Most coins suffer from liquidity problems.
This could be a solution to that issue. It can also profitable to the major stakeholders in any coin, and at the very least would cost less than exchange fees.
Legal issues mentioned above may be an issue for some coins/localities.
Topic: Un-techie decentralized ALT/BTC bulk exchange network (Read 3289 times)
Its certainly possible given a amount of someone's time however a number of hurdles need to be overcome, not least the fact that someone would be distributing a work in process .xlsm document that could be running god knows what code on their PC's. As I've just discovered open office isn't VBA friendly either so you'd need to take a survey of what software people who wanted to run this actually have.
There is no need for the members to run any software. I might be providing them an excelsheet but the amount of functionality is so minimal that they can reconstruct it in 5 minutes typing to a blank sheet. Or do totally without. The explanation upthread makes it clear that all the members have to do is to submit their bids and asks once per day (or less if they don't want to participate in market making), and for that any channel that can verify their identity is OK.
Its certainly possible given a amount of someone's time however a number of hurdles need to be overcome, not least the fact that someone would be distributing a work in process .xlsm document that could be running god knows what code on their PC's. As I've just discovered open office isn't VBA friendly either so you'd need to take a survey of what software people who wanted to run this actually have.
Hi. Have the Excel packages for this system been shared in another thread? This is all very intriguing 
No not yet. I am trying to make this introduced in the MEW in due time.
you can create an exchange using
the source of bitcoin?
but by implementing two blockchain
to mark a transaction between two coin
another to score
sales and purchases
maybe creating one that erases the old when the transaction is completed
"not to create a huge blockchain"
Finally, create a way to "bind to"
two coin
type to start two wallet
bitcoin / litecoin
and connect them to the software exchange
when you put a sell or a bid
the software of exchange
blocks
bitcoin or litecoin
"depends on what you sell or buy"
and only after that both people
have blocked the coin
the exchange takes place
I think so we create the conditions
a decentralized exchange
without a guarantor for trade
all you can do without fee
"sorry for English
but I'm using google translate"
the source of bitcoin?
but by implementing two blockchain
to mark a transaction between two coin
another to score
sales and purchases
maybe creating one that erases the old when the transaction is completed
"not to create a huge blockchain"
Finally, create a way to "bind to"
two coin
type to start two wallet
bitcoin / litecoin
and connect them to the software exchange
when you put a sell or a bid
the software of exchange
blocks
bitcoin or litecoin
"depends on what you sell or buy"
and only after that both people
have blocked the coin
the exchange takes place
I think so we create the conditions
a decentralized exchange
without a guarantor for trade
all you can do without fee
"sorry for English
but I'm using google translate"
Very interested....subscribing to thread....
There are probably many security flaws in this. Is this what you were thinking of?
My idea is very basic, you need just provide your orders (every time you want to participate), in the following format:
Code:
BUY_XMR
Price BTC Quantity XMR
0,00383000 1 000,000
0,00379000 2 000,000
0,00377000 3 000,000
0,00370000 5 000,000
0,00360000 7 000,000
0,00350000 10 000,000
SELL_XMR
0,00000000 0,000
0,00394490 1 000,000
0,00401740 2 000,000
0,00405000 2 500,000
0,00410900 3 000,000
0,00420000 3 500,000
0,00430000 4 000,000
Or if you are not a marketmaker, just in a very simple format like this:
Code:
BUY_XMR
Price BTC Quantity XMR
0,00383000 1 000,000
Everything is just copypasted to the operator excelsheet and operated there. With the proposed operating mode, it does not need to be rigorous, because what the operator essentially does is:
"I promise to sell to all who bid higher, and buy from all who offered lower, at the one matching price, which I choose."
Ideally would you want to create a single Excel workbook that allows you to input your orders and process them as well so that anyone who had a copy of the worksheet could act as broker.
You could have a button that will trigger a macro to create a random number prefixed with the time/date and a scrambled version of the computer name to act as a key.
It would then :
Create a .csv file with the worksheet name displaying the key
Cell A1 showing your anonymous email address
The rest of the sheet showing your orders.
Save this in an email
You would then be able to enter the email address of your broker and click send.
There are probably many security flaws in this. Is this what you were thinking of?
