Very portable Bitcoin

[we6jbo]

I know that I can take my US Dollars down to the corner and buy a pack of gum and I can take another USD and give it to the homeless guy. I'm just wondering if there is a way to make bitcoin so portable that strings won't prevent me from using the currently that I take for granted.

[theGECK]

One disadvantage of ⓑ is that it is not physical. You can't transfer it without a device that has internet connectivity, or will have in the near future. If we see the rise of overlapping networks (like cell networks now) that provide easy and nigh-unlimited access to the internet, then we could see this type of portable use of Bitcoin.

[mail2345]

There is a mobile client port that should work for almost any mobile device with internet:
http://tcatm.github.com/bitcoin-js-remote/

However it requires a server(which is being ported to Android along with a native client).

In these cases, bitcoin will obviously need a mobile device with internet.

I had an idea though.
A QR code that can be redeemed for bitcoins(via some online service).
It can't be verified without a computer, so maybe only for smaller transactions, paying for something without having a computer on your end, or an automatic bitcoin trading machine.

[we6jbo]

There is a mobile client port that should work for almost any mobile device with internet:
http://tcatm.github.com/bitcoin-js-remote/

However it requires a server(which is being ported to Android along with a native client).

In these cases, bitcoin will obviously need a mobile device with internet.

I had an idea though.
A QR code that can be redeemed for bitcoins(via some online service).
It can't be verified without a computer, so maybe only for smaller transactions, paying for something without having a computer on your end, or an automatic bitcoin trading machine.

In the case of the android client (and let's just assume that this was a fully functional client) would there be a way in which the client would only have Internet access on Friday but the user of the android client would trade bitcoins to other people with the same android client on a day to day base. I think it would be possible because when I was listening to the security now podcast I remember hearing that Bitcoin is decenteralized in which case all you should need is the current database and your wallet.

I was thinking that someone could develope a small electronic card. On the card would be a Ethernet port, a IR port and just enough memory and a SDCard to operate a portable Bitcoin client and hold the wallet and database. When someone holding the card points it at another card and enters the address and funds to transfer, the card would zap the nessisary data across the IR port and onto the other persons card. Then that person would be able to upload their newly aquired funds to the Internet and buy stuff on the market.

[Mike Hearn]

It can be done, technically, but it opens you up to fraud. Android devices are all designed on the assumption of nearly 24/7 internet connectivity, so it's unlikely to be worth bothering with.

The future is connectivity, all the time, everywhere. I don't think it's worth worrying about the offline case much.

[malditonuke]

The future is connectivity, all the time, everywhere. I don't think it's worth worrying about the offline case much.

Agreed. Even Somalia has cell phones.

[caveden]

One disadvantage of ⓑ is that it is not physical.

That's probably bitcoin's greatest advantage! Easy to transfer, easy to stock/protect...

You can't transfer it without a device that has internet connectivity, or will have in the near future.

Well, if both parties trust each other, they can transfer offline. Trust is needed due to the risk of double-spending.

And, as people said, with 3G and smartphones, internet is available mostly everywhere. We shouldn't really worry.

[bitjet]

Technically there is NO reason why one could not issue promissory notes backed by bitcoin. It would require that the issuer keep those bitcoins in reserves. So there is no reason way someone could not create their own bitcoin 'bank'.

[[Tycho]]

I see this like QR-code-based systems.
When you order, for example, a hamburger in your starbucks, a waitress gives you your bill with QR-code containing bitcoin address, amount and description. You point your smartphone camera at this code and bc client sends money to the shop. After seeing that this transaction is confirmed, they can let you go

[bitjet]

I see this like QR-code-based systems.
When you order, for example, a hamburger in your starbucks, a waitress gives you your bill with QR-code containing bitcoin address, amount and description. You point your smartphone camera at this code and bc client sends money to the shop. After seeing that this transaction is confirmed, they can let you go

While this seems perfectly reasonable the big problem here is that it takes forever for the network to verify. You could be in and out with your burger and the system wouldn't know you just tried to pull a fast one for 10 minutes, by that time you are already gone with the goods. At what point can we expect confirmation to be much faster?

[mail2345]

Use it for paying small amounts and hold temporary records(maybe just names or even faces) that are cleared upon payment confirmation.
Pull a fast one, and you stay in their records and join the DO NOT ALLOW list.

[bitjet]

Use it for paying small amounts and hold temporary records(maybe just names or even faces) that are cleared upon payment confirmation.
Pull a fast one, and you stay in their records and join the DO NOT ALLOW list.

I don't think I would be comfortable with a retail establishment taking down all my info. What we need is fast transfer rates. Besides it would be a lot of extra work for a retail place to maintain a database like that. Fake names would be easy to use. It would make more sense for an establishment to only use bitcoin from trusted individuals. I gues you could swipe your credit card as a retainer. I dunno this almost cries for a 3rd party escrow type of service... IE bank. Which I dont want to deal with.

