By Mac Hill

During the 2016 presidential election, The New York Times’ election coverage incorporated two moving “needle” gauges to convey the uncertainty in their election forecasts. The needle, as The Times (Wartik, 2017) has taken to calling it, was an object of both “obsession and derision,” sparking social media hashtags and a host of memes (Figure A). The reaction to and evolution of the needle points to a desire among users and designers for more complete visualizations, ones that can convey moments of doubt alongside confidence. My final project focused on developing experiential visualizations, similar to the needle, that gave users full pictures of the uncertainty in information.

Figure A: Responses to The New York Times’ Election Needle on Twitter

Data and its collection, however, removes information from the original phenomenon that it represents and can be difficult for individuals without specific training or expertise to understand.

The needle, and information visualizations in general, gives users the opportunity to explore information quickly and on their terms. With the rise of big data, almost every aspect of modern life—shopping patterns, web searches, voter behavior—has become a data source, involving some form of data collection, to the point that missing data makes as much news as the data itself (Bach, 2018). The pervasiveness of big data provides unique opportunities for us to explore the world around us through numbers and statistics. Data and its collection, however, removes information from the original phenomenon that it represents and can be difficult for individuals without specific training or expertise to understand. This gap in understanding provides a unique problem space for designers to explore methods for making information accessible.

My investigation centered on “research through design” or exploring issues through reflective making (Martin and Hannington, 2012). I created a series of visual studies that explored different informational contexts, types of uncertainty, and techniques for engaging users’ cognitive processes. These studies pointed to several experiential principles designers can use to convey uncertainty in information visualizations.

The needle, and information visualizations in general, gives users the opportunity to explore information quickly and on their terms. With the rise of big data, almost every aspect of modern life—shopping patterns, web searches, voter behavior—has become a data source, involving some form of data collection, to the point that missing data makes as much news as the data itself (Bach, 2018). The pervasiveness of big data provides unique opportunities for us to explore the world around us through numbers and statistics. Data and its collection, however, removes information from the original phenomenon that it represents and can be difficult for individuals without specific training or expertise to understand. This gap in understanding provides a unique problem space for designers to explore methods for making information accessible.

My investigation centered on “research through design” or exploring issues through reflective making (Martin and Hannington, 2012). I created a series of visual studies that explored different informational contexts, types of uncertainty, and techniques for engaging users’ cognitive processes. These studies pointed to several experiential principles designers can use to convey uncertainty in information visualizations.

Motion can function as a metaphor for uncertainty.

Designers can use simple motions to convey uncertainty to a user. Motion functions as a type of metaphor, for instance, a back and forth motion is analogous to indecision (Figure B). Designers should consider the connotations of different motions, as well as whether the motion relates uncertainty or is open to other interpretations (for example, expanding elements can suggest growth rather than change). Visualizations can also push the limits of interpretation by challenging a user’s ability to latch onto concrete data points and force a user to make broad generalizations.

Figure B: The graphic uses bouncing balls to represent the participants in a poll and the uncertainty involved in the relationship between a poll and the total population.

Familiar contexts and metaphors make data relatable.

Designers must consider how a user will relate to the data or information being conveyed. Providing a relatable context makes visualizations more useful and easier for a user to employ existing knowledge structures when interpreting visual elements (Figure C).

Figure C: Dashboard structures are familiar to users and make it easy for users to interpret information experientially.

Allowing a user to tailor content to their particular situation, for example, by dictating a location for a weather-related map, is one method for making the context relatable to a user. A tailored visualization acts as a concrete and specific tool for the user, rather than just an abstract resource. Furthermore, when designers use metaphors to relate complex information, these metaphors have to be familiar enough that a user can interpret them quickly without getting distracted. Dashboards and simple tool structures, like scales, work well, as users are already used to reading them for information.

User control in the analysis process brings the user
closer to the represented phenomenon.

Allowing a user to control the stages or components of analysis brings the user closer to the initial phenomenon and makes changes in the data and moments of inference clearer (Figure D). This strategy allows a designer to scaffold information, making it easier to understand and build on initial insights with new information. For instance, designers can use sliders that tailor how much or how little information is seen, step-by-step walkthroughs of data analysis, or rollovers to give a user control over the analysis involved in a visualization.

Figure D: The user can control the visualization and the analysis process.

Designers have a unique opportunity to bring design techniques like narrative and metaphor into new contexts, expanding the range of forms and methods for conveying information.

After exploring the topic through research and making, I found that experientially based visualizations can convey the uncertainty involved in complex information to non-expert users. Including uncertainty through familiar contexts, metaphors, and structures gives users a fuller picture of information and empowers a user to make well-informed decisions. The topics of uncertainty and information visualization both provide a great deal of design and design research opportunities. The current demand for information visualizations, especially those that appeal to a wide audience, makes research into the subject especially timely. Designers have a unique opportunity to bring design techniques like narrative and metaphor into new contexts, expanding the range of forms and methods for conveying information. Furthermore, both uncertainty and information visualization have the potential to segue into numerous subject areas beyond the scope of this investigation, such as public health and education. In a way, this problem space is representative of the future of design; it is an inherently interdisciplinary subject matter that requires designers to collaborate with researchers from a variety of fields.

References

Bach, A. (2018, March 21). Opinion | Missing: Criminal Justice Data. The New York Times. Retrieved from https://www.nytimes.com/2018/03/21/opinion/missing-criminal-justice-data.html.

Martin, B., & Hanington, B. M. (2012). Universal methods of design: 100 ways to research complex problems, develop innovative ideas, and design effective solutions (Digital ed.). Rockport Publishers. Retrieved from https://catalog.lib.ncsu.edu/record/NCSU2690134.

Wartik, N. (2017, December 14). NYT Needle Returns to the Spotlight. The Internet Notices. The New York Times. Retrieved from https://www.nytimes.com/2017/12/14/reader-center/nyt-needle-election.html.

By Krithika Sathyamurthy

Toys are artifacts that encourage children’s expression, fantasy, interest, exploration, education, cognitive development, and gender-role learning (Kursat, Nuri et al., 2013). Since the existence of toys, many of us throughout childhood have dreamt of the ability to speak to our toys or somehow wish them into existence. According to the NY Times contributing writer, James Vlahos (2015), toys can fulfill this timeless dream with the help of Artificial Intelligence or A.I. Advances in A.I. and speech recognition are transforming toys and complexifying the toy market landscape (Vlahos, 2015). The modern toy industry includes establishments primarily engaged in manufacturing dolls, toys, and games (Hung, Iqbal, Huang, Melaisi, & Pang, 2016). As smart toys become more prevalent in the emerging market, they could play a significant role in redefining and expanding an understanding of what constitutes a doll.

Smart toys today can draw upon the insights of child’s play to generate intelligent responses to dialogue and queries.

Smart toys for young children have reached a sophisticated level of development that utilizes the latest technological advancements in artificial intelligence and robotics (Vlahos 2015). A smart toy is defined as a device with a physical toy component that connects to a Cloud computing system, augmenting its functionality through networking and sensory technologies (Hung et al., 2016). Hung et al. (2016) establishes that a smart toy, in this context, can be effectively considered an Internet of Thing with AI and can provide Augmented Reality experiences to users. These toys often use complex sensory technologies to garner information from children and then process this information through cloud-based platforms, which results in real-time interactions (Mascheroni and Holloway, 2017). In other words, smart toys today can draw upon the insights of child’s play to generate intelligent responses to dialogue and queries (Vlahos, 2016). These responses could, in many ways, guide and mold a child’s understanding of their identity at a subconscious level.

Vlahos (2015) describes how, in the 20th century, toy makers boasted products like Dolly Rekord, who spoke nursery rhymes to children in the 1920s, and Chatty Cathy, Mattel’s 1959 release whose eleven phrases included “I love you”. The Chatty Cathy doll primarily requested care from her owner (Hilu, 2016). When technology gave dolls the ability to speak, it contributed to the expected criteria of girls giving care in doll play (Hilu, 2016). In 1958, Barbie gained her voice through a pull string that triggered eight phrases. More recently, a study by Hilu (2016) explains that the toy industry began to incorporate microchips into talking dolls in the 1980’s. As a consequence of this innovation, Mattel released microchip enabled Teen Talk Barbie in 1992, whose programmed phrases included “Math class is tough”, which upset many people (Vlahos, 2015). When commenting on this controversy, Vlahos (2015) refers to May Halim, an assistant professor of psychology who studies gender identity. Halim notes that giving a voice to Barbie only increases her potential impact, and Barbie’s messages could ultimately influence how kids define being a girl.

