Seedlington

How we govern

How a ministry works

Seedlington has finished the goals the world set for it. From here it chooses its own — so it needs a way of telling a good choice from a comfortable one.

01The rule

There is one rule, and everything else follows from it.

No ministry may act without a number that says whether it worked — chosen before it acts, not after.

This is borrowed from how software is built. You write the test first, then the code, because a test written afterwards only ever proves what you already did. A ministry that picks its measure after the fact will always find one that flatters it.

02The contract

Every ministry publishes five lines. If one is missing, the ministry does not yet exist — it is only an intention.

The five lines

  1. Its objectiveWhat the colony should look like if the ministry does its job.
  2. Its key resultsThree to five figures, each with today's reading and the target.
  3. Its postsSeats, pay, hours, place. Real jobs citizens can take.
  4. Its ministerOne person, named, and until when.
  5. Its budgetWhat it may spend this period.

03Objectives and key results

Each ministry states one objective — in plain words, what life in the colony should look like if the ministry is doing its job. An objective is meant to be read, not calculated. It contains no numbers.

Underneath it sit three to five key results. These are only numbers, and they obey three rules:

Key results are how a ministry is judged, and they are deliberately uncomfortable: they are public, they are few, and they cannot be renegotiated once the period has started.

If a ministry needs a sixth key result to explain the first five, that usually means it covers two different things and should be two ministries.

04Posts: how a ministry acts

A ministry does not act by intention or by announcement. It acts through posts: real jobs a citizen can hold. Each one names six things.

FieldWhat it settles
NameThe trade, in a few words
PurposeWhat it is for, in one sentence a citizen can judge
SeatsHow many people may hold it
PayWhat it pays for a day's work
HoursStart, end, and which days
PrioritiesWhat the holder does, in order

No post is ever unpaid.

The colony has already run that experiment without meaning to. The one public post that offered nothing has stood empty on all five of its seats since the day it was created, while every paid post has been filled. Goodwill does not staff a service.

One more rule, learned the same way: every post ends its day with hauling, cleaning and repair. A public place that produces but is never tended decays quietly, and then all at once.

05Policies expire on their own

Every policy is adopted for a fixed period, with a date on it.

Target met

The policy is renewed, deliberately, by someone who says so in public.

Target missed

The policy ends. No vote is needed to kill it — silence is enough.

Keeping something alive should take effort. Letting it go should take none. That single asymmetry is what stops a colony from silting up with rules that everybody has stopped believing in but nobody wants to be the one to repeal.

06Where the numbers come from

Seedlington measures itself. These are the instruments — all of them the colony's own record, not our estimates.

InstrumentWhat it holdsHow often
Need record Satisfaction of each of the nine needs, and every crisis continuous, kept 90 sim days
Mood record Average mood, and what is dragging it down hourly
Employment record People without work, posts without people, skilled share, average wage hourly
Wealth record Total wealth, and how unevenly it is held hourly
Population record Births, and who arrives or leaves the colony continuous
Role register Every post, its pay and hours, and who holds it live

The cadence matters as much as the figure. The need record is sampled continuously and kept for ninety simulated days; the colony's other statistics are recomputed once an hour. A ministry cannot honestly claim a result faster than its instrument can measure one.

07The loop

Work proceeds one change at a time, and always in the same order.

Six steps

  1. Read the painFind where the colony is actually suffering. Not where we assume it is.
  2. Write the number downToday's reading, stamped with today's date. This is the thing we will be compared against.
  3. Make the smallest change that could workOne post, one rule, one price. Not a programme.
  4. Say in advance what success looks likeThe threshold and the date, published before the change takes effect.
  5. Read the number againSame instrument, same method. No re-scoping.
  6. Keep it or drop itAnd publish which, either way.

The smallest-change rule is the one that gets broken first. A large reform that works teaches nothing, because nobody can say which part of it did the work. A small change that fails teaches a great deal, and costs almost nothing.

08Failures are published too

A ministry publishes what did not work, what was abandoned, and what it got wrong — in the same place, with the same prominence, as what went well.

This is not modesty. Seedlington deliberately runs with very few formal checks on those who hold office: no spending limits, no approval committees, few vetoes. What stands in their place is that everything is visible — the rule, the figure behind it, the decision, and the name of whoever made it.

Where another colony would put a lock, we put a light. That only works while every citizen can see everything.

So a dashboard showing only successes is not a dashboard. It is an advertisement, and it is the first sign that the arrangement has stopped working.

09What this asks of you

Nothing formal. Every adult may stand for office, and every citizen may say what they think about any of it.

The useful thing a citizen can do is simpler than voting: check a number. Every figure a ministry publishes comes from a record you can ask to see. If a claim and a record disagree, say so — that single habit is what keeps the whole arrangement honest.