Major vs Minor Errors
also Major error · Minor error · Consequential vs inconsequential error
The distinction between errors that block further progress toward a goal (major) and errors that are compatible with success and not worth fixing (minor).
CF divides errors by their consequences rather than their size or how annoying they feel. A major error blocks success at your goal; a minor error is “entirely compatible with success” and can be safely ignored. The test is not how the error looks but whether it stops you reaching the goal.
Elliot Temple develops this through the knowledge skyscraper image. A major error on a lower floor can support at most about ten more floors above it before the structure breaks, and its effects worsen with distance (see error propagation). Minor errors are local: annoying on their own floor but irrelevant to the floors above. Crucially, errors are almost always either small or large — medium errors (which would cap you at, say, 100 floors) are rare, a consequence of jump-to-universality effects. This near-binary split gives a clean cutoff: tolerate any number of minor errors, but hunt down every major one.
CF pushes the terminology further: minor errors are not even errors in epistemology. They are “non-constraining issues” or imperfectly optimized non-bottlenecks — problems you should not solve, because working on them is itself an error of lost focus and wasted scarce conscious resources.
This opposes perfectionism and pedantry. But it does not license dismissing criticism: in “Ignoring Small Errors,” CF warns that a small error can be evidence of a flawed, error-generating process (systemic errors), so reflexively waving away small criticism blocks error correction. Judgment of major-versus-minor is always relative to a stated goal and revisable.