Proposal: Fully Decentralised Exchange Mechanism for All Cryptocurrencies & Fiat

[bluemeanie1]

here we go again,

"“Alongside the distributed infrastructure exchange that we’re building, we’ll provide our own proprietary software that will add benefit,” "

http://www.coindesk.com/uk-firm-promises-skype-style-bitcoin-exchange/

Don't jump to conclusions... I think that their idea here is awesome.

They're saying that the network, like bitcoin is open. And that one client software (like bitcoin-qt) is open, and another client software (Like armory) is to be closed and sold so that they can make some money.

That's a great way to monetize open source.

I've been in the open source world for a while and I've seen many of these ideas come and go.  Many of them are listed here:

http://www.zdnet.com/blog/open-source/11-open-source-business-models/5371

I have, however, NEVER seen any of these 'professional open source companies' ever produce something that is qualitatively different than any other commercial company.  The large majority of them never even managed to 'harness the community' of open source developers.  Rarely, if ever, has another group successfully forked a 'commercial open source' code base and turned it into something useful- so this option primarily exists as a marketing device.

Consider this: did anyone ever attempt to commercialise Bitcoin code?  is there a 'commercial' client floating around?

[bluemeanie1]

To clarify, the design provides:

    • Fully decentralised history of trading data; volumes traded and inter currency rates going back over time.
    • Fully decentralised history of bid/ask data; a set of the bid/ask history going back over time.

These can be tonnes of data (over time) if transactions are frequent. Will the system users pay for downloading these data? From whom? - a decentralized system of data brodcasters should be established just like a system of escrow agents.

   • System will also include live bid/ask data.

The above is good for depth of market (DOM) analysis, but you will most importantly need:

   • Live trading data (in ticks) - data of actual transactions happening is needed to plot charts.

It would be okay if the timestamping were not just in seconds. If the system is to succeed, in future you need at least milliseconds or even microseconds (for both historical and real-time events (ticks).

I created a system that can accomplish this, and there is no proprietary hooks in it at all.

https://docs.google.com/file/d/0BwUFHE6KYsM0ZkxLVmFwbXQ3ck0/edit?usp=sharing

[Loozik]

I created a system that can accomplish this

I hardly understand your white paper (I am a non-techie with poor English), but if you really created such a system then big WOW for you  

The reason why I focused on a data feed matter is my futures and forex trading experience.

There are market data vendors that charge monthly from USD 100 (I use those cheap data vendors) http://www.iqfeed.net/index.cfm?displayaction=data&section=fees through USD 1-3k http://www.barchartmarketdata.com/solutions_datafeed_pricing.php up to USD 10k per month for high quality data (like Bloomberg).

Transmitting quality data in acceptable latency mode can be resource (capex on servers and opex on electricity and lines) consuming. The decentralized exchange system should be designed to provide for remuneration to full data nodes (market data distributors). Just like now Bitcoin system is designed to provide remuneration to miners.

[bluemeanie1]

I created a system that can accomplish this

I hardly understand your white paper (I am a non-techie with poor English), but if you really created such a system then big WOW for you  

The reason why I focused on a data feed matter is my futures and forex trading experience.

There are market data vendors that charge monthly from USD 100 (I use those cheap data vendors) http://www.iqfeed.net/index.cfm?displayaction=data&section=fees through USD 1-3k http://www.barchartmarketdata.com/solutions_datafeed_pricing.php up to USD 10k per month for high quality data (like Bloomberg).

Transmitting quality data in acceptable latency mode can be resource (capex on servers and opex on electricity and lines) consuming. The decentralized exchange system should be designed to provide for remuneration to full data nodes (market data distributors). Just like now Bitcoin system is designed to provide remuneration to miners.

thanks!

yes, Im somewhat familiar with those kind of information services.  I have worked on Wall St. for a period pre-9/11.

There are some very complex problems when considering the issue of time series over a p2p network.  As a matter of fact Bitcoin itself descended from a 'distributed timestamping' algorithm.  So with any market you need to find a way to have all nodes agree that transactions appear in a specific ORDER.  This is required for simple account transfers, or more complex arrangements such as an order book.  My system solves this problem WITHOUT mining.

when the system gets to the point where it's usable, I'll announce on here and perhaps we can talk about supporting higher level use cases.

[Loozik]

thanks!

yes, Im somewhat familiar with those kind of information services.  I have worked on Wall St. for a period pre-9/11.

