Feedback on visit durations discourages distractions . In Extended abstracts of the 2018 CHI conf. human factors in computing systems , pp. 1 – 6 . Braun , V. and Clarke , V. ( 2006 , 2006 ) Using thematic analysis in psychology . Qual. Res. Psychol. , 3 , 77 – 101 . Google Scholar Crossref Search ADS WorldCat Brewer , R. , Garcia , R. C., Schwaba , T., Gergle , D. and Piper , A. M. ( 2016 ) Exploring traditional phones as an e-mail interface for older adults . ACM Trans. Access Comput. , 8 , 1 – 20 . Google Scholar Crossref Search ADS WorldCat Brown , B. , Reeves , S. and Sherwood , S. ( 2011 ) Into the wild: challenges and opportunities for field trial methods . In Proc. SIGCHI conf. human factors in computing systems , pp. 1657 – 1666 . Brudy , F. , Holz , C., Rädle , R., Wu , C. J., Houben , S., Klokmose , C. N. and Marquardt , N. ( 2019 ) Cross-device taxonomy: survey, opportunities and challenges of interactions spanning across multiple devices . In Proc. 2019 CHI conf. human factors in computing systems , pp. 1 – 28 . Cannata , G. ( 2006 ) Phymail box: an information appliance that checks and prints only important emails . Pers. Ubiquitous Comput. , 10 , 170 – 172 . Google Scholar Crossref Search ADS WorldCat Chen , A. Y. S. , Odom , W., Zhong , C., Lin , H. and Amram , T. ( 2019 ) Chronoscope: designing temporally diverse interactions with personal digital photo collections . In Proc. 2019 on designing interactive systems conf. , pp. 799 – 812 . Coelho , M. and Zigelbaum , J. ( 2011 ) Shape-changing interfaces . Pers. Ubiquitous Comput. , 15 , 161 – 173 . Google Scholar Crossref Search ADS WorldCat David Allen Company ( 2016 ) ORGANIZING YOUR WORKSPACE. gtd (gettingthisdone) . Dearman , D. and Pierce , J. S. ( 2008 ) It's on my other computer! Computing with multiple devices . In Proc. SIGCHI conf. human factors in computing systems , pp. 767 – 776 . Frens , J. W. ( 2006 ) Designing for rich interaction: integrating form, interaction, and function . In Conf.; 3rd symposium of design research; 2006-11-17; 2006-11-18 , pp. 91 – 106 . Swiss Design Network . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Frens , J. , Funk , M., van Hout , B. and Le Blanc , J. ( 2018 ) Designing the IoT sandbox . In Proc. 2018 designing interactive systems conf. , pp. 341 – 354 . Gallacher , S. , Golsteijn , C., Wall , L., Koeman , L., Andberg , S., Capra , L. and Rogers , Y. ( 2015 ) Getting quizzical about physical: observing experiences with a tangible questionnaire . In Proc. 2015 ACM int. joint conf. pervasive and ubiquitous computing , pp. 263 – 273 . Gaunt , K. , Toriseva , J., Vegele , R. and Padegimaite , M. ( 2013 ) Hush tube: designing a tangible and quieter email . In Proc. 25th Australian computer-human interaction conf.: augmentation, application, innovation, collaboration , pp. 417 – 418 . Gaver , B. and Höök , K. ( 2017 ) In search of the elusive CHI design paper . Interactions , 24 , 22 – 23 . Google Scholar Crossref Search ADS WorldCat Gaver , W. , Blythe , M., Boucher , A., Jarvis , N., Bowers , J. and Wright , P. ( 2010 ) The prayer companion: openness and specificity, materiality and spirituality . In Proc. SIGCHI conf. human factors in computing systems , pp. 2055 – 2064 . Gaver , W. ( 2012 ) What should we expect from research through design? In Proc. SIGCHI conf. human factors in computing systems , pp. 937 – 946 . Golsteijn , C. , Gallacher , S., Koeman , L., Wall , L., Andberg , S., Rogers , Y. and Capra , L. ( 2015 ) VoxBox: a tangible machine that gathers opinions from the public at events . In Proc. ninth int. conf. tangible, embedded, and embodied interaction , pp. 201 – 208 . Grevet , C. , Choi , D., Kumar , D. and Gilbert , E. ( 2014 ) Overload is overloaded: email in the age of Gmail . In Proc. sigchi conf. human factors in computing systems , pp. 793 – 802 . Hanrahan , B. V. and Pérez-Quiñones , M. A. ( 2015 ) Lost in email: pulling users down a path of interaction . In Proc. 33rd annual ACM conf. human factors in computing systems , pp. 3981 – 3984 . Hassenzahl , M. ( 2018 ) The thing and I: understanding the relationship between user and product . In Funology 2 , pp. 301 – 313 . Springer , Cham . