XXV.6 November - December 2018
Page: 90
Digital Citation

Scaling interaction—From small-scale interaction to architectural scale

Fatemeh Moradi, Mikael Wiberg, Mikael Hansson

back to top 

Smart homes, intelligent environments, interactive architecture, and human-building interaction (HBI) [1,2] all come with the promise of interactivity at the scale of architecture. However, typical interaction design works at the level of the small scale and at the level of the individual—just think about the typical use case with a person in front of a laptop, or one where a person is fiddling on his or her mobile phone. At this scale, user interfaces are typically designed for an individual to operate a computational device. But what if we want to scale interaction design from the device to the building, and from the individual to the public? Put differently, how can we reimagine interaction design at architectural scale, while enabling people to interact with such large-scale interactive systems?

back to top  Insights


In this article, we elaborate on scaling as a design challenge for HCI. We do this to imagine innovative solutions that might enable human-building interaction (HBI) [2]. We also do it to explore how interaction design might operate across different scales—from small-scale designs to interaction design at architectural scale.

back to top  What is Scale? Drawing on Architecture

Any architectural drawing or blueprint typically includes some information concerning its scale. This information is essential in order to understand the size and dimensions of the building; it also allows for small-scale interaction (e.g., sketching and remodeling the blueprint) with the building as a large-scale construction. As such, this information relates the small drawing to the building's construction, and it enables the scaling of interaction—from sketching on the table to the effect of these edits on the overall design and layout of the full-size building.

From a practical viewpoint, the notion of scale also works the other way around. Scale enables the illustration and representation of a huge building to fit on a piece of paper, providing an overview. This in turn enables a person to play around with the drawing of a building, and to study and experiment with its form, program, and general layout. Through this use of scale, the architect can work at the individual level, with an intended outcome at architectural scale.

Scaling is also something that has recently been addressed by HCI researchers. As pointed out by Dalton et al. [1], Molly Wright Steenson [3], and Mikael Wiberg [4], HCI does in fact have some roots in architecture. So it comes as no surprise that HCI researchers are increasingly drawing on architectural notions such as scale. For instance, in their article "Does HCI Scale?" Barry Brown, Susanne Bødker, and Kristina Höök [5] refer to scale in terms of "how technology is used in large networks of interconnected systems, with billions of users, across diverse contexts." They discuss three different scales: the number of users, the different contexts of use, and the multitude of systems and technologies. In this article we add a fourth perspective. We discuss how the interaction model might scale—from the level of the individual and the artifact to public spaces and interactive solutions at architectural scale.

back to top  Scaling—Reimagining IXD in the Realm of Human-Building Interaction

If we now consider this on a more concrete level in terms of design solutions, we are faced with a central design challenge: How do we unite the individual's ability to interact with an interface with a larger canvas for interaction that might be at architectural scale? Here we consider something at architectural scale to be, for instance, a room or a building—a built environment that we can inhabit.

Throughout the history of interactive systems, digital artifacts have typically been physically embodied and structured within architectural spaces. The device paradigm—in which we conceptualize and design interactive systems as a "tool" or within some computational container (rather than imagining interactive systems as completely blended with the architecture)—remains strong. Some exceptions include early work on intelligent environments, as well as more recent work on so-called interactive architecture [1].

ins02.gif Final implementation of the Light Dice.

Still, less is known about how to scale the interaction model from the individual to the building. One basic reason might be that for traditional interaction design, input and output are typically implemented at the same scale, usually that of the individual. For example, the multitouch display on a mobile phone is large enough to work as an interface for input (say, to fit a representation of a keyboard, with keys large enough for the fingers to click on), as well as large enough for displaying information (text and graphics can be displayed and read). With one such implementation, the interaction model is at the scale of the individual, and the model is also symmetrical in terms of the scale of the input and output modalities. However, for any HBI solution to work, the input modality needs to be at individual scale, whereas the output might need to be at architectural scale. As such, an HBI interaction model might need to be asymmetrical to allow for individual interaction at architectural scale. We think of this as an approach to implement scaling in HCI.

To illustrate these ideas of small-scale interfaces that allow for playful, large-scale interaction with the atmosphere of a building, we will now introduce our design project the Light Dice. This prototype was inspired by the ideas of speculative design [6]; we introduce it here as an example of individual interaction at architectural scale.

back to top  Designing the Light Dice—Playful HBI with an Alternative Light Switch

Traditional wall-mounted light switches might be one of the most common and lightweight ways for individuals to interact with the built environment. A light switch enables an individual to change the lighting conditions, and thus the atmosphere, of a room with a single finger. As such, it is a lightweight design of a simple interface that allows for HBI.

For any HBI solution to work, the input modality needs to be at individual scale, whereas the output might need to be at architectural scale.

So how can we reimagine this ubiquitous button—the traditional light switch—from the viewpoint of interactive systems design and scaling interaction? Here we present one such design project, where we have implemented an asymmetrical interaction model to allow for playful interaction (at an individual level) with two interactive dice that serve as a light switch for a room. By staying close to this idea of unifying different scales, the Light Dice show what the scaling of interaction might offer as we move from HCI to HBI.

