Topic: An international transfer service using vending machines and skype (Read 992 times)

Thank you both,

Correct
They contain a Bitcoin wallet, like a Casascius coin.  This is the only reason two machines are necessary really.  By using the tokens to separate the fiat > bitcoin transaction and the transfer it allows users both ends to walk away with part of the transaction as 'physical Bitcoin'.


Yes.  And that I think is what got me most excited about the idea was that of using a machine to verify the value on the token (having the machine check the RFID public key against the blockchain)  and to send that same value from a Bitcoin client which does the inverse the other end.  Whilst you're talking to someone, pop in one Bitcoin and it pops out however many thousands of miles away A simple means to let money go where people want it to without the learning curve.

This had crossed my mind and I think you're right.  Apparently even just buying GBP for Bitcoin is likely this year to start coming under the auspices of the Financial Services Authority meaning one will have to have passed an exam to trade let alone to be responsible for the international aspect.

Then of course the question of whether there's a business case for it.  My guess is, as I sort-of knew but as Stephen Gornick put clearly, it also needs to be local for the recipient.  So without forking out for a massive network abroad to serve one community here the odds of it working from that perspective are getting thinner by the minute.

I think that is much more sensible than what I had in mind.  In fact it could just be an informal network of individuals rather than an organisation that can be held responsible for the international aspect.  For instance if/when a means of receiving/sending with a dumb phone with SMS is worked out it could be a matter of using smartphones this end to sell them the BTC and giving them a contact in their family's city with whom they could meet the other end with their old Nokia to receive their Rupee or whatever.

THEN, as you say, if ever there becomes sufficient activity between two particular locales it is easier to make the business case for investing and setting up an automated solution, rules permitting!

Thanks again, tf

For bitcoin to be an alternative to traditional remittance transfers a convenient, inexpensive method for the recipient to obtain cash is needed.  The solution is convenient only when the recipient doesn't have to travel far and it is inexpensive when no intermediary is assessing fees to provide the convenience.

The average cost for this was about 10%, but the World Bank is attempting to effect a lowering of that to 5% through "enhanced information, transparency, competition and cooperation with partners":
 - http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTFINANCIALSECTOR/0,,contentMDK:22383199~pagePK:210058~piPK:210062~theSitePK:282885,00.html

Even 5% is more than enough to generate a nice little side income for people to trade bitcoins face-to-face.   So before there is a need for some machine to automate this process, what should be seen first is a growing number of individuals offering this trading.   When there is a sufficient amount of that activity, then steps to automate it at an even lesser cost start to make sense.

Hi thoughtfan,

There are a lot of different things proposed, but mostly your idea seems to boil down a combination of things that have been discussed already... Please correct me if I am misunderstanding:

- Bitcoins <-> Fiat, and a vending machine that spits out physical tokens
- A physical token that can be converted back into BTC
- Using the Bitcoin network to transfer funds (verifying the receiving address by Skype instead)
- A self-contained sending/receiving machine that can interact with a Bitcoin wallet (or token), and transfer funds (like a PC, running a Bitcoin client?)

I have a question:

Are the proposed tokens just backed by Bitcoin (no value stored on the token itself)? or do they actually contain a Bitcoin wallet (private/public key for the Bitcoin network, like a Casascius coin)?

Personally, I think that Bitcoins might be a bit too dangerous from a regulatory perspective if you are starting a vending machine business like this. I believe it can be done but you would need quite a bit of information on each of your customers on both ends for money laundering regulations, and live with the risk of a shutdown as you'd be the easiest target for Bitcoin unfriendlies, bank lobbyists, and western union lobbyists. The Bitcoin network needs to gain trust and more user acceptance first.

Otherwise, there is nothing stopping you from implementing this!

I've got an idea that is so complex the odds are there's at least one critical flaw in it meaning it wouldn't fly.  However, there are so many separate elements to it I think there's a reasonable chance something useful could come out of it so let me throw it into this ideas pot.  Of course I'd greatly appreciate any thoughts...

The way it is envisioned is as a service for low-tech folk in international communities (let's say a London-Bangladeshi district) who regularly wish to send money home (anonymously) - and to talk to their loved ones.

It does so using a combination of two pairs of identical machines set up, one of each in a local store (serving said community) at the sending end, the other at the far-flung receiving end.  It also needs a new physical bitcoin token.

Remember this is all concept-stage stuff although I acknowledge some very similar elements are already established (I am a fan of casascius coins and I've seen some cool stuff in terms of vending machine prototypes too).

Let's start with the tokens:

They are plastic - lets say like poker chips with both a visually readable private key and an RFID with the public key buried inside that can only be accessed by smashing it!  They  could also have the public key on a QR code on the token.* This would mean they could work like casascius coins but also be verifiable as valid by a vending machine checking the public key against the blockchain.

So...

The first machine is a simple cash (and maybe card?) vending machine that
Sending end: dispenses physical Bitcoin tokens for sender local fiat payment;
Receiving end: dispenses receiver local fiat for physical tokens.

The other machine is a PC running Skype and a Bitcoin wallet with a touchscreen monitor and headset and also has a BTC token coin slot and dispenser (and maybe a camera in the dispenser tray too).

So the process is:
Sender and receiver arrange to 'meet' in their respective stores at a certain time.
Sender, using the vending machine, pays for and obtains BTC tokens (which can also be taken away to keep/save, to trade with as a physical token or destroyed to put in a wallet as per casascius);
Sender Skypes the receiving machine and whilst chatting (having identified the receiver using the oldest facial/voice recognition technology in existence!) drops a token into sending machine which sends token value to receiving machine which dispenses token.  (Sender uses the receiver dispenser camera to fisually verify tokens being dispensed the other end).
At the end of the conversation the receiver takes tokens to vending machine and receives local fiat.  The receiver may also decide to keep some coins and not to convert all to fiat.

Machines are locked and set up with equal value amounts of Bitcoins in Skype wallet, tokens and cash according to storekeepers' requirements.

A the sender end when the Bitcoin wallet in the Skype machine is near enough empty the Storekeeper sends Bitcoin to it which then dispenses the tokens, which the storekeeper then puts back in the cash vending machine and takes out the cash (using it to replenish his own Bitcoin wallet by another means).

At the receiving country end the storekeeper needs to charge the vending machine with cash which releases the tokens that get put back into the Skype machine which sends the Bitcoin to the storekeeper's wallet (which he'll then need to sell for fiat for next time).

Profit for both storekeepers (and machine owners) is in the exchange rate and sale of tokens.

Obvious disadvantages I can see:
People who use a machine that serves a concentrated ethnic community in London can only send to one vending machine (or one per city) the other end so relatives have to travel to the 'meet'.
Shopkeepers at both ends have to keep on the machines three times the value of the transactions that can be made before having to reset them.
Amounts to be sent have to be rounded down to token values which may require a number denominations.
Machines with mechanisms such as these need maintaining which would require people both ends to be able to come in and do the maintenance and to be available for call-out when something stops working.

What do you think folks?

*I seem to recollect reading here about the pros and cons of using a QR code on a coin in relation to the casascius coins.