Gender forcefully defines everyday behavior, especially in its use in marketing, which further entices us to buy into gender constructs.

Gender forcefully defines everyday behavior, especially in its use in marketing, which further entices us to buy into gender constructs. Gendered toys are assumably the most well-known and visually comprehensible examples of segmentation in society (Mascheroni and Holloway, 2017). A visit to any local toy store glaringly reveals the “pinkification” of store sections and the toy industry at large. Smart toys, however, bring a new dimension to how the industry employs segmentation (Mascheroni and Holloway, 2017). Studies have extensively researched definitive identifiers like colors, packaging, logo, etc., in traditional toys that support gender stereotypes (Mascheroni and Holloway, 2017). Still, there is little research on how the reinforcement of gender norms in smart toys pattern child-toy interactions.

Hilu’s (2016) work is idiosyncratic amidst a sea of research that discusses visual examples of gender segmentation in traditional toys. The study, instead, examines how doll play has been historically constructed and explores how microchips and microprocessors only reinforced and redirected existing girlhood practices. Hilu states, “Computer talking dolls exhibit tension between positioning girls as productive and technologically equipped, on the one hand, and conforming to a nurturing femininity believed to be natural to girlhood, on the other” (2016). Mascheroni and Holloway (2017) display to a similar sentiment when describing the Hatchimal toy. They characterize a Hatchimal as a technological or “interactive puppy” that requires care (Mascheroni and Holloway, 2017). The hatchimal facilitates gender differences by tying ideas of girlhood and motherhood to nurturing. Both these studies consider how behavior or phrases spoken in interactive dolls consist primarily of requests for care and expressions of affection.

Talking dolls could shape voice itself as they mediate the boundary between meaningful sound and meaningless noise in the exercise of voice control.

Even though Hilu’s study does not specifically talk about smart toys, the paper explains how voice control plays a significant role in talking doll technologies. Hilu (2016) claims that, in the case of talking dolls, the technological mediation of girlhood is achieved through the disciplining of the voice. This disciplining, she says, is a toy’s way of managing girls’ speech and the sounds of their voices. Furthermore, she argues that talking dolls could shape voice itself as they mediate the boundary between meaningful sound and meaningless noise in the exercise of voice control. I have noticed that many children, especially in early childhood, use noises or onomatopoeic words as form of communication. With limited scripts, Hilu points at how talking dolls could encourage girls to think of their voices as a means of activating a corresponding response from the doll, rather than treating voice as a medium for expression.

Even amidst the advances in smart toys, there are still recurring gendered functionalities that remain largely unchanged (Mascheroni & Holloway, 2017). Two examples of explicitly gendered smart toys in today’s toy market include Mattel’s Hello Barbie and Genesis Toys’ My Friend Cayla. Hello Barbie was introduced as “the first fashion doll that can have a two-way conversation with girls” including speech recognition and cloud computing technologies (Hung et al., 2016). Hello Barbie also has a complex script, with 8,000 lines of dialogue (Smiley, 2016). Vlahos (2015) visited Mattel’s campus and observed their child-testing specialist introduce a Hello Barbie prototype to a little girl around the age of seven. Vlahos considered Barbie’s ability to remember answers and use them for conversation topics days or weeks later as one the toys most “unnerving powers.” Hello Barbie has since sparked a lot of controversy over privacy concerns (Smiley, 2016). As a result, extensive research has been pursued on the privacy and security flaws of this doll. There is little research, however, on how the conversational interface of Hello Barbie—such as the scripted lines—could be problematic in girls’ play.

Other studies have included research in how child-robot interaction could shape the way children think and understand. Vlahos (2015) refers to developmental psychologist Jean Piaget, who wrote the pioneering book The Child’s Conception of the World, in which he asks, “Does the child attribute consciousness to the objects which surround him, and in what measure?” Smart toys can blur a child’s understanding of something alive and not-alive, animate and inanimate, human-operated and autonomous (Spektor-Precel and Mioduser, 2015). Druga, Williams, & Resnick (2017) echo these thoughts by referencing the research of Sherry Turkle, who wrote the highly influential 1984 book The Second Self. Turkle (1991) argued that computers, as objects that exist somewhere between the animate and the inanimate, prompt people to reexamine their own minds. These studies implicate that a child’s understanding of whether or not a toy is real could play a role in how much children trust and believe a toy’s response.

By asking uncertainty to sit at our table, we, as designers further question the role objects and systems play in our everyday life.

While literature concerning children’s understanding of traditional toys is vast, little is known about children’s perception of smart toys and how these interactions could impact an understanding of gender in early childhood development. More specifically, there are still gaps in how conversational and gestural interfaces in a smart toy affect an understanding of gender in early childhood development. We can use design as a lens to illuminate the social implications of gender construction in smart toys. Design is also a powerful tool that can help transform our uncertainty about future child-toy interactions into tangible challenges that we can overcome. By asking uncertainty to sit at our table, we, as designers further question the role objects and systems play in our everyday life. Embracing uncertainty offers a different perspective to gender-biased thinking in interface design practices. By developing a critical sensibility in our designs, we will move beyond these traditional applications and, instead, shape our desired future in child-toy interactions.

References

Cagiltay, K., Kara, N., & Aydin, C. C. (2014). Smart Toy Based Learning. In Handbook of research on educational communications and technology (pp. 703-711). Springer, New York, NY.

Druga, S., Breazeal, C., Williams, R., & Resnick, M. (2017). Hey Google is it ok if I eat you?: Initial explorations in child-agent interaction. IDC 2017 – Proceedings of the 2017 ACM Conference On Interaction Design and Children, 595-600. Association for Computing Machinery, Inc.

Hilu, R. (2016). Girl Talk and Girl Tech: Computer Talking Dolls and the Sounds of Girls’ Play. The Velvet Light Trap, (78), 4-21. doi: 10.7560/VLT7802

Hung, P. C., Iqbal, F., Huang, S. C., Melaisi, M., & Pang, K. (2016, July). A glance of child’s play privacy in smart toys. In: Sun X, Liu A, Chao H-C, Bertino E (eds)  International Conference on Cloud Computing and Security (pp. 217-231). Springer International Publishing, Cham. doi: 10.1007/978-3-319-48674-1_20

Mascheroni, G., & Holloway, D. (2017). The Internet of Toys: A report on media and social discourses around young children and IoToys. DigiLitEY. http://digilitey.eu.

McReynolds, E., Hubbard, S., Lau, T., Saraf, A., Cakmak, M., & Roesner, F. (2017, May). Toys that listen: A study of parents, children, and internet-connected toys. In Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems (pp. 5197-5207). Association for Computing Machinery, Inc.

Plowman, L., & Luckin, R. (2004). Interactivity, interfaces, and smart toys. Computer, 37(2), 98-100. doi: 10.1109/MC.2004.1266302

Smiley, L. (2016, September). When toys talk (and listen). The California Sunday Magazine. https://story.californiasunday.com/when-toys-talk-and-listen

Spektor-Precel, K., & Mioduser, D. (2015). 5-7 Year Old Children’s Conceptions of Behaving Artifacts and the Influence of Constructing Their Behavior on the Development of Theory of Mind (ToM) and Theory of Artificial Mind (ToAM). Interdisciplinary Journal of e-Skills and Lifelong Learning, 11, 329-345. http://www.ijello.org/Volume11/IJELLv11p329-345Spektor1973.pdf

Tanaka, F., & Kimura, T. (2009, September). The use of robots in early education: a scenario based on ethical consideration. In Proceedings of the 18th IEEE International Symposium on Robot and Human Interactive Communication (pp. 558-560). https://doi.org/10.5898/JHRI.1.1.Tanaka

By Bree McMahon

The ability to hold collaborative conversations is crucial to supporting sustainable and diverse communities—whether physical or virtual—including effective real-time and ongoing conversations. Communities provide a sense of belonging through intimate connections and are an important sociological characteristic of human existence (Anderson, 2006). Communities are sustained through sharing, participation, and developing relationships. In order for these activities to occur, users need to participate in particular conversations that establish shared values and goals (Tunstall, 2008). Conversation assists individual and group negotiations that lead to shared understanding of why the community exists. Designers have a role to play in “raising or widening the circle of participation” which will foster the growth of any community (Putnam, 2001). Social interaction and participation not only builds relationships, but are also crucial to the survival of a community. Relationships are also developed through conversation. Therefore, communities that foster conversation are more likely to have a sense of solidarity and a stable foundation (Anderson, 2006).