In designing the Light Dice system, we discussed ways of moving beyond the traditional and functional on/off light switch. We considered interaction with a light switch as something we could play around with, that might actually invite further interaction. During the design process, we experimented with form factors. We finally decided on two interactive and networked dice, each with a plus and a minus sign. Each die has a built-in accelerometer, a microcontroller, and wireless networking; by rolling these dice (the playful mode of interaction) or combining them in a deliberate way (e.g., placing the two dice with the plus signs facing upward), an individual or even a group of people can interact with them at individual scale while seeing the effect at architectural scale—how it affects the lighting in the room. Here, the combination of two plus signs would brighten the light; two minus signs would dim it; and one plus and one minus would reset the light to the initial value.

Although this is a simple example, we hope the Light Dice project illustrates how interaction design can scale by working with asymmetrical interaction models that allow for individual interaction while the impact of these interactions—the output—is at architectural scale.

back to top  Discussion and Some Implications for Scaling Interaction

As previously mentioned, from a research point of view, we are interested in adding a fourth dimension to scaling in HCI, where interaction that occurs at the small scale also affects the large scale. By working with asymmetrical interaction models, we see an opportunity in bridging from the individual to the built environment—that is, to scale the interaction. Still, there are many design challenges ahead of us. For instance, what works as a proper interface for input in the built environment? In this article we have focused on reimagining the traditional light switch, but what else might work, both at an individual level and as part of a public space? Further, should anyone be allowed to interact with everything on an individual level when the output is at architectural scale? Just consider thermostats in public spaces. Some thermostats allow for an individual to change the temperature only a couple of degrees up or down, but few such systems implemented in public spaces allow for any individual to completely shut the heating system off or turn it up all the way. As we start this journey of scaling interaction, such questions involving control, roles, and responsibilities come into the picture as well.


In this article, we have also elaborated on how the notion of scale could work to bridge from the small-scale interface to human-building interaction. First we addressed the need for such a bridge given the current trend toward the design of smart homes and intelligent environments. Following from that, we presented a design project, the Light Dice, aimed at illustrating what the scaling of interaction could look like in practice. We focused on playful interaction in this particular context to move away from purely functional user interfaces. It was also an opportunity to present interaction design that not only operates on architectural scale but also enables individual playful interaction with the atmosphere of a building. The dimensions we focused on in this design project concern novel ways of interacting with lighting as an important dimension of the perceived atmosphere of a room.

Through this work with the Light Dice, we have also arrived at a couple of implications for moving forward. First, we think it is not necessary to scale the user interface to architectural scale to enable HBI. Instead, the user interface can be kept small, but the effect of the interaction can operate at the scale of architecture (scaling the interaction rather than the user interface). Second, the interaction model for successful HBI needs to be imagined as asymmetrical rather than symmetrical to allow for the scaling of individual interaction to architectural scale. Finally, our advice is to start with established, well-known, and small-scale interaction models typically found in buildings (e.g., light switches) and then try to reimagine these in an interactional moment—in short, try to imagine how everyday HBI can be fun, playful, and easy, and still have large-scale effects.

back to top  References

1. Dalton, N.S., Schnädelbach, H., Wiberg, M., and Varoudis, T., eds. Architecture and Interaction—Human Computer Interaction in Space and Place. Springer, 2016.

2. Alavi, H., Lalanne, D., Nembrini, J., Churchill, E., Kirk, D., and Moncur, W. Future of human-building interaction. Proc. of CHI Extended Abstracts. ACM, 2016.

3. Steenson, M. Architectural Intelligence. MIT Press, 2017.

4. Wiberg, M. The Materiality of Interaction. MIT Press, 2018.

5. Brown, B., Bødker, S., and Höök, K. Does HCI scale? Scale hacking and the relevance of HCI. Interactions 24, 5 (2017), 28–33.

6. Dunne, A. and Raby, F. Speculative Everything. MIT Press, 2013.

back to top  Authors

Fatemeh Moradi is an assistant professor in informatics with an orientation toward digital fabrication at Umeå University, Sweden. Her research is focused on the Internet of Things, ubiquitous computing, and ways of integrating interactive technologies into our built environment.

Mikael Wiberg is a full professor of informatics at Umeå University, Sweden. He has held positions as chaired professor in HCI at Uppsala University and as research director for the Umeå Institute of Design. His research interests focus on the materiality of interaction and ways of integrating architectural thinking with interaction design. He is the author of The Materiality of Interaction: Notes on the Materials of Interaction Design from MIT Press.

Mikael Hansson is a master's student in human-computer interaction in the Department of Informatics at Umeå University, Sweden. He is the designer of the Light Dice system presented in this paper.

back to top 

©2018 ACM  1072-5520/18/11  $15.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2018 ACM, Inc.

Post Comment

No Comments Found