A critique of Design

In the beginner’s mind there are many possibilities, but in the expert’s there are few

– Shunryu Suzuki

On the front wall of the workshop I’m standing in, one diagram, printed across three reams of paper covered the entire 8’ by 10’ area. This diagram was constructed with boxes upon boxes stacked in both directions like a long container ship, each the size of my palm containing a quote painstakingly transcribed from user interviews. Running across the bottom side of the diagram were colorful post-its containing sharpie-scribbled insights. This was the result of tens if not hundreds of hours of work by the Design Lead and User Researcher. It rivaled the Water Lilies at full scale, both for its exquisite details and staggering scale. Instead, we were trying to design AI technology for healthcare.

I am no stranger to models and diagrams, having deployed them countless times in my career, and still I marvel at our sophistication at wielding them these days. Yet, I can’t shake the feeling that we’ve been doing this all wrong. Design was supposed to be the panacea that solves the ills of untamed technology development (certainly an exaggeration). However, as certain as Design found our roles in technology teams and the development process, it sometimes feels like what we do perpetuates the same issues that we proclaimed so loudly in the first place.

I ponder about the usefulness of this diagram in helping us design technology to help raise standards of care. In a similar vein, it will be useful to interrogate our design process to figure out if things have truly gone awry.

Let’s consider a prototypical design workflow. The first step is to survey the field that we are designing for. From there, we construct our user personas, cultural diagrams, user journey maps, mental models, et cetera, that tightly pack and slice the observed nuances into different dimensions. Armed with that information, the Designer – through explorations, wireframes and prototypes put in front of prospects – searches through the solution space. This prunes out false starts and secondary options, approximately, and the design is then presented to other stakeholders as a candidate to be built. Our practice today still broadly models the Double Diamond method first proposed a few decades ago.

I want to pick at three things here. First, notice that while this process can be described as two diamonds, in practice it ends up feeling like a funnel. In the beginning, there is one problem statement and infinite possibilities, both good and bad. At each step, as we gain clarity through our methods, the possibilities shrink correspondingly. With the increased emphasis on research – both through a feeling of pragmatism and providing value to other stakeholders – our solution space collapses early and quickly. Solution finding can then feel like threading a needle through a complex weave pattern. Our solutions overfit real, perceived and made-up constraints and inherit the problems that we set out to solve. I’m not arguing for no research here, but being overly dogmatic places us squarely in the faster horse territory.

Second, as complexity and abundance plays an ever growing part in our daily existence, Design solution finding can seem like, to borrow a video game analogy, designing for a single player experience in a multi player context. Solution spaces in software projects are rarely greenfield or neatly partitioned for our manipulation. After all, we are no longer designing a new widget or coffee maker but must take into consideration how it works in the broader context as it pertains to data attention and societal impacts.

This may read like a contradiction to my first point, but what I’m arguing for here is to consider the effects of the design. Solution spaces today are ecosystems. Our methods help us make sense and map out spaces but rarely consider how our interventions alter them. We have to believe our design output to be substantial as to embark on the effort, and by corollary, would certainly leave an imprint. Even as we improve our research acumen, our methods in solution finding do little to consider second, third and multiple order effects. In many organizations today, metrics and results fall squarely onto Product or Business teams. We need to develop methodologies and processes to evaluate the commercial and ethical success of our designs as we do today for usability. It would be in our best interest, and our collective stature and influence, to hold ourselves accountable.

Third, the current state of designer tools to explore solution spaces are at best a weak simulacrum of what we are creating designs for, and increasingly so as technology and its surrounding ecosystem gets more complex. The interaction between prototypes and its affected users are superficial at best; this gulf makes it difficult to properly predict and simulate the effects of new technology. Our labor ends up either being anemic to the possibilities, deviant from real-world constraints or even worse, resembling prolonged imagination and wishful thinking.

Let me be specific. In the job of a Product Designer today, almost all the tools that we use are focused on visual fidelity. To unfairly take one, Figma, as an example. The primary user interface contains of overlapping panels with controls to tweak visual properties: x, y positions, colors, fill, stroke and really, anything you want. In comparison, the prototyping capabilities are barely more sophisticated than what one could produce using slideshows and even then much more attention is paid to visual transitions than data states. But truly consider what the designer is supposed to be designing. Software is the manipulation and dispersion of information and data that enables outcomes or workflows between one or multiple parties. An online marketplace is the dispersion of information from selling parties, buying parties, and pricing. Even in something as venerable as that, the design impacts trust, judgement, exchange of money and other concerns such as fraud and scams. The user interface is merely a thin veneer.

Consequently, it appears as though science fiction writers and dreamers have taken on the mantle of designing disruptive innovation instead. Let’s take Augmented Reality as an example: how does Design guide us in bringing its promise and possibilities into reality? Instead, we have generations of fiction writers, artists, technologists, and makers that speculate on versions of its effects and impact. People who are building augmented reality turn to them for guidance and inspiration. Are we designers merely concerned with where the knobs are and how to curl your fingers to activate a specific functionality, or solve for an esoteric use case and claim that as innovation.

Let me flip this discussion around. Wicked problems, as defined by Rittel and Webber, are classes of problems where its denominations and effects are understood but difficult to define and tackle. Design theorist Richard Buchanan connected Design to these issues in education, social equality, climate change, et cetera, and postulates that our iterative design process is a spearhead in those fronts. Here, I find that Design currently practiced is necessary but insufficient. Our methods can be useful for inquiry, discussion and exploration but any material shift can only come with the help of other expertise. As a thought exercise, take the cliché of building an app to address mental health issues. It may be a good starting point for software designers to chip at the iceberg, but there are limits to what we can do alone. It is through working with healthcare providers, social psychologists, software engineers, social workers and other professionals collaboratively that may even help move the needle.

What I’m trying to get to is the and question – Design and what. Is Design the primary or the supplementary skillset? What the Double Diamond insinuates is that by following this designerly process, teams can yield more innovative products that satisfy their target audience. The reality of building software today is that professional designers play the focused role of dictating user interfaces. There are obvious gaps here. As designers, we have to recognize that our tools may be limiting our reach and constricting our influence; that for things to be designed, the person doing the designing must be able to meaningfully interrogate the subject matter.

One may argue that this is not a critique of Design, but of design needs within a capitalist context. We are most useful and lucrative when creating incremental improvements for commercial products. Another may argue that the term Design is overwrought since it covers disparate disciplines. Any discussion that seeks to unite them under a single umbrella muddies the discussion since every one qualitatively different from each other.

There may be hints of truth here but none of them leave me gratified. After all, I take after Herbert Simon’s interest in that there are better and worse ways in ordering human affairs and we as designers strive to make the best of them. To design today is to pair human intent and technology in ways that meaningful better our lives. We often hear of engineers tinkering into new inventions, or of scientists pursuing deviant results into new discoveries. I’m interested in how designers can use our skills, methods and more importantly, our judgement, to shape and mold a tangible and better future for us all.

Being Design educated has served me well personally and professionally over the years. As the mobile epoch matures and technology broadly looks forward to new paradigms, it would be a shame if we let our roles languish and our methods calcify.

A big thank you to En, James and Jayme for proofreading and providing me with invaluable comments.