herzmeister
Legendary
Offline
Activity: 1764
Merit: 1007
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 19, 2013, 10:07:39 AM |
|
I recently wrote this up. Some ideas are similar, but I want to enable real daytrading with real depth rather than OTC only. BTC escrow also may not be necessary.
1. There is a fiat ledger database. It can be a distributed. It only contains internal bookkeeping information. Much like Ripple.
2. Any exchange website can use this database. Such a website can also run in Tor. 3. I'm new, I want to buy bitcoins, I register and log on to one of those exchange websites. 4. It looks and works like any other exchange. First, I have to deposit some fiat.
5. This website does not manage its own bank account however. 6. Instead, it tells me to wire-transfer the money to one or more random guy bank accounts. 7. After a while, the random guys confirm that they received the fiat funds. 8. I now have my fiat displayed in my exchange account. 9. Everything looks and feels just like MtGox and other exchanges. No OTC crap. I can instantly buy and sell BTC for fiat. It's just local bookkeeping and number juggling for the website after all at this point. 10. I've made some gains (or not) and want to cash out some or all of my fiat. 11. The network picks one or more appropriate candidates who want to deposit fiat to buy bitcoins. 12. From their point of view, the random guy of point 6 is now me.
13. I have to send a confirmation to the network, probably manually, that I've received my funds.
14. Mission accomplished.
Remarks:
A) The fiat ledger database contains all the necessary bookkeeping, but should keep private information like bank accounts secret.
B) However, cheaters could be blacklisted and their bank account details could be published.
C) Deposits and withdrawals may take quite a long while depending on current market parameters as there'd essentially be internal waiting lists that would have to be matched. Spectacular price crashes and bubble bursts can have dramatic effects in this regard.
Note: There's no one here who receives money to "not return it". Those who receive fiat are those who want to cash out anyway.
The only problem here is that there would have to be some confirmation mechanism about that fact. There's no public API for wire transfer data AFAIK to automate that. The receiver could complain they didn't receive any money. If they make that believable to the network, they could theoretically receive money twice.
Another problem rather is that those who put money into the system could revoke the transaction later, if their bank allows chargebacks. Here the blacklisting and publishing of their identity could be applied. The network could offer a bounty here, which could be funded by trading fees.
|
|
|
|
edmundedgar
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 19, 2013, 10:14:16 AM |
|
It's one of two things that would help mitigate the problems. The other is an alt-coin that controls its own money supply to keep its value pegged to a fiat currency - a kind of parallel crypto-dollar.
That could be Liberty Reserve? Perhaps that is the best quasi-crypto version of the dollar (also a euro version exists, I think). I was thinking we'd have to make a thing like Bitcoin but with a way to find out its own exchange rate and print or destroy money, but maybe Liberty Reserve would be good enough for practical purposes. I guess the minimum you'd need would be: 1) An API allowing transactions to be made in real time, and confirmed by third-parties. (Not sure if they have the latter, but if not maybe they could be persuaded.) 2) Non-reversible payments.
|
|
|
|
BTC Books
Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 84
Merit: 10
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 19, 2013, 10:26:15 AM |
|
Personally, I have no interest in daytrading. There are plenty of forex-type bitcoin exchanges for that - Gox, for example. The issue is making it easy for new people to get bitcoin; and easy for those who want to, to get fiat. Period.
For a P2P system like waxwing describes, I would be absolutely in favor of integrating delays in order matching to eliminate daytrading and HFT bullshit.
And I would never participate in anything with even a whiff of Ripple to it. That thing is designed to own all the bitcoin and issue IOUs in the form of XRP, near as I can tell. It's not open source, it's premined, it's completely centralized, and I want no part of it.
|
Dankedan: price seems low, time to sell I think...
|
|
|
waxwing
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 19, 2013, 10:39:39 AM |
|
I recently wrote this up. Some ideas are similar, but I want to enable real daytrading with real depth rather than OTC only. BTC escrow also may not be necessary.
1. There is a fiat ledger database. It can be a distributed. It only contains internal bookkeeping information. Much like Ripple.
2. Any exchange website can use this database. Such a website can also run in Tor. 3. I'm new, I want to buy bitcoins, I register and log on to one of those exchange websites. 4. It looks and works like any other exchange. First, I have to deposit some fiat.
5. This website does not manage its own bank account however. 6. Instead, it tells me to wire-transfer the money to one or more random guy bank accounts. 7. After a while, the random guys confirm that they received the fiat funds. 8. I now have my fiat displayed in my exchange account. 9. Everything looks and feels just like MtGox and other exchanges. No OTC crap. I can instantly buy and sell BTC for fiat. It's just local bookkeeping and number juggling for the website after all at this point. 10. I've made some gains (or not) and want to cash out some or all of my fiat. 11. The network picks one or more appropriate candidates who want to deposit fiat to buy bitcoins. 12. From their point of view, the random guy of point 6 is now me.
