The Demand for Fuzzy Science and Explicit Design: Rationalism is Not Universal

The article was initially published on 5/24/17 at www.humanitarianspace.com. It has been relocated here for continuity. 

Whenever I explain Design to non-designers, I essentially describe the scientific method - observe, hypothesize, test, repeat. Many designers work this way, although they might think creativity is something more magical, and are less inclined to describe their work as sequenced trial and error via prototyping.  Also, let's face it, clients will pay for magic but not for experimentation. The differences between science and design are more cultural than procedural. Science is better tuned to the needs of validation and design facilitates more generative insight, but the largest difference is that the scientific method is often stuck in the culture of science - we tend to think of the scientific method within fields like biology or physics, and thus resort to reasoning and intuition for day-to-day matters. Thus we think we use the scientific method all the time, but actually, we rely mostly on the power to reason.  

At the Thresholds of Reason

Unfortunately, human reasoning is not universal. Rather, it is situated in disciplines, as each discipline enforces a particular kind of language and patterns of cognition. The lawyer does not interpret the world the same way as the doctor or the engineer. Yet if forced to work together on a worldly problem, they will each insist on using reason (common sense) to engage problems of magnitude. Yet as each person in the room is attuned to a different way to frame and engage a given problem, it seems that direct experimentation would be more valuable. So why the fear to experiment for results? 

There is a widespread predisposition to assign the scientific method to the profession of science, wherein science is only for science-y things, and likewise,  we assign design methods to the design professions like architecture. Why is the method of science only reserved for biology but not for daily action? By extension, we are prone to work through other disciplinary problems in discipline-centric methods- and those methods are all the same, but get diluted by disciplinary language and habit - so why don't we step outside of our discipline with our methods? Most MBAs for example, never do scientific experimentation, although it is the premise of all worldly knowledge. They might embrace a design workshop (as business has to embrace superficial elements of design), which is a good way to synthesize ideas and create opportunities, but it also is a high-risk endeavor because it lacks any direct role for validation - it is a synthesis of assumptions. In contrast to generating learning opportunities, we rely upon the idea of the "expert" who has knowledge based on previous experience and we expect that knowledge to cross over. 

In contrast, we resort to rationalism and rationalism relies heavily upon our subjective interpretation of previous experiences - and our memories are not the best way to record the world because memory is highly subjective. Design is an approach to understanding the world through observation and experimentation, yet it also considers and takes advantage of personal subjectivity, cultural patterns, and the emergent outcomes of group interaction.  It does not merely accept the implicit assumptions and detriments of subjective memory. Design attempts to leverage those things actively avoided within Science. We could all probably use more science in our lives.

Dangers of Expertise
Dropping expertise within approaching a problem is important because expertise is really the opposite of science and design.  Expertise makes the assumption that experimentation is no longer necessary because the expert has all the necessary information already or can quickly filter available information.  Experts supposedly already did all the hard work to understand a particular worldly pattern. Expertise can be reasonable - in any given problem, we all find a point in which additional experimentation is redundant, and an outlier result will have little statistical significance. Yet what happens when we combine this statistical argument with the psychology of a group? In this context, the value of expertise is diminished because of a necessity to all reach a common understanding. Without a process of experimentation, people will talk about things they understand as individuals with poor translation to others. They will impose known patterns upon new problems. They will fail to experiment.

Patterns of Cognition: Habits
Later, studying robotics at Carnegie Mellon University, this way of thinking was entirely beneficial to advance the state of human robot interaction.  How do you build stronger human relationships with technology? It takes more than mere manipulation of pixels or a human factors design assessment. Yet when I was a graduate student in economics and law, this way of thinking was not helpful.  In fact, economics never touches the same time of reasoning. Economists are looking regularly at the circulation and balance of inputs and outputs - a far more linear and sequential type of thinking. Law likewise was a headache to study, as legal reasoning relies more upon the accounting of evidence to identify conflicts of logic. The reasoning of pattern relations - whatever they teach in art school - has far less to do with sequencing or logic. These different paths to engage and interpret the world establish radically different understandings of a given problem.

The notion of "patterns of reason for knowledge construction" is perhaps a critical element in how we all approach problems. When I was an art student at the Art Academy of Cincinnati, I developed a way to see and understand the world that was not explicitly rational or empirical. It was more akin to pattern recognition at large, the ability to identify and symmetries/asymmetries amid abstract collections of information.  Studying historical and contemporary art, I became attuned to the relationship between cognition and body mechanics and to the relationship between physical materials and economic production. I learned to look at the associative relationship between materials and ideas.  For example, I recall seeing a photo with a dogwood tree, and by association bring up a legend about how the dogwood tree is cursed by God to grow crooked. This story implies personal meaning, and thus as an artist, I am inclined to use the tree to indirectly communicate something about religion. Compared to basic arithmetic, this way of thinking makes no sense, and yet, moving through the world by association is an important part of the human experience. 

