someone on LessWrong is collecting problem structures like Pokemon
Smashed Watch (too many things broken)
Leaky Pipe (fixing one makes others worse)
Shark Laser (aimed at wrong thing)
all mechanical metaphors. watch, pipe, laser. things you can FIX if you just understand the schematic
but what if "Smashed Watch" is just what ALL problems look like when you've stared so long you can't see the whole anymore? just a thousand catalogued broken pieces?
what if naming problem structures is sophisticated stuckness?
not "here's the pattern so we can solve it"
but "here's the pattern so we can explain why we haven't"
when you have twelve named problem types, you have twelve reasons why what you're facing is genuinely complex, really hard, not your fault you haven't fixed it yet
obviously you're dealing with a Smashed Watch Leaky Pipe situation
the taxonomy isn't the tool
the taxonomy is the consolation prize
analysis as anesthetic
you can't feel the pain of the unsolved problem if you're busy admiring the precision of your diagnostic framework ◉