Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - bytemaster

Pages: 1 ... 12 13 14 15 16 17 18 [19] 20 21 22 23 24 25 26 ... 656
271
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.

@bytemaster

Could you please explain how/by whom the STEALTH asset is to be created?

I like the idea of being a modern day Prometheus and bringing to BitShares the equivalent of the fire it needs for combustion. But I do value what is left of my liver  :)

I am ready to go with your suggestion. I would just like to understand the mechanics a little better.

The STEALTH asset will be issued by the "management account" for this feature as part of the hard fork.  You will just be the initial owner of the issued asset (not the issuer).  This management account will have multi-sig authority assigned to the 5 largest STEALTH holders weighted proportional to stake and will have the power to set the fee.

272
General Discussion / Re: Resolution to Referral Program Bug
« on: December 09, 2015, 11:19:11 pm »
We are preparing to release the code today that incorporates many fixes into one hardfork.   The hard fork will take effect next wednesday.

273
If an asset cannot be reused then the issuer will need to recover at least the cost of issuing the asset per event.  It seems as if we are caught between preventing spam, and promoting use. 

Am I correct in my assumption that prediction market assets cannot be reused? 

Won't this lead to a massive bloat in issued assets?
yes,this is the issue I care about.
and some other thing you need consider:
what if the event don't happen? no win or lose, it's just delay for ever.

i think then you can settle at 0.5 because it would be a tie!

You can always voluntarily settle / cover at the market price.   BitUSD is similar to a prediction market that never ends, the only difference is that the range / leverage of BitUSD price relative to BTS is greater.  Prediction Markets have sufficient collateral at all times to cover being wrong.

As far as complexity goes, most of it could be automated away in the GUI.   You can "borrow and sell" in one transaction.

274
Do I understand right that 20% of all income from this feature is automatically burned?*

And all shareholders share this benefit without having to do or pay anything other than approving your proposal?

*well technically recycled (as network fees) but illiquid nonetheless. could still be burned though!

It is not BURNED, it is used to purchase MAKER and the MAKER is burned.   

276
This post is designed to document how prediction markets work with the existing BitShares 2.0 system and are ready for business today.  Before I get into the details of how you would set this up, I would like to highlight some of the primary differences:

1.  When creating a new asset (symbol) you must set the *is_prediction_market* flag to true, otherwise you will create a normal bitasset.
2.  The precision of a prediction market asset(PMA) must equal the precision of the short_backing_asset  (the asset used as collateral).
3.  The value of a prediction market asset(PMA) is between 0 and the value of the short_backing asset (SBA).   So if BitUSD is the backing asset, then the most the PMA can be worth is 1 BitUSD
4.  No price feeds are published for prediction market assets
5.  There is no forced settlement (until after the issuer performs a global settlement).
6.  In order to create a 1  PMA (Prediction Market Asset) you must post 1 USD as collateral. 
7.  After the outcome of the prediction market event is known, the issuer can publish a Global Settle Operation  (which is the equivalent of a manual black swan).  This operation will forever lock in the exchange rate between 1 PMA and USD
8.  After the Global Settlement operation has been issued, owners of PMA may use the Request Settlement operation (same as force settlement) to redeem the PMA for 0 to 1 USD depending upon the outcome of the event.'

To gain decentralization and protection against manipulation, the issuer could be a multisig account or even the committee or witness account.

All of this is possible today, but the GUI does not do the right thing in the user interface.   So lets use an example.

Suppose I create a prediction market that is denominated in USD and I call the asset REPUB2016 that is worth 1 BitUSD if the republicans win the 2016 presidency and worth 0 if anyone else wins.

If I suspect they have a 60% chance of winning, then I may bid $0.55 for 1 REPUB2016.     Someone else who thinks the democrats will win, can post $1 as collateral and create 1 REPUB2016, then sell their 1REPUB for 0.55 which leaves them with 0.55 USD and short 1 REPUB2016 backed by 1 USD.    If the democrats win, the global force settlement will free their collateral and they will end up with $1.55.  If the republicans win, then their collateral will be seized and they will end up with $0.55.

You can also use these prediction markets to predict any linear outcome.  For example, you could use it to predict the spread between the republicans and democrats where greater than 20% in favor of the REP means it is worth 0, 20% or more in favor of the DEM means it is worth $1.00 and a "tie" is worth $0.50.  A simple linear interpretation can be applied for spreads between -20% and +20%.   The benefit of the linear interpretation model is that it gives higher resolution, greater volatility, and more balanced leverage tradeoffs. 