[BitterTea]

I don't think I would be comfortable with a retail establishment taking down all my info. What we need is fast transfer rates. Besides it would be a lot of extra work for a retail place to maintain a database like that. Fake names would be easy to use. It would make more sense for an establishment to only use bitcoin from trusted individuals. I gues you could swipe your credit card as a retainer. I dunno this almost cries for a 3rd party escrow type of service... IE bank. Which I dont want to deal with.

I don't expect Bitcoin will ever be suited for large, instantaneous transactions. For small transactions, I think it's fairly safe to accept them without confirmations. For larger transactions, I think it will be an individual's preference between either waiting for confirmations, or using some sort of centralized service backed by Bitcoin, like a bank.

[we6jbo]

These are difficult questions but I have come up with one idea that might work though it's kinda sketchy.

The bitcoin user deposits all the BCoins into a trusted service account called Bitcoin_validation. Bitcoin_validation counts all of the Bitcoins that were deposited and then returns them back to the bitcoin user with an additional identification. The Bitcoins that were returned to the bitcoin user may be used only at stores which participate in Bitcoin_validation service. When you go out to eat and you give the restaurant Bitcoins, you also give them the Bitcoin_validation ID and the restaurant calls the Bitcoin_validation service with the difference in the bill which is then deducted by the Bitcoin_validation service and they then report if te user has enough Bitcoins for the bill. On the other hand if you spend your Bitcoins at a place that does not participate in the Bitcoin_validation then that nullifies your credentials because Bitcoin_validation service now can not validate how much money you have left and they know this because Bitcoin naturally keeps a transaction trail.

[bfever]

Several other threads discuss possible scenario's for "fast" transfers, so that you do not have to wait for a confirmation.

In my opinion, a client like the official client now on a smartphone, is a bit of overkill: it downloads and stores the entire chain, and extracts the unspent transactions into a "wallet" file (together with the private information of each wallet bitcoin address). I find it also a major security risk (easy target for pick-pockets) and you'll need backup's in case your phone stops working, etc.

I'm working on a smartcard approach as follows:

• The smartcard will contain just the keypair (private+public key) of one of your normal bitcoin addresses, and the signing algorithm (= no exposure of the private key needed).
• A new client that will be able to talk to the smartcard (using a standard smartcard reader) and that will ask to sign the transaction.
• Using a double signature principle, "fast" transactions are possible like this: A (that's me) sends B BTC to C (my favourite store) with 2 signatures needed to be able to claim the coins; A waits for enough confirmations of this transaction (= 1 hour); A goes to the store C; at the payment desk, A inserts his/her smartcard and the B bitcoin transaction can now be used to transfer the needed amount to the store (smartcard has private key of A, local client has private key of C).
  

Simple, quick and secure without the need to trust anyone else but yourself (and the bitcoin network of course) !

[BitterTea]

These are difficult questions but I have come up with one idea that might work though it's kinda sketchy.

The bitcoin user deposits all the BCoins into a trusted service account called Bitcoin_validation. Bitcoin_validation counts all of the Bitcoins that were deposited and then returns them back to the bitcoin user with an additional identification. The Bitcoins that were returned to the bitcoin user may be used only at stores which participate in Bitcoin_validation service. When you go out to eat and you give the restaurant Bitcoins, you also give them the Bitcoin_validation ID and the restaurant calls the Bitcoin_validation service with the difference in the bill which is then deducted by the Bitcoin_validation service and they then report if te user has enough Bitcoins for the bill. On the other hand if you spend your Bitcoins at a place that does not participate in the Bitcoin_validation then that nullifies your credentials because Bitcoin_validation service now can not validate how much money you have left and they know this because Bitcoin naturally keeps a transaction trail.

I don't find this to be a very sound idea. If you're going to trust your money to a third party, I think a Bitcoin bank sounds much more convenient. Validating that an individual owns some number of Bitcoins does not solve the double spending problem, which is the entire reason for transaction confirmations. If there were Bitcoin banks, you could keep whatever percentage of your total owned Bitcoin in a checking account, which could function almost exactly like it does today. The only differences would be that it is possible to audit a Bitcoin bank via the block chain, and these mechanisms are only necessary/beneficial when time is of the essence.

[we6jbo]

Bitcoin banks would centralize money and could become a target for hackers. In addition, they would evade the built in trust in the network. There would also need to be some form of reporting for example I could make a huge transaction to my alternative bitcoin address and claim that the bank stole my money. Then again everything that I mentioned about Bitcoin banks could also be applied toward Bitcoin_verification. So I think the idea of banks and verifications services should still be tested but I think ultimatly atleast for me some record system will need to be in place. Otherwise it's just cash.