08.15.18
Lilly Smith | Interviews

Chain Letters: Paul Pangaro



This interview is part of an ongoing Design Observer series, Chain Letters, in which we ask leading design minds a few burning questions—and so do their peers, for a year-long conversation about the state of the industry.

It‘s August, while it's still crazy hot and we'd all like to be poolside, it's time to think about heading back to school. This month, we examine design and education, and look at how education will shape the design discipline of the future.

Paul Pangaro's career spans teaching, research, startups, and consulting. He is currently Chair and Associate Professor of the MFA Interaction Design program at the College for Creative Studies. In January Pangaro will begin a new role at Carnegie Mellon University as Professor of the Practice in the Human-Computer Interaction Institute. Pangaro has worked with and within software startups in New York, Boston, and Silicon Valley in product and technology roles. As a consultant, Pangaro has been hired by Du Pont, Nokia, Samsung, Instituto Itaú Cultural (São Paulo), Ogilvy & Mather, 8 Inc, Intellectual Ventures, and PoetryFoundation.org. His published papers explicate designing for conversation from his research and his implementations of software products and organizational processes (pangaro.com).

How much emphasis do you place on “real life” project-based learning as compared to theoretical design study?

When possible I prefer not to separate theory from practice. With clients, the theory is always close but rarely mentioned (it tends to unnerve them). In teaching, equal emphasis on theory gives students a grounding that they carry not only into their school projects but into their careers (they have told me this).

For example, designing services that require human-to-human interaction will involve conversation whenever a participant learns, considers, or decides (which is nearly always). A formal model of conversation, derived from a theory — how conversation works, what makes it successful, how a given channel may be improved — informs the design process deeply. Students (and consultants) who can model conversations have more confidence in their design proposals and more easily communicate them to others (they tell me this also). I believe that their designs are also better for it.

Having a model of what constitutes ‘successful conversation’ leads the design process toward greater, and more humane, outcomes. That is a corollary to my main point, that a good theory helps tame the context in which design is to take place.


Colloquy of Mobiles installation 2018, Interaction Design Masters Program, College for Creative Studies, Detroit. Full-scale replica of Gordon Pask’s original design from 1968 by TJ McLeish. Photos by Paul Pangaro.

In November, AIGA is publishing “Designer 2025,” the culmination of research led by Meredith Davis and other leaders across multiple sectors, which will identify seven emerging trends with long arcs of influence and describe key competencies necessary to compete in the coming decade. All this begs the question: How can teachers prepare their students for jobs that don’t exist yet?

Full disclosure, I’ve been involved with one of the seven trends, the one about ‘making sense’ of the design context as an inseparable part of the design process. Co-written with Hugh Dubberly, we claim that bringing designers to understand and apply certain frameworks prepares them for some, at least, of the unknown challenges of the future. These are foundational frameworks that can be applied in novel situations, including jobs that don’t yet exist.

One framework declares, in part, that designers are inextricably part of the system they are designing in, which itself is a ‘complex adaptive system.’ (Joi Ito, director of the MIT Media Lab, is writing frequently about this, which derives from second-order cybernetics.) Designers’ viewpoints — their knowledge, beliefs, and personal values — form a ‘language for seeing.’ This language is inseparable from a designer’s capability to frame a given context in a way that leads to viable problem definitions and their concomitant solutions. ‘Problem-finding' and ‘framing’ are relatively recent though not new ideas in design; what I believe is new here is the emphasis on creating and evolving new languages in order to develop truly novel outcomes (some call this ‘innovation’).
In my experience, an understanding of the processes of design and the means for expanding techniques and capabilities are a matter of practice and critique, tightly coupled.

There are so many resources available online—YouTube tutorials, MOOCs, an endless supply of design blogs. Does this indicate that learning been democratized, in that there are many ways to get to the same end destination—a position in the design field? Or, has decentralized learning created a lack of a cohesive set of standards?

In my experience, an understanding of the processes of design and the means for expanding techniques and capabilities are a matter of practice and critique, tightly coupled. You don’t get that from online delivery (and I don’t believe AI will help). There is no substitute for the design studio, populated at all hours, rich with making and rich with intense conversations about goals and means. Here is democratization, when professors and students and external critics are all enmeshed.

Perhaps the demand for some types of designers (for example, in interaction design) is such that someone with little actual experience can get into a job, I don’t know. But once there, if talented, the practice and critique is necessary (though not sufficient) to make that individual a good designer.

