by Ali Anwar '17 and Eshan Tewari '17
Matías Duarte (@MatiasDuarte) was the distinguished alumni speaker at the 25th annual Magnet research convention on January 12, 2017. Having graduated the Magnet program in 1992, Duarte went on to study, in tandem, Computer Science and Fine Art and Art History at the University of Maryland (terp.umd.edu). In the years that followed, he went on to direct many facets of user experience and product interface design with companies like Danger, Helio, and Palm. Duarte was hired by Google in 2010 as the Director of the Android User Experience (AllThingsD), and today is a Vice President working on the company’s Material Design interface (The Verge).
In preparation for this interview, we read through a few of Duarte’s old interviews, speeches, and press statements. In the course of this research, one particular non sequitur from his 2014 interview with The Verge stuck in our heads—simply, “Mobile is dead.” This is about the last thing one would expect to hear from the mouth of Google’s then-Director of Android User Experience, so naturally we had to ask him to clarify. Duarte’s answer not only elucidated the meaning of this odd phrase, but provided a deep insight into his vision for the software design and distribution landscape of tomorrow.
“You guys have seen all those companies like IBM, Nokia, Corning, put out these videos [saying] like, ‘This is the future of technology!’” He proceeded to describe the kinds of glitzy tech ads that we’re all used to watching on TV, promising to one day put manipulable computer screens on every car windshield and restaurant tabletop. “That stuff is coming,” he continued, “but the way we think about . . . developing software isn’t mature enough for that yet.” Today, Duarte explained, different devices (PCs, tablets, phones) are treated like isolated objects. Software that aspires to be cross-platform must be deployed to each of these device in a different form. As more and different devices emerge, Duarte predicts that this already troublesome approach to software development will become “untenable.”
The current model of software development, he says, is a vestigial organ dating back to a previous era of computing: “Originally, very few people had computers, and they could only engage with computers for a few hours at a time in a few places. Software tended to be large and monolithic . . . Most people were using the same software because there was so little of it.” By contrast, “Now we live in this world where . . . you've probably got two computers more powerful than what we put men on the moon with for Christmas. There’s a ton of software out there, a lot of it is highly specialized . . . The world has changed a lot, a lot, a lot, but a lot of the . . . technical infrastructure behind how we deliver software and even how we think of ‘What is software?’ hasn’t changed.” The core problem, to his mind, is that, “We still have a model of delivering software which says . . . the software [should be] tailored for one particular platform and one particular configuration. Adapting to different contexts or screen sizes is a problem left to the developer,” as opposed to being handled by operating systems.
By way of a solution to this problem, Duarte has one overarching piece of advice for the developers of tomorrow: recenter the focus of software design from devices to users. “What I would like to see is a world where . . . there’s multiple devices, and there’s an operating system that runs across them, and the user, therefore, runs across those devices.” He envisions, “a world where software is associated with people, rather than devices.” With this idea in mind, “mobile is dead” makes perfect sense: Duarte welcomes the demise not of mobile technology itself, but of the archaic notion of mobile devices as some distinct category of software executor. In his ideal future, a phone will be the same, from a software perspective, as any desktop, laptop, or tablet . . . or even windshield or tabletop.
The Road to Material Design
In the early stages of Android — before Duarte was hired — there was a considerable amount of employee churn, rendering fragmented operating systems and design schema. Duarte’s objective, however, was to frame a team around purely design philosophy: a design which would span across products.
Rallying a team around design was, at the onset, quite an uphill battle due to the overwhelming product focus. To address challenges of this nature, Duarte advised us to “find allies, look for the lowest hanging fruit or the most egregious problem, play the long game.” In the Android user interface, that lowest hanging fruit was the menu button.
The menu button was a defining, quintessential component of the Blackberry user interface, and it also found its way into the original Android user interface. The menu button remained there due to the fact that many Android devices may not have touchscreens. What ensued, however, was a perversion of the simplicity of the menu button; what once was a means of scrolling and clicking became a “dump for contextual features,” whose true potential and intentions could only be harnessed by advanced users.
Duarte found Apple's reasoning behind not wanting a two-button mouse credible and salient to the issue of the menu button — “the presence of two buttons leads people to treat them as equals . . . [we should] instead relegate it to two-finger click and advanced motions.” Moreover, Duarte emphasized the importance of research: “Nobody can keep themselves honest enough . . . anything that you can’t measure, you’re not going to be good at — user experience research is key.” Duarte was rather astounded that the team did not already include a user experience researcher; when he hired a researcher, he found that the menu button was one of the top user issues across countries. As a result of his research, there are only three top level meta-actions in Android today — home, back, and switch application.
The Art of Design
Duarte's art history education bore a tremendous influence on the entirety of his design career. On a macro scale, through scrutinizing works of art in their historical and cultural contexts, Duarte learned to understand the “organs of perception” and what drives people to “perceive things in certain ways.” Studying works of art spanning across history was not only indispensable to designing products that conveyed a very particular message (for instance, embedding the essence of the Google brand into the new “Super-G” logo), but also for nuances in design to appeal to different cultures across the globe.
For example, the core aesthetic values of cultures that base their writing in iconography are entirely different from those of romanized western cultures. Romanized writing systems require many “icons” to convey a word, and the icons you see in phones are far more concise and value dense. For pictography based cultures, however, phone icons are more convoluted than language, and can even be seen as an insult to the beauty of the language. Design must also be tailored to account for reading direction (left-to-right vs. right-to-left) in a culture, the connotation of colors in a particular culture, and a myriad more factors.
The Magnet Foundation
Finally, we talked with Duarte about his own experience studying in the fledgling Magnet of the late 1980s. For this last segment of our interview, we also got the opportunity to speak to Duarte’s old friend Anne “Wiz” Wisniewski — former Magnet network administrator and self-described “mother of the Magnet” during his matriculation — who shared her own experience working with the program in those early years.
Wisniewski agrees: “When Matias was in the Magnet at the very beginning, there were definitely a lot of social outcasts that were part of the group, and it was a group that didn’t judge. I mean, that was the coolest thing about the Magnet, because all these kids were in this new boat and we were trying to see if it’s going to float or if it’s not going to float. Nobody rattled that boat.”
Asked about the role the Magnet played in his future success, Duarte added: “The way that the Magnet, for example, introduced computer science through STELLA helped me understand system thinking and simulation thinking. Then, you know, it was like later that we got into ‘Here’s programming’ and ‘Here’s functions,’ and ‘Here’s, you know, data structures’ and stuff — which is where most people kind of start. I feel like that, even that conceptual understanding of thinking about systems [gave me a] foundation of taking that perspective on things.”