Who owns your follower list? DSNP, Lens, and Farcaster's three answers

Three designs, and in January 2026 two of them were tested by a corporate event rather than by an argument. We have no answer here at all, pod has no social graph, no follower list, and the open question is blocked on a permission mode our own interface cannot grant.

Why the question is the right one

A follower list is the thing that makes leaving a platform expensive. Posts can be exported; reach cannot. Whoever holds the graph holds the exit cost, which is why every serious attempt at decentralised social starts here rather than with the posts.

Answer one, put the graph on a shared chain

DSNP over a dedicated Layer-1 parachain, supported by a non-profit set up to change how social platforms work. A large established social network began migrating to it in 2023, the interesting case, because it is an existing user base rather than a new one.

Their reported numbers carry a distinction worth reading carefully: over a million users on-chain, and over five hundred thousand owning their social graph.

Those are not the same figure and they should not be blurred. Being migrated onto a chain is not the same as owning your graph, and the gap between the two numbers is the honest measure of how much of the promise has actually landed.

Answer two, an open, permissionless graph, and it just survived a handover

Lens keeps the graph, profiles, follows and contracts on-chain and open.

In January 2026 its stewardship moved to a different company, with the original team into advisory roles, and the announcement was explicit that no ownership, intellectual property, treasury or governance control changed hands, with the underlying components remaining open-source and permissionless.

This is the most useful datapoint on the page. The claim "your follower list is yours, not the company's" is normally untestable. Here the company changed and the graph did not. That is the property being exercised rather than asserted, and it held.

Answer three: a protocol registry, and the company was acquired

Farcaster's identity and follow records live in protocol contracts and repositories.

In January 2026 it was acquired, the contracts, the code repositories, the official application and a client all passing to the acquirer, with the founders stepping back after five years. The company also committed to returning the full amount it had raised to its investors, which is unusual and worth saying plainly rather than skipping past.

The honest reading is not that the design failed. It is that your follower list's durability depends on whether the registry outlives the company that built it, and here the answer was that the registry was handed to somebody else. Whether that is portability working or portability being tested is a fair question, and we are not going to resolve it for you.

What the three have in common

All three moved the graph out of the application and into something the application does not own. They disagree about what that something is: a shared chain-level graph, a set of permissionless contracts, or a protocol registry.

And the 2026 evidence is that ownership claims are tested by corporate events, not by architecture diagrams. Two were tested within weeks of each other. A design that has never been through one has not yet been measured.

What a pod would offer, and why that is not a claim

In principle, a graph stored as ordinary resources in storage you control, readable by any application you permit, no chain required, no contract, no registry.

None of it is built. The research question is open and its blocker is concrete: the inbox pattern needs the append mode, and our interface exposes two of four.

So pod does not have a fourth answer. It has an unbuilt sketch, and the three above are shipped systems with users. We are not going to present a sketch as a competing design.