Did the design field have a cohesive set of standards before online resources existed? Typography comes to mind and likely product (a.k.a. industrial) design, to some extent. But with digital, prior knowledge feels severely hampered in its scope and effectiveness. Designer 2025 is intended to address this massive technological shift and hopefully the social shifts that come with it. Far from having standards about them, we are not yet even prepared to teach Big Data and AI as materials of design; we better be soon.

When you reflect on your body of work to date, which has the most personal value for you? Why?

I’d like to offer an example of methodology and an example of making.

In regard to methodology, I believe design is a set of conversations. When practitioners speak of their work as ‘design thinking’, their main pillars are ethnography, brainstorming, and prototyping. I think of these as conversations to agree on goals and conversations to agree on the means for achieving the goals (that’s two types of conversation). But who designs the conversation — who even determines who is to be in the conversation? Where is the vigilance to assure we have the right depth and diversity? Surely that is an act of design and surely it matters a great deal to the outcome. “Conversations to design the design” is a third class. So I claim the “design of conversations for design” has to be explicit, also. But these three classes of conversation are not enough to bring innovation — for that you must design new language, as I suggested earlier. So, evolving this necessary and sufficient set of ‘conversations for design’ has felt like a valuable effort.

In regard to making, while I’ve designed and coded a number of interfaces for ‘conversation with content’, I’ve had the most personal satisfaction from leading a project to recreate an example of conversational machines from 1968, Gordon Pask’s Colloquy of Mobiles. Imagine an exhibit in which human-scale mobiles interact through light and sound, cooperating and competing and sometimes not. It’s astounding to me that so long ago Pask came to conceive of machines talking to machines and, oh yes, allowing humans to sometimes intervene. This was long before computers were ‘personal’ and even longer before we spoke to our devices every day. He saw our future and chides us to remember that we are biological, embodied, and analog. His Colloquy, by the way, is another example of the power of theory.


Colloquy of Mobiles installation 2018, Interaction Design Masters Program, College for Creative Studies, Detroit. Full-scale replica of Gordon Pask’s original design from 1968 by TJ McLeish. Photos by Paul Pangaro.

Talk about a professor that improved your design approach. How did they do it?

Having just spoken about Colloquy I have to invoke Gordon Pask, who began studying interaction by making machines that interacted with humans as well as with other machines, from which he derived his rigorous, scientific theory of conversations. He was a ‘maker’ long before that term existed. By looping between tinkering and theorizing, he constructed an unparalleled body of work whose impact on interaction design has yet to be felt.

From Dori Tunstall: We harness the computing power of digital technologies to often perform work that we no longer value having humans do. The anxiety that drives the plot of the Matrix, Terminator, or Blade Runner series is the reversal of the human as master and technology as slave relationship. If you were to design a new system of technologies based on a non-hierarchical relational model between humans and technologies, upon what phenomenon would you base that model? Please speculate on how it would change your current creative practices.

First, I would begin not with a new system of technologies but with a framework of humanism, grounded in ethical values and a definition of humane interactions. Let’s do that collaboratively and well. An issue I have with Silicon Valley, which carries some of the presumptions of computer engineering, is that technology can and should steer us toward solutions. For me: shared values first, then conversations for design in which technology is part of, but not delimiting, a solution space of new possibilities.

Second, do we really want a non-hierarchical relationship between humans and technologies? Perhaps I misunderstand what you mean. Would that mean that we cede something to “technologies” and, if so, what? An autonomous execution of a task like driving, ok — but does that mean we let it decide how it decides? A question was posed to me recently: If a machine makes better decisions than a judge about how long to return someone to jail, wouldn’t we want to use that machine? To which my answer, and I’m paraphrasing what I actually said was, “WTF do you mean by ‘better’?” If we let a machine decide for us, especially when there is no transparency about how the proprietary algorithm functions, in what ways do we think of that as “better”? Don’t we need to define that first? Without a declaration of our shared values we are likely to lose them.

Lastly, at the risk of sounding staid: If I follow the path of my first comment, then I should be OK with my design (creative) practices so long as I’m vigilant about questioning them along the path of designing. But I do want to repeat the point that we should be designing for ourselves and our values. You’re asking a fundamental question that may be difficult for one person to answer well; certainly this person (me). Can we have a conversation about that. When are you available and who else might we include?



Posted in: Chain Letters, Education




Jobs | October 10