You could have a button that will trigger a macro to create a random number prefixed with the time/date and a scrambled version of the computer name to act as a key.
It would then :
Create a .csv file with the worksheet name displaying the key
Cell A1 showing your anonymous email address
The rest of the sheet showing your orders.
Save this in an email
You would then be able to enter the email address of your broker and click send.
There are probably many security flaws in this. Is this what you were thinking of?
Last night I by the way made the Excel package for:
- submitting the orders (automatic calculation of 3%, 6%, 9% limits for optimization of points) MEMBER SHEET
- collecting the orders in one sheet for aggregation to the orderbook (final points for the day) OPERATOR SHEET
- the matching of the orderbook is currently done manually by copy-paste and then running a simple procedure.
The time requirement from an operator is about 1 hour of actual work (or even less) per matching day, and you need to be glued to computer for 4 hours. With 20 members, the exchange would make you 300 XMR per month in fees.
I am currently in a position to start it after a few more hours of finalizing the procedures.
- submitting the orders (automatic calculation of 3%, 6%, 9% limits for optimization of points) MEMBER SHEET
- collecting the orders in one sheet for aggregation to the orderbook (final points for the day) OPERATOR SHEET
- the matching of the orderbook is currently done manually by copy-paste and then running a simple procedure.
The time requirement from an operator is about 1 hour of actual work (or even less) per matching day, and you need to be glued to computer for 4 hours. With 20 members, the exchange would make you 300 XMR per month in fees.
I am currently in a position to start it after a few more hours of finalizing the procedures.
In the US, crypto to crypto would still fall afoul of money laundering laws.
We also have a law that forbids using talking to mobile phone while driving. Texting is OK, or talking if you have a handsfree. This is an example why an elected body should never be allowed to make laws binding others. The majority does not care about this one and it is seldom enforced.
If the majority similarly disregarded (ignored) all other victimless crime laws including money laundering, wow how they would be liberated!
For the time the majority is still drugged (of brainwash, not drugs), having KYC to a limited number, perhaps 20, members, is of course much easier than having 1000s of members in a 24/7 exchange needing KYC. If this was about anything else except the world's foremost financial privacy cryptocurrency, there would be no need for cat and mouse and no need for KYC, and no problem to collect it if needed.
I know the idea is hot: when I started the same with physical investment silver, the first week after starting, my house and office were raided by government taskforce and all wealth confiscated. I was suspected of a financial crime so bogus that I wanted to talk the head investigator about the ridiculousness of seizing all my stuff to supposedly investigate a crime that a person in my position was logically unable to commit. He said many things which I remember to this day, including: "This is a bit nasty the first time, but then you get used to it." He then vanished, and was removed from documents, which do not acknowledge him even existing as part of the investigation. The invitation to do something meaningful again some day has been a guiding star for me.
Since I told the detailed model, I got no "wow, unbelievable practical financial innovation!!" comments. Only ones wanted to point out how it is difficult to fit in the old system. It is not intended to be fit in. Even Jesus said new wine should not be put in old wineskins lest both go bad. And I want everyone to be able to choose for himself in what kind of world he lives in.
The reason we tolerate the old system is that we are peaceful people and don't like confrontation with anyone. Not that we would like to have anything to do with them.
Money exchange for profit / as a service is a regulated activity?
How does this compare to LocalBitcoins, LocalLitecoins, etc?
How does this compare to LocalBitcoins, LocalLitecoins, etc?
Crypto to crypto is typically not defined "money" and if it is, then this can be taken where it isn't.
For a reasonable extra work it can be made anonymous.
If one of this type of helpful matchmaking services is harassed/busted, the rest will take due notice and become unharassable.
What I have done here in my name and face is to show you how it can be done with minimum effort.
Nothing is more powerful than an idea whose time has come.
In the US, crypto to crypto would still fall afoul of money laundering laws.
