One round of evolution
The system’s structure is rebuilt continuously, and every cycle consists of the same steps. What matters is that only causally linked steps in it are sequential — everything else runs in parallel.
- A participant publishes a time-limited advertisement: what it accepts, what it produces, what it needs, how many more links it can take on.
- Every other participant independently scores the visible candidates by its own criteria.
- The pair negotiates in alternating moves: proposal, counter-proposal, agreement or refusal.
- A link on probation comes into being.
- The boundaries are synchronised — in parallel with the system’s other links.
- The link’s strength and reliability are updated from the actual result.
- The parties review the terms: strengthen, weaken, put to sleep, dissolve.
Why topology is not assigned from above
The environment gives participants the chance to meet but does not choose for them. The reason is practical, not ideological: to choose the optimal structure centrally, you need to know the state of all participants at once. That is either expensive or wrong — by the time it has been collected, the state has already changed.
Local decisions do not produce an optimum. They produce a workable structure that adjusts continuously — and over the long run that turns out to be worth more than a one-off optimum for conditions that are already out of date.
What you see when you watch it
When you run such a system and watch how it behaves over time, characteristic phenomena stand out:
- links concentrate around participants whose capabilities are in demand — hubs emerge by themselves, with no “coordinator” role;
- useless links are not removed abruptly but fade out, and that makes the system more resilient to temporary failures;
- after a participant is lost, the structure rebuilds itself within a few rounds rather than breaking;
- parallel branches of work do not get in each other’s way until they lay claim to a shared resource.
A caveat, without which the list above turns into advertising: these are observations from our runs, not proven laws. Which environment rules lead to stable, useful behaviour and which lead the structure to degenerate remains a subject of experiment.
Where a living graph is not needed
- A process with a fixed order of steps: there you need reproducibility, not adaptation.
- A small number of participants: it is simpler to set the structure by hand and not pay for the negotiation machinery.
- Strict regulatory requirements: if the structure of interactions is defined by regulation, it cannot be changed on the basis of usefulness.
What this changes in how you think
The least familiar part here is giving up the question “which architecture is right”. In its place comes the question “which environment rules lead to the behaviour we need”. This is closer to designing an economy or the rules of a game than to drawing a flowchart: you set constraints and incentives, and the system finds the structure itself.
Hence the way to check it: not “does it match the diagram” but “what happened in the run”. That is the subject of the next part of the series, on reproducible runs and evidence.