There are some very complex problems when considering the issue of time series over a p2p network.  As a matter of fact Bitcoin itself descended from a 'distributed timestamping' algorithm.  So with any market you need to find a way to have all nodes agree that transactions appear in a specific ORDER.  This is required for simple account transfers, or more complex arrangements such as an order book.  My system solves this problem WITHOUT mining.

when the system gets to the point where it's usable, I'll announce on here and perhaps we can talk about supporting higher level use cases.

FANTASTIC 

[Kouye]

To clarify, the design provides:
    • Fully decentralised history of trading data; volumes traded and inter currency rates going back over time.
    • Fully decentralised history of bid/ask data; a set of the bid/ask history going back over time.
    • System will also include live bid/ask data.

Bitcoin client needs heavy storage (this is just the begining) and consumes quite his share of RAM, handling just the Tx...
Any thought about this ?

[Loozik]

Will there be a dark pool mechanism?

[crazynoggin]

We do need a decentralized exchange on a network such as Tor eventually though.

[No_2]

Ok, I can now confirm that two of the MetaLair team (one of them being me) will be attending the BTC London conference on July 2nd.

Hope to see some of you there.

[boomerlu]

The design of the system will be such that to transfer fiat, e.g. USD for BTC you will need to do the following steps:

1. Download and install the wallet-client.
2. Find a bid/ask that matches yours.

how do you determine who has 'first dibs' on a bid/ask?

Pro rata? Back to the days of the open outcry... or Eurodollars

[No_2]

The design of the system will be such that to transfer fiat, e.g. USD for BTC you will need to do the following steps:

1. Download and install the wallet-client.
2. Find a bid/ask that matches yours.

how do you determine who has 'first dibs' on a bid/ask?

Pro rata? Back to the days of the open outcry... or Eurodollars

Bids/asks are satisfied between piers on the network. Therefore a bid/ask is satisfied when a pair of wallet-exchange clients with matching bid/asks agree and announce this to the network to process.

Because the system is fully decentralised the paradigm of getting 'closer' to the exchange to reduce trading time no longer applies, every entity on the network will have a different view based on their locality to other nodes with some receiving data earlier than others. The design is fully decentralised so is able to accommodate for this.

[bytemaster]

I have been working to develop a set of criteria by which to evaluate all of the half dozen ideas out there on a p2p exchange and here is what I have:
1. Decentralized
All parties in the system exist on equal footing, no party has special privileges, and there are no barriers to entry beyond an average PC and internet connection.   The loss of any party does not harm those that remain.   
2. Trustless
No trust shall be required of any party in the system.   No party has the ability to default.
3. Limited Liability
No party to the system is required to publicly engage in illegal or highly regulated activities or take any legal risks in excess of direct exchange of Bitcoin for cash among friends and family.
4. Scalable
The system must scale to handle any volume without compromising the other criteria of the system.
5. Diverse
The exchange should enable most common investment vehicles including, shorts, put and call options.   It should enable trading in any tangible asset class.
6. Aggregated
A single bid can be matched against many asks in an atomic exchange.  Users attempting to exchange a large amount of money should be able to do so in a single transaction.
7. Atomic 
No exchange or transaction is ever in a partial, incomplete, or reaches an invalid state. 
8. Fiat Escrow
Exchange of fiat for credit with the exchange via the traditional banking system shall not depend upon trusting any single party including buyer, seller, or escrow agent.   The escrow system should not be vulnerable to collusion of either the buyer or seller with the escrow agent.
9. The Price is Right
The exchange must not use any price information not derived from actual bids and asks provided by users of the system. 
10. Zero Sum
The exchange must neither create nor destroy value and every proffit by one party is matched by a loss of another party.  No party is ever in debt to the system (see Trustless)
11. Viral
The exchange should offer compelling benefits  which exceed the risks  for everyone (regardless of their need to exchange currency) to participate, share, and promote.  This will generate the deepest possible market and most liquidity, broadest public support, and greatest demand in spite of any regulation.
12. Private
The exchange should provide at least as much privacy as afforded by Bitcoin for all users involved.   
13. Fast
The exchange should allow the average individual to deposit, trade, and withdraw within 24 hours.  Trades executed within the system should be confirmed at least as fast as Bitcoin transactions.
14. Cross-Chain Trades
The exchange must support escrow-free cross-chain trading between compatible crypto-currencies.
15.  Secure
The exchange must have a level of security on par with Bitcoin.
16. Open Source
All hardware and software components of the exchange must be open, auditable, and reproducible by any average developer.
17. Passive Order Execution
Orders may be executed without the interactive participation of the user or their computer.