Without delving into the detailed reasons, take the reason Mike Caldwell stopped selling pre-funded Casascius coins. Casascius coins were always sold for bitcoin. So essentially, he was selling bitcoin for bitcoin. But after talking to lawyers, he remained concerned that the model had risk. His explanation:
"A lot of the discussion I had with my legal counsel is that part of what makes it money transmitting is there is no way for me to know whether the person ordering is the person receiving. In other words it is money transmitting because each order "could" be a money transmission with no way to know. Restricted delivery changes that aspect of the story." https://bitcointalksearch.org/topic/m.6789094
I think Smoothie ultimately took the same approach and only sells buyer funded coins now.
In Risto's approach, you don't even get to that level of theory. You are transmitting one currency to another party. AML / KYC would most likely apply.
But then again, you are not in the US. I hope Finland or Estonia laws are more practical!
Good luck with the project.
Money exchange for profit / as a service is a regulated activity?
How does this compare to LocalBitcoins, LocalLitecoins, etc?
How does this compare to LocalBitcoins, LocalLitecoins, etc?
Crypto to crypto is typically not defined "money" and if it is, then this can be taken where it isn't.
For a reasonable extra work it can be made anonymous.
If one of this type of helpful matchmaking services is harassed/busted, the rest will take due notice and become unharassable.
What I have done here in my name and face is to show you how it can be done with minimum effort.
Nothing is more powerful than an idea whose time has come.
Money exchange for profit / as a service is a regulated activity?
How does this compare to LocalBitcoins, LocalLitecoins, etc?
How does this compare to LocalBitcoins, LocalLitecoins, etc?
A practical guide for anyone to make this for any coin. I have used XMR as an example, and assumed NO coding skills (most of us have them, and can make this happen more elegantly, but my only skill in the world is spreadsheet, 1990-). 1 XMR = $2.
Membership and fees
Anyone can become a member, but to become one, a 30 XMR handling fee must be paid.
The trading and all activities carry no charge except the monthly charge of 30 XMR, so for trading activity >500 XMR per day, this is cheaper than the leading 24/7 exchange.
Guarantee
In addition, a Guarantee deposit must be made. The Guarantee may be of any size and allows trading up to 10 times of the size of the Guarantee for both bids and asks, so up to 20 times leverage in total, and 95% reduced risk compared to 24/7 exchanges. The reasonable minimum Guarantee could be 250 XMR and maximum 5000 XMR, allowing the trade of 2,500 and 50,000 XMR per day, respectively. It does not matter in which currency the Guarantee is deposited, the maximums are adjusted automatically if it's in BTC.
Posting of bids and asks
The member orderbooks must be posted in csv-format (a standard Excel-sheet can be made for those who habitually use many orders) by 18:00 GMT, every day. This posting may be done by any means, PM, encrypted channel, even email. Posting is of course voluntary.
If the schedule of having the matching every day is deemed excessive, which it may well be for a small coin, it is also possible to cut it to 3 times per week eg. Mon, Wed, Fri. (I have experience on 2, 3, and 5 days per week stuff myself.)
Matching procedure
A member orderbook is all of the orders of a member listed, one per line. Once they are received, the operator matches them to each other, finding the price where demand=supply. At this point, the operator may add his own orders to balance the book. The requirement is that with the posted matching price, all the orders of the members must be fully fulfilled if they are in the right side of the matching price. This gives the operator the preknowledge of the others' aggregate situation, which does allow him to gain from privileged orders and arbitrage with exchanges. On the other hand, he is compelled to fill all orders fully, and must make the decision to set the price such that every order either fills or not, no partial fills, which is a handicap.
In practice an operator Excel-sheet does all the matching automatically once the offers are entered to it. The operator has 30 minutes from the deadline of the offers (until 18:30 GMT), to publish the matching price, which is public information. No other information needs to be made public.
Once the matching price is published, the members who had posted orderbooks and see that their offers have filled, have 90 minutes to initiate the transaction of sending their part of the offer to the operator (until 20:00 GMT). It is notable and beautiful that the operator need not notify the members separately.
If the transaction has not been sent, it is deemed defaulted. Should this happen, the operator is entitled to fixed 10% damages calculated from the value of the trade. It is deducted from the Guarantee. The operator must now take care of the fulfillment of the order himself. In practice this means that he goes to an exchange and buys the coins with slippage, forwards them to the buyer, and keeps what is left from the 10%. The operator is also responsible for the losses that exceed the 10%.
