0 Members and 1 Guest are viewing this topic.
Here is what I have implemented as an intermediate step:1) If delegates publish a feed, then the feed will be used rather than the average. In this way if there is a service interruption on the feed, then the network can still function like it does today.2) This puts the "shareholders" in control over whether to use feed or average.3) I have restricted the creation/execution of new shorts to the median / average price.Result... those who have USD have priority in selling over those who would like to short. This adds liquidity for those with USD by removing competition from those looking to short. Next we are going to open up a second bitasset so that "short demand" can spread between two BitAssets and the bitassets can trade against one another. I believe this should implement your proposal without the arb. bot.
Agent86, I know you are very passionate and bright guy who has convinced me of things in the past (like the approval voting) but I think you are entirely wrong in your assessment.If we limit shorts then you are resorting to price fixing. The peg is supposed to have some variance based upon supply and demand and I think you are judging the system entirely too soon. Right now the demand for BitUSD is low and the demand for leverage is high and overall risk is high. We will see what happens and monitor the correlation in price movement. In your limit scenario I can see some benefits... you reduce selling pressure below the price feed. This would give USD sellers a priority over shorts which in turn would help make USD more liquid. But right now USD is liquid at a price that is correlated to USD. But this introduces a requirement for a price feed and the purpose of this experiment is to attempt something without a price feed.I think you are being alarmist.
If the network is the USD Buyer of Last Resort and buys it at a price of 10% below the price feed and sells it at 1% below the price feed... it would end up "printing XTS" to buy the USD and burning it when it sells the USD. High demand to short USD would result in the network buying it up with inflation and then selling it back at a profit resulting in long-term deflation. In this case 100% of the "inflation" would be locked in collateral of the short position.
I just want to make it clear that we will use price feeds if necessary to establish the peg.
BitUSD will eventually get to USD parity because otherwise this network will be worthless.
The network will end up with USD on its balance sheet proportional to the "surplus short demand" and the XTS created to buy this USD is actually locked away in the collateral of the shorts. Thus we can safely say that printing XTS to perform this market making algorithm will not create additional XTS in circulation.
The primary tool we have at our disposal is to implement a market maker algorithm into the blockchain based upon the median feeds of the delegates.It would automatically buy BitUSD at .95 and sell BitUSD at 1.05 and then we would limit shorts to 1.05. The impact this would have on the network:1) When demand to sell BitUSD is high the network is buying at .95 with new BTSX2) Shorts would be unable to sell at .95... so they would have to first buy at 1.053) This establishes a 10% initial fee for any shorts and gives longs a priority in selling.4) As the price fluctuates between those who want into or out of BitUSD the network makes money.The network will end up with USD on its balance sheet proportional to the "surplus short demand" and the XTS created to buy this USD is actually locked away in the collateral of the shorts. Thus we can safely say that printing XTS to perform this market making algorithm will not create additional XTS in circulation. We can quibble about the spread.This process would function much like it does today except it would make the "shareholders" more money from the built in arbitrage bot. The built in "bot" doesn't need to predict which way the price will ultimately go... only that it will eventually change directions. What is the risk from running this BOT on the network? Little that I can see right now.1) An attacker could print USD at will provided they were willing to buy at 1.05... but they would only be able to sell back at .95 unless there was real demand.2) An attacker would be unable to print arbitrary XTS because they would have to buy BitUSD high and then tie up the XTS in the collateral and their USD would only have guaranteed demand at .95 so when they "cover" they would get back less than they started with.The primary down side is the requirement of a price feed. I think if you start the BOT off with wide market maker functionality, that eventually other players will enter the market and provide tighter market maker functionality at which point in time the "feed" almost never factors into the equation. If this is what it takes to bootstrap the peg I am willing to do it. Thoughts?
Quote from: bytemaster on August 28, 2014, 09:34:07 pmHere is what I have implemented as an intermediate step:1) If delegates publish a feed, then the feed will be used rather than the average. In this way if there is a service interruption on the feed, then the network can still function like it does today.2) This puts the "shareholders" in control over whether to use feed or average.3) I have restricted the creation/execution of new shorts to the median / average price.Result... those who have USD have priority in selling over those who would like to short. This adds liquidity for those with USD by removing competition from those looking to short. Next we are going to open up a second bitasset so that "short demand" can spread between two BitAssets and the bitassets can trade against one another. I believe this should implement your proposal without the arb. bot. This looks acceptable to me in order to prevent rampant manipulation because of the thin orderdepth yet. How would you call the new "bitUSD-short pegged" and "bitUSD-market unlimited"? Also, in which version will this be included, I suppose the delegates and users will have to download a new version, for example 0.5.0 first, right?