From what I can determine this exchange fails in the following categories:
Trustless - you are trusting the escrow agent not to run with your money
Diverse   - I am not sure how this system could possibly enable call/put/shorts
Aggregated - this system is subject 'dust bids' and requires potentially a hundreds of escrow transactions to fill a single large bid.
Passive Order Execution - all parties must be online to execute trades with their private key
Atomic  - After you find a match you are at the mercy of the counter-party for the timeout period. 
The Price is Right -  Not all trades are fungible (every counter party has different risks) and if you can pick your counter-party then this system relies upon 'reporting' of transactions.   
Fast - the trading of fiat is much slower and an after-thought.  While you wait for escrow to clear (or not) you are subject to exchange rate fluctuation
Cross-Chain-Trades - should not require escrow.

In many ways this system is very similar to Market Coin.     

[trilightzone.org]

Ok, I can now confirm that two of the MetaLair team (one of them being me) will be attending the BTC London conference on July 2nd.

Hope to see some of you there.

First of all  you guys have a great initiative !

As for the escrow trust issue;

How about introducing a 2 level system of escrows ? It would basically add some extra trust to the escrow issue like this:

- Level 2 escrows deposit X amount of BTC to Level 1 escrows (well known reputable members of the BTC community)
- Level 2 escrows do their business as usual but if a Level 2 escrow fails / decides to run with the valuables then the Level 1 escrow uses their deposit to
  cover the damages up to some height
- Level 2 escrows show (proof) how much they have deposited as a garantee with a Level 1 escrow in case of issues, so people can judge for themselves
  the amount of the transaction they want to do with them

Well, this is a bit of quick brainstorming but feel free to consider it as another possible way to increase escrow trust in general.

[boomerlu]

The design of the system will be such that to transfer fiat, e.g. USD for BTC you will need to do the following steps:

1. Download and install the wallet-client.
2. Find a bid/ask that matches yours.

how do you determine who has 'first dibs' on a bid/ask?

Pro rata? Back to the days of the open outcry... or Eurodollars

Bids/asks are satisfied between piers on the network. Therefore a bid/ask is satisfied when a pair of wallet-exchange clients with matching bid/asks agree and announce this to the network to process.

Because the system is fully decentralised the paradigm of getting 'closer' to the exchange to reduce trading time no longer applies, every entity on the network will have a different view based on their locality to other nodes with some receiving data earlier than others. The design is fully decentralised so is able to accommodate for this.

So you're saying there is no concept of a central order book. Any quote you're looking at may or may not be there, and may or may not accept your trade request (and can use any criterion it wants to pick a match).

[jubalix]

3. Setup an escrow and execute.

sooo how does this bit work, the gov's will regulate this part heavily, and how can I trust the escrow, how is this different to gox


what I want is a way to put in my Iban/sepa/BSB wire or whatever and that money is held some how securely in this distributed system, in such a way that I can recall it at any time, before exchanged

then your talking.

Eg bitcoin works because there is not counter party risk within the bitcoin system....I don't need to trust an escrow.

[hamiltino]

How about have a system where multiple escrows deal with 1 single fiat transaction? So we don't put all our eggs in one basket.

[bytemaster]

The ability to claw back funds is designed to force dependency on courts.   The only thing the gov needs to do is to legalize clawback of all crypto currency related trx and thus give scammers a license to steal.


This is why we need a crypto currency that has no insurer yet tracks the value of usd.  Ie. bitshares

[hamiltino]

The ability to claw back funds is designed to force dependency on courts.   The only thing the gov needs to do is to legalize clawback of all crypto currency related trx and thus give scammers a license to steal.


This is why we need a crypto currency that has no insurer yet tracks the value of usd.  Ie. bitshares

Ok, i should have read the article, didn't get the definition of claw-back right. Sorry.

[No_2]

We now have a twitter account:

    @MetaLair

Hope to catch some of you at BTC London tomorrow.

[gambit1]

Having an investment level at £25k might filter out less serious investors but is a barrier to investment by people with more modest resources. £25k is a big punt when there is no guarantee of return.