Scientists attempt to identify and negate their values, yet there is sufficient evidence that all science is situated in the subjectivity of the scientist.  Many scientists accept this.  Good designers also take stronger responsibility for their own values and situated knowledge. Yet both can also discover other paths of inquiry and reason.  In the end, the only ones who fail to adapt to new problems are those who rely explicitly on their own situated reason.  The world is too complex for experts.

The Art Academy of Cincinnati - Education to be Radical, Relentless, & Radiant

The article was initially published on 5/19/17 at www.humanitarianspace.com. It has been relocated here for continuity. 

I was deeply honored to give the commencement speech to the graduating class of 2017 at the Art Academy of Cincinnati. These last few days, I am now continually reflecting upon the unique and powerful proposition this school makes to the world. There is no other school like it. 

The only other college to which I can compare it is the mythical Black Mountain College of the 1960s that produced revolutionary minds such as John Cage.  To plagiarize someone else’s story, the Art Academy (AAC) doesn’t merely graduate artists or designers, it graduates the critical but hard to find team member of every successful business: 

"there are three kinds of people you want to launch a business: the person with the idea, the person with the financial sense,  and the person who makes you say 'what the fuck?' The last is the person who can rip ideas apart, remix them, and flip everything upside down to generate breakthroughs that no one else can see."  

Blackmoutain College w/ Buckminster Fuller

The last kind of person is particularly hard to find. Many schools can teach people to become accountants or to be entreprenuers but no school teaches students to be intellectually rebellious and operationally radical.  Except for the Art Academy of Cincinnati. No joke. It is even in their mission statement.

Everyday books about Innovation, Design, and Economic Disruption churn through billions of dollars in annual publishing sales. Parallel to the publishing industry, countless institutions argue they offer an education that will transform students into innovators who will change our world.  But do these industries actually generate the change-makers we seek?

In the last ten years, I’ve been fortunate to spend time at the world’s best universities as a speaker, student, or instructor including Oxford University, MIT, Harvard, Cornell, and Carnegie Mellon University – and these are indeed great schools.  Their students are brilliant and the faculty are more than competent. The programs are well funded and the students are nearly guaranteed the security of a well-paying job upon graduation.  These schools also attract people who already have a history of success - when Elon Musk attended Stanford, he had already earned degrees in Physics and Economics. Yet I have never encountered another school that transforms unknown students into true innovators.  In fact, when I recently taught Design Thinking at an East Coast top-tier MBA program, my students complained the entire time about the lack of clear directions and the constantly shifting parameters within the course requirements.  I have since learned that this complaint is exceedingly common within MBA Design degrees. These programs are forcing square people through intellectual circles and many graduates come out very little changed. 

2017 Commencement Address

Do all art schools impact students to think so differently?  I'm not sure... there are many art schools in the world. My sister is a student at SCAD. I have friends as RISD. When I was a teenager, I lusted for the attention of the San Francisco Institute of Art (SFAI) and the School of the Chicago Institute of Art (SCIA).  Unfortunately, in 1999, I had so little money for college, I did not even have the 50 dollars to apply to any of those programs let alone all of them.  With little hope to attend any college, I drove my broken-down ‘91 Geo Prism to the Art Academy of Cincinnati for a Portfolio Review Day in mid-October, to present my high school artwork to various colleges.  San Francisco was there, as was Chicago, and at least a dozen others.  Chicago offered a partial scholarship on the spot, which was incredible… yet, as I did not have the money to apply, let alone to live in Chicago, it held more symbolic meaning than opportunity. I was nonetheless motivated at that moment to find a way to go to art school.

Weeks later I happened to cross paths with some artists, Aaron Butler and Christopher Daniel.  Aaron worked at the Art Academy of Cincinnati and pioneered the experimental music group, Dark Audio Project, while Chris was a metal sculptor who went on to found the extraordinary and thriving Blue Hell Studio. They both held Art Academy ties, and with their encouragement, I decided to do everything possible to earn a scholarship. I applied only minutes before the deadline, in person, submitting my application in a massive wooden box crafted from an old PA system pulled from a dumpster in Kentucky (at Aaron’s suggestion that I make the physical application somehow stand out). As a mediocre student in high school, I had only applied to one other school at the time – the globally exceptional design school of the University of Cincinnati, DAAP – and I was not accepted.  The Art Academy took a chance on me, offered a scholarship to cover more than half of tuition, and I will be forever grateful.  Notably, after later graduating from the Art Academy, I received a full scholarship to DAAP for graduate school.