Quote from: Gentso1 on August 28, 2014, 08:42:51 pmQuote from: bytemaster on August 28, 2014, 08:03:21 pmQuote from: MolonLabe on August 28, 2014, 08:01:06 pmFrom libertarianism to democracy to oligarchy to tyranny, even faster than I expected. that is an interesting point of view... and I see your parallels.I am for the feed and against the network owned arb bot. Its a free market and this would be a bad step. The whole, is it ok to do bad things for the greater good kind of thing, their has to be another way. Why not just use the feed and let the market decide the value of bitUSD no arb bot needed.USE THE FEED TO LIMIT VALID SHORTS TO ONLY ABOVE THE FEED. No 95-105% range and no arb bot is needed.Sometimes I don't know how you guys come up with this stuff and start wasting time on it before thinking it through. Sorry for being annoyed.
Quote from: bytemaster on August 28, 2014, 08:03:21 pmQuote from: MolonLabe on August 28, 2014, 08:01:06 pmFrom libertarianism to democracy to oligarchy to tyranny, even faster than I expected. that is an interesting point of view... and I see your parallels.I am for the feed and against the network owned arb bot. Its a free market and this would be a bad step. The whole, is it ok to do bad things for the greater good kind of thing, their has to be another way. Why not just use the feed and let the market decide the value of bitUSD
Quote from: MolonLabe on August 28, 2014, 08:01:06 pmFrom libertarianism to democracy to oligarchy to tyranny, even faster than I expected. that is an interesting point of view... and I see your parallels.
From libertarianism to democracy to oligarchy to tyranny, even faster than I expected.
Quote from: bytemaster on August 28, 2014, 09:34:07 pmHere is what I have implemented as an intermediate step:1) If delegates publish a feed, then the feed will be used rather than the average. In this way if there is a service interruption on the feed, then the network can still function like it does today.2) This puts the "shareholders" in control over whether to use feed or average.3) I have restricted the creation/execution of new shorts to the median / average price.Result... those who have USD have priority in selling over those who would like to short. This adds liquidity for those with USD by removing competition from those looking to short. Next we are going to open up a second bitasset so that "short demand" can spread between two BitAssets and the bitassets can trade against one another. I believe this should implement your proposal without the arb. bot. what happens if you publish a feed but don't update it?
Quote from: Agent86 on August 28, 2014, 09:33:27 pmQuote from: toast on August 28, 2014, 09:30:24 pmYou can still short below the peg, it just costs a premium. You can match your own short and sell undervalued. I think all this is doing is forcing shorts to say "oh, I am effectively paying a 10% fee? I didn't want that!". I think BitUSD will still be undervalued.You can't match your own short. This is not allowed by the system. The system matches every buy order with the lowest price sell order.You can match your own short... you just have to buy up the order book first FYI... I appreciate your feedback and challenges. They help strengthen everyone here.
Quote from: toast on August 28, 2014, 09:30:24 pmYou can still short below the peg, it just costs a premium. You can match your own short and sell undervalued. I think all this is doing is forcing shorts to say "oh, I am effectively paying a 10% fee? I didn't want that!". I think BitUSD will still be undervalued.You can't match your own short. This is not allowed by the system. The system matches every buy order with the lowest price sell order.
You can still short below the peg, it just costs a premium. You can match your own short and sell undervalued. I think all this is doing is forcing shorts to say "oh, I am effectively paying a 10% fee? I didn't want that!". I think BitUSD will still be undervalued.
QuoteBM, can you answer my question about whether there is an equivalent thing or person that performs the function of the arb bot that you described, as part of the structure of the stock market other than just traders? This is for my education please.As part of the structure of a "market" not that I am aware, but I think you can view it like a business.We all know that if the market peg holds then every deviation from the peg is a profit opportunity. We all expect someone to perform the role of providing liquidity. How do we expect these individuals to do that job? If a DAC hard-codes business rules into a blockchain, and we all know that providing liquidity is a profit generating business, then we can encode the same rules that a trader would execute into a blockchain and the network can make money the same way.The benefit a trader has is they can maintain a dynamic spread, increasing/decreasing as necessary to account for volatility. I think that if you hard-code a bot with very wide buy/sell prices around a trusted feed then the bot would almost never execute, but everyone would "feel better" just knowing it is there to back things up.
BM, can you answer my question about whether there is an equivalent thing or person that performs the function of the arb bot that you described, as part of the structure of the stock market other than just traders? This is for my education please.
With the whole market price depending on a feed but not on depth this might allow a few delegates to slightly manipulate the price which will allow someone with large capital to abuse the market. I also think this is not a significant problem for now but it could be exploited.
