ENGL 2105 : Workplace-Based Writing and Research

Home | AI | Calendar

Kinds of Problems

What follows further explores section 1 of the workflow. My goal is to help you organize the problems you brainstorm along a continuum from solvable to unsolvable given your resource constraints -- time, will, money, knowledge, job description, station in life, etc.

"There's an app for that." App-based solutions will become more and more common now that anyone who can clearly express what they want can get an AI to build a prototype in seconds for free, no knowledge of coding required. Keep that in mind as you go along.

So once you have brainstormed a list of problems, as many as you can think of, try to describe each using the categories below, realizing that not all problems fit neatly and sometimes what you think belongs one place might be more accurately placed somewhere else.

To help you organize your thinking, you might use a matrix similar to this one:

Problem Type Knowldge required Duration Frequency Impediments ROI Population
  1. Deterministic -- if X then Y :: outcomes are 100% predictable
  2. example: Domino A falls into domino B, B falls
  3. method: Direct observation, logic
  4. terminology: Cause to effect
  5. possible solution: Increase seperation between dominos A and B

    Purely deterministic problems only occur when everything relevant is known with certainty, under carefully controlled conditions. Outside of basic logic, certain kinds of mathematics, certain kinds of physiological conditions, and closed circuits, most problems identified as deterministic are not. People dislike complexity and are loath to make decisions that have uncertain outcomes. But people are also impatient and so are inclined to ignore uncertainty, fail to see complications, over-simplify, and jam a square peg in a round hole.

  6. Less obviously deterministic -- if X then Y :: outcomes are 99.9% predictable

    The difference between dominoes and teeth is just the complexity of the system, the number of irrelevant variables that might nevertheless come into play. If you don't have good dental hygiene, you will get gum disease ifyou live long enough. Purely deterministic situations, where outcome are 100% predictable, require closed systems.

  7. Probabilistic (Stochastic) -- probability of X :: outcomes are less than 100% predictable but within an acceptable margin of error nevertheless

    A Matter of Persepctive

  8. Complex -- a combination of deterministic and probabilistic, outcomes are predictable but with a wide margin of error
  9. Wicked -- systemic, many contributing factors, diverse ambiguously interrelated and unpredictable consequences, many unknowns, some perhaps unknowables, the strong possibility of both failure and unforeseen consequences for any attempted intervention
  10. Physiological
  11. Psychological -- problems that exist in the suffers' head, imaginary but not unreal
  12. Temporary delusions -- problems caused by ways of thinking rather than brain function per se
  13. Circumstantial -- Context is the controlling factor. People who find themselves there have that problem. Therefore, they don't once they leave, though they may suffer after affects. People who have never been there don't understand it. People who could never be there don't think it exists.
  14. Pseudo -- non-problematic situations that someone frames as a problem in order to profit
  15. Proxy -- a distraction in order to keep people busy so they don't notice what the real problems are
  16. Unimaginable -- Black swans - no one saw it coming and when it arrives, everything changes.

    In the physical world, most events are probabilistic (there's 99.7% chance of a 6.7 earthquake in California in the next 30 years) and their consequences can be measured (# of lives and buildings and infrastructure destroyed). Decision problems stem from figuring out what to do with the numbers. What does a 20% chance of rain look like and what do you do about it? Do you carry an umbrella? Postpone the picnic? Move ahead but scope out nearby shelter? Or focus instead on 80% chance of no-rain? What if 20% becomes 50%? What if picnic is instead wedding? Or rain is wintery mix and picnic is bicycle ride? If you live in California, what do you do with the current estimations?

    In the social world, the probabilities can be complicated by psychology. The numbers, in other words, can be both hard to generate accurately and misleading.

    Politics is applied psychology exacerbated by marketing, ie propaganda

    In both the physical and the social world, once we focus on subsequent events (events that happen after the fact but aren't causally related, correlations in other words and in some cases just things that could happen because something else did, with no measurable link) levels of certainty drop dramatically. Anyone who KNOWS in these contexts is delusional.

    Time frames

    Knowing when a problem is likely to occur gives you insight into how to market and to whom to market

    Durations

    Knowing the duration of the problem gives you insight into customer service

    Frequency

    Knowing frequency gives you a sense of market potential and population target. The more people who suffer, the more people you can sell to, but the more flexible the solution needs to be. If the problem you are solving is unique to an individual, you can tailor it specifically for them. But then you only get to sell one of them, so you need to be able to make it fast and cheap.

    Population

    This is your market but also perhaps your competitors and impeders

    Context

    You need to know this to know where and how to test the efficacy of your solution. A complicated process will fail if it has to be executed in a busy, noisy environment. An unprotected camera won't work underwater.

    Sources of errors

    Things to consider when evaluating the value of a solution

    To sell a solution, people have to believe they have a problem and they have to feel the effects of that problem intensley and immediately. The results of not flossing your teeth won't materialize until 30 years from now. Teaching a flossing habit has to overcome the percieved delay. The people you are trying to sell to need to crave a solution. They also need to believe that what you offer solves their problem, that they can afford it -- three easy payments -- and that there's no risk involved -- money back guarantee.

    The more complex your description, the harder the sell. You need to simplify and if necessary over symplify. Do the work to be certain you've solved the problem but hide the work from your intended population. Implementation needs to be painless and flawless. There's no such thing as user error, as someone once said.