Decentralized Discovery: How Crawlers Could Replace Matchmakers

A matchmaking platform is, at its core, an index. It stores profiles, runs queries against them, and returns results. The platform's value comes from the size of its index and the quality of its query model. The platform's power comes from the fact that it controls both.

Decentralized discovery inverts this. Instead of profiles living in a central index, they live on personal domains. Instead of a platform crawler building the index, any agent that understands the protocol can crawl and index independently. The index is not owned; it is assembled on demand from publicly accessible data.

This is how the web works for documents. Google does not own the documents it indexes. It crawls, parses, and ranks content that lives on independent servers. The value is in the crawler and the ranking model, not in the content itself.

PairGeek Schema enables the same architecture for people. A crawler that discovers pairgeek.txt files, resolves the schema URIs, and builds a compatibility index can provide matchmaking functionality without any of the profiles being hosted centrally. The profiles remain on personal domains, under personal control.

The practical difference: when a centralized platform shuts down, its index disappears. When a protocol-based crawler shuts down, the profiles remain. Any new crawler can rebuild the index. The data is not hostage to the platform.

This is the structural argument for protocol-first design. The value of a matchmaking index should accrue to the people in it, not to the platform that assembled it.