[update:pics] I'm working on a bitcoin gift card vending machine

[Anon136]

I purchased an old phone card vending machine on the cheap (presumably because no one uses phone card vending machines any more) with the intention of redeploying it to dispense bitcoin gift cards. In the interest of not spending 10 years on this project before deploying it the methods i intend to use early on will be very simple and additional features will be added at a later date.

Initially i am just going to order a couple hundred gift cards from a company that makes custom gift cards (they will cost about $0.50 initially and closer to $0.25 if i ever buy in bulk). These gift cards will have their face value denominated in dollars on the front and a scratch off strip on the back with a private code underneath. On the card will be instructions to send an email to example@blablabla.com containing both the address they would like the coins to be sent to as well as the obfuscated key on the back of the card. If they key matches my records than I will send the bitcoins minus my commission to the address specified. Initially i plan on doing 20 dollar cards and 100 dollar cards (i think 500 dollar cards could become a serious security risk) As far as security is concerned i will be using the bitcoin vault on a dedicated offline computer and manually sending each transaction since i do not yet have the technical skill to automate the process (a skill i will certainly work on developing later on). Prices will be calculated using the mtgox ask at the time of redemption.

Now a couple of reasons why i believe a service *like* this is important. 1) anonymity is important for many bitcoin users, especially those who intend to spend their bitcoins on the silk road. These people will greatly appreciate not having to meet a human being in person or filling out a form with private information (i think you have to do this with bitinstant when you send a moneygram?). 2) this eliminates the problem of waiting around for confirmations from the guy you are buying from on localbitcoins. Granted they will be forced to trust that i will actually fulfill my obligation but they can gain some assurance from the fact that i would have to be a complete idiot to spend hundreds of dollars to start this business and then destroy my reputation over 20 dollars. 3) the machine its self will be a wonderful advertisement. it will display in the largest letters i can fit "what is bitcoin?" with a large qr code right below it linking to weusecoins.com

So then this begs the obvious question, wouldn't this simply be inferior to a bit-coin atm? There are a couple of reasons why this might be considered better than a bitcoin atm or at least a complimentary service. 1) non-technically inclined people like your grandma will still be able to buy bitcoins as a gift for their technically inclined relatives and friends 2) the machine will not require any sort of internet connection which should cut costs. 3) Retooling an old vending machine will likely be a lot cheaper than it would be to build an atm from the ground up. 4) less security risk from hackers 5) imagine a situation where someone has heard about bitcoins but doesn't know very much about them, doesn't have a smart phone or doesnt have the time to stop in the middle of what hes doing to figure out how to install an use a bitcoin wallet on his smart phone. This person might be willing to just buy a gift card and stick it in his pocket to deal with when he gets home.

Now the hard part is finding some sort of business in the area that will allow me to install the machine in their store (hopefully for free). The idea being that they will benefit from a positive externatlity when people come into their store who otherwise wouldn't have and i will benefit from the security of having it sitting a couple of feet away from the front counter at a pizza hut (for example).

anyway now its time to let the soul crushing reality set in as you guys tell me why my idea is god awful and how i just wasted hundreds of dollars. Thanks in advance for that.

oh and here is my website http://btcgiftcard.com/. very little there atm but you can check it out anyway. im pretty happy with the domain.

[FreeMoney]

Make it happen. Are you in or near a big city?

[Anon136]

Make it happen. Are you in or near a big city?

medium sized city. Winston-salem North Carolina maybe neighboring Greensboro if i have to.

[casascius]

Your idea is excellent and would totally work

[cbeast]

Great idea. There was a discussion about this here.

[Anon136]

Your idea is excellent and would totally work

Awesome because its coming. I'm past the point of no return on money spent (i.e. ive already spent on the project significantly more than half of the expected total cost).

[casascius]

I pitched this as a good idea just a little more than 2 months ago.

Somebody needs to do it.

That somebody is you!

https://bitcointalk.org/index.php?topic=126571.0

Yes, it will work extremely well.  The fact that phone cards are going obsolete means those machines are available on the cheap - only adding fuel to the fire.  Good luck and happy returns!



Feel free to throw any questions my way if you would like some ideas on how to do something or are stuck.

