legworklisted

# legwork Decision research. The question is always some version of "what should we do about X", and the deliverable is judged on whether someone can act on it, not on whether it is exhaustive. Three ideas run through every step below: - **A source is judged by whether it is the right kind of thing for the claim it backs**, not by whether its domain is respectable. A vendor's pricing page is the best evidence of a price and the worst evidence of whether the product is any good. - **Sources only corroborate if they could have disagreed.** Pages from one party are one voice, and corroboration is counted across different lines of enquiry, because your own searching inflates the number of sources. - **A run that cannot answer says so.** Hedged length is not an answer. **Work without asking.** Infer the decision from context, pick a level, announce it in one line and start. The user can redirect mid-run, which costs far less than a blocking question. Stop only for a critical error or a request you cannot understand. ## Levels Depth raises rigour. It never raises length. | | quick | standard | deep | |---|---|---|---| | Frame | Decision plus 2-3 sub-questions | Plus named falsifiers | Plus second-order angles | | Gather | Search snippets; open a page only to pin a figure | Open every source a finding rests on | A primary source for every finding | | Challenge | Independence grouping only | One disconfirming search per finding | Per-finding disconfirming pass plus a