What changed
What the paper said, and what happened after.
Twelve records. Each one keeps the design apart from the later network, the failure, or the thing that replaced it.
- CurrentTerra: the peg was the design, and the design failed.The paper's thesis was that a dollar token could be held by arbitrage against the system's own token. Practice showed that defence and crash can be the same trade.
- CurrentDiem: a careful design that did not become a network.The paper is a permissioned payment system and a reserve. The project was wound down. A design is not a licence.
- CurrentPlasma: the exit only works if the data was published.The paper's thesis was a tree of chains, with the parent as the court. The hard part was proving you could leave when the operator withheld data.
- CurrentArbitrum: one name, two eras.The paper asserts a result and lets a challenger force a step-by-step dispute. The later network is an Ethereum rollup with its own client, data posting and upgrade keys.
- CurrentOuroboros: the paper is the slots. The network is later.The design divides time into slots and samples leaders from a stake snapshot. A live chain adds clients, parameters, delegation and an exit. Those are not the proof.
- CurrentFilecoin: storage was proved. Retrieval was a market.The paper's hard part is showing that a provider still has the data. Getting that data to a user, quickly, is a different market.
- CurrentInterledger: connectors, not a new chain.The paper packets a payment across systems that do not share a ledger. The connector who fronts the value is the trust assumption.
- CurrentHelium: the proof was coverage. The customer was someone else.The paper's mechanism rewards hotspots that can show they hear one another. Hearing is not paid traffic.
- CurrentCorda: a shared record for parties who already have a contract.The design is not an open membership network. It is a way for known firms to share a state they can both check.
- CurrentPolkadot: shared security was the paper. Slots were the market.A relay chain checking many chains is a design. Who rents a slot, and for how long, is a later rule.
- CurrentCosmos: many chains, and no one chain's security covers the rest.The paper's point is interoperability without forcing every application onto one validator set. The cost is that a weak zone is not saved by the hub.
- CurrentEOSIO: speed came from a short list of producers.The paper trades open membership at the block level for a delegated set. Who sits in that set is the system.