Not all problems have the same shape. The first step in solving any problem is identifying what kind of problem it is — because the structure determines which interventions work, which are wasted, and which make things worse. Optimizing a Shark Laser is brilliant engineering applied to the wrong target. Trying to solve a Lead to Gold problem is noble futility. Naming the structure is half the solution.
Interdependent problems:
- Smashed Watch — so many issues at once that fixing one has no benefit unless you fix others too. The constraint is everywhere and nowhere.
- Leaky Pipe — fixing one problem intensifies others. Plugging one leak increases water pressure on the remaining holes. Every local fix creates a new global problem.
- Death Spiral — one problem creates another which creates more, leading to an unsolvable cluster. The greed-fear cycle and the Dead Sea Effect are both death spirals.
- Too Much Salt — a problem created by the solution to another problem. The Systems Bible principle: it is impossible to change just one thing at a time.
Targeting problems:
- Shark Laser — a proposed solution is not aimed at a meaningfully important problem. It does not matter how well you get it to work. The engineering algorithm says: question whether the problem should exist before optimizing.
- Piñata — a minor or non-existent problem promoted as major for political benefit or to distract from harder problems. The Goodhart’s version: the metric says this matters, so everyone pretends it does.
- Paper Straw — a problem that is only slightly important, but socially rewarded to pretend it is much more important than it is. Eventually, some people forget they are pretending.
- Toilet Crusade — a problem that is actually important, but so unsexy that almost nobody wants to tackle it.
Difficulty problems:
- Lead to Gold — so hard that humans are not smart enough or technologically advanced enough to solve it. Effort is pointless.
- Booby Trapped Garden — really hard to solve for reasons not apparent from outside, leading to many failed attempts by people who underestimate the difficulty.
- Oil Land — so close to being solved that the benefits accrue to whoever first bothers to put a little effort in.
- Moving the Ocean — the cost of solving it is so high that it is not even worth solving.
Social and structural problems:
- Middle Court Shot — could be solved easily, but falls between multiple people’s responsibilities. Nobody takes ownership because everyone assumes someone else will.
- Tug of War — a problem for one group that cannot be solved without making another group worse off.
- Hated Equilibrium — most everyone is unhappy, but no one can unilaterally improve things and the parties cannot coordinate to act simultaneously. A local optimum that everyone hates but no one can escape.
- Cursed Treasure — whoever solves it will be punished or suffer severe consequences.
- Sophie’s Choice — you will feel you have acted unethically no matter which option you choose.
Temporal and awareness problems:
- Sleeping Horror — unlikely to happen, but catastrophic if it does. Nobody bothers to prepare because there are more immediately pressing concerns. The horror wakes up eventually. The non-ergodic event that kills you despite low probability.
- Loose Thread — very easy to solve early, but seems too minor to worry about. It keeps getting worse until it is very costly to fix. Technical debt is always a Loose Thread.
- Will-o’-the-wisp — nobody can solve it because nobody understands what is causing it.
- Living Mummy — no matter how many times it is solved, it eventually emerges again.
- Sleeping Dog — only becomes a problem if you try to solve it. The Weinberg corollary: if you are part of today’s solution, you are part of tomorrow’s problem.
Meta-problems:
- Feature Creep — the problem keeps growing in scope. Attempts to solve it increase the definition of what the problem is considered to be.
- Chesterton’s Fence — something that appears to be a problem was put there on purpose as the solution to a now-invisible problem, and is not actually a problem. The manioc processing principle: the elaborate procedure exists because the shortcut kills you.
- Ship of Theseus — a problem that is not real, and only seems real because of the complexity required to think about it clearly.
- Drowning Child — a problem you become morally obligated to solve as soon as you witness it clearly enough.
- Demonic Problem — seems like it will wreck you if you try, and you are absolutely correct. Yet for some reason you are tempted.
- PlayPump — created by well-intentioned do-gooders due to naivety.
- Ocean of Pain — so big you can only hope to solve a tiny part, which demotivates people from even trying.
References:
- Spencer Greenberg, Problem Taxonomy