If I were doing this myself, I would deem it important for redemption to be nearly instant, because I would recognize that someone buying bitcoins from a vending machine is probably doing so to fund an impulse purchase, and expects convenience in exchange for the purchase price paid.  This is one of those instances where having and securing a modest hot wallet is a necessary risk to take.

I have a card printing company in China I use and they do a good job - let me know if you want contact info.  I met them at a Las Vegas trade show, have a wide selection of products and processes, reasonable prices, and their turnaround time is usually under a month.  I use them for RFID cards for payroll processing.

[Anon136]

I pitched this as a good idea just a little more than 2 months ago.

Somebody needs to do it.

That somebody is you!

https://bitcointalk.org/index.php?topic=126571.0

Yes, it will work extremely well.  The fact that phone cards are going obsolete means those machines are available on the cheap - only adding fuel to the fire.  Good luck and happy returns!

Feel free to throw any questions my way if you would like some ideas on how to do something or are stuck.

If I were doing this myself, I would deem it important for redemption to be nearly instant, because I would recognize that someone buying bitcoins from a vending machine is probably doing so to fund an impulse purchase, and expects convenience in exchange for the purchase price paid.  This is one of those instances where having and securing a modest hot wallet is a necessary risk to take.

I have a card printing company in China I use and they do a good job - let me know if you want contact info.  I met them at a Las Vegas trade show, have a wide selection of products and processes, reasonable prices, and their turnaround time is usually under a month.  I use them for RFID cards for payroll processing.

Awesome, i do have one question. What is the best way to go about learning how to automate the payment/transaction process. i.e. whats the best way to learn how to make the payments virtually instantaneous. Right now im working on learning the xhtml and css that ill need just to build the website. But after that the next step will be the previously mentioned.

[cbeast]

I misunderstood the OP. The term "gift card" threw me off. I thought you were going to provide bulk VISA/MC non-reloadable cards. Since BitInstant seems to have gone silent about their debit card, it would be great to see another form of liquidity. I suppose a Casascius style card would be great in certain markets, maybe in haute couture.

[cbeast]

You might even try a pawn shop to sell them for people that want to send money overseas.

[casascius]

Awesome, i do have one question. What is the best way to go about learning how to automate the payment/transaction process. i.e. whats the best way to learn how to make the payments virtually instantaneous. Right now im working on learning the xhtml and css that ill need just to build the website. But after that the next step will be the previously mentioned.

Don't go this alone.  This MUST be done professionally.  It WILL be attacked by hackers and needs to be secure from day 1, and that's just impossible if you're learning web design.

I would suggest putting up a bounty for creating an open source platform for prepaid card redemption.  There are enough skilled and motivated people on the forums that a little bit of money goes a LONG way.  Break your project into discrete goals that you're willing to pay $100 to $200 a piece for, and see if you can get contributors (I'd contribute to this).  Ultimately it's going to have to be written for a platform you're comfortable with administering.

I will pledge 10 BTC/$200 (which ever is worth the most) right now for someone who writes and publishes under an open source license, a basic card redemption platform using the tools of their choice, that does the following:

1. Displays a redemption web page
2. Gets a card number
3. Looks up the card's fiat value in a database
4. Gets and validates the Bitcoin address to receive the bitcoins, OR uses a bitcoin address that has already been pre-designated for the card (e.g. if the card has a private key printed on it).
5. Maintains a constant feed of a current exchange rate
6. Enforces a minimum exchange rate to a sensible value of an admin's choice (you don't want a MtGox error or MITM attack to make the site believe BTC = 0.01 USD and allow someone to take nearly 1000 BTC for a $10 card)
7. Kicks off bitcoind to do a redemption transaction and save the transaction ID.  Before doing so, the app should query bitcoind to ensure connectivity, responsiveness, and sufficient balance to do the transaction.  Before attempting the transaction, the app should also update a flag the database to indicate that the transaction is attempted, so if it fails to record success, it can't be tried again other than with administrator intervention.

I realize that a real production version will have to be way more robust than this, but this will be an excellent head start, and probably something that a 10 BTC / $200 bounty may be enough to inspire.  Also I won't be running it: I will just be paying the bounty when others indicate that the goal has been achieved.

If anyone wants to contribute to the bounty, simply indicate here and send to 1BowSerbsenLBELbwR5aZE1aqy7ooGCydT.  All contributions will either be awarded or refunded, never kept by me.