A designer interested in designing for conversation or communities will engage with uncertain and unstable factors.

A designer interested in designing for conversation or communities will engage with uncertain and unstable factors. For example, when working with multiple participants, the conversation will always be unique to that specific grouping. Tools should be designed to accommodate varying scenarios with diverse participants. If conversations occur online, it is also likely that participants are geographically dispersed, which contributes to uncertainty and complexity when it comes to design. Users and participants are unpredictable in many ways; thus designers are responsible for anticipating the needs of community members. To understand how users interact with their digital communities, designers may turn to Elizabeth Tunstall, as well as Etienne Wenger, Nancy White and John D. Smith.

Elizabeth Tunstall’s Dimensions of
Online Communities

People experience their communities through five dimensions: agency, life goals, historical consciousness, organizational structure, and relationships.

Elizabeth Tunstall, a design anthropologist, identified five aspects through which people experience their communities. Also referred to as “dimensions,” the five aspects serve as a means to support this investigation. Tunstall was inspired by Benedict Anderson’s concept of imagined communities, which addresses the impact of digital communities. According to Anderson, members may “live in the image of their communion” without knowing, meeting, or hearing their fellow-members. People experience their communities through five dimensions: agency, life goals, historical consciousness, organizational structure, and relationships. Agency refers to a person’s ability to control, or at least influence, decisions about the things that impact their communities and themselves. Life goals represent the opportunity for people to articulate what matters most to them. Historical consciousness is a person’s ability to openly express their history, where they come from, and who they are to find a sense of belonging in their community. Organizational structure speaks for itself as it provides people with an understanding of how they fit into and contribute to the greater whole or the community. Lastly, relationships are the basic units of the community and how people establish trust, understanding, and reliability (Figure A).

Figure A: Diagram based on Elizabeth Tunstall’s Five Dimensions of Online Communities

Wenger, White and Smith Community Orientations

Within online platforms, participants experience their community in many different ways and will engage in a variety of activities. The activities are categorized into nine orientations: meetings, projects, access to expertise, relationships, context, cultivation, individual participation, content, and open-ended conversations (Figure B). Orientations are standard patterns of activities through which members experience the feeling of being a community. Tools serve as a means of support for the orientations and activities. Communities are flourishing when significant orientations complement aligning technologies. Orientations are not mutually exclusive, although communities may have primary or secondary orientations. The varying degrees of orientations contribute to the style, personality, and distinction of unique communities. As a community grows and evolves, it is likely that the orientations will change over time, and therefore so will the configuration of the supporting technologies (Wenger, White, & Smith, 2009).

Figure B: Diagram based on Wenger, White and Smith’s Nine Community Orientations

Conceptual Framework (The Matrix)

I devised the conceptual framework to guide the design and selection of tools for specific online communities.

Using the methods mentioned above, I devised the conceptual framework to guide the design and selection of tools for specific online communities. Additionally, the framework provides the means to understand the implications of various tools. Users experience their community through five dimensions outlined by Tunstall (2008). To take part in experiences, they participate in a variety of activities (Wenger et al., 2009). When it comes to digital communities, specific tools facilitate various interactions. The tools needed by users are determined by the designer, but also by the type of activity that needs to occur, as well as the dimension that the activity supports.

Wenger et al. (2009) categorized activities that occur within online communities into nine broad orientations that represent typical patterns of events and connections. Specific tools often found in digital platforms support the activities within each orientation. Activities and supporting tools are not unique to each orientation; rather they can be used across online platforms to accomplish a variety of tasks. In place of using the orientations to guide tool selection and design, I categorized the supporting tools into five buckets that represent a broad range of functions across all nine community orientations: Messaging tools, Knowledge Sharing tools, Validation and Input tools, Searching tools, and Networking tools.

The conceptual framework created using Tunstall (2008) and Wenger et al. (2009) acknowledges that dimensions of a community have orientations of activities that implicate technology selection in the form of tools.

The five tool buckets and Tunstall’s five dimensions of online communities form a matrix that enables a designer to consider features and functions that are needed to best support all aspects of experiencing an online community. This framework could be applied to the design of other online communities; however, the selection of tools should also consider the overall function and mission of the community in question. Overall, the conceptual framework created using Tunstall (2008) and Wenger et al. (2009) acknowledges that dimensions of a community have orientations of activities that implicate technology selection in the form of tools. Consequently, the tools support the dimensions of the community.

The conceptual framework I developed is particularly important and relevant for designing digital habitats because it recognizes the many and diverse needs of online communities and their members, but also calls for intentional design that meets those needs (Figure C).

Figure C: Conceptual framework matrix

References

Anderson, B. (2006). Imagined communities: Reflections on the origin and spread of nationalism. London: Verso Books.

Putnam, R. D. (2001). Bowling alone: The collapse and revival of American community. Simon and Schuster.

Tunstall, D. (2008). Unpublished workshop held at NCSU College of Design Department of Graphic Design.

Wenger, E. (1998). Communities of practice: Learning, meaning, and identity. Cambridge university press.

Wenger-Trayner, E., & Wenger-Trayner, B. (2015). Introduction to communities of practice: A brief overview of the concept and its uses. Grass Valley, CA: Wenger-Trayner. Retrieved from http://wengertrayner.com/introduction-tocommunities-of-practice/.

Wenger, E., White, N., & Smith, J. D. (2009). Digital habitats: Stewarding technology for communities. CPsquare.

By Rachael L. Paine

Wewere charged to rethink and redesign the culture of graphic design in the context of contemporary practices, application, and usage (Gonzales Crisp, 2017). My investigation led me to look at cultural attitudes of ownership, self-aggrandizement, and control. Is there potential to revolutionize these attitudes to create a space of shared purpose and focused action?

My research led me to hypothesize that a designer’s schema and knowledge patterns can limit innovative leaps forward in the design process. I propose developing processes that remove this rigidity through the relinquishing of control and ownership of ideas. I began designing methods which would implement the rotation of thinking throughout design processes.

To radically introduce these new attitudes, I designed a method of rotation through the design process (Figure A). Starting with a prompt, designer-1 would develop an idea and pass it to designer-2, who would then invest time researching and mapping concepts surrounding the idea. Designer-2 would pass the mapping to designer-3, who would develop a framework within which to situate design iterations. One additional round of passing, and finally designer-4 would be producing a prototype for the original prompt. This proposed process would result in concepts and work that the individuals would be unable to produce independently, serving as a creative form of groupthink.

Figure A: Under the radicalized conditions, I am proposing a forced dispersal of ownership. Methods will be implemented to rotate thinking throughout the design process.

Revolutionized design processes have the power to eliminate the overly-emotional, work hoarding, competitive, hierarchical climate in design groups.

As designers become accustomed to such processes, old attitudes will be replaced, creating normalized conditions (Figure B). A dispersal of ownership will become the norm, and designers in groups will no longer work in a state of fear that someone might steal an idea, get the best opportunity, or win the favor. Creativity and flow of information will serve everyone equally.

Figure B: Under the normalized conditions, I am proposing a database of shared knowledge. Ideas, research, and mapping will be collected into a knowledge database. Processes will be implemented for random distribution of that knowledge to stem new design innovation.

Ideas, research, and mapping will be collected into a knowledge database. Processes will then be implemented for random distribution of that knowledge to stem new design innovation.

