Dear Ranjan, Johann, and Fil,

This is a quick response to your posts on acting before analysis.
Action may take place before analysis in design, but your examples are
of a different kind than policy design, the subject of the thread.

The example of the cows involves a single individual designing a system
to manage a problem under his own control on his own property. It does
not entail systems of the kind one deals with in maritime security.

Policy design typically involves large-scale systems with many actors.
Problems arising in policy design generate the difficulties that Rittel
and Weber identified in their work on wicked problems. Policy design
involves goals, plans, and strategies together with enabling laws,
treaties, and regulations, followed by programs and tactics to implement
them. Policy design generally involves multiple actors, often from
different stakeholder groups, sometimes from different organization and
even different regulatory situations. The planning horizon is extensive,
and you generally can’t try rapid prototypes in the field.

While you can learn a great deal by observing students at work when
they think and design, you can’t learn much about policy design. You
learn about creative proposals and ways of working that you may later
bring to bear on policy design.

Whether a student develops a great proposal or a bad one in solving
thorny problems for a course, nothing happens to real human beings.
Imagined consequences affect imaginary people in the imagined situation
whose problems we imagine trying to solve. Depending on the subject,
design students create mental models that build cities, restructure the
work-flow of a hospital, transform fuel efficiency, feed orphaned
calves, or make new chairs. But nothing more than thinking and modeling
happens until students take their mental models into the world, doing
the hard work of bringing the idea from mental model into living
application. This happens only rarely, in special programs such as the
Aalto Design Factory or the WhizKid Games project we developed here at
Swinburne to work with autistic children.

Policy design is also different to the world of designing artifacts or
small-scale systems accessible to control by a single individual. In the
world of policy design, small changes that are nearly invisible to most
of us can have major effect. Let me give you two cases of real-world
policy design that are in the news with respect to social welfare in the
United States. In social security planning, changes to the basis on
which the government calculates cost-of-living adjustments (COLA) can
improve the lives of retired people or tip them into poverty. One debate
before the United States Congress involves such a change. Some people
propose a new method called “chained CPI” to calculate adjustments
(Let’s leave the technical details to the side. It’s a new way to
calculate COLA that will save money by reducing payments to the
retired.) The proponents of chained CPI represent it as a minor
budget-saving proposal. It’s minor for them, but it would effectively
end cost-of-living adjustments through a series of economic austerity
tricks. Millions of Americans depend on social security for 90% of their
income. This minor adjustment makes a huge difference to them. For some,
it will mean living on a diet of bread or cat food. For many retirees
with ongoing mortgage payments or those who live in rental properties,
it will mean homelessness. A similar issue involves proposed changes to
the US Medicare program. Over half of all Medicare beneficiaries have an
income of less than $22,000. For many, some proposed changes mean no
medical attention and no prescription drugs. This is a case of policy
design. Seemingly minor figures buried in a large plan can mean reduced
life quality for millions. Proposed changes in Medicare would likely be
a death warrant for several hundred thousand whose coverage will change
so drastically that they do not have access to the health support they
require to live.

This thread began in a response on policy design. This is a specific
and limited instance of design. Student design activities or small-scale
design projects are not cases of policy design. These require more than
an art school perspective, particularly when policy covers such
large-scale issues as public health systems, maritime security, or
pension indexing, large-scale systems involving multiple actors and
multiple overlapping legal or regulatory regimes.

You and Jonas pointed to situations where we can act before analyzing.
I agree. I’ve often taken action before undertaking a full analysis.
This is entirely reasonable in situations where the stakes are moderate
or low, especially when I can rapidly reverse the effects of a wrong
decision, monitoring the effects of my decision to follow the
consequences. This is often necessary when the effects of inaction are
likely to be worse than the effects of a wrong decision. Designers do
this all the time. Expert designers working in context have a background
of mental models, heuristics, and experience – sometimes including
research – to permit skilled judgment in practice.

Policy design situations are generally too large and complex to permit
action before analysis. Time horizons, budgets, and stakeholder
constraints make consequences far more significant than any other kind
of project we address.

Action before analysis works most of the time for what most of us do.
This is especially true for those who analyze and reflect on the their
actions. Starting with analysis is important when people design policies
affecting many people, especially when we design policies for other
people that may not affect us.

In this thread, I’m not describing all the kinds of design we
consider on this list or in our work. I focus on policy design.

Yours,

Ken

Professor Ken Friedman, PhD, DSc (hc), FDRS | University Distinguished
Professor | Dean, Faculty of Design | Swinburne University of Technology
| Melbourne, Australia | [log in to unmask] | Ph: +61 3
9214 6078 | Faculty www.swinburne.edu.au/design