Since
@onceuponatime is paying we will do things how he wants.
I just wanted to chime in on the regulatory issue. Those who have followed us for a long time know I am very careful and have become increasingly so. That said, based upon the SWARM working paper (
http://www.scribd.com/doc/255347578/SWARM-Working-Paper-Distributed-Networks-and-the-Law) produced by Members of policy group Coin Center, law firm Perkins Coie as well as Harvard and MIT we now have a much clearer idea on what constitutes a security and high risk.
My proposed solution of implementing the feature and giving onceuponatime 100% of the STEALTH asset that gets bought back by the network over time. If he is the only owner, then it is clearly not a security. It just gives him a different way to claim the fees to his account (by order) as well as a way of dividing up control (by dividing his STEALTH among multiple accounts). So from the perspective of the law, no security has been offered to the public and everything is merely an accounting system on the blockchain for a single user.
If onceuponatime wished to sell his revenue stream he could do so by transferring the stealth asset. This is much more powerful than transferring control over an account and/or requiring a bot to automatically forward payments. It creates a far more trust free transfer of revenue (or fraction thereof) to a 3rd party. Once again, this wouldn't be a security by any stretch of the imagination when selling to private parties because there was no PUBLIC offering.
At this point I have believe I have demonstrated that we can design the feature to buy back a STEALTH asset while being so far away from regulatory issues that it wouldn't make sense to do it any other way.
All of that said, I believe that the STEALTH asset could be sold to the public AFTER the feature has been implemented and accepted by the network. This is based upon the Howey Test.
The tests for a security require *ALL* of the following properties:
1. Investment of money - token buyers pay money for their tokens. 80%
2. Common Enterprise - the funds received by the sale of STEALTH are not pooled, they become the private property of the seller. Thus clearly not a common enterprise. 20% risk
3. Reasonable Expectation of Profit - buyers purchase it for speculative purposes in the belief that the value of the token will rise. 90% risk of being a security.
4. Derived Mainly from the Efforts of Others - the value of STEALTH depends upon users of an existing system. No effort is required nor promised of other individuals for the token to receive automatic repurchases from the use of the system. In other words, the value of the token does not depend upon onceuponatime to take further actions. 10% risk
The combined risk for STEALTH is therefore less than 2% chance of being classified as a security and that is with me over estimating and then rounding up.
The penalty risk is $35,000 (what satoshi dice paid + disgorgement of profits). When discounting the penalty for the risk you can price the cost at $500. Any profits would be calculated after recovering his $45,000 investment.
The risk is so low (in my estimation) that CNX will probably use this model to fund future features.
So I am going to argue strongly that we implement this feature as a STEALTH asset, and then leave it to onceuponatime to determine whether or not to SELL the asset except for the purpose of claiming fees from the automatic buyback.