Art Academy of Cincinnati

Visiting AAC this last weekend was not only nostalgic – it was inspirational.  The Art Academy is a weird place. It consistently takes chances on people like me. It is a community of outsiders. It pushes them to build expertise on the ability to make something new  – which is not typical, considering most degree programs demand students acquire knowledge on a longstanding subject or methodology. It pushes students to invent new models of production, new identities as artists, and to take life to the frontier of possibility.  Graduates of the Art Academy of Cincinnati do not need books on creative problem solving, they need wicked problems where all others have failed.  If the Art Academy has a flaw, it is a simple fact that they do little marketing or high-profile partnering, and consequently, the world knows little about this school amid an insatiable demand.  The Art Academy of Cincinnati is not a diamond in the rough – it is a silent A-bomb in the exosphere.

My life has changed much since I attended the Art Academy. I am writing this blog entry while on a flight to San Francisco. Tomorrow morning, I will run a series of design strategy workshops for a Venture Capital firm in Silicon Valley to explore new investment models for Artificial Intelligence. Since attending the Art Academy, I have lived in multiple countries, built companies, and am fortunate that my abilities to tackle entrenched problems in new ways are continually in demand. When I think of the year I started college, 2000, my life is now very different from the future that was most likely ahead.  Though I have my fair share of life challenges, I have a wonderfully creative and satisfying life. It has been a hard journey, but I credit the faculty and students of the Art Academy of Cincinnati. While most colleges chart a path for your future, the Art Academy provided a compass to guide me through the deep woods of the unknown.

Advancing New Economic Models by Design

The article was initially published on 5/24/17 at www.humanitarianspace.com. It has been relocated here for continuity. 

A few weeks ago I had to opportunity to spend a couple days at the Urban Planning Department of Cornell University. I was impressed by the graceful way this group was able to move fluidly between rigorous quantitive analytics and participatory public processes.  Yet for all the brilliance I found among faculty and students, it became clear to me how much urban planning education lacks sufficient focus on design methods.

I am not referring to design as urban design or architecture. I am referring to the ability to translate a series of complex social and technical processes into physical form.  I am referring to the ability to translate ambiguity into action and to oversee the transmission of that action to generate results. This is not the same as project management or the mere practice of the profession.  Rather, given the massive range of assets are available in urban planning for engagement and analysis, the discipline completely lacks a rigorous methodological framework for what actions to implement by consequence of the planning process.  If the discipline of urban planning stops with pitching the plan - then planners deserve to be disappointed when their work does not reach fruition.

This realization explains much about the disappointments of the planning profession - such as the constant repetition of "off the shelf" solutions such as green roofs, walkable streets, and historic main street development initiatives. These tactics are fine - but why such a small range of possible outputs in a world of more than 2.5 million cities, towns, and villages?  Basic statistical intuition suggests that a profession dedicated to building new futures and generating new economic development initiatives would capture a broader range of possible solutions.

To consider the urban planning process is to recognize that it remains rooted in a Waterfall design methodology - which has been proven to drive up costs and reduce stakeholder participation.  Most socio-technical systems have long since discovered that Waterfall methodologies fail to consider the variability of human actors, and thus tend to fail.  While organizations continue to search for replicable solutions utilizing scientific research designs and clinical trial models, the assets of localized place-based development go ignored or fail to scale.

Private sector technical sectors have shifted toward lean frameworks, agile methods, and other systems rooted in rapid feedback to avoid the high risk approach of waterfall planning. Unfortunately this understanding has yet to see the light in American politics where sweeping legislative action is the norm - not iterative improvement and variation. Urban planning, a field long aligned with design, has an opportunity to update to the 21st century - but it needs to start in education.  Design is more than architecture, it is the execution of ambiguity into meaningful consequences. 

Unlocking Machine Learning with Human Centered Design

Human Centered Design (HCD) is at its core, a process for eliciting the practice of human values and embedding those new values into a new artifact, process, or service.  It is perhaps challenging to consider how HCD may have a role within something as quantitative as Machine Learning, yet HCD can be an important component to the formation downstream benefit for machine learning services. The value to injecting HCD into the design and use of machine learning solutions is valuable to contribute to unknown opportunities and risk the threat of future-tense machine learning decisions.

Quick Intro to Machine Learning
Machine learning is best described as a methodology for a computer program to learn to write new programs without human assistance.  There are two primary domains of machine learning - supervised and unsupervised. Unsupervised machine learning utilizes various statistical functions to identify patterns in information and to extrapolate decisions from that information.  The reference to Big Data is often a use of unsupervised algorithms that are, at their core, a regression analysis. Supervised Machine Learning also leverages statistics, but to do so, relies upon a body of data curated by an individual.  The computer program builds an understanding of the data to replicate it or to use that understanding as a filter on other data.  For example, if 2000 images of a tree are shared with the program, it will use those 2000 images to build a concept of a tree, and will then be able to identify a tree from a suite of images which may or may not contain trees.