The operator now has 2 more hours (until 22:00 GMT) to initiate the transactions to send the coins to the buyers.
In this model the operator does not need much trading capital himself. He can easily make the system even more trustworthy for members to join in, if the Guarantee funds or a majority of them are held in an escrow or multisig.
Market-making bonus
To enable huge orders to be fulfilled, which is not really possible in 24/7 exchanges due to reasons previously outlined, there must be massive volume in both buy and sell side. Market-making is the activity where an owner of both the tradable things lists them both for sale simultaneously. As the price fluctuates, he is able to increase both stacks slowly. Here doing this is almost risk free, the system actually rewards the members for doing this (in addition to the fact that the trading fee is zero and buying low, selling high is profitable).
This does not need anything from the member's part. The matching system gives the members market-making points using the following formula (this is tailored for XMR with its low historical volatility, for some other coin the bounds should be higher):
- The depth at the point where the member's bid and ask total are of equal size and 3% apart in price;
- The depth at the point where the member's bid and ask total are of equal size and 6% apart in price;
- The depth at the point where the member's bid and ask total are of equal size and 9% apart in price.
To illustrate what this means, I take the numbers from the most voluminous exchange for XMR today, Poloniex:
- Depth 3% = 1,200 XMR
- Depth 6% = 4,600 XMR
- Depth 9% = ~7,000 XMR.
With all due respect towards both the exchange and XMR, the liquidity is not in professional level yet. A trade of ~7,000 XMR (0.2% of XMR) moves the price by 4.5%. With the matching system in place, it is likely that the competing market makers eager to make a buck and earn bonus, will be able to collectively offer 30,000-60,000 XMR (1-2% of XMR) in the 9% category, even more after getting used to the system and its benefits.
The bonus is calculated as follows: For each category, there is 1.0000 points up to the grabs per day, so total 90 points per month. Half of the monthly fees will be given back to the members in the proportion of points they earn, so the most eager market makers can earn their whole fee back and then some.
Each day and each category, the 1 point will be distributed in proportion to the depths that the members offered. If it is an uneventful day, the members feel comfortable to offer large walls, if it is an interesting and volatile day with insecurity where the price may end up, they are more hesitant to have large orders. In both cases the point total is 1, and it is divided in the proportions of the depths.
It is easy to envision that during some days a member is ready to list the maximum he is allowed in both bids and asks, between 9% of each other, to grab points and thus maximize the profit for the day that otherwise is dull. It should be noted that the percentage limits for each member are calculated from his orders only, so the percentage brackets may and will overlap, making the price-finding process all the more robust and not prone to even large shocks from external supply and demand.
Benefit to dealers and the public
If someone is in the business of selling XMR to outsiders, by joining this system he can make the price risk zero, by just fixing his selling price to the matching price beforehand, which the customer has learnt to accept in other markets such as precious metals (london gold & silver fixing), mutual funds (buy and sell prices quoted daily but decision must be made before the quote).
The public history of XMR/BTC matching prices will certainly show lower volatility than the records in any of the 24/7 exchanges, and it will quickly gain dominance as a public service based on which OTC trades will be conducted all over the world, with a negotiable premium +/- based on the mutual agreement of the participants.
Membership and fees
Anyone can become a member, but to become one, a 30 XMR handling fee must be paid.
The trading and all activities carry no charge except the monthly charge of 30 XMR, so for trading activity >500 XMR per day, this is cheaper than the leading 24/7 exchange.
Guarantee
In addition, a Guarantee deposit must be made. The Guarantee may be of any size and allows trading up to 10 times of the size of the Guarantee for both bids and asks, so up to 20 times leverage in total, and 95% reduced risk compared to 24/7 exchanges. The reasonable minimum Guarantee could be 250 XMR and maximum 5000 XMR, allowing the trade of 2,500 and 50,000 XMR per day, respectively. It does not matter in which currency the Guarantee is deposited, the maximums are adjusted automatically if it's in BTC.
Posting of bids and asks
The member orderbooks must be posted in csv-format (a standard Excel-sheet can be made for those who habitually use many orders) by 18:00 GMT, every day. This posting may be done by any means, PM, encrypted channel, even email. Posting is of course voluntary.
