LIVE FEED — JUL 28, 2026
Uncategorized

The Frame Problem: Knowing What Stays the Same

When something changes, most things do not.

By · July 7, 2026 · 2 min read

When something changes, most things do not. A cup moved leaves the room, the laws, and ten thousand facts untouched. Representing this vast constancy, and reasoning efficiently despite it, is the frame problem.

The core idea

The frame problem is the challenge of specifying what does not change when an action occurs. Listing every unaffected fact is impossible, yet assuming everything changes is absurd. The default of persistence must be handled without stating it exhaustively.

Why it holds

The difficulty is combinatorial. The number of facts that remain true after any action is essentially unbounded. No explicit enumeration of non-effects can keep pace.

A deeper reading

The turn is that common sense hides enormous machinery. What seems trivial to a person, knowing the world mostly stays put, conceals a deep representational demand. The obvious is computationally heavy.

The tension within

Attempts to formalize persistence face subtle failures. Simple defaults produce paradoxes when actions interact or when exceptions cascade. The naive rule that things stay the same is harder to state than it sounds.

Where it fails

The implication is that reasoning about change requires reasoning about stability. Intelligence must encode the inertia of the world as much as its dynamics. Knowing what not to reconsider is itself a skill.

The larger point

The frame problem is the difficulty of representing what remains unchanged when actions occur. Persistence is vast and cannot be listed, yet must be assumed. Common sense conceals a heavy demand: knowing what to leave alone. Seen this way, the concept is less a fact to be filed than a lens through which other facts arrange themselves. Its worth lies not in any single application but in the pattern of thought it makes available. To hold it clearly is to see a whole family of problems as variations on one theme, and to recognize the same shape recurring where it was not expected.