Quote from: Shentist on August 28, 2014, 09:00:59 pmdon't hurry. it is to early to make a decision. it is normal that in the beginning every one expects that btsx is rising, but maybe in the future everything will balance out. the importent thing is - 1 bitUSD == 1 USDYep... the price is holding up fairly well thus far.
don't hurry. it is to early to make a decision. it is normal that in the beginning every one expects that btsx is rising, but maybe in the future everything will balance out. the importent thing is - 1 bitUSD == 1 USD
Quote from: Agent86 on August 28, 2014, 08:54:50 pmUSE THE FEED TO LIMIT VALID SHORTS TO ONLY ABOVE THE FEED. Reasonable.And now the next issue:Feeds. Can a few delegates manipulate it ?
USE THE FEED TO LIMIT VALID SHORTS TO ONLY ABOVE THE FEED.
Quote from: GaltReport on August 28, 2014, 08:46:53 pmIs their a real-world equivalent of this bot? Not being that knowledgeable about exchanges and market, it might be more palatable if there is an existing exchange function in the real world performing the equivalent function that you could explain. To the lay it just sounds like price controls.The real world equivalent of this bot is what provides the ultimate peg for BitUSD to USD. I would gladly run such a bot myself, but these bots require capital and early on no one wants to take the risk. Coding something into the network that we expect people outside the network to do merely provides liquidity and enforces the peg. If you do it right the market maker in the network would eventually no longer be competitive with the market makers outside the network and thus its price targets would never get hit. In fact you could "pretend" such a market maker exists right now with a spread that is wider than the current BitUSD/USD peg. It hasn't hit so you don't even see its involvement. So all we are ultimately talking about is decreasing the spread of the current "imaginary" market maker.
Is their a real-world equivalent of this bot? Not being that knowledgeable about exchanges and market, it might be more palatable if there is an existing exchange function in the real world performing the equivalent function that you could explain. To the lay it just sounds like price controls.
I mean you include a bid... it doesn't get matched.. price goes up.. bid is now out of range.
Quote from: bytemaster on August 28, 2014, 08:30:02 pmQuote from: emski on August 28, 2014, 08:26:28 pmWhy not just hardcode max/min buy price for bitUSD 0.95 and 1.05 based on feeds ?Any block including transaction not satisfying the above rule is invalid.Price can change "after the fact" and that kind of filtering would just close the market.Its a single block. It is either valid or not. If the transaction cannot be included => the user should retry.
Quote from: emski on August 28, 2014, 08:26:28 pmWhy not just hardcode max/min buy price for bitUSD 0.95 and 1.05 based on feeds ?Any block including transaction not satisfying the above rule is invalid.Price can change "after the fact" and that kind of filtering would just close the market.
Why not just hardcode max/min buy price for bitUSD 0.95 and 1.05 based on feeds ?Any block including transaction not satisfying the above rule is invalid.
Quote from: bytemaster on August 28, 2014, 07:43:53 pmIt would automatically buy BitUSD at .95 and sell BitUSD at 1.05 and then we would limit shorts to 1.05. When you say "buy BitUSD at .95" you mean "buy BitUSD at 95% of the market price provided by the feed", right?
It would automatically buy BitUSD at .95 and sell BitUSD at 1.05 and then we would limit shorts to 1.05.
Quote from: GaltReport on August 28, 2014, 07:49:59 pmI get an assert exception trying to publish feed.as I see your delegate is not active... thats why.... (it is standby)
I get an assert exception trying to publish feed.
Quote from: bytemaster on August 28, 2014, 07:43:53 pmThe primary down side is the requirement of a price feed.Why is this a downside? I mean why having a price feed is bad per se?
The primary down side is the requirement of a price feed.
Since 0.4.9, I've been completely unable to access the BitUSD market... I can only guess as to what is happeing, but it sounds like BitUSD is getting heavily shorted.I have a possible explanation: It would appear the BTSX market has bottomed, and many traders feel that a strong rebound in price is now imminent. The Chinese have been calling for 0.18 CNY for days, and their target has finally been hit. It's an important technical level, and strong support has been shown. It seems to me that traders with a high level of confidence in a large price increase seem determined to short at almost any price, and people just can't place bids fast enough to keep up with it. We just don't have enough players in the game -- yet.It's speculation of course, but I wanted to mention this so that people can see there isn't necessarily anything "wrong" with the way the market is operating -- we just need more people to be active in this market. This doesn't happen overnight, especially with the technical issues we've seen. With more participants, I can't see why the peg wouldn't hold.After the wallet is working better, I am sure all will be well in time. I have every bit of confidence that these bugs in the wallet will be ironed out, so I'm not too concerned about that.
Published. Also I heard Xeroc had a way to autopublish if the price swings +/- 5%. Can anyone shed any light on that?
We have many tools at our disposal to make this happen.