I designed an interface prototype for the radicalized condition (Figures D-G). The interface would serve as a remote-access co-working space that forcefully eliminates the ownership of ideas. The space will have a set of rules, ethics, and customs, and will implement Robert’s Rules of Orders to bring certain motions before the group such as “randomly rotate artboards” amongst members, to all “join on one design,” or to “request additional project information” (Figure H).

Figure C: A motion animation of the radicalized co-working space.

Figure H: Although any idea can be proposed for consideration, a few common motions for group work are built into the co-working environment.

Figure G: When a motion is made, web chat is enabled for discussion and voting procedures.

Figure F: At certain intervals, artboards will be randomly rotated. Designers can make a motion to rotate artboards earlier than the allotted time.

Figure E: A designer can choose to swap workspaces with another designer or take elements off their artboard.

Figure D: A designer can see what all the other designers are working on at any time.

Robert’s Rules of Order (Robert, Honemann, & Balch, 2011) is a widely used manual of parliamentary procedure. This book is designed as a set of operational rules to be implemented by organizations. These procedures include guidelines for making motions, conducting debates, facilitating votes, and coming to a place of unanimous consent (Figure I). The original creator of the rules, Henry M. Robert, stated, “Where there is no law, but every man does what is right in his own eyes, there is the least of real liberty.”

Figure I: The digital co-working space is a deliberate assembly of designers and Robert’s Rules of Order lays the foundation for the “rules of play” for participating in the shared design practice.

So, we approach a paradox of sorts. We create a co-working space where no one has individual authority to their own time, ideas, resources, or work, and must submit to the agreed upon rules of the group. But perhaps we uncover real freedom of innovation. When the fear is gone and we release the white knuckling grip of climbing the ladder of creativity, what might we master in turn?

In time, the radical will cease to be radical. Sharing ideas will be conventional. To support the normalized condition, I designed a second interface (Figures K-N). The goal of the software, NODE, will be to take the ego out of ideas. A user will have the opportunity to both give and take from the knowledge database. A user will start a new project by following a set of prompts and sifting through collected ideas.

Figure J: A motion animation of NODE prototype in action.

Figure K: The goal of the software will be to take the ego out of ideas.

Figure L: The software will allow you to either GIVE to the collaborative knowledge database or TAKE from it.

Figure M: The user will be able to choose one element of user-centered design (productivity, comprehension, usability, accessibility, sustainability, serenity) and an artifact type (mobile, web, print, video, illustration, installation) as a starting point.

Figure N: Once you’ve chosen your initial design objectives, the software will be populated with pertinent node maps.

As designers, can we create deliberate processes that grant continued creative pursuit, allowing space for ideas to unfold? I believe we can. There is unlimited potential in trying to do so. If even one designer is influenced to transition their thinking from “I need to have all the answers” to “my curiosity will spark creative ideas in others,” then we are surely moving the culture of graphic design in the right direction.

References

Gonzales Crisp, D. (2017, August 16). Design as a cultural artifact [class syllabus]. Content posted to GD502_FA17 Syllabus Google Docs Spreadsheet.

Robert, H. M., Honemann, D. H., & Balch, T. J. (2011). Robert’s rules of order: Newly revised (11th ed.). Philadelphia, PA: Da Capo Press.

By Jessye Holmgren-Sidell

As designers, we have, and should embrace, the powerful opportunity to construct customizable interfaces that help restrict government access and restore user autonomy.

Weare all activists now,” says cybersecurity counsel Jennifer Granick in her 2017 TED Talk. “And that means we all have something to worry about from surveillance.” She goes on to explain, in detail, how the American government collects our online data “easily, cheaply, and without warrant” (Granick, 2017). In 2013, Edward Snowden exposed thousands of classified NSA documents detailing the surveillance measures used on United States citizens. As designers, we have, and should embrace, the powerful opportunity to construct customizable interfaces that help restrict government access and restore user autonomy. And yet, there is still little protection in place to stop data collection from happening through digital platforms. To incorporate surveillance protection in the current User Experience (UX) design process, we must design for user safety rather than just efficiency, change the frequently hostile language and imagery we use to represent security, and communicate directly with security experts.

The US government acquires our data through online services and mobile applications like Facebook, Amazon, LinkedIn, and Google. What is not so apparent is that users often willingly provide access to that information. Ame Elliott, Design Director for nonprofit security organization Simply Secure, explains that UX designers create interfaces that utilize “the path of least resistance” (Elliott, 2018). LinkedIn, for example, asks new members to share their address book with just a simple click; it is far easier to hit the large “share” button than find the (much less apparent) “X” to skip that part of the registration. “The truth is people have no interest in using applications or websites,” says UX expert Paul Boag. “They are tools for a goal. [Users] want to use your website or application for the smallest amount of time” (Boag, 2016). In many cases, the path of least resistance forces users to reveal personal information. And the consequences can be disastrous.

There is no guarantee how companies will utilize or protect collected data and that uncertainty threatens user safety. In some cases, they pair shared information with machine learning to tailor experiences. LinkedIn generates specific job postings and suggested connections for members; Facebook’s algorithm caters ads and news to users based on recorded interests; Amazon utilizes user search history to better recommend products for its customers. In all of these cases, machine learning improves or, at least, streamlines user experience.

In March of 2018, however, The New York Times and The Guardian revealed that Cambridge Analytica “accessed data of about 50 million Facebook users” (McKinnon, 2018). Researcher Aleksandr Kogan designed a personality-quiz app for the social media platform that asked users for access to their profile pages. He then sent that recorded data to Cambridge Analytica to make 30 million voter targeting profiles. Facebook maintains the quiz breached none of their systems, but journalist Robinson Meyer explains, “It’s almost like Facebook was a local public library lending out massive hard drives of music, but warned people not to copy any of it to their home computer” (Meyer, 2018). A warning is not encrypted protection. Social media asks users to share parts of their lives online, but with the understanding that users control who views those shared moments. By following the path of least resistance and allowing a supposedly harmless quiz to access their profiles, millions of people involuntarily compromised their data. Facebook allows security measures to be outweighed by streamlined user experience.

Before designing for the path of least resistance, we should understand that path’s real purpose. LinkedIn requests users to share their address books to help connect them with employers and opportunities, but LinkedIn is also a service that needs members. By sending out invitations to everyone in a user’s address book, it reaches potential new clients who will have to register on the platform to accept the invitation. We also need to recognize what the path is bypassing. Users share their entire address books to avoid individually selecting who can view their profiles. That would be tedious and time consuming. In doing so, however, they give up control and lose autonomy over the process. Finally, we must consider the path’s consequences. These can range from users inundating everyone they know with LinkedIn friend requests to giving a “voter-profiling company” the data needed to target them during an upcoming election. With these considerations in mind, we can re-configure the path of least resistance to incorporate user safety, even if that just means making the “X” out option bigger.

Online security iconography and verbiage focuses so much on keeping threats out, that it forgets to let users in.

Security services frequently use negative language and imagery to represent their products. Ame Elliott calls this practice “the language of no” and maintains that it deters potential users from installing protective software (Elliott, 2018). Cybersecurity company Symantec, for example, offers defense methods that “Protect against tomorrow’s attack” and “Sharpen your responses after an attack and prevent the next one.” The website’s aggressive tone implies that users are responsible for security attacks because they were not “sharp” enough to recognize obvious threats to the system. Proficio, another cybersecurity service, represents incident response with a cross-hairs icon; the US Department of Homeland Security uses a picture of a lock to link to its cybersecurity overview page. These graphics attempt to scare clients into secure behavior—do not open that link, do not download that file, or attack is imminent. Online security iconography and verbiage focuses so much on keeping threats out, that it forgets to let users in.

