The protocol is built around typed, opt‑in subnetworks. Developers define hierarchical type identifiers for their applications and record metadata about them as relons in the datachain. The neuromorphic hypergraph can then be queried to discover which subnetworks exist and what types they use. But discovery stops at metadata. The actual payloads are only accessible to nodes that have explicitly opted into the subnetwork.
Node operators control participation through two lists in their ANNODE properties. The provider list specifies which types the node can handle locally, each associated with a script that processes incoming requests. The whitelist specifies which types the node is willing to accept and forward. If a type isn't whitelisted, requests for it are rejected. This opt‑in model ensures nodes only bear resource costs for applications they explicitly choose to support.
When a node originates a request, it propagates through the network to eligible peers. Each node appends its identifier to a route list. When a provider node handles the request by invoking its configured script, the response travels back along the reverse of that route. Nodes maintain performance statistics per peer and per type, allowing them to probabilistically skip slow or unresponsive peers. Successful return routes are cached for future use, building efficient paths over time.
The Alt Data Network works alongside the other protocols. 1Schema records subnetwork metadata in the hypergraph for discovery. ANTOR handles large file transfers when an alt data response returns a manifest. The separation is intentional: each protocol focuses on what it does best.
Some things worth discussing:
- How typed subnetworks are defined and discovered
- The provider and whitelist configuration
- Flood routing and return‑path tracking
- Adaptive forwarding based on peer performance
- Route learning and caching
- How scripts are implemented and invoked
- Private subnetworks and access control
- Integration with 1Schema for discovery
- Using Alt Data and ANTOR together
- Application examples and implementation patterns
Check out the paper "The Alt Data Network: Application-Specific Payload Distribution Across Opt-In Peer-to-Peer Subnetworks" published at https://anne.media/alt-data-network-pay ... stribution for further insights and share your findings.