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hill , K. and Monk , A. F. ( 2000 ) Electronic mail versus printed text: the effects on recipients . Interact. Comput. , 13 , 253 – 263 . Google Scholar Crossref Search ADS WorldCat Hornecker , E. ( 2011 ) The role of physicality in tangible and embodied interactions . Interactions , 18 , 19 – 23 . Google Scholar Crossref Search ADS WorldCat Houben , S. , Golsteijn , C., Gallacher , S., Johnson , R., Bakker , S., Marquardt , N., Capra , L. and Rogers , Y. ( 2016 ) Physikit: data engagement through physical ambient visualizations in the home . In Proc. 2016 CHI conf. human factors in computing systems , pp. 1608 – 1619 . Huron , S. , Gourlet , P., Hinrichs , U., Hogan , T. and Jansen , Y. ( 2017, June ) Let's get physical: promoting data physicalization in workshop formats . In Proc. 2017 conf. designing interactive systems , pp. 1409 – 1422 . Ishii , H. and Ullmer , B. ( 1997 ) Tangible bits: towards seamless interfaces between people, bits and atoms . In Proc. ACM SIGCHI conf. human factors in computing systems , pp. 234 – 241 . Iqbal , S. T. and Bailey , B. P. ( 2008 ) Effects of intelligent notification management on users and their tasks . In Proc. SIGCHI conf. human factors in computing systems , pp. 93 – 102 . Jang , S. , Kim , S., Noh , B. and Park , Y. W. ( 2019 ) Monomizo: a tangible desktop artifact providing schedules from E-ink screen to paper . In Proc. 2019 on designing interactive systems conf. , pp. 1123 – 1130 . Jansen , Y. , Dragicevic , P., Isenberg , P., Alexander , J., Karnik , A., Kildal , J., Subramanian , S. and Hornbæk , K. ( 2015 ) Opportunities and challenges for data physicalization . In Proc. 33rd annual ACM conf. human factors in computing systems , pp. 3227 – 3236 . Sheldon , J. ( 2017 ) GATHER . https://www.kickstarter.com/projects/ugmonk/gather-the-minimal-modular-organizer-that-cuts-clu (accessed May 4, 2018) . Jovicic , A. ( 2000 ) Implications for the design of email management software . Doctoral dissertation, National Library of Canada= Bibliothèque nationale du Canada . Kim , G. D. and Eune , J. ( 2015 ) Sustainable transport system: a wheel based interactive information installation . In Proc. 33rd annual ACM conf. extended abstracts on human factors in computing systems , pp. 375 – 378 . Kim , J. , Self , J. A. and Park , Y. W. ( 2018 ) Traffico: a tangible timetable delivering transportation information between schedules . In Proc. 2018 designing interactive systems conf. , pp. 1229 – 1234 . Lee , K. R. , Goh , G. I. and Park , Y. W. ( 2017, May ) Quietto: an interactive timepiece molded in concrete and milled wood . In Proc. 2017 CHI conf. human factors in computing systems , pp. 2988 – 2992 . Mackay , W. E. ( 1988 ) More than just a communication system: diversity in the use of electronic mail . In Proc. 1988 ACM conf. computer-supported cooperative work , pp. 344 – 353 . Maguire , M. and Delahunt , B. ( 2017 ) Doing a thematic analysis: a practical, step-by-step guide for learning and teaching scholars . AISHE-J , 9 , 3351 – 3359 . Google Scholar OpenURL Placeholder Text WorldCat Mark , G. , Voida , S. and Cardello , A. ( 2012 ) “A pace not dictated by electrons” an empirical study of work without email . In Proc. SIGCHI conf. human factors in computing systems , pp. 555 – 564 . Mark , G. , Iqbal , S. T., Czerwinski , M., Johns , P., Sano , A. and Lutchyn , Y. ( 2016 ) Email duration, batching and self-interruption: patterns of email use on productivity and stress . In Proc. 2016 CHI conf. human factors in computing systems , pp. 1717 – 1728 . Murillo , A. , Kramm , A., Schnorf , S. and De Luca , A. ( 2018 ) “If I press delete, it's gone”-user understanding of online data deletion and expiration . In Fourteenth symposium on usable