We can change the “language of no” to the “language of yes.” “You don’t need to be a cryptographer to work in security…You don’t need to be technical,” says Elliott (2018). Indeed, designers are integral to the cybersecurity field because it seems so technical and unapproachable. And just because security involves technology does not mean our designs have to be technical or cryptic. We have the opportunity to help create products and services that encourage users to secure their data without resorting to scare tactics. TunnelBear, for instance, is a virtual private network that uses fun and friendly imagery to explain its functionalities. Images show the mascot, a cartoon bear, physically blocking users’ faces to protect them from online surveillance. “Browse privately with a bear,” the website reads. “It’s easy to enjoy a more open Internet.” The language is humorous, with no mention of “attacks” or “threats” to the system. TunnelBear makes security approachable and inviting, a practice we can and should utilize more frequently.

Our expertise and research can help prevent security experts from making assumptions about users’ behaviors.

In order to make cybersecurity understandable, however, we must first communicate directly with security experts. According to Sara “Scout” Sinclair Brody, the Executive Director of Simply Secure, “Neither security nor usability are binary properties. There’s a lot of grey area when it comes to whether something is secure or insecure” (Sinclair Brody, 2016). She explains that security experts ask, “Is this the most secure solution possible?” while designers ask, “Is this secure enough for my user, while not being restrictive?” (Sinclair Brody, 2016). It is, therefore, critical that we know how security experts are incorporating users and their needs into product development. As designers, we conduct interviews to understand how users want to move through an interface and then create personas. Our expertise and research can help prevent security experts from making assumptions about users’ behaviors.

Likewise, security experts can help make our solutions “as secure as possible” to ensure that we protect the users for whom we are designing. We need to understand security jargon to properly translate that information for non-expert consumers. We should be asking security experts questions to facilitate collaboration between our two fields. Sinclair Brody (2016) explains that designers must know to which security threats our shared project is most vulnerable and how its software will protect against those threats. We can then consider how users put themselves at risk and design personas that reflect those specific actions. This has already been put into practice by security and usability designer Gus Andrews, who created personas with a range of privacy concerns and potentially risky behaviors. He intended for them to “communicate user needs” to security experts in the terms those experts provided (Andrews, 2015). By utilizing personas like Andrew’s and continuing conversations with security experts, our designs will keep users secure without restricting their experience or following the path of least resistance.

We can avoid creating for the path of least resistance by determining the path’s real purpose, what it is bypassing, and the full extent of its consequences.

As the US government continues to collect citizens’ data without warrant—as LinkedIn uses the path of least resistance to remove autonomy, as Facebook “quizzes” convey data to voter-profiling companies without permission—we must integrate surveillance protection in our user experience designs. We can avoid creating for the path of least resistance by determining the path’s real purpose, what it is bypassing, and the full extent of its consequences. Additionally, we can improve users’ relationships with cybersecurity services by incorporating positive language and imagery, while also communicating directly with security experts on joint projects. In implementing these changes, we will create a safer, more pleasant online environment for users, thereby optimizing users’ experiences. Designers not only have the opportunity to alter the way people perceive cybersecurity, but the responsibility to invest our user-centric methods in protecting the public from surveillance.

References

Andrews, G. (2015, April 14). User Personas for Privacy and Security. Retrieved from https://medium.com/@gusandrews/user-personas-for-privacy-and-security-a8b35ae5a63b

Boag, P. (2017, July 19). Users always choose the path of least resistance. Retrieved from https://boagworld.com/marketing/users-will-always-choose-the-easiest-option-so-if-we-want-a-competitive-advantage-we-must-focus-on-simplicity/

Elliot, A. (2017, February 28). Pre-Work Talk Berlin 02/2017 – Designing for Trust. Retrieved from https://www.youtube.com/watch?v=lOt_mc9FRDg&list=PLgKQebNo0trgNxpfvAF2u6KkybOeJju8l&index=5

Granick, J. (2017, April). How the US Government Spies on People Who Protest – Including You.  TED. Retrieved from https://www.ted.com/talks/jennifer_granick_how_the_us_government_spies_on_people_who_protest_including_you

Meyer, R. (2018, March 20). The Cambridge Analytica Scandal, in 3 Paragraphs. The Atlantic. Retrieved from https://www.theatlantic.com/technology/archive/2018/03/the-cambridge-analytica-scandal-in-three-paragraphs/556046/

McKinnon, J. D. (2018, March 20). FTC Probing Facebook Over Data Use by Cambridge Analytica. The Wall Street Journal. Retrieved from https://www.wsj.com/articles/ftc-probing-facebook-over-data-use-by-cambridge-analytica-1521561803

Sinclair Brody, S. (2016, July 5). Talking Across The Divide: Designing For More Than “It’s Secure”. Retrieved from https://simplysecure.org/blog/talking-across-divide

By Ashamsa Mathew

Throughout my explorations, I acted as a critic, using the system I developed to take on a rhetorical position.

We were tasked with creating the beginnings of a sovereign identity system in a technological future for an imagined state (Peterson, 2017). Throughout my explorations, I acted as a critic, using the system I developed to take on a rhetorical position.

I chose Jammu and Kashmir (J&K), a northern state in India, as my contemporary faction. J&K has been engaged in a struggle for peace for many years. In my scenario, the state breaks away from India to become a part of Pakistan, in the hopes of establishing stability.

History

To contextualize this imagined situation, we must first understand the state’s complex history. The end of the British rule in India was a time of both independence and separation. In 1947, part of India elected to break away from the nation and form Pakistan. Every princely state was given the option to remain part of India, join the newly formed Pakistan, or become an independent nation. While the Maharaja of J&K was making this decision, the Pushtoon tribe and citizens from the western districts invaded the state with support from Pakistan. Initially, the Maharaja fought back independently, but eventually requested military assistance from India. The government agreed to help under the condition that the the J&K Maharaja pledge temporary allegiance to India. The Maharaja signed the Instrument of Accession on October 26, 1947. The Indian government accepted the accession, but stated that it was provisional, as only the people, and not the Maharaja, could decide J&K’s fate. The public would formally make the decision once the state was freed from the invaders.

Thus began a series of wars which turned the beautiful state into a bloodbath. A special United Nations Commission for India and Pakistan (UNCIP) negotiated the withdrawal arrangements to establish peace, but Pakistan refused to remove its forces, stating India would stay once it had left. Likewise, India refused to leave until Pakistan withdrew its troops.

Since India acquired J&K as a provisional accession, the nation gave J&K certain privileges. The state had its own constitution under Article 370, and its flag held the same level of significance as the Indian national flag. Indians from other states could not purchase land or property in the state. Additionally, “under Article 370 of the Constitution of India, according to which no law enacted by the Parliament of India, except for those in the field of defence, communication and foreign policy, will be extendable in Jammu and Kashmir unless it is ratified by the state legislature of Jammu and Kashmir” (Makkad, 2012).

The Separation

After years of bloodbaths, rape, and countless other horrors, the war has ended. The UN finally negotiates a peace treaty in J&K, to which both India and Pakistan agree. The public must now decide if they want to remain independent or join one of the two countries. Monitored by the UN, the people of the state cast their vote.

Figure A

The world is shocked by the final verdict (Figure A), although J&K’s population is not surprised. They explain their reasoning, noting that:

•  The majority of the populace are Muslims, meaning the community is a pro–Islamic nation.
•  Over years of war, India conveniently forgot that J&K was a provisional accession and took away the autonomy the state enjoyed.
•  The Indian army was not a noble protector. They raped women, harmed citizens, and committed countless other horrific acts.
• The state is not strong enough to function on its own, especially after decades of conflict.

Figure B

The state agrees to be a part of Pakistan, provided Pakistan’s government allows the public to maintain their secularist ideologies and does not discriminate against minorities on the basis of education, employment, and religion.

Flag

J&K and Pakistan claim to accept each other’s differences and present a united front; sadly, in my scenario, this symbolism is only theoretical.

The new state flag (Figure C) symbolizes unity. J&K and Pakistan claim to accept each other’s differences and present a united front; sadly, in my scenario, this symbolism is only theoretical.

Figure C

• Moon –  The moon acknowledges Pakistan’s Islamic ideals.
• Lotus –  The lotus represents the newly joined nation’s 3 main religions and signifies how their differences create a beautiful entity. The lotus is also the state flower.
• Stars –  The 3 stars represents J&K’s 3 regions. There are 22 districts in the new, shown by the 2 seven pointed stars and 1 eight point star.
• Red –  The red background represents the martyrs who shed their blood to protect the people of the J&K.