Of Machine Models and Human Cognition
To build an HCD approach to Machine Learning, it is important to first distinguish how machine learning is similar and different from Human Learning.  First, both the human brain and the ML program do have a core similarity - they are massive engines of statistical pattern recognition.  Our brains engage and understand the world through pattern recognition. To see an apple, know it is an apple, and understand what an apple does is in fact a massively complex undertaking.  The connection between sight and concept requires our brain to internalize and make sense of about 8,000 different points of information... how light bounces off the surface into our eyes to inform geometry, color, shape, texture, and subtle visual implications of weight, density, and so on that are difficult to articulate.  When our brain has sufficiently identified enough information bits to create a pattern - the apple pattern - it has built a conceptual model.  Our brains are massive repositories of conceptual models, and we use these models to discover new ones. "That is not an apple but it looks like an apple."

Machine learning programs are similar (as most are based on our brain's neural architecture). They use many different methods to analyze information and identify a pattern, like "tree."  Yet their methods are various different from ours, such as rapid quantitative measurements between corners, measurements of curvature, measurement of gaps - and thus generate a very different kind of model.  The ML model is not interpretable to the human - nor does it need to be. The result is the total information required by a computer to make inferences about other possible models.  A computer, also, has limits and it is difficult to make inferences. Thus a human can rapidly intuit a new situation "apple by a tree suggests apple tree," the computer will only see an apple and see a tree,  unless enough information is presented about apple trees to make the leap.

The Learning and Reproduction of Human Values
Machine Learning asserts a transactional value by  using found pattern within a set of information to predict how that pattern is extended against a new inflow of information.  We can use ML to facilitate market analysis, reduce risk in complex situations, and optimize organizations. Yet what does this ability inform... optimize how? Reduce risk to do what?  HCD is a process that requires the designer also conducts pattern analysis - but a key distinction is in the domain of qualitative patterns.  How does a user feel about a given situation over time?  How can that situation shift over time to elicit other feelings?  The intersection of opportunities between HCD and Machine Learning are vast when you consider the bigger context of the data.

Within supervised machine learning, the assembly of training data could be done in bulk or it can be curated.  A designer can pay close attention to the data... what kind of trees are presented? What health are they in?  Can multiple sets of training data be presented that contain healthy trees, sick trees, and variations in environmental conditions?  By curating the data, the human agent can tune the specific types of patterns to be determined, and thus generate greater value later.  Not all data is equal.

Likewise, as the outputs are generated from an ML program - such as new sets of data (often relayed in a dashboard or a customized suggestion), the human experience of that new data point can be catalogued and leveraged to tune the ML program.  While there is such a thing of "Human in the Loop," to aid the precision of the algorithm, another way to thin about this is "Value in the Loop," to reward the program for generating particular kinds of experiences. While abstract, the program does not need to understand the human value - it will build its own conceptual model to make sense of the behavior.  These values can tune the algorithm to new directions over time, wherein we can expect the algorithm to generate new data points very much unlike the original yet of great use.

Geographic Variance in Modelling
A critical aspect of being human is the variations that exists between humans for perception in the world.  Within small groups, language can take on a range of meanings, while across geographies, the variance of meaning ascribed to concrete things can be quite large. Machine learning generates one kind of solution for one kind of person and then generate a new solution for another kind of person... it can be granular, like Netflix, to suggest customized outputs.  Yet what about machine learning applications for broad data sets that are irregular?

Within the design community, I advocate close consideration of the role of place and data. There is a tendency for developers to look to large open data sets. Vast databases exist with training data (like this popular one at MIT).  Yet consistent with my above comments, what are the values which drive the labelling of this data?  For example, if you apply this training data to images of Mogadishu, it will generate results like "earthquake."  Not only is this inaccurate - but the correct label will vary depending on who and where the labeler resides.  A Somali in Nairobi will give a different answer than a design student in Boston to describe a set of data.  Yet most Somalis in Nairobi will supply more similar labels.

Consequently, it is not enough to parse data by meaning.  One must also ground the meaning, and to do this, I suggest the use of classic lat/long coordinates.  If we can build rich databases of place-based data and place-based identification, we can do more than build intelligent softwares, we can build softwares that are flexible to the global shifts in meaning and identity which are traditionally at odds with the demands of computation.  To fuse machine learning with the vast ocean of human value creation and reproduction is a great opportunity for the future.