5 ms·

> Or, if you are planning to use the note immediately or near term in commerce, you might have sufficient confidence that our re-key service won't shut down (bu

> Or, if you are planning to use the note immediately or near term in commerce, you might have sufficient confidence that our re-key service won't shut down (but yes, this obviously implies some trust if you hold notes without re-keying of claiming). Does this mean that your private servers have the ability to re-key notes? What prevents someone from scanning the chip or storing the user key that they generated, handing the note to someone else, and then walking around the corner and immediately using the scanned information to re-key it? Without understanding more about the technical details, this feels like unless a user is using an app to re-key immediately at the time of transfer that it's significantly less secure than normal cash. I guess I'm not completely clear on what you're doing with user keys. --- > Because with a phone alone they need to store their keys locally. But they have to do that here too, or is the assertion that they don't need to back up the user key necessarily and that it's not secret? Is the user key just plain-text readable? If the user key is accessible from the chip just via scanning (not cutting), and isn't a user-defined secret that only they know (which seems like would require some form of backup), then does that mean an attacker can just walk past my wallet with a reasonably powerful RFID reader and then rekey every single one of my bills?

ccamrobertson 4y ago
> Without understanding more about the technical details, this feels like unless a user is using an app to re-key immediately at the time of transfer that it's significantly less secure than normal cash. We will have a security overview doc up soon, we ran out of time to have it ready today. Long story short, there are pending key rotation states and the latest holder can always re-key the note again. But it is indeed distinct from cash; we can't just state a note has value like governments do, by fiat. We have to provide users with sufficient information to assess the veracity of that claim relative to the functions of Bitcoin and the information we provide on the note. > But they have to do that here too, or is the assertion that they don't need to back up the user key necessarily? Not necessarily; in the condition where a user is confident they are the only ones that can claim, they can place the notes in a safe the same way they might printed Bitcoin keys or Opendimes (I would recommend a fireproof safe).
- throwaway92394 4y ago
  > We will have a security overview doc up soon, we ran out of time to have it ready today. I mean this respectfully - why would you release a cryptographic method of transferring money - without going into detail of how it actually works? That's kinda the entire point of cryptocurrency, that we have a way to be mathematically confident the money is safe - but you didn't tell us the math. Especially on HN were a lot of the audience is technical enough to want to and be able to verify it to some degree. I see you have patent US10896412B2, which honestly I have to ask how this is any different from any other hardware wallet? > A physical cryptocurrency may comprise a physical medium and an attached processor. I read some of the details (admittedly not all) and I'm still unsure how this is different from any other hardware wallet. AFAIK this is just a hardware wallet that exposes it's public key, then exposes it's private key when you cut the wire? Then we still need your signature to transfer the crypto, which is so double spend is prevented? Also, if your servers go down or you're hacked or rm -rf dir/ * happens, will all the notes become unusable? Are we relying on you to maintain servers indefinitely?
  - ccamrobertson 4y ago
    > I mean this respectfully - why would you release a cryptographic method of transferring money - without going into detail of how it actually works? That's kinda the entire point of cryptocurrency, that we have a way to be mathematically confident the money is safe - but you didn't tell us the math. Especially on HN were a lot of the audience is technical enough to want to and be able to verify it to some degree. As noted, the app will be open source as will the (brief) Bitcoin script for audit. For the sake of simplicity 2-of-2 multisig that downgrades to 1-of-2 multisig over time should capture the essence of how these notes work. > Also, if your servers go down or you're hacked or rm -rf dir/ * happens, will all the notes become unusable? Are we relying on you to maintain servers indefinitely? Nope. Users can always claim funds after the expiration date of January 3rd 2029 using the user key stored on the note.
    - throwaway92394 4y ago
      > For the sake of simplicity 2-of-2 multisig that downgrades to 1-of-2 multisig over time should capture the essence of how these notes work. > Users can always claim funds after the expiration date of January 3rd 2029 using the user key stored on the note. Interesting, thank you. That does seem more reasonable to me. Best of luck with your project.

ccamrobertson 4y ago

> Without understanding more about the technical details, this feels like unless a user is using an app to re-key immediately at the time of transfer that it's significantly less secure than normal cash. We will have a security overview doc up soon, we ran out of time to have it ready today. Long story short, there are pending key rotation states and the latest holder can always re-key the note again. But it is indeed distinct from cash; we can't just state a note has value like governments do, by fiat. We have to provide users with sufficient information to assess the veracity of that claim relative to the functions of Bitcoin and the information we provide on the note. > But they have to do that here too, or is the assertion that they don't need to back up the user key necessarily? Not necessarily; in the condition where a user is confident they are the only ones that can claim, they can place the notes in a safe the same way they might printed Bitcoin keys or Opendimes (I would recommend a fireproof safe).

throwaway92394 4y ago
> We will have a security overview doc up soon, we ran out of time to have it ready today. I mean this respectfully - why would you release a cryptographic method of transferring money - without going into detail of how it actually works? That's kinda the entire point of cryptocurrency, that we have a way to be mathematically confident the money is safe - but you didn't tell us the math. Especially on HN were a lot of the audience is technical enough to want to and be able to verify it to some degree. I see you have patent US10896412B2, which honestly I have to ask how this is any different from any other hardware wallet? > A physical cryptocurrency may comprise a physical medium and an attached processor. I read some of the details (admittedly not all) and I'm still unsure how this is different from any other hardware wallet. AFAIK this is just a hardware wallet that exposes it's public key, then exposes it's private key when you cut the wire? Then we still need your signature to transfer the crypto, which is so double spend is prevented? Also, if your servers go down or you're hacked or rm -rf dir/ * happens, will all the notes become unusable? Are we relying on you to maintain servers indefinitely?
- ccamrobertson 4y ago
  > I mean this respectfully - why would you release a cryptographic method of transferring money - without going into detail of how it actually works? That's kinda the entire point of cryptocurrency, that we have a way to be mathematically confident the money is safe - but you didn't tell us the math. Especially on HN were a lot of the audience is technical enough to want to and be able to verify it to some degree. As noted, the app will be open source as will the (brief) Bitcoin script for audit. For the sake of simplicity 2-of-2 multisig that downgrades to 1-of-2 multisig over time should capture the essence of how these notes work. > Also, if your servers go down or you're hacked or rm -rf dir/ * happens, will all the notes become unusable? Are we relying on you to maintain servers indefinitely? Nope. Users can always claim funds after the expiration date of January 3rd 2029 using the user key stored on the note.
  - throwaway92394 4y ago
    > For the sake of simplicity 2-of-2 multisig that downgrades to 1-of-2 multisig over time should capture the essence of how these notes work. > Users can always claim funds after the expiration date of January 3rd 2029 using the user key stored on the note. Interesting, thank you. That does seem more reasonable to me. Best of luck with your project.