Emblem

This new identification system subtly promotes discrimination by playing upon people’s preconceived perceptions of certain religions and regions.

The government assigns emblems to every citizen based on his or her religion, gender, and region (Figure D). The mark masks any favoritism by combining all symbols into one tattoo. This new identification system subtly promotes discrimination by playing upon people’s preconceived perceptions of certain religions and regions. Subconsciously, people behave differently toward individuals based on their respective emblems.

Figure D

The personal emblem functions similarly as fingerprints or social security numbers (Figure E). The ink mixes with DNA and can only be applied and removed by government officials. The tattoo gives people access to their phones and official information, but also makes it easier for the government to track citizens’ activities.

Figure E

The state emblem depicts all regions, religions, and genders in an attempt to promote peace and harmony (Figure F). The government prints the image on all its documents and certain officials have it tattooed on their skin to provide them access to confidential information and restricted areas.  

Figure F

Propaganda

The government uses personal emblems to control what information citizens receive through the media. This feature allows the government to push its own propaganda, altering people’s opinions by playing on their emotions and fears.

Aim: To convert the public to Islam and maintain control over the country through these shared beliefs.

Strategy: Manipulate the public’s emotions to instill a pervasive sense of fear.

I have designed the propaganda system in this imagined scenario to be implemented well into the future, when the internet and apps primarily utilize virtual and augmented reality.

Figure G

The government subtly fulfills its agenda using human psychology.

Figure G is the news as seen by a man with a significant woman in his life, such as a daughter, wife, or sister. The government wants to prevent women from going outside late at night. To promote this agenda, it shows this user a news story about how a Hindu girl was raped while out walking late. The VR landscape also projects a 5% drop in rape amongst Islamic women, implying that there is a correlation between that statistic and conservative Islamic ideals. The male user views his government-generated news feed and fears for the women in his life and does all that he can to prevent them from going out at night. The government subtly fulfills its agenda using human psychology. This way, they are never seen as the villains.

Figure H

Figure I

Figure J

Figure K

As I designed this sovereign identity system, I began to question how we might avoid government surveillance as the world moves towards a technology-driven future.

My proposed scenario is a far cry from democracy and secularism. As I designed this sovereign identity system, I began to question how we might avoid government surveillance as the world moves towards a technology-driven future. Even today, we face countless cybersecurity scandals, from social media giant Facebook misusing private user data to the US government listening to citizens through wiretapping devices. As designers, I believe we possess the tools and training to collectively design systems that are both inclusive and protective. The question is, how we maintain a balance between the two.

References

Peterson, M. (2017, October). A modest proposal. Content included on poster template.

Makkad, R. K. (2012, March 27). Why special status to j & k? – Constitution – Constitutional Law. Retrieved from www.lawyersclubindia.com/forum/Why-speical-status-to-j-amp-k–54474.asp.

By Ellis Anderson

Definitions

Improvisation can be described as the process of thinking and acting on your feet in response to new and shifting phenomenon, often times linked to chance results when faced with sudden provocation (Kamoche & Cunha, 2001). While there are multiple existing definitions of improvisation, Stephen Leybourne outlines a number of key improvisational constructs as seen in both jazz music and agile project environments (Leybourne, 2009):

Processes

Improvisation uses social and technical structures to allow for cohesion within the group, creating an inclusive experience with an informed audience, who also operate within certain expectations.

Improvisation is most commonly associated within a group context such as jazz ensembles. Jazz is inherently an inventive and collaborative process where all members of the group move in unison towards a collective goal (Bastien & Hostager, 1988). Its unique brand of collaboration requires that its members be especially attuned to each other’s voices and input in order to maintain momentum and cohesive direction. Improvisation uses social and technical structures to allow for cohesion within the group, creating an inclusive experience with an informed audience, who also operate within certain expectations (Sawyer, 2000).

In jazz, social structure can be understood as a form of egalitarian ‘etiquette,’ which provides an interactional framework for collaboration and creative production (Sawyer, 2000). In another way, social structure is defined as predetermined roles and responsibilities within the group, a ‘collective mind’ towards experimentation among team members (Kamoche & Cunha, 2001). Trust is also a significant component to the success of improvisation. In jazz improvisation, trust is defined in “three dimensions: ‘consistency trust’ (people will do what they said they would); ‘competence trust’ (having faith in other’s abilities) […]; ‘goodwill trust’ — this refers to openness and goal congruence” (Kamoche & Cunha, 2001). Much like the idea of cohesion that Magni, Proserpio, Hoegl, & Provera (2009) propose in their study of how behavioral integration shapes individual improvisation, trust is key to improving directional focus or “togetherness” within the group. In addition to social structure, technical structure provides another layer of variable control within the jazz environment. Technical structure is defined through compositional frameworks, song structures and the use of licks or “musical grammar” (Bastien & Hostager, 1988), historically located clichés, and “culturally shared scripts for conversation” (Sawyer, 2000).

Outcomes

Team cohesion ultimately impacts an individual’s comfort and feeling of acceptance among their teammates.

In their study of information system developer teams, Magni et al. looked at how different industries employ improvisational techniques into their workflow. In their analysis, they found that behavioral integration and cohesion are directly tied to an individual’s ability to respond with solutions to emerging uncertainty (Magni et al., 2009). In their research, they describe cohesion as the social bond between team members towards a unified goal, a sense of community, or a feeling of loyalty to the group (Magni et al., 2009). Team cohesion ultimately impacts an individual’s comfort and feeling of acceptance among their teammates. Behavioral integration is defined as the ability for all members of a group to freely interact, share and absorb information, cooperate, collaborate, and receive feedback from other members (Magni et al., 2009). Their research findings ultimately support their original hypothesis; the level of behavioral integration and team cohesion directly corresponds to an individual’s ability to improvise.

Keith Sawyer’s research in group creativity mirrors many of these same findings, highlighting common themes found in “effective creative teams” (Sawyer, 2007). These themes closely resemble the behaviors found in improvisational groups. Sawyer relates group creativity to Mihaly Csikszentmihalyi’s idea of “flow” or a “particular state of heightened consciousness” (Sawyer, 2007). Sawyer, a student of Csikszentmihalyi, adapts this idea towards his own concept of “group flow,” and provides guidelines for establishing group flow within creative teams (Sawyer, 2007):

Innovation in improvisation comes from a specific kind of relationship between actors within a group, not from a predetermined end goal.

It’s important to note that innovation in improvisation comes from a specific kind of relationship between actors within a group, not from a predetermined end goal. “[Collaborative musical performance] is the paradigmatic example of a form of human interaction in which the processes of engagement, innovation, and ensemble coordination—rather than outcome—are the goals of interaction” (Healey, Leach, & Bryan-Kinns, 2005). As Healey et al. (2005) suggest, if future tools and technology seek to increase participants’ capacity for innovation, they should be created with the intention to replicate the experience of interaction within a collaborative music ensemble.

However, some issues may arise from this model. The variability and potential for mistakes inherent with an improvisational approach to generating new ideas may not be appropriate when used in industries requiring maximum consistency and reliability (Kamoche & Cunha, 2001). Another difficulty is that improvisation functions on a delicate balance of structure and flexibility, which may be hard to implement. The model requires that team members are able to improvise, act quickly, and respond to the group’s momentum/direction; each individual needs to possess an ability to perform within this framework in order for it to succeed (Kamoche & Cunha, 2001). Perhaps the greatest issue with this model is the evaluation of ideas created in the process. Improvisation becomes a useful strategy for generating novel responses to changing criteria but places a stronger emphasis on the generative process itself over specified outcomes (Healey et al., 2005).

Designers must establish some level of improvisation, behavioral integration, and/or cohesion in order to respond to their collaborators in an equitable fashion.

