If witcoin.com had a way to let external sites access and use witcoins (which are bitcoins stored in a database) you would have easycoin. All we would have to do is let people buy and sell witcoins which are tied to a user account. This is a dangerous area though because you get into money transmitting laws...
The reason we did it that way was to allow micro transactions without needing to pay fees on the bitcoin network. You can withdraw them as bitcoins at any time. You cant pay .00000001 bitcoins without paying a fee but you can pay .00000001 witcoins to vote on something or donate/support other people without it touching the bitcoin network.
Databasecoins are a different animal to bitcoins. One way to prove you have all the actual coins onsite is to transfer them to a bitcoin address provided by an auditor who then transfers them back to your site. Then the site has to trust an auditor with the customers coins.
For auditing, you dont need to move all the coins, just prove you have control of the address they reside in. There are probably hundreds of ways this could be achieved. Some kind of zero-knowledge proof on the key? Anyway I remmeber reading a number of already posted schemes to handle this problem.
This part of the post is directed at nobody in partucular. Just a general comment.
One of the issues I see with a lot of these 'lets make a php website and store the bitcoin as a database field' is that it completely invades everyones privacy. Sure, I bet you all promise
not to tell anyone who i sent my coins to, including the porn merchants who may also have an account on your site (that i wish to keep private), but why should I have to believe you? What happens if a court orders you to reveal the transaction history? Are you liable to comply?
This is why, a few posts ago in this thread, I linked a bunch of research into their problem already and some open source projects that are implementing it.
Nobody is going to trust your service long term, big scale, if it doesnt use some kind of chaumian blinded signature scheme for the tokens. But I suspect almost no current 'bitcoin bank' type services even know what that is.
If people grab an anonymous open id and use that for witcoin then fund the account with bitcoinhttp://openid.anonymity.com/
How does that invade their privacy ? Sure we can give the feds the anonymous open id but it wont do any good
Just on that, does anyone think that we will always be able to dodge 'know your customer', aka The Patriot Act (US) and Anti-Money Laundering and Counter-Terrorism Financing Act 2006 (Australia)?
They can't regulate bitcoin, but they can regulate you. and your merchants. I wonder how long it will take.
Back to your question. What I think I want is the digital equivalent of the following system:
1. I have a verified account with a bitcoinbank, according to my local rules and regulations. They can help me manage my account balances, etc. Maybe they datamatch with the ATO to profile sus activity. Who knows.
2. to make payments, I always withdraw cash from the bank, then give it to the merchant. The merchant then give it to his bank, and neither bank knows that its the 'same money'. Maybe I post half my paychecque each week to wikileaks. Maybe I spend it on hookers and blow.
3. if the merchant needs to verify me because of the product type or location, we do that process separately.
The problem with most digital versions of this process, is that it's not possible to have both step 1 AND step 2. They (VISA, paypal, Commonwealth bank transfers, etc) know exactly
who I gave the money too. This is what blinded tokens try to solve.
Actually there are quite a few other problems too, like how can I prove that the bitcoinbank has all my tokens, and how can I prove that they are not a malacious bank and will steal from me, authorise transactions without my signature, etc etc.
I don't know the solution to all the problems, but so far in my travels open-transactions has done the best job of addressing them (and others).
Final thoughts; I don't think there is a perfect solution, just better solutions and different tradeoffs.