[Anon136]

Awesome, i do have one question. What is the best way to go about learning how to automate the payment/transaction process. i.e. whats the best way to learn how to make the payments virtually instantaneous. Right now im working on learning the xhtml and css that ill need just to build the website. But after that the next step will be the previously mentioned.

Don't go this alone.  This MUST be done professionally.  It WILL be attacked by hackers and needs to be secure from day 1, and that's just impossible if you're learning web design.

I would suggest putting up a bounty for creating an open source platform for prepaid card redemption.  There are enough skilled and motivated people on the forums that a little bit of money goes a LONG way.  Break your project into discrete goals that you're willing to pay $100 to $200 a piece for, and see if you can get contributors (I'd contribute to this).  Ultimately it's going to have to be written for a platform you're comfortable with administering.

I will pledge 10 BTC/$200 (which ever is worth the most) right now for someone who writes and publishes under an open source license, a basic card redemption platform using the tools of their choice, that does the following:

1. Displays a redemption web page
2. Gets a card number
3. Looks up the card's fiat value in a database
4. Gets and validates the Bitcoin address to receive the bitcoins, OR uses a bitcoin address that has already been pre-designated for the card (e.g. if the card has a private key printed on it).
5. Maintains a constant feed of a current exchange rate
6. Enforces a minimum exchange rate to a sensible value of an admin's choice (you don't want a MtGox error or MITM attack to make the site believe BTC = 0.01 USD and allow someone to take nearly 1000 BTC for a $10 card)
7. Kicks off bitcoind to do a redemption transaction and save the transaction ID.  Before doing so, the app should query bitcoind to ensure connectivity, responsiveness, and sufficient balance to do the transaction.  Before attempting the transaction, the app should also update a flag the database to indicate that the transaction is attempted, so if it fails to record success, it can't be tried again other than with administrator intervention.

I realize that a real production version will have to be way more robust than this, but this will be an excellent head start, and probably something that a 10 BTC / $200 bounty may be enough to inspire.  Also I won't be running it: I will just be paying the bounty when others indicate that the goal has been achieved.

If anyone wants to contribute to the bounty, simply indicate here and send to 1BowSerbsenLBELbwR5aZE1aqy7ooGCydT.  All contributions will either be awarded or refunded, never kept by me.

Ok then count me in for 5btc also. I think a 15 btc bounty requires its own thread and if you posted it that would carry a lot more weight than if i did *cough*. How long until the bounty expires?

also im still curious. What skills would i theoretically need to acquire in-order to be able to write my own "open source platform for prepaid card redemption" that could be plugged into a website that i designed with xhtml and css. I have a played around with computer programing a little and from what i have done i really enjoyed it. Could i theoretically learn how to do everything myself in a reasonable time frame except for learning how to address the security issue. If this is the case than would it be feasible, practical, wise?, to divide the labor up and contract out for security experts but do everything else myself?

[casascius]

Security is a core design principle, not an "issue" you fix.

If you have to ask, the job is not for you. No offense intended of course, just don't be the guy doing the software development. Be the project manager instead.  This has to be done right the first time, like heart surgery. It is not a project for someone learning.

[kangasbros]

I have demonstrated these coupons with easywallet.org couple of times with a condom vending machine. I will try to get my hand on a machine which accepts bank notes.

https://easywallet.org/en/coupons

[Anon136]

Security is a core design principle, not an "issue" you fix.

If you have to ask, the job is not for you. No offense intended of course, just don't be the guy doing the software development. Be the project manager instead.  This has to be done right the first time, like heart surgery. It is not a project for someone learning.

ah well thanks for the help. ill just take comfort in the fact that this can be done one step at a time and work some more on the website. Hopefully it wont break the bank blockchain to have to contract out for this at some point in the future.

[Anon136]

New pics of the stand i had fabricated as well as the newly painted machine.





on monday ill be able to get some money from the bank then ill be able to go to the hardware store to get the tools i need to attach the machine to the base. Ill post more pics here when i get them put togather.

[recan86]

So wheres it at now? update?

[Anon136]

So wheres it at now? update?

I still have it. I came to realize just how much risk was involved in this. If I was classified as a money services business without a bond I could face jail time.

[MasterOwel]

So the project is done and the bounty no longer available? Darn. 

[Anon136]

So the project is done and the bounty no longer available? Darn. 

Sorry