Solution Space
also Restricted Solution Space
The set of answers a problem permits; the tighter its constraints, the more candidate answers can be ruled out as errors trivially.
A problem’s solution space is the set of answers it allows. CF treats restrictions on that space not as mere bookkeeping but as error-detection mechanisms: every constraint that excludes a region of the space converts the answers in that region into immediately detectable errors. Saying “the answer can’t lie between 5 and 50” means any value there is wrong by inspection; saying “the answer must be a whole number” makes any fraction a spotted error, and even enables cheap correction by rounding at each step so small errors never compound.
CF’s distinctive move is to treat shrinking the solution space as a deliberate strategy for reliable thinking. The smaller the space, the more of it is ruled out for free, leaving only a handful of candidates to scrutinize. The limiting case is a binary question — yes or no — where almost everything is excluded and detection of wrong answers becomes trivial. This is why yes-or-no questions are generally easier to reason about and get right. They are also less ambitious and individually less valuable, but CF endorses the small-steps tradeoff: ten easy, low-error steps beat one large risky leap. People stall by attacking a whole undivided problem at once instead of breaking it down.
This connects solution-space restriction to CF’s broader account of error correction and mechanistic thinking: useful constraints, like error bars, are valuable precisely because they make most mistakes cheap to catch.