return the good or service, and then be assured of a refund from an unknown and untrusted actor
Escrow systems are inherently a second layer atop a currency. The "issuing authority" of a currency does not engage in such dealings with the users of the currency; these arrangements are made by the seller (or a palatable third party on the seller's behalf) or by the buyer (or a third party contracted by the seller, e.g. a credit card company).
With the "un-named payment system", escrow is entirely possible and has been successfully set up in many instances, most notably on popular darknet markets. A type of protocol-layer escrow is also possible, by the way, through the use of multi-signature transactions.
payment system that requires an internet-connected device
This is exactly what the OP article is proposing to build, and what these comments are discussing.
no third parties with the ability to store and process
SPV and other "light" wallets that remain cryptographically verifiable and decentralized without needing a constant, high-bandwidth connection exist and are, in fact, the default option for most people.
This is exactly what the OP article is proposing to build, and what these comments are discussing.
If you read the white paper, which apparently nobody else did, they say (paraphrased) "Look, this thing is online but we recognize that not everyone has a computer or cellphone so part of the system is a physical card like a debit card that poor people can use".
Comments
Escrow systems are inherently a second layer atop a currency. The "issuing authority" of a currency does not engage in such dealings with the users of the currency; these arrangements are made by the seller (or a palatable third party on the seller's behalf) or by the buyer (or a third party contracted by the seller, e.g. a credit card company).
With the "un-named payment system", escrow is entirely possible and has been successfully set up in many instances, most notably on popular darknet markets. A type of protocol-layer escrow is also possible, by the way, through the use of multi-signature transactions.
This is exactly what the OP article is proposing to build, and what these comments are discussing.
SPV and other "light" wallets that remain cryptographically verifiable and decentralized without needing a constant, high-bandwidth connection exist and are, in fact, the default option for most people.
If you read the white paper, which apparently nobody else did, they say (paraphrased) "Look, this thing is online but we recognize that not everyone has a computer or cellphone so part of the system is a physical card like a debit card that poor people can use".