Ten subsystems. Two fabrics. One uncomfortable open problem. This page is the live diagnostic surface for the Body-as-Network thesis — every card maps a biological element to a network primitive it provably is.
The mapping earns its place if it transfers a shared constraint, not a surface resemblance.
A dedicated line carries an action potential from a precise origin to a precise destination. Latency is low. Delivery is targeted. A synapse is a router with a weight.
Connection-oriented, ordered, fast. This is how the body coordinates anything in real time — a routed packet from a node to a node, not a broadcast.
A hormone is released into the bloodstream and diffuses everywhere. There is no addressee. Every cell hears the message; only those with the matching receptor act on it.
Publish–subscribe over a shared medium. The body’s global mutable state — a circulating concentration that, given time, every part reads from the same pool.
A hormone is not received by every cell it reaches — only by cells expressing the matching receptor. Ligand–receptor specificity is content-addressed, schema-checked input.
Receptor up- and down-regulation is autoscaling and rate-limiting. Subtypes are API versioning. A viral spike protein is a malformed message that passes type-check and is admitted as valid.
Trans-membrane voltage patterns propagate cell-to-cell through gap junctions, storing the target morphology of tissues. The pattern says what to build with the proteins the genes spec.
Software-defined biology. The genome is the source image; the bioelectric pattern is the running coordination — rewritable at runtime, where the agency lives.
Lymph nodes scattered through the body authenticate every passing cell. Antibodies catalogue past threats. The blood–brain barrier is an allow-list at a network boundary.
Distributed authentication and anomaly detection. Self vs non-self is a continuously-running classifier with a long memory and adversaries actively trying to mimic the type signature.
A versioned build artifact, read on demand. One gram theoretically holds hundreds of exabytes; recovered intact from remains hundreds of thousands of years old.
The highest-density, longest-lived storage medium known. Industry is no longer inventing a new medium — it is reaching for the biological one.
The bloodstream is the medium messages travel through — hormones, immune cells, oxygen, signal molecules. The heart is the pump that keeps the fabric live.
A shared transport layer with bounded latency and capacity. Every service is reachable through it; partition it and the system fails fast.
Sits exactly on the seam between the two fabrics — translating neural packets into endocrine broadcasts and back. The body’s hardest piece of integration code.
The adapter where shared-state and message-passing meet. Engineers know this seam: the oldest tension in distributed systems, solved here in wetware.
A statistical partition of internal from external states, interacting only through sensory and active states. Blankets nest: organelle inside cell inside tissue inside organ inside organism.
A scale-free architecture of selves within selves. Each level is a node whose internal states are the next level’s network. The body provably is a multi-scale inference network.
How does a swarm of asynchronous parts resolve into a single experiencing subject? Broadcast gives agreement, not unity. The binding of experience may not be classical at all.
A thousand servers reading the same config agree; they do not thereby become one server. Unity demands a non-classical sharing of state — one joint state with no copies and no channel.
Engineered systems that imitate the architecture of biological intelligence — neuromorphic chips, edge-AI meshes, attention + memory.
Engineered + biological components, each doing what it is best at. BCIs already read and write neural signal. Cortical Labs sells living neural processors.
Engineered intelligence delivered in substrates that are themselves alive — synthetic cells, programmable tissues, molecular machines with intelligence built in.
We will not reach unified machine minds by adding GPUs. We will reach them by changing the physics of the hardware — and the most successful unified mind we have ever found will be the reference design.