← Previous: An ordinary day
The whole picture 02 Sandbox
Next: Three systems →

The whole picture · about 12 minutes

Whole-system sandbox

Elsewhere, the website explains subsystems one by one. Here they share one screen. Introduce an event and follow what each component passes to the next, in what order and over how many months. Switch off a tile and the event breaks down where the missing link is needed. This is the shortest route to seeing why no subsystem works alone.

Three things before you start

How to read this without mistaking it for a forecast

The numbers illustrate proportions

Indicators run from zero to one hundred and move a few points per step. No economic model or dataset underlies them: they show direction and allow comparisons between runs. Absolute values have no real-world meaning.

Nothing is random

Identical settings and enabled tiles always produce the same sequence. This is deliberate: the aim is to compare configurations, not simulate chance. A real system would, of course, involve uncertainty.

Events are not victories

One of the six ends in a draw rather than a success, with an explanation. A sandbox in which the designer always wins would be advertising, not a tool.

Ten subsystems, six events

The board

Select a tile to disable a subsystem. Choose an event and follow its path. Months advance automatically, the indicators update at each step and the log records what happened and what was missing.

Introduce an event

System settings

These are parameters, not immutable laws. The proposal assumes they will be set and revised through experience when initial assumptions prove mistaken.

Month0

What becomes visible from this height

Four things you cannot see by looking at one subsystem

No subsystem commands another

At no point does one component order another to act. Each starts because it receives something it needs from the previous one. That is the proposed distinction between a system and a hierarchy: there is no single top to take over.

The knowledge map is the bottleneck

Disable it and try any event: the sequence breaks in at least three places. That is why the Tree of Knowledge room was built before voting, and why the early years would largely be spent constructing the map.

Enter the Tree of Knowledge room →

Money is a mechanism, not a prize

Disable Informatium and many sequences still advance, but voluntary activities stop happening. Checking, reviewing, judging and documenting failures are not civic duties here. They are work supported by income streams, intended to make participation worthwhile.

Decisions are a small part of the system

The Voting Portal appears once or twice in each event. Most steps involve measuring, checking, producing and accounting. What is often imagined as “another way to vote” is largely another economy, with voting attached.

What is deliberately left out

A dependency illustration, not a predictive model

The sandbox shows sequence and dependencies: what each component needs and what fails without it. It does not quantify outcomes. None of the numbers comes from data. There is no economic model, population model or measured timescale. Claiming that it “raises trust by twelve points” would describe a coloured bar, not reality.

The hardest events are absent: war, regional collapse, a deliberate attack on the knowledge map, or a dispute involving the people who build verification tools. The last is among the project’s most serious gaps. This sandbox does not pretend to answer it.

Most importantly: none of these six events has happened in practice, because the system operates nowhere. The sequences were written to illustrate how components connect, not drawn from observations.

A system is revealed not when everything works, but by what fails when one component breaks.