13. I have to send a confirmation to the network, probably manually, that I've received my funds.
14. Mission accomplished.
Remarks:
A) The fiat ledger database contains all the necessary bookkeeping, but should keep private information like bank accounts secret.
B) However, cheaters could be blacklisted and their bank account details could be published.
C) Deposits and withdrawals may take quite a long while depending on current market parameters as there'd essentially be internal waiting lists that would have to be matched. Spectacular price crashes and bubble bursts can have dramatic effects in this regard.
Note: There's no one here who receives money to "not return it". Those who receive fiat are those who want to cash out anyway.
The only problem here is that there would have to be some confirmation mechanism about that fact. There's no public API for wire transfer data AFAIK to automate that. The receiver could complain they didn't receive any money. If they make that believable to the network, they could theoretically receive money twice.
Another problem rather is that those who put money into the system could revoke the transaction later, if their bank allows chargebacks. Here the blacklisting and publishing of their identity could be applied. The network could offer a bounty here, which could be funded by trading fees.
If I understand you correctly, the main difference between what you're suggesting and what I'm suggesting is you're looking to have exchanges to create deep order books for liquid trading between BTC and USD held on those exchanges, or more accurately, the USD will be held in a distributed ledger. That would mean you need to perform the same magic as Satoshi did with the blockchain (verifying the ledger), which means proof of work again (to prevent the double spend attack). Isn't that right? Because effectively to take USD off that ledger means a transaction, which means you're having to create a whole crypto again (USDcoin or whatever). Possibly I misunderstood. (I was just looking for a way to let people bring USD into the BTC world, rather than trading back and forth, so I wasn't thinking about a ledger for it, just peer to peer buying/selling).
|
PGP fingerprint 2B6FC204D9BF332D062B 461A141001A1AF77F20B (use email to contact)
|
|
|
waxwing
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 19, 2013, 10:46:24 AM Last edit: May 19, 2013, 10:56:40 AM by waxwing |
|
This is really quite good. Here's a missing piece:
Integrate knowledge of ACH and SEPA into the routing scheme of the (P) client. Use bitcoin transfers to hop between the two when necessary, so that wire transfers are never needed. Then you've got the whole world into the system with no transaction having a base cost of more than a dollar. What individual members of set S choose to charge is up to them, of course.
EDIT: I would participate in this, and contribute to a kickstarter-type campaign.
Thanks for the support. I would definitely participate too! But I think github is the right place to start a project like this and take it from there. I have coding experience of sorts, but it goes back a decade or more, and even worse I have zero experience of P2P or network programming, so others would have to take a leading role in the implementation. On the other hand, I luckily will have a lot of free time over the next 6 months so I would definitely be in a position to take part. I'm struggling to get your idea, however. How does knowledge of bank transfers inside the software help us to avoid the need for doing the wire transfers? Probably being dense here... EDIT: Just to add, my original thought was to keep the bank transfer process completely out of the core of the software. There could be a mutable layer on top describing the payment with certain data (the amount, the currency etc.), to make it easier for the user, but it would probably be good if the software kept a certain "deniability" of the actual outside world transaction, for security reasons.
|
PGP fingerprint 2B6FC204D9BF332D062B 461A141001A1AF77F20B (use email to contact)
|
|
|
herzmeister
Legendary
Offline
Activity: 1764
Merit: 1007
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 19, 2013, 11:07:50 AM |
|
That would mean you need to perform the same magic as Satoshi did with the blockchain
Ripple or colored coins.
|
|
|
|
waxwing
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 19, 2013, 01:28:50 PM |
|
That would mean you need to perform the same magic as Satoshi did with the blockchain
Ripple or colored coins. Good point, I'm pretty sure colored coins could solve that. It's a pretty complex architecture though.
|
PGP fingerprint 2B6FC204D9BF332D062B 461A141001A1AF77F20B (use email to contact)
|
|
|
Peter Lambert
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 19, 2013, 01:42:44 PM |
|
Unfortunately I don't understand how ripple accomplishes this, it requires a network of friends to develop a trust relationship, so for those who have never used it before and don't have friends that use it, its worthless?
Or am I just missing something.
You don't need friends to develop trust lines, you just have to have some sort of business which you can trust. For example, one gateway which many people use is BitStamp. BitStamp only sends out ripple IOUs for bitcoins or USD which they are holding because people deposited those funds at their exchange. So if somebody gives you a BitStampBTC, you know there is an actual bitcoin sitting at BitStamp which you can withdraw if you have an account set up with BitStamp. You could also trade that BitStampBTC for USD without ever having an account with BitStamp.
|
Use CoinBR to trade bitcoin stocks: CoinBR.comThe best place for betting with bitcoin: BitBet.us
|
|