Though the scope of my research is relatively limited, it would appear that the cross-application of improvisational models to various industries has yet to be exhaustively explored or studied. Methodologies that incorporate both structure and flexibility are needed to cope with the increasing pressure to develop outcomes with speed and responsiveness in fast-changing environments such as new product development and technology (Kamoche & Cunha, 2001). Industries that seek to move away from traditional hierarchies and move towards “flexibly-structured learning entities” stand to benefit the most from improvisation (Kamoche & Cunha, 2001). Design is one such example. As designers seldom produce work in a vacuum, they must engage in some form of team collaboration with various stakeholders and potential end-users during the creative process. Designers must establish some level of improvisation, behavioral integration, and/or cohesion in order to respond to their collaborators in an equitable fashion.

Group improvisation is the proactive recognition of equality and respect for collaborators. It is a mutual acceptance of each member’s  ability to inspire everyone’s production. It is also a method for coping with shifting requirements and uncertain futures (Kamoche & Cunha, 2001). The benefits of adopting improvisation are clear. Improvisation can serve as a reactive/reflexive tool to anticipate turbulence and respond with novel solutions. However, the designer must first understand the subtleties of improvisational structures when applying them in their own practices if they wish to produce something that is inclusive, innovative, and responsive.

References

Bastien, D. T., & Hostager, T. J. (1988). Jazz as a process of organizational innovation. Communication Research, 15(5), 582-602. doi:10.1177/009365088015005005

Healey, P. G., Leach, J., & Bryan-Kinns, N. (2005). Inter-play: Understanding group music improvisation as a form of everyday interaction. Proceedings of Less is More—Simple Computing in an Age of Complexity.

Kamoche, K., & Cunha, M. P. E. (2001). Minimal structures: From jazz improvisation to product innovation. Organization studies, 22(5), 733-764. doi:10.1177/0170840601225001

Leybourne, S. A. (2009). Improvisation and agile project management: A comparative consideration. International Journal of Managing Projects in Business, 2(4), 519-535.

Magni, M., Proserpio, L., Hoegl, M., & Provera, B. (2009). The role of team behavioral integration and cohesion in shaping individual improvisation. Research Policy, 38(6), 1044-1053. doi:10.1016/j.respol.2009.03.004

Sawyer, K. (2007). Group genius: The creative power of collaboration. New York: Basic Books.

Sawyer, R. K. (2000). Improvisational cultures: Collaborative emergence and creativity in improvisation. Mind, Culture, and Activity, 7(3), 180-185. doi:10.1207/s15327884mca0703_05

By Shadrick Addy

Historically, African Americans have been subjected to racial discrimination and prejudice, an injustice that is still sewn within American society. With the rise of technology, machine learning promised to improve decision making by removing human bias (Baer & Kamalnath, 2017). Yet, even with the wide use of automated programs in several American industries, people of color are still not given an equal chance at attaining the American dream. The historical implication of racial discrimination and prejudice is missing in the training data of machine learning algorithms. The absence of these historical data points, the transfer of human biases during training, and automated programs’ inability to account for future events are key contributors to algorithmic bias. Building an inclusive future driven by artificial intelligence demands that we understand the cultural implications of algorithmic bias and establish new procedures for training machine learning algorithms.

Defining Bias

Mitchell, a computer scientist and Professor at Carnegie Mellon University, describes bias as, “any basis for choosing one generalization over another, other than strict consistency with the observed training instances” (Mitchell, 1980). Mitchell argues that “learning involves the ability to generalize from past experience in order to deal with new situations that are related to this experience” (Mitchell, 1980). He suggests that “the inductive leap needed to deal with new situations seems to be possible only under certain biases for choosing one generalization of the situation over another” (Mitchell, 1980). Michell’s analysis shows that there is a need for bias in learning generalizations and that removing it is a useless goal. I agree with Mitchell and acknowledge that algorithmic bias in automated programs is inevitable. In addition, removing bias from machine learning algorithms is difficult because of the human bias that informs how developers, stakeholders, and users make decisions regarding machine learning applications.

For people of color, who have historically faced inequality, the presence of human bias in algorithms further enforces discriminatory barriers.

Tobias Baer, a partner and researcher at McKinsey & Company, and Vishnu Kamalnath, a specialist in the North American Knowledge Center, assert that algorithmic bias is one of the biggest risks because it compromises the very purpose of machine learning. The research states that “artificial intelligence is as prone to bias as humankind” (Baer & Kamalnath, 2017). For people of color, who have historically faced inequality, the presence of human bias in algorithms further enforces discriminatory barriers. Because the algorithms make predictions based on past correlations, they amplify the effects of historical prejudice against marginalized populations by reinforcing human bias found within the data set used for training. As designers, we must recognize that computer programs are encoded with human prejudice, misunderstanding, and bias (O’Neil, 2016), in order to establish new procedures to address algorithmic bias.

Algorithmic Bias in Employment

Automation has become an essential part of the hiring process for American businesses (O’Neil, 2016). 60 to 70 percent of prospective U.S. workers’ chances of getting a job are contingent on personality test results (O’Neil, 2016). Furthermore, computer programs can now parse through resumes and rank applicants based on how well they match the criteria for a job position (Abdel-Halim, 2012). Automation in hiring practices is a burgeoning industry, grossing $500 million annually and growing by 10 to 15 percent a year (O’Neil, 2016). For employers, the benefits of automation often outweigh its negative social implications. Yet, hiring processes that use these automated programs reinforce biases against African Americans by wrongly associating name, race, and other social identifiers with an applicant’s inability to fulfill the responsibilities of a position.

False correlations between data points have traditionally led employers to assume the race of applicants simply by looking at the names on their resumes. Can an applicant’s name, for instance, serve as a clear indicator of his race or gender? Could his name also serve as an indicator of his ability to fulfill the responsibilities of a potential job? The answer, of course, is no, for numerous reasons. Anyone can change their birth name through legal proceedings to one commonly associated with another race or gender. Therefore, using names as indicators can result in misidentification. Names are not clear indicators of an applicant’s race, gender, or merit because no empirical evidence establishes a correlation between the data points.

The same human bias present in conventional applicant screening is unconsciously programmed into automated applicant tracking software.

A 2003 study done by researchers from the University of Chicago and MIT revealed that applicants with white-sounding names are 50 percent more likely to receive callbacks than those with black-sounding names (Bertrand & Mullainathan, 2003). The study discovered that white names on a higher quality resumé received 30 percent more callbacks than African American names. The same human bias present in conventional applicant screening is unconsciously programmed into automated applicant tracking software. Resumé analyzing programs, however, are only one sub-category of many automated programs prone to algorithmic bias during job hiring processes.

Personality tests are still being used to determine who gets hired for a job, despite research showing they are poor predictors of job performance (O’Neil, 2016). In 1971, the Supreme Court ruled that job responsibility-determining intelligence tests were discriminatory and illegal (O’Neil, 2016). Yet, these illegal tests are still being used in other forms, to eliminate as many job applicants as possible (O’Neil, 2016). Personality tests and many other job screening programs have become ubiquitous hiring standards. An applicant who is rejected due to bad scores at one business is likely to face a similar fate at another. Cathy O’Neil, the author of Weapons of Math Destruction, explains that while employers had bias during traditional hiring practices, the biases varied at each business (O’Neil, 2016). In contrast to traditional hiring approaches, automated programs are more likely to repeat biased results because systems often share the same training data (and prejudices) amongst businesses. Most job applicants are unaware of the algorithmic correlations that influence the results of the automated programs. The obscurity of the correlations restricts many people of color with financial difficulties from challenging biased algorithmic predictions—predictions that could have severe implications on the livelihood of Americans workers.

Algorithmic Bias in Law Enforcement

The legal system widely implements these predictive algorithms. Research by Garvie, Bedoya, and Frankle (2016) shows that face recognition in law enforcement affects over 117 million American adults. Further investigations revealed that 16 states let the FBI use face recognition technology to compare the faces of suspected criminals to their driver’s license and ID photos (Garvie et al., 2016).  For people of color, who have long faced facial discrimination and inequality, predictive policing and other uses of machine learning in the criminal justice system can lead to devastating consequences.

Extreme care must be taken to address wrongful predictions that could result from algorithmic bias in automated applications used in law enforcement.

