'Bitcoin Box' Can Process Payments With No Web Connection

[fenghush]

If the channel is not encrypted regardless if it's bluetooth, the internet or nfc the data exchanged can be tampered.

[redsn0w]

Why is Bluetooth insecure for Bitcoin transactions? I don't understand. Transactions are already sent signed, so it's not like someone could tamper them, or can they?

I don't know if it is possible "tamper" a transaction when it was sent, maybe you can modify it before but not  during/after the sending.

OK, so sending the transaction via Bluetooth is secure. Is there something I'm missing? Maybe the other ones who said it's insecure can answer me.

I would like to  read a reply from those users that have told it is not secure , can you explain better ? Thanks.

[nextblast]

That's a good invention.
No one's gonna steal your coin in this way!
But how does one send coin if the wallet has long been offline? (No information of TX in)

[Elwar]

After reading more about this, I found the title confusing. I assumed the merchant could somehow process a payment without a web connection.

But the merchant needs a web connection and the user can sign a transaction and pay without a web connection, sending the signed payment over Bluetooth.

But if you have a cell phone, it is very likely that you already have an Internet connection.

If it could be done where the merchant does not need a connection, that would be revolutionary. I have no idea how it would be done though.

[enryk]

"A prototype cryptocurrency point-of-sales (POS) terminal called Bitcoin Box can process transactions without the need for an Internet connection. Instead it relies on Near Field Communications (NFC) and Bluetooth to enable payments."

I think this is a great news, and in the next 5-6 years the bitcoin will be more easily to use.


Source : http://www.coindesk.com/bitcoin-box-can-process-payments-web-connection/

I think this will change almost everything, as it will allow merchants in poor parts of the world without a net connection to accept payments as well, and later like in a bank get it converted to fiat.
However the volatility would have to be stable at that point.

[Elwar]

I think this will change almost everything, as it will allow merchants in poor parts of the world without a net connection to accept payments as well, and later like in a bank get it converted to fiat.
However the volatility would have to be stable at that point.

The title is misleading...the merchant needs a net connection.

[redsn0w]

I think this will change almost everything, as it will allow merchants in poor parts of the world without a net connection to accept payments as well, and later like in a bank get it converted to fiat.
However the volatility would have to be stable at that point.

The title is misleading...the merchant needs a net connection.

As I understand :


The buyer doesn't need an internet connection , but the seller (merchant) should have an internet connection. 

Here a demonstration :


- http://andyschroder.com/BitcoinFluidDispenser/

[kenbytes]

what a brilliant idea! Can´t wait to see that feature!

[coa032]

Well, this is still first version of bitcoin box. Over the years they will probably find the way for security issues and stuff like that, but anyway great news.

[R2D221]

If the channel is not encrypted regardless if it's bluetooth, the internet or nfc the data exchanged can be tampered.

Yes, but Bitcoin transactions are signed, so any tampering between the sender and the receiver would result in an invalid transaction, right?

Or maybe the hacker just replaces the transaction with an entirely different one, which makes no sense to me, since they can just send it from the beginning anyway.

[Nagle]

There could be a danger of a MITM attack, though. The sender must be sure they are sending to the correct address.

That's the basic problem with authenticating to a new device via an omnidirectional radio signal. You don't really know who you're talking to, and you have no shared secret prior to authenticating.

[bitllionaire]

so ,if I have understood it

This transmits the tx to the network instead of waiting to receive it, isn't it?

[redsn0w]

There could be a danger of a MITM attack, though. The sender must be sure they are sending to the correct address.

That's the basic problem with authenticating to a new device via an omnidirectional radio signal. You don't really know who you're talking to, and you have no shared secret prior to authenticating.

So this system is not "secure" .... Maybe they will change way  , I don't know.

[fenghush]

If the channel is not encrypted regardless if it's bluetooth, the internet or nfc the data exchanged can be tampered.

Yes, but Bitcoin transactions are signed, so any tampering between the sender and the receiver would result in an invalid transaction, right?

Or maybe the hacker just replaces the transaction with an entirely different one, which makes no sense to me, since they can just send it from the beginning anyway.

I wouldn't be 100% confident to say no attacks are possible, remains to be seen, tested, audited properly.

[roadbits]

and hope we will see much more variety of Process Payments to experiment with in the near future.

[Q7]

I think this will change almost everything, as it will allow merchants in poor parts of the world without a net connection to accept payments as well, and later like in a bank get it converted to fiat.
However the volatility would have to be stable at that point.

The title is misleading...the merchant needs a net connection.

As I understand :


The buyer doesn't need an internet connection , but the seller (merchant) should have an internet connection.  

Here a demonstration :


- http://andyschroder.com/BitcoinFluidDispenser/

From there I presume once the transaction between two parties is done via nfc is completed, the merchant will then need to broadcast to the network for it to get written to the blockchain. But what if before that was done, can the sender perform a double spend?