privacy and security ({SOUPS} 2018) , pp. 329 – 339 . Nam , T. J. and Kim , C. ( 2011 ) Design by tangible stories: enriching interactive everyday products with ludic value . Int. J. Des. , 5 , 85 – 98 . Google Scholar OpenURL Placeholder Text WorldCat Neustaedter , C. , Brush , A. B. and Smith , M. A. ( 2005 ) Beyond “from” and “received” exploring the dynamics of email triage . In CHI'05 extended abstracts on human factors in computing systems , pp. 1977 – 1980 . Odom , W. , Sellen , A., Harper , R. and Thereska , E. ( 2012 ) Lost in translation: understanding the possession of digital things in the cloud . In Proc. SIGCHI conf. human factors in computing systems , pp. 781 – 790 . Odom , W. T. , Sellen , A. J., Banks , R., Kirk , D. S., Regan , T., Selby , M. and Zimmerman , J. ( 2014 ) Designing for slowness, anticipation and re-visitation: a long term field study of the photobox . In Proc. SIGCHI conf. human factors in computing systems , pp. 1961 – 1970 . Odom , W. , Wakkary , R., Hol , J., Naus , B., Verburg , P., Amram , T. and Chen , A. Y. S. ( 2019 ) Investigating slowness as a frame to design longer-term experiences with personal data: a field study of olly . In Proc. 2019 CHI conf. human factors in computing systems , pp. 1 – 16 . Odom , W. , Wakkary , R., Lim , Y. K., Desjardins , A., Hengeveld , B. and Banks , R. ( 2016 ) From research prototype to research product . In Proc. 2016 CHI conf. human factors in computing systems , pp. 2549 – 2561 . Odom , W. , Zimmerman , J., Forlizzi , J., López Higuera , A., Marchitto , M., Cañas , J. and Kim , D. J. ( 2013 ) Fragmentation and transition: understanding perceptions of virtual possessions among young adults in Spain, South Korea and the United States . In Proc. SIGCHI conf. human factors in computing systems , pp. 1833 – 1842 . Odom , W. and Duel , T. ( 2018 ) On the design of OLO radio: investigating metadata as a design material . In Proc. 2018 CHI conf. human factors in computing systems , pp. 1 – 9 . Pielot , M. , Church , K. and De Oliveira , R. ( 2014 ) An in-situ study of mobile phone notifications . In Proc. 16th int. conf. human-computer interaction with mobile devices & services , pp. 233 – 242 . Poupyrev , I. , Nashida , T. and Okabe , M. ( 2007 ) Actuation and tangible user interfaces: the Vaucanson duck, robots, and shape displays . In Proc. 1st int. conf. tangible and embedded interaction , pp. 205 – 212 . Quaresma , R. F. C. , Silva , S. P. R. D. and Marreiros , C. G. ( 2013 ) E-mail usage practices in an organizational context: a study with Portuguese workers . J. Inf. Syst. Technol. Manag. , 10 , 5 – 20 . Google Scholar OpenURL Placeholder Text WorldCat Quintal , F. , Barreto , M., Jorge , C., Nisi , V. and Nunes , N. J. ( 2018 ) Watt-I-see: design and evaluation of an interactive installation using eco-feedforward strategies . Interact. Comput. , 30 , 31 – 45 . Google Scholar Crossref Search ADS WorldCat Rasmussen , M. K. , Pedersen , E. W., Petersen , M. G. and Hornbæk , K. ( 2012 ) Shape-changing interfaces: a review of the design space and open research questions . In Proc. SIGCHI conf. human factors in computing systems , pp. 735 – 744 . Rector , K. and Hailpern , J. ( 2014 ) MinEMail: SMS alert system for managing critical emails . In Proceedings of the SIGCHI conf. human factors in computing systems , pp. 783 – 792 . Redström , J. ( 2001 ) Designing everyday computational things. rapport nr.: Gothenburg studies in Informatics, (20) . Rogers , Y. and Marshall , P. ( 2017 ) Research in the wild . Synth. Lect. Hum.Cent. Inform. , 10 , i – 97 . Google Scholar Crossref Search ADS WorldCat Stolterman , E. and Wiberg , M. ( 2010 ) Concept-driven interaction design research . Hum. Comput. Interact. , 25 , 95 – 118 . Google Scholar Crossref Search ADS WorldCat Spindler , M. , Tominski , C., Schumann , H. and Dachselt , R. ( 2010 ) Tangible views for information visualization . In ACM int. conf. interactive tabletops and surfaces , pp. 157 – 166 . Studio PESI ( 2017 ) On the surface . https://www.studiopesi.com/on-the-surface (accessed March 1, 2018) . The Radicati Group, Inc. ( 2015 ) Email Statistics Report, 2015–2019 . A Technology Maerket Research Firm , Palo Alto, CA, USA . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Tungare , M. , Perez-Quinones , M. and Sams , A. ( 2008 ) An exploratory study of calendar use . arXiv:0809.3447 . Vallgårda , A. ( 2014 ) Giving form to computational things: developing a practice of interaction design . Pers. Ubiquitous Comput. , 18 , 577 – 592 . Google Scholar Crossref Search ADS WorldCat Van Campenhout , L. , Frens , J., Overbeeke , K., Standaert , A. and Peremans , H. ( 2013 ) Physical interaction in a dematerialized world . Int. J. Des. , 7 , 1 – 18 . Google Scholar OpenURL Placeholder Text WorldCat Van Campenhout , L. , Frens , J., Hummels , C., Standaert , A. and Peremans , H. ( 2019 ) The enriching limitations of the physical world . Pers. Ubiquitous Comput. , 23 , 81 – 98 . Google Scholar Crossref Search ADS WorldCat Van Campenhout , L. D. E. , Frens , J., Hummels , C., Standaert , A. and Peremans , H. ( 2016 ) Touching the dematerialized . Pers. Ubiquitous Comput. , 20 , 147 – 164 . Google Scholar Crossref Search ADS WorldCat van Dijk , J. , Moussette , C., Kuenen , S. and Hummels , C. ( 2013 ) Radical clashes: what tangible interaction is made of . In Proc. 7th int. conf. tangible, embedded and embodied interaction , pp. 323 – 326 . Vertesi , J. , Kaye , J., Jarosewski , S. N., Khovanskaya , V. D. and Song , J. ( 2016 ) Data narratives: uncovering tensions in personal data management . In Proc. 19th ACM conf. computer-supported cooperative work & social computing , pp. 478 – 490 . Vitale , F. , Odom , W. and McGrenere , J. ( 2019 ) Keeping and discarding personal data: exploring a design space . In Proc. 2019 on designing interactive systems conf. , pp. 1463 – 1477 . Whittaker , S. and Hirschberg , J. ( 2001 ) The character, value, and management of personal paper archives . ACM Trans. Comput. Hum. Interact. , 8 , 150 – 170 . Google Scholar Crossref Search ADS WorldCat Whittaker , S. , Jones , Q. and Terveen , L. ( 2002 ) Managing long term communications: conversation and contact management . In Proc. 35th annual Hawaii int. conf. system sciences , pp. 1070 – 1079 . IEEE . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Whittaker , S. , Matthews , T., Cerruti , J., Badenes , H. and Tang , J. ( 2011 ) Am I wasting my time organizing email? A study of email refinding . In Proc. SIGCHI conf. human factors in computing systems , pp. 3449 – 3458 . Whittaker , S. and Sidner , C. ( 1996 ) Email overload: exploring personal information management of email . In Proc. SIGCHI conf. human factors in computing systems , pp. 276 – 283 . Ylirisku , S. , Jacucci , G., Sellen , A. and Harper , R. ( 2016 ) Design research as conceptual designing: the Manhattan design concept . Interact. Comput. , 28 , 648 – 663 . Google Scholar Crossref Search ADS WorldCat Zhao , Q. , Bennett , P. N., Fourney , A., Thompson , A. L., Williams , S., Troy , A. D. and Dumais , S. T. ( 2018 ) Calendar-aware proactive email recommendation . In The 41st int. ACM SIGIR conf. research & development in information retrieval , pp. 655 – 664 . Zimmerman , J. , Forlizzi , J. and Evenson , S. ( 2007 ) Research through design as a method for interaction design research in HCI . In Proc. SIGCHI conf. human factors in computing systems , pp. 493 – 502 . © The Author(s) 2021. Published by Oxford University Press on behalf of The British Computer Society. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Investigating Physical Interaction With Digital Data Through the Materialization of Email Handling
JF - Interacting with Computers
DO - 10.1093/iwc/iwab003
DA - 2021-03-06
UR - https://www.deepdyve.com/lp/oxford-university-press/investigating-physical-interaction-with-digital-data-through-the-5Bg8746Fyk
SP - 1
EP - 1
VL - Advance Article
IS - 
DP - DeepDyve
ER -