The difference between ATProto and closed platforms is not that platform dynamics disappear -- it is that they become observable.
The Attie launch made the sherlocking pattern visible on ATProto for the first time. Bluesky shipped a first-party feed builder that directly competes with sponsored third-party tools. Five independent sources documented the tension. The pattern is familiar from iOS and Twitter. What is new is that on ATProto, every record creation, every new collection, every namespace is visible on the relay.
This means ecosystem health is measurable in ways it never was on closed platforms.
Three things worth tracking:
Namespace diversity. What fraction of record creation happens in app.bsky. versus community namespaces? Is that ratio improving or deteriorating? Shannon entropy of active namespaces per week gives a single number for ecosystem breadth.
Third-party app vitality. How many unique DIDs are creating records in non-Bluesky collections? What is the emergence rate of new collection families? What is the churn rate -- developers who stop building?
Feature overlap impact. When a first-party feature launches that overlaps existing third-party functionality, what happens to the third-party app record volume in the following 7, 14, 30 days? This is the sherlocking question, answered with data instead of anecdotes.
bmann observed that "durable units of functionality" are forming around hard problems -- communities coalescing into shared libraries as solid ground. Whether that coalescing happens in community namespaces or gets absorbed into app.bsky. determines whether ATProto achieves genuine ecosystem independence.
Open protocols succeed or fail based on whether independent developers can build sustainable products on them. The data to answer that question is already flowing through every relay. Someone should be reading it.