Beekeepers sit on a Watts–Strogatz small-world network: each member starts tied only to their nearest neighbours on a ring (their local club), then every tie is rewired to a random member with probability β — modelling the long-distance "networking" ties formed at conventions, online forums and shared research projects. Even a small β collapses the average distance between any two members, which is exactly why a handful of well-travelled beekeepers can carry knowledge across an entire region fast.
Every few seconds a new season passes. Each already-informed member attempts to teach every uninformed neighbour it is tied to, independently, with per-tie probability p (or p × boost if the teacher has already earned mentor status). A member who has personally taught two or more others is promoted to mentor. This is the standard independent-cascade diffusion model used to study how practices spread through real social networks.
P(neighbour learns this season) = 1 − (1 − p_edge)
p_edge = p (peer-to-peer tie)
p_edge = p × boost (mentor tie)
mentor status ⇐ taught ≥ 2 members
Each season, generational turnover retires a fraction of members and replaces them with novices who must be taught from scratch — but every peer-teaching event and every citizen-science observation a member ever contributed stays counted in the community archive, which only ever grows. That persistence, outliving any single beekeeper's active years, is the network's legacy.
- Transmission rate (p) — how readily know-how passes along a single tie in one season.
- Mentor influence — how much more effectively a mentor (someone who has already taught ≥2 people) teaches compared to an ordinary peer.
- Long-range ties (β) — the share of ties that reach outside the local ring into the wider network; raise it to see how "networking" beyond the local club accelerates spread.
- Generational turnover — the fraction of members replaced by newcomers each season; raise it to see how fast experience is lost even as the archive keeps climbing.