We can send slatepacks through Bluetooth for this.
For the communication between phones, yes. We could also use QR codes or Near Field Communication (NFC). But we still need internet access to the blockchain to prove there is no double spend, unless we hardware chips to protect against double spends.
I donāt think there is a problem if we have internet access at 1 of 2 devices. Itās only question of trust. This solution can be called Grin Tethering
Another option includes mesh networks built on protocols such as the Reticulum network, which can use LoRa.
Unsigned.io is producing their own open source transceiver based on radio transmission, sharing much of the ethos of the Grin community. These can be created with off the shelf components.
A few years back Gotenna introduced TxTenna as a way to broadcast signed Bitcoin transactions off grid. It still requires an online TxTenna user to confirm however they may be multiple āhopsā away. It doesnāt look like these are in sale any more, likely because of supply chain issues but Amazon has plenty of options for LoRa development boards, such as from LilyGo, Heltec and Unsigned.
Crazy idea: Larry asks Steve to open a mobile hotspot
Since a more complex way is appreciated:
- Steve choose the offline Button inside the app
- This triggeres a BLE scan for nearby devices
- Larrys app recognized that a BLE connection is requested and can accept the connection
- Steve choose his amount and receipient wallet - submit it
- the app encrypts Steves data (wallet address, pw) and send it via BLE to Larry
- Larrys app forward Steves encrypted data to an axios/ajax call
- Axios/Ajax target would be an online onchain wallet with RPC
- A Service waits on top of the RPC API and decrypt Steves data and authorize Steve
- The transaction is triggered
I love the idea of transacting via Bluetooth or LoRa, BUT these are still ignoring the fact that the transaction is not secure until it gets broadcast and mined into the blockchain.
Trompās suggestion of payment channels with long dispute times addresses it for one very specific scenario, if Larry and Bob were precisely in the situation where they knew they would be offline together and prepared in advance by opening a direct 1:1 channel with a dispute time longer than their trip time⦠aside from that fairly specific scenario, offline TXs are not possible.
I am doing this at my app, 1st mobile device has internet connection and has running Grin node, 2nd device is offline and using Bluetooth to access node on 1st device to broadcast txs in case of sending. It will be very useful to use with hardware wallet (later).
Another way to translate slatepacks remotely.
Letās say your friend is living across the street from you, he has no internet/slow internet connection, how to send GRIN to him? With Light or Sound! He needs camera/microphone at his device and ability to send signal back.
Inspired by Morse/Semaphore system. We just need to introduce support for lowercase chars.
This system can be implemented everywhere across internet too to make slatepacks more user-friendly and funny.
You can make that a separate topic on the forum, or continue it here:
Would a one-time-usable bearer kind of thing work? Larry would compose the one-time-bearer-thing by crafting a new address, transferring the 20 coins to it, then securely transmitting to Steve the secret key for this āpouch.ā It could be written on the back of an envelope ā thereās no hurry. I call these lightweight accounts āpouchesā and I donāt think anyone else does. Anyway, would a pouch-based approach meet the needs? ādouble spendingā becomes a race to be the first to empty the pouch. Steve and Larry are the only actors in the scenario with the secret key, unless it gets intercepted or something. Assuming Larry isnāt a jerk about it, Steve is the only one who can empty the pouch and he can do it at his convenience, when he has internet access. I doubt you can do the scenario without Steve having to trust Larry. If the two of them are in the same place and Larry is the only one with a working device and Larry is no super-hacker who would be suspected of having a fake lookalike app, Steve can watch Larry craft the pouch before he gets his copy of the key. I realize I may have just described a way to implement āunidirectional payment channelā although not securely. And if Steve gets hit by a bus before he can empty the pouch, Larry can ārevertā the pouch by emptying it back to his main account. Thoughts?