Only business events that produce provable results and leave a submittable evidence package can be considered as qualified closed loops for PoB discussions.
When the boundaries are blurred, PoB degenerates into another integral. An effective closed loop must simultaneously satisfy: the final state has occurred, the evidence can be reviewed by a third party, the participants can be distinguished, and the event is uniquely traceable in the system - if one of the four is missing, it can only be counted as process noise or intention inventory.
Four hard conditions (not related to PoW/PoS, but related to the results)
The result is established: not planned, not "in progress", not a unilateral statement. Evidence exists: It can point to specific documents, on-chain records, bank/custodial statements, acceptance signatures, etc. Roles can be distinguished: At least it can distinguish between leading, collaborating, benefiting, and reviewing (even if the subsequent attribution ratio is to be determined). Events can be numbered: Bind to Deal/Task/node to prevent the same result from being packaged and submitted multiple times.
This is completely different from PoW's "meeting difficulty target" or PoS's "signature weight": the "difficulty" of PoB lies in real-world proof costs and cross-verification, not cryptographic puzzles.
Six types of typical closed loops and examples of “valid evidence”
The following examples illustrate what materials often form part of a proof package; the exact list is based on current network rules and Proof Desk templates.
The types of closed loops are different, and the shape of the evidence package is different, but the underlying layer remains the same: Final State × Evidence × Distinguished Role × Unique Event.
Analogy with TradFi’s “deal” concept
In PE/VC, Deal Completion looks at signatures, fulfillment of closing conditions, closure of funds and legal documents - not "meeting the founder a few times". The effective closed-loop logic of PoB is the same: Only events that pass the final state threshold will enter the performance and accounting semantics, otherwise no matter how exciting the due diligence is, they will not be included in the portfolio return.
First write down "what counts", then the exclusion rules of 8.3 will have teeth. Otherwise, the system will continue to be drained under the pressure of "it counts as a contribution", and ultimately the nodes that are serious about closing the loop will be harmed.