If the schedule of having the matching every day is deemed excessive, which it may well be for a small coin, it is also possible to cut it to 3 times per week eg. Mon, Wed, Fri. (I have experience on 2, 3, and 5 days per week stuff myself.)
Matching procedure
A member orderbook is all of the orders of a member listed, one per line. Once they are received, the operator matches them to each other, finding the price where demand=supply. At this point, the operator may add his own orders to balance the book. The requirement is that with the posted matching price, all the orders of the members must be fully fulfilled if they are in the right side of the matching price. This gives the operator the preknowledge of the others' aggregate situation, which does allow him to gain from privileged orders and arbitrage with exchanges. On the other hand, he is compelled to fill all orders fully, and must make the decision to set the price such that every order either fills or not, no partial fills, which is a handicap.
In practice an operator Excel-sheet does all the matching automatically once the offers are entered to it. The operator has 30 minutes from the deadline of the offers (until 18:30 GMT), to publish the matching price, which is public information. No other information needs to be made public.
Once the matching price is published, the members who had posted orderbooks and see that their offers have filled, have 90 minutes to initiate the transaction of sending their part of the offer to the operator (until 20:00 GMT). It is notable and beautiful that the operator need not notify the members separately.
If the transaction has not been sent, it is deemed defaulted. Should this happen, the operator is entitled to fixed 10% damages calculated from the value of the trade. It is deducted from the Guarantee. The operator must now take care of the fulfillment of the order himself. In practice this means that he goes to an exchange and buys the coins with slippage, forwards them to the buyer, and keeps what is left from the 10%. The operator is also responsible for the losses that exceed the 10%.
The operator now has 2 more hours (until 22:00 GMT) to initiate the transactions to send the coins to the buyers.
In this model the operator does not need much trading capital himself. He can easily make the system even more trustworthy for members to join in, if the Guarantee funds or a majority of them are held in an escrow or multisig.
Market-making bonus
To enable huge orders to be fulfilled, which is not really possible in 24/7 exchanges due to reasons previously outlined, there must be massive volume in both buy and sell side. Market-making is the activity where an owner of both the tradable things lists them both for sale simultaneously. As the price fluctuates, he is able to increase both stacks slowly. Here doing this is almost risk free, the system actually rewards the members for doing this (in addition to the fact that the trading fee is zero and buying low, selling high is profitable).
This does not need anything from the member's part. The matching system gives the members market-making points using the following formula (this is tailored for XMR with its low historical volatility, for some other coin the bounds should be higher):
- The depth at the point where the member's bid and ask total are of equal size and 3% apart in price;
- The depth at the point where the member's bid and ask total are of equal size and 6% apart in price;
- The depth at the point where the member's bid and ask total are of equal size and 9% apart in price.
To illustrate what this means, I take the numbers from the most voluminous exchange for XMR today, Poloniex:
- Depth 3% = 1,200 XMR
- Depth 6% = 4,600 XMR
- Depth 9% = ~7,000 XMR.
With all due respect towards both the exchange and XMR, the liquidity is not in professional level yet. A trade of ~7,000 XMR (0.2% of XMR) moves the price by 4.5%. With the matching system in place, it is likely that the competing market makers eager to make a buck and earn bonus, will be able to collectively offer 30,000-60,000 XMR (1-2% of XMR) in the 9% category, even more after getting used to the system and its benefits.
The bonus is calculated as follows: For each category, there is 1.0000 points up to the grabs per day, so total 90 points per month. Half of the monthly fees will be given back to the members in the proportion of points they earn, so the most eager market makers can earn their whole fee back and then some.
Each day and each category, the 1 point will be distributed in proportion to the depths that the members offered. If it is an uneventful day, the members feel comfortable to offer large walls, if it is an interesting and volatile day with insecurity where the price may end up, they are more hesitant to have large orders. In both cases the point total is 1, and it is divided in the proportions of the depths.
It is easy to envision that during some days a member is ready to list the maximum he is allowed in both bids and asks, between 9% of each other, to grab points and thus maximize the profit for the day that otherwise is dull. It should be noted that the percentage limits for each member are calculated from his orders only, so the percentage brackets may and will overlap, making the price-finding process all the more robust and not prone to even large shocks from external supply and demand.