Algorithmic bias in facial recognition can lead to wrongful accusations, resulting in devastating consequences for individuals and their families. “Someone could be wrongfully accused of a crime based on erroneous but confident misidentification of the perpetrator from security video footage analysis” (Buolamwini, 2018). Buolamwini, a graduate researcher at the MIT Media Lab and founder of the Algorithmic Justice League, research shows that an underrepresented demographic group in benchmark datasets can nonetheless be subjected to frequent targeting. A 2013 study done by the New York Civil Liberties Union reveals that African Americans and Latinos made up only 4.7 percent of the city’s population, yet, they accounted for 40.6 percent of the stop-and-frisk checks by police (O’Neil, 2016). The presence of human bias in law enforcement means that extreme care must be taken to address wrongful predictions that could result from algorithmic bias in automated applications used in law enforcement.

Addressing Algorithmic Bias

A diverse development team creates an environment that reduces human biases.

A balanced representation of African Americans during program development is the first step to establishing new procedures to remedy algorithmic bias in machine learning. A diverse development team creates an environment that reduces human biases. African Americans programmers can work on development teams to create algorithms that are objective. Another approach to reducing algorithmic bias is establishing procedures that encourage developers to solicit feedback from a diverse population when programming algorithms. User testing with a diverse group can help prevent problematic algorithmic models.

Designers can play an essential role in establishing new procedures to address algorithmic bias. For example, they can work with developers to create interfaces that give agency to users (Borenstein, 2016) and provide transparency on how correlations are made by predictive algorithms. Transparency in algorithmic predictions can help a person determine if a model is working against their interest (O’Neil, 2016). A strong partnership between designers and developers will lead to the development of inclusive procedures to address algorithmic bias. To strengthen their collaborative relationships with developers, designers must develop a deeper understanding of artificial intelligence systems, their affordances, and how humans might use, misuse, and abuse these affordances (Borenstein, 2016).

As a dog is loyal to its trainer, algorithmic bias often serves the interest of its developers.

As we move towards an inclusive future powered by artificial intelligence, machine learning will have both positive and negative implications upon multicultural American society. However, the severity of the negative impact of algorithmic bias requires that new procedures be established to address the cultural implications of machine learning. A necessary step in establishing such procedures starts with the diversification of software development teams. As a dog is loyal to its trainer, algorithmic bias often serves the interest of its developers. A diverse team of developers and designers ensures that there is equal representation in decisions that are made when training algorithms—decisions that could reinforce human biases in machine learning algorithms. If the cultural implications of machine learning are not recognized and addressed, the mantra “garbage in garbage out” could soon become synonymous with “bias in bias out.”

References

Abdel-Halim, M. 12 ways to optimize your resume for applicant tracking systems. Mashable Website. https://mashable.com/2012/05/27/resume-tracking-systems/. Updated 2012. Accessed April 9, 2018.

Baer, T., & Kamalnath, V. Controlling machine-learning algorithms and their biases. McKinsey & Company Website. https://www.mckinsey.com/business-functions/risk/our-insights/controlling-machine-learning-algorithms-and-their-biases. Updated 2017. Accessed April 7, 2018.

Bertrand, M., & Mullainathan, S. (2004). “Are Emily and Greg More Employable Than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination,” American Economic Review, American Economic Association, vol. 94(4), pages 991-1013, September.

Buolamwini, J. & Gebru, T. (2018). Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification. Proceedings of the 1st Conference on Fairness, Accountability and Transparency, in PMLR 81:77-91

Garvie, C., Bedoya, A., & Frankle, J. (2016). The Perpetual Line-Up: Unregulated Police Face Recognition in America. Georgetown Law, Center on Privacy & Technology.

Mitchell, T (1980). The need for biases in Iearning generalizations. Technical Report CBM-TR-117, Rutgers University.

O’Neil, C. (2016). Weapons of math destruction: How big data increases inequality and threatens democracy. Broadway Books.

Borenstein, G. Power to the People: How One Unknown Group of Researchers Holds the Key to Using AI to Solve Real Human Problems. Medium Website. https://medium.com/@atduskgreg/power-to-the-people-how-one-unknown-group-of-researchers-holds-the-key-to-using-ai-to-solve-real-cc9e75b1f334. Updated 2016. Accessed May 25, 2018.

By Grace Anne Foca

Individuals, particularly those diagnosed with chronic diseases such as type 2 diabetes, want control and to feel “heard,” especially if it is clear that the doctors and dietitians do not have the time to customize dietary regimens to fit their individual needs.

When medical professionals diagnose patients with type 2 diabetes, they typically recommend websites, mobile applications, and information sources that provide nutrition plans with broad food categories, rather than a list of specific brands that an individual can consume to remain compliant with the recommended daily calories and food groups. Not every person diagnosed with type 2 diabetes is the same; therefore, directionally appropriate content, delivered with static information design and content through digital health services, are not sufficient tools to help people adapt to a life-changing health condition. Each person diagnosed with type 2 diabetes has different pre-existing health conditions and glucose levels that vary throughout the day, creating the need for customization—one that food and beverage companies are only just beginning to address. Individuals, particularly those diagnosed with chronic diseases such as type 2 diabetes, want control and to feel “heard,” especially if it is clear that the doctors and dietitians do not have the time to customize dietary regimens to fit their individual needs (Ball et al., 2016). The purpose of this investigation is to understand how the design of a customizable food shopping application can support dietary adherence in adults newly diagnosed with type 2 diabetes by assisting and integrating their food purchasing decisions related to food selection, food substitution, and food combination. I conducted discussions with medical professionals that confirmed the types of appropriate nutrition plans and the socioeconomic barriers to living with type 2 diabetes.

The project that emerged from my research suggests ideas for interface design approaches that could help people adapt long-term strategies for type 2 diabetes management with greater ease.

Based on the lists of doctor-recommended mobile applications and websites, I conducted a comparative analysis that identifies gaps in food shopping services, both in-store and online, for people with type 2 diabetes. The project that emerged from my research suggests ideas for interface design approaches that could help people adapt long-term strategies for type 2 diabetes management with greater ease. Rather than follow a  “one-size-fits-all” nutrition plan, this project would provide enhanced knowledge of appropriate and desirable food choices. With more specific and applicable knowledge at a patient’s fingertips, it is likely that greater dietary compliance will result with fewer daily swings in blood glucose levels. Ultimately, the result could be a longer and greater quality of life for those with type 2 diabetes. The food shopping application, named TYPE2U, focuses on three phases during the first few weeks of a newly-diagnosed person living with type 2 diabetes. These phases include: 1) making food choices for the first time since the diagnosis; 2) making lifestyle adjustments according to blood glucose level testing; and 3) making mistakes that lead to meal planning as a method of increasing control over food choices.

Receiving Initial Diagnosis & Making Lifestyle Adjustments

The user, Sam, is going on her first food shopping trip since her diagnosis, and she uses TYPE2U to create a customized food shopping list that adjusts based on her lifestyle choices, current location, and blood glucose levels. Sam is also receiving blood glucose level updates on her smartwatch that TYPE2U uses to adjust the food recommendations on the shopping lists TYPE2U generates for Sam. She receives notifications throughout the day, informing her of high and/or low blood glucose levels, following up with questions about her food choices, and then giving her advice for how to behave in the future.

Figure A: Receiving Initial Diagnosis & Making Lifestyle Adjustments

Making Mistakes & Taking Control

Sam makes mistakes with her food choices this week, she just could not resist having some junk food. TYPE2U shows her how her food choice mistakes led to irregular blood glucose levels in her weekly progress report, which displays a data visualization available on smartwatch and smartphone. The application wants to help Sam get back on track and prompts Sam to confirm if she would like to activate the meal planning feature of TYPE2U. Meals suggested to Sam are part of a large database that collects data from Sam’s social media and search engine activity.

Figure B: Making Mistakes & Taking Control

References

Ball, L., R. Davmor, M. Leveritt, B. Desbrow, C. Ehrlich, and W. Chaboyer (2016). “The Nutrition Care Needs of Patients Newly Diagnosed with Type 2 Diabetes: Informing Dietetic Practice,” Journal of Human Nutrition and Dietetics 29, no. 4, pages 487–94, August.