Topic: Instant payments (Was: Why bitcoin won't make it) (Read 4790 times)

Imagine a number of instawallets are already widely accepted.
The vending machine would simply say "instawallets accepted".
You insert your instawallet-banknote into the machine and it goes "no, i don't accept that one".
After some time, that's what's gonna be the practical day-to-day use.
I fail to see a problem with that.
The customer will have a strong motivation not to use too exotic an instawallet service, but at the same time the vending machine will have a strong motivation not to be too picky about the instawallet services to accept.
The market will take care of the rest.

This problem CAN BE SOLVED!

1) Transaction fees
2) Have the buyer send an ip-address of the bitcoin node that first confirms the payment. Then the seller can quickly connect to that node and verify the payment. Of course, this would require a modified bitcoin client program, but it should be a small tweak to the code.

Any thoughts?

"How long until there's Bitcoin equivalent to the SEC?"

...or taking it to the next level(s)--every nation mints there own physical bitcoins using there own designs back by their fiat currency.

The BTC SEC will be controlled by ALL members of this forum where the person with the most posts is president, newbies voter on important issues, Jr. Members write the newsletters, Sr. Members lobby, and the trolls keep the whole system in check.

have you checked  the new android POS application yet ?

Nothing stops you from building a p2p network for messaging. There's really no need to build stuff into the client. And even if you think that being built into the client is a must, you can make your own client with the p2p exchange.

Hopefully we can get back on topic now.

Imagine how glad the customer will be when she has to skim through a hundred of these labels just to find out whether or not you accept megawallet. She will be even happier to see a simple slot for coins.

Bitcoin will not fail because of the lack of merchants accepting Bitcoin.

Nor will Bitcoins fail because of the lack of 'myBitcoin.com'-like sites.

Bitcoin will fail because of: http://forum.bitcoin.org/index.php?topic=25438.0

Only Satoshi can fix it

I hope that the customers would actually spread their favor over a lot of fancywallets. Keeping an up-to-date list of maybe a couple hundred websites i trust, would be no problem at all. And even if one of them turns out to be crooked, i'd be happy that this reduces my actual damage.

Imagine such a vending machine. On the label it would say "I accept fancywallet, mybitcoin, instawallet, wallywallet...", just like merchants today accept VISA, MASTER, AMEX, DINERS...

Hey, the more i think about it, the more i love that idea. Really.

Yep, it may take three to five seconds to propagate throughout the network. A well connected bitcoind should get it much sooner.

Are you looking at this from a merchant's perspective, from a bank's perspective or from a bank customer's perspective? The customer doesn't deal with the payment processor. For the merchants it's no different than putting your faith into mtgox. For the banks - this non-profit would need to earn their trust before they put a bit of funds in it. There's no need to transfer all of the funds to the payment processor, only an amount that is needed for the time that it takes the network to confirm all "instant" payments made from such a bank. In other words as the bank's client requests an insta-transfer, the bank connects to the payment processor and requests the same amount of funds to be transfered to the recipient specified by the bank's client. If the bank's account that is managed by the payment processor has enough money, the payment processor issues a transaction, if not - an error is returned. The bank can transfer some funds to the payment processor whenever it feels like and those will be used to issue transactions on behalf of the bank.

And if you bought an exclusive license to a digital product? A good example would be exclusive web-site templates - some of those cost thousands. Another is software. I suppose the wait doesn't matter that much in these cases, and people would be fine with waiting for some time, but only if bitcoin was the only way to pay for these things. More likely than not people would prefer paypal's instant service to waiting for a while. Besides, I don't think 6 confirmations takes just 10 minutes - it took an hour or so when I checked last (was some time ago, though).

I suppose it is, but IIRC even 0 confirmations takes time because of the decentralized nature of the network.

You would trust instawallet.org, but what would you do if a considerable share of customers was using fancywallet.com, bitwallet.org and a hundred more instawallet lookalikes?

Because they unilaterally changed my password on me 2 weeks ago (after I had already changed my password after Mt.Gox), and still won't let me access my account, nor do they respond to my support requests. At this point I believe that they've stolen from me.

Thank you!

Forget China, try getting money back and forth from the US to Europe.  They think we (the US) are in the banking stone ages because we still use checks/wire payments and not IBAN (which is true).  That's what got my interested in Bitcoin initially.  Bitcoin has the opportunity to change the (multi-billion dollar) industry I work in.

Right. I'm thinking grocery store checkout line, not vending machine. That's a bit different (has its own thread, too).

That comes straight from the book, and i unfortunately have to tell you that it is outright wrong.
It may be acceptable for an over-the-counter trade with two real persons involved. It is not very likely that you would buy a coke from me, jump out of the shop, and hit that enter button on your laptop just to make a double-spend, that's true. Even if that may happen from time to time, the loss will be minimal for me as a vendor.

But with vending machines, it's a completely different thing.
If i sell something valuable with a vending machine, it may be very profitable for the attacker to do just that, double- or triple-spend at different machines. Think about prepaid mobile phone cards, they can be purchased from vending machines. Think about train tickets. Cigarettes. Even gold, now that this new goldbar-ATM-thingy is up and running 


But i am working on it , as i have mentioned earlier in this thread

You're going about this the wrong way.

At retail, and with the typical retail transaction amounts, it's perfectly fine to accept a transaction with 0 confirmations. The risk on this has a LOT of zeroes after the decimal point, and in any case is far lower than the risk of accepting someone's credit card. The double-spend attacks are pretty much impossible to accomplish while standing in line at Walmart.

A web of trust, for example. Best common practice as another. The market, a third.

That's what instant payments are all about. If you bought a car, a 10 minute wait for a confirmation would be no problem.
If you bought a coke at a vending machine, 10 minutes is not acceptable.



That's why you would have to rely upon the instant wallet service. For minor amounts, i could imagine that i'd be willing to trust instawallet.org in exchange for being able to sell via a vending machine.



No, it's the same as relying upon another third party. This third party is in no way affiliated with the customer and has nothing to gain from submitting false positives for minor amounts. They do have a lot to gain from building a strong reputation with merchants, which will lead to higher acceptance of their service.



That's the mtgox approach i've been talking about. The big issue i have with that is centralization. My approach would allow for a list of accepted service providers, not one single provider.

Quis custodiet ipsos custodes?

Your idea fails here. First you can only trust a minor amount of coins which makes buying expensive things impossible, second if you're a merchant you have no way of finding out if you really were sent coins or not, it's the same as relying on the customer saying "I've sent you the coin, you'll receive it later, can I have my goods now?"

The best approach I could come up with to date is to create a central bitcoin payment processor for banks. This payment processor would have to be a non-profit organization and would ensure an even playground for all the banks wrt instant payments. The banks would store some money with this payment processor and as soon as a bank issues a transfer using the payment processor's api, the payment processor can confirm a transaction. So basically this would be a bank for banks.

In Gox We Hack Trust