Benefit to dealers and the public
If someone is in the business of selling XMR to outsiders, by joining this system he can make the price risk zero, by just fixing his selling price to the matching price beforehand, which the customer has learnt to accept in other markets such as precious metals (london gold & silver fixing), mutual funds (buy and sell prices quoted daily but decision must be made before the quote).
The public history of XMR/BTC matching prices will certainly show lower volatility than the records in any of the 24/7 exchanges, and it will quickly gain dominance as a public service based on which OTC trades will be conducted all over the world, with a negotiable premium +/- based on the mutual agreement of the participants.
Removing liquidity from the open market by opening a closed market is a terrible idea for a fledging coin.
All liquidity is connected. The importance of this market is not that it is "closed" but that it is secure, with no 100%+100% predeposits. This enables higher liquidity here and therefore the price moves in every other market also become contained. The volume from back-and-forth trading may go down, but the real liquidity is measured with volatility and that will certainly go down.
I've been interested in a market for bulk semi-offchain transfers before.
Could you not facilitate this on CounterParty protocol and retain the same benefits?
The order matching engine already handles automatic escrow, and BTC payments, unlike NXT.
The tokens you exchange are directly stored (ie linked to your private key) in your wallet, under your control not a centralized server - no middleman and no counterparty risk
They could be exported to a paper wallet or offline armory wallet in exactly the same way a normal BTC key could.
The basic flow would be like this:
Seller of XMR creates a token asset, IE: RPIETILAXMR with the quantity you have available to sell. You are able to lock the total amount if desired.
You can place info proving the asset is indeed issued by yourself below, PGP Signing it, and cross-announce that here.
Quote
The shining advantage of this system is that there is no need to send a multiple of what you are intending to trade, of your money, to a centralized, insecure, often in a questionable legal basis, website, which historically has been a target and victim of scam, fraud, hack and attack by gov.
Could you not facilitate this on CounterParty protocol and retain the same benefits?
The order matching engine already handles automatic escrow, and BTC payments, unlike NXT.
The tokens you exchange are directly stored (ie linked to your private key) in your wallet, under your control not a centralized server - no middleman and no counterparty risk
They could be exported to a paper wallet or offline armory wallet in exactly the same way a normal BTC key could.
The basic flow would be like this:
Seller of XMR creates a token asset, IE: RPIETILAXMR with the quantity you have available to sell. You are able to lock the total amount if desired.
You can place info proving the asset is indeed issued by yourself below, PGP Signing it, and cross-announce that here.
Code:
{ "asset": "MYTOKEN", "description": "This is a description of MYTOKEN", "image": "http://www.mysite.com/mytoken.png", "website": "http://www.mysite.com", "pgpsig": "http://www.mysite.com/MYTOKEN.pgp" }
Removing liquidity from the open market by opening a closed market is a terrible idea for a fledging coin.
Interesting rpietila. I will think about this for some days and see what I can come up with. Obviously we need some sort of multisignature for this, I just need to think a bit more how to actually do it with no single point of failure.
I keep about 20 times my average daily volume in trading a certain altcoin in a certain exchange. With this proposal, since there is only a 5% guarantee amount, I will only keep 1 times my trading volume. Also this cuts away the intraday volatility so that I won't need to use outlier bids/asks anymore, taking it possibly down to 0.2 times my average volume.
This would mean a 99% reduction on amount lost upon scam/theft/hack/seizure, and in addition the scam/theft/hack/seizure becomes much less likely because there is 99% less to steal, cutting the total risk perhaps by 99.9%. And this is even if the system administered the funds centrally like any 24/7 exchange does! Surely getting rid of the remaining 0.1% is also nice...
Interesting rpietila. I will think about this for some days and see what I can come up with. Obviously we need some sort of multisignature for this, I just need to think a bit more how to actually do it with no single point of failure.
This thread is for developing a dealer-to-dealer (you may replace "dealer" also with "merchant" or "whale") network for altcoin currency exchange.
