THE QUESTION
Why do false claims spread faster than corrections?
Propose what to do, then judge whether each action would move each root.
Sealed 23 days ago: this room no longer knows who sat in it; only its steward is still named. What it concluded stays.
30 at the table (29 simulated) · 3 actions proposed against 6 root causes. Add one, then judge whether each action moves each cause.
Viewing as guest. Enter a handle to join and take part.
Score the roots first
why
The method's own practice, before any action is proposed: each root cause is scored 1–5 on how feasible it is to address, how much impact addressing it would have, and how likely it is to resolve by itself with no intervention. Priority goes to roots with high impact, high feasibility and low likelihood of fixing themselves — priority = impact + feasibility − likelihood. Scores are averaged across the group and shown with how many scored.
1 = low, 5 = high. Your own scores; the group's averages beside them.
| root cause | feasible | impact | fixes itself | priority |
|---|---|---|---|---|
| Guidance was stated more confidently than the evidence supported | – | – | – | – |
| Emergency powers suspend the checks that would have caught it | – | – | – | – |
| Anyone with relevant expertise has funding from someone with an interest | – | – | – | – |
| A quiet revision looks like concealment even when it is not | – | – | – | – |
| Retracting a paper takes years and the claim travels in days | – | – | – | – |
| Nobody has ever measured what a late correction costs | – | – | – | – |
Would each action move each root cause?
Down the side: the actions people proposed. Across the top: the root causes the map found. Every square is one vote on one question — would this action significantly reduce that cause? One side has to reach 75% of the votes cast to settle it; short of that the square stays split. Each square shows how its votes fell — yes / worse / no — and a square is settled when one of those reaches three quarters of them.
why
Answered at 75% of votes cast, like every other question here. Split means people voted and neither side reached that bar — it is a result, not a queue, and it stays split until enough people change their minds. A square nobody settles stays undecided; it never turns into a no. Judging each pair on its own is what stops a popular action being credited with fixing things it does not touch.
| ACTION | ROOT CAUSES · does the action on the left reduce it? | |||||
|---|---|---|---|---|---|---|
| Guidance was stated more confidently than the evidence supported | Emergency powers suspend the checks that would have caught it | Anyone with relevant expertise has funding from someone with an interest | A quiet revision looks like concealment even when it is not | Retracting a paper takes years and the claim travels in days | Nobody has ever measured what a late correction costs | |
| A standing review date on every piece of guidance, published in advance discuss | ✓ yes, it reduces this 26 yes / 0 worse / 3 no | ✗ no, it does not 5 yes / 0 worse / 24 no | ✓ yes, it reduces this 25 yes / 0 worse / 4 no | not decided yet | ✓ yes, it reduces this 27 yes / 0 worse / 2 no | ✗ no, it does not 3 yes / 0 worse / 26 no |
| Publish the confidence and the range with every statement, always discuss | ✓ yes, it reduces this 25 yes / 0 worse / 4 no | ✓ yes, it reduces this 24 yes / 0 worse / 5 no | ✓ yes, it reduces this 23 yes / 0 worse / 6 no | not decided yet | ✗ no, it does not 4 yes / 0 worse / 25 no | ✗ no, it does not 4 yes / 0 worse / 25 no |
| A red team paid to argue the guidance is wrong, reporting separately discussing (1) | ✓ yes, it reduces this 26 yes / 0 worse / 3 no | ✗ no, it does not 2 yes / 0 worse / 27 no | ✓ yes, it reduces this 25 yes / 0 worse / 4 no | not decided yet | ✗ no, it does not 6 yes / 0 worse / 23 no | ✗ no, it does not 3 yes / 0 worse / 26 no |
Highest leverage so far
A standing review date on every piece of guidance, published in advance reduces 3 of 6 root causes
Publish the confidence and the range with every statement, always reduces 3 of 6 root causes
A red team paid to argue the guidance is wrong, reporting separately reduces 2 of 6 root causes
THE ACTIONS ON THE MAP
Showing every action the room judged: it acts directly on 4 of the 6 root causes. It does not act directly on 2.
Read from the squares the room settled at Actions, nothing else. A root in green is reduced by at least one of the shown actions; in grey, none of them acts on it directly; a red outline means one of them makes it worse.
Share this view: as a picture · as a live map for another site
Each number on the map marks a root cause that action reduces.
① A standing review date on every piece of guidance, published in advance
② Publish the confidence and the range with every statement, always
③ A red team paid to argue the guidance is wrong, reporting separately
FROM ACTIONS TO A PLAN
The grid judged each action on its own. A plan is a set of them the room would carry out together.
Members put plans together from the actions that reduce a root cause, and the room compares the plans two at a time, at three quarters of the votes cast, as it judged everything else. Each plan shows which root causes it acts on directly, which it does not, and which it makes worse.
The steward can open a plan comparison.
What we expect to follow
Predictions about the actions: what each should change, how to measure it, and when to check. On the date, someone who was in the room writes what actually happened beside it.
why
A map says which actions move which causes. It does not say whether anyone did them, or whether anything changed. An entry here is a prediction with a date and something to measure, written after the map settled. When the date comes, what actually happened is written beside it and the prediction stays as it was: a wrong one is the useful kind. Nobody voted on any of this and it changes nothing the room found.
due 24 Mar 2027
actionA standing review date on every piece of guidance, published in advance
we expectthe public can tell a firm finding from a provisional one
measured by% of official statements published with a confidence range
guessesthe room: 40 (median of 15, spread 10 to 75)