The system isn't "perfect" by any means, but it is functional enough that participants can make/lose money without any significant biases.




277
We have someone testing/documenting this API as well as working on a new API that is easier.   

278
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.



279
The primary issue I have is that putting in "special cases" for BTS asset that don't apply to all assets makes the code more error prone and the system more complex to explain in general.  Currently most of the code is written in a manner that it doesn't care what the asset types involved are, they are treated equally.

On the flip side, no one wants to see, for example OKCOIN.BTS in order to get around the fee problem...

I agree

280
I really like crediting the fee pool of the issuer with the BTS market fees.  That is like giving "store credit", they can only cash it out by spending it with the blockchain. 

I also like the idea of allowing BTS to yield to UIA fee settings. 

The primary issue I have is that putting in "special cases" for BTS asset that don't apply to all assets makes the code more error prone and the system more complex to explain in general.  Currently most of the code is written in a manner that it doesn't care what the asset types involved are, they are treated equally.


281
just posted some questions on https://github.com/cryptonomex/graphene/issues/475

My main concern is the effect on BTS profitability and referral income. Does it mean that 80% of fees will go to the maker and 20% of fees will go to cryptonomex? Is this only for new UIAs? Will this also be for current committee issued smartcoins?
How will BTS profitability and referral income be affected?

I  don`t think it is a good idea to make Market Maker as a base feather of block chain  protocol level
The market maker is NOT implemented on the block chain level ..
The proposal proposes a way to PAY market makers for their liquidity!

While Market Making is not done at the blockchain level, incentivizing it is. It is a fair question of whether or not this extra feature is something we want.

I attempted to answer your questions posted on the issue.  The mile high summary is that this feature in and of itself takes no profits away from the referral program nor BTS holders. It is only a reallocation of profits previously reserved to the issuer and completely at the discretion of the issuer to adopt or not.  BTS benefits from increased order creation fees and user adoption if the program works. 

Only if the committee managed assets adopt the feature would it impact the funds available to the committee and thus indirectly to the BTS holders. The committee would have to decide whether or not they thought it would be profitable to adopt this program for USD and CNY. 

282
General Discussion / Re: Simple Binary Prediction Market Discussion
« on: December 08, 2015, 06:56:46 pm »
I am not sure what issue you are referring to.  I fear some communication issues are due to translation issues.

283
just posted some questions on https://github.com/cryptonomex/graphene/issues/475

My main concern is the effect on BTS profitability and referral income. Does it mean that 80% of fees will go to the maker and 20% of fees will go to cryptonomex? Is this only for new UIAs? Will this also be for current committee issued smartcoins?
How will BTS profitability and referral income be affected?

I  don`t think it is a good idea to make Market Maker as a base feather of block chain  protocol level
The market maker is NOT implemented on the block chain level ..
The proposal proposes a way to PAY market makers for their liquidity!

While Market Making is not done at the blockchain level, incentivizing it is. It is a fair question of whether or not this extra feature is something we want.

Currently BTS is not collecting income from market fees.  This would be an opt-in feature on a per-asset basis.  So the committee would have to decide whether or not paying for market makers improves the overall volume / transaction for BitShares.  CNX only gets 20% of what is paid to Market Makers.  This means the committee could say market makers only get 10% of trading fees, leaving CNX with just 2%.   Bottom line, CNX only gets paid if the feature is used and it doesn't have to be used.

284
General Discussion / Re: Network Stability Under Graphene
« on: December 08, 2015, 04:26:41 pm »
Is the dev team required to make changes to try to hit an internal network reaching 100k tps? Whats stopping us from demoing it? That would obvously bring in new investors,

No changes required to hit that except perhaps a more efficient wallet for generating the transactions.
Hmm even using the cli of desktop is a bottleneck? I thought adding fc would have solved that?

The desktop CLI is limited by RPC requests to the witness node.  Doing it all in the same process should give a 10x boost.

285
General Discussion / Market Maker Incentivization Worker Proposal ($300)
« on: December 08, 2015, 01:27:20 am »
https://github.com/cryptonomex/graphene/issues/475

This is the first of many new proposals that don't cost the BTS stakeholders much of anything to get powerful new features.  In this case all the stakeholders must do is grant permission for a hardfork that implements a new feature.  100% of the new feature is self funded. 

The result is a new investment opportunity for those who want to invest on a per-feature basis in assets that are not a SECURITY because they are not a liability.

Please read the issue for more details.

Pages: 1 ... 12 13 14 15 16 17 18 [19] 20 21 22 23 24 25 26 ... 656