Desirable qualities for an exchance service (especially from a professional standpoint):
Important:
- Less initial outlay of funds is better
- The less funds the exchange holds, the better
- The more liquidity, the better
- The less effort needed to execute a large trade, the better
- No single point of failure
Nice to have:
- Ease to join
- Ease of use
- Low fees
Matching engine:
Members post their orderbook as a signed and encrypted message to the matching engine once per day at a set time. The matching engine solves the clearing price, publishes it, and posts to all the members, how much they need to send. The matching engine can act as the decentralized escrow, or directly give the address of the counterparty.
Clearing price:
Every day the engine takes together the members' orderbooks. If the bids and asks in the members' orderbooks overlapped, there forms a one price that clears all bids higher than that, and all asks lower than that. If the price is ambiguous (a range), the clearing price is the midpoint of the highest and lowest such price.
If bids and asks no not overlap, there are no trades. In this case the "clearing price" is the midpoint between highest bid and lowest ask.
There is no need to publicly disclose the orderbooks that the members posted. The clearing price can not be manipulated, because this would lead to some of the orders that should fill, not filling. So if none of the members makes an issue about a non-fill, it can be concluded that everyone has had their fills correctly, and therefore the clearing price must be published correctly also. In this sense it is like a hash.
Fees:
Depending how much organization/setup work is needed, this can work on very low fees. I had a similar system working purely on excelsheet basis in silver since 2008 and it still does in an altered form.
Because installing and learning the system takes some time, and it works best with frequent need to trade, a fixed monthly fee is perhaps in order, regardless of trade activity. This encourages professionals and large traders, whereas keeping the occasional traders away just makes it much easier to maintain and less risky for pros.
The fee could even be waived for those who provide market-making ie. frequently or constantly send large orders that are near each other in both bid and ask side. The matching engine can easily keep track of the market-making "points" on a monthly basis and credit them to the member automatically. I will post this separately because it is a detail.
Guarantees:
The shining advantage of this system is that there is no need to send a multiple of what you are intending to trade, of your money, to a centralized, insecure, often in a questionable legal basis, website, which historically has been a target and victim of scam, fraud, hack and attack by gov.
Yes, there will be a need to send a fraction of it. There is no way around that all offers must be covered, but with a little bit of forethought, the capital requirement can be decreased by a lot.
Let's assume that the matching engine is a decentralized exchange, capable of doing escrow. In this case, all BTC or ALT for the filled trades must be sent in 1 hour following the matching. After the necessary time for confirmations, it will be sent to the new owners. A member only needs to trust the engine for the trades that already happened, as much as he typically needs to trust any escrow, and only for 1 hour. This is way better than having both the ALT and BTC funds at risk day and night in shady online places.
But the engine also needs to trust you. You cannot run away with the money because nothing will be sent to you unless you send first, but you can still default sending if you feel it is expedient. (It should not be, because: you are just trading ALT/BTC at the exact price that was just determined to be the market price. So even if you made a mistake for example entering an ask instead of bid, or something as common and stupid, it can be rectified in the next matching, or in a 24/7 exchange, for typically a 1-2% loss.)
Anyway, for this case you must post a bail to the system, covering you in the event that you do not pay on time, and the counterparty's money needs to be used to source the coins from the market instead. I believe for a mature altcoin 5% should be enough in this case, which means that for a deposit of 1,000 LTC, you could send orders up to 20,000 LTC. Now that is leverage, and reduction of risk.
If the engine does not act as an escrow but rather directly provides you the address of the counterparty, then the bail must be larger to cover for the event that party A sends his goods but party B runs with the money without sending. This can be mitigated by forcing an incremental send, where an increment is maximum the size of the bail. Although the standing bail becomes larger this way, the temporary risk for a member (in the event that you have sent the whole sum to the escrow engine, and the operator runs with the money) is smaller. Which way to proceed depends on many things.
The bail posted by one member can easily serve multiple-duty: if the owner of the bail A says that member C is also guaranteed by the bail, this can be enforced every time that the bail is not directly used by A himself.
The idea is that there is no need to send escrow payments back and forth, or keep coins in limbo (and possibly at risk) to guarantee your possible future trades. One bail payment that is a fraction of your maximum daily trade size, and stays constant until you want to exit the network, is enough.
