Before the work, one question:
Is there a prior state you are restoring?
Yes, and you can name it. This is a fire. Route to
@medicine-wheel/gap-analysis
— evidenced baseline, observation, difference, verifiable steps.
No, nothing existed that you are returning to. You are creating. Route to structural tension — the outcome named as a state that will exist, held against current reality.
Looking for creative problem solving? It lands here first, then goes one way or
the other. The package that carries that name
(@medicine-wheel/creative-problem-solving)
is a signpost pointing at this one.
[!WARNING] Experimental alpha. Part of the Medicine Wheel Developer Suite, which is under active development. APIs change between patch versions and all packages move in lockstep — pin exact versions. See ALPHA.md.
Problem-solving is not the degraded case. It is the correct case for a specific situation: something worked, it stopped working, restore it. Creating is correct for a different one: bring into being a state that has never existed.
They produce different structures. Problem-solving oscillates back to a baseline — which is exactly right when you are below the baseline. Creating advances toward an outcome. Using the wrong one is how rigorous work goes in circles: chart an oscillating goal and you build a very careful machine for returning to where you started.
This package answers the routing question and nothing else.
A feeling of urgency is not evidence. Under pressure every situation feels like a fire. One question separates them:
Is there a prior state you are restoring?
If you can name it and it genuinely existed — fire. Gap analysis is the right instrument. If you cannot, you are creating, and difference-closing language will mislead you, because there is no baseline to close toward.
import { readOrientation, worthSaying } from '@medicine-wheel/creative-orientation';
const reading = readOrientation({
outcome: 'catch cross-service contract breaks',
restores: null,
});
reading.orientation; // 'advancing'
reading.route; // 'structural-tension'
worthSaying(reading); // true — the phrasing and the situation disagree
reading.advice;
// "This outcome is phrased as removing something, but nothing is being
// restored. Elimination framing here optimises against a list of known
// failures rather than toward the state you actually want."
The baseline claim decides the route. Vocabulary only ever contributes signals,
because reduce latency is a correct outcome during an incident and a symptom
worth surfacing as a product vision. A blocklist would be wrong in precisely
the situation where that vocabulary is right.
Nothing here blocks a caller. A package that refuses is a package people learn to route around, and routing around it teaches nothing. The reading is returned, the disagreement is named, and the caller proceeds — because sometimes the caller knows something the check does not.
creative-orientation ......... the gate
├── gap-analysis ......... the fire path
└── structural-tension ... the advancing path
@medicine-wheel/brainstorming calls this on every outcome statement it emits,
including its own questions. That is the load-bearing use: an elimination-shaped
option label, once selected by a human, stops looking like an agent’s bias and
starts looking like a requirement.
First release. The surface is small on purpose — the judgement of whether a claimed baseline is real stays with a person, and everything around that socket is deterministic so it can be tested and cannot be forgotten under load.
MIT