Desirable qualities for an exchance service (especially from a professional standpoint):
Important:
- Less initial outlay of funds is better
- The less funds the exchange holds, the better
- The more liquidity, the better
- The less effort needed to execute a large trade, the better
- No single point of failure
Nice to have:
- Ease to join
- Ease of use
- Low fees
Matching engine:
Members post their orderbook as a signed and encrypted message to the matching engine once per day at a set time. The matching engine solves the clearing price, publishes it, and posts to all the members, how much they need to send. The matching engine can act as the decentralized escrow, or directly give the address of the counterparty.
Clearing price:
Every day the engine takes together the members' orderbooks. If the bids and asks in the members' orderbooks overlapped, there forms a one price that clears all bids higher than that, and all asks lower than that. If the price is ambiguous (a range), the clearing price is the midpoint of the highest and lowest such price.
If bids and asks no not overlap, there are no trades. In this case the "clearing price" is the midpoint between highest bid and lowest ask.
There is no need to publicly disclose the orderbooks that the members posted. The clearing price can not be manipulated, because this would lead to some of the orders that should fill, not filling. So if none of the members makes an issue about a non-fill, it can be concluded that everyone has had their fills correctly, and therefore the clearing price must be published correctly also. In this sense it is like a hash.
Fees:
Depending how much organization/setup work is needed, this can work on very low fees. I had a similar system working purely on excelsheet basis in silver since 2008 and it still does in an altered form.
Because installing and learning the system takes some time, and it works best with frequent need to trade, a fixed monthly fee is perhaps in order, regardless of trade activity. This encourages professionals and large traders, whereas keeping the occasional traders away just makes it much easier to maintain and less risky for pros.
The fee could even be waived for those who provide market-making ie. frequently or constantly send large orders that are near each other in both bid and ask side. The matching engine can easily keep track of the market-making "points" on a monthly basis and credit them to the member automatically. I will post this separately because it is a detail.
Guarantees:
The shining advantage of this system is that there is no need to send a multiple of what you are intending to trade, of your money, to a centralized, insecure, often in a questionable legal basis, website, which historically has been a target and victim of scam, fraud, hack and attack by gov.
Yes, there will be a need to send a fraction of it. There is no way around that all offers must be covered, but with a little bit of forethought, the capital requirement can be decreased by a lot.
Let's assume that the matching engine is a decentralized exchange, capable of doing escrow. In this case, all BTC or ALT for the filled trades must be sent in 1 hour following the matching. After the necessary time for confirmations, it will be sent to the new owners. A member only needs to trust the engine for the trades that already happened, as much as he typically needs to trust any escrow, and only for 1 hour. This is way better than having both the ALT and BTC funds at risk day and night in shady online places.
But the engine also needs to trust you. You cannot run away with the money because nothing will be sent to you unless you send first, but you can still default sending if you feel it is expedient. (It should not be, because: you are just trading ALT/BTC at the exact price that was just determined to be the market price. So even if you made a mistake for example entering an ask instead of bid, or something as common and stupid, it can be rectified in the next matching, or in a 24/7 exchange, for typically a 1-2% loss.)
Anyway, for this case you must post a bail to the system, covering you in the event that you do not pay on time, and the counterparty's money needs to be used to source the coins from the market instead. I believe for a mature altcoin 5% should be enough in this case, which means that for a deposit of 1,000 LTC, you could send orders up to 20,000 LTC. Now that is leverage, and reduction of risk.
If the engine does not act as an escrow but rather directly provides you the address of the counterparty, then the bail must be larger to cover for the event that party A sends his goods but party B runs with the money without sending. This can be mitigated by forcing an incremental send, where an increment is maximum the size of the bail. Although the standing bail becomes larger this way, the temporary risk for a member (in the event that you have sent the whole sum to the escrow engine, and the operator runs with the money) is smaller. Which way to proceed depends on many things.
The bail posted by one member can easily serve multiple-duty: if the owner of the bail A says that member C is also guaranteed by the bail, this can be enforced every time that the bail is not directly used by A himself.
The idea is that there is no need to send escrow payments back and forth, or keep coins in limbo (and possibly at risk) to guarantee your possible future trades. One bail payment that is a fraction of your maximum daily trade size, and stays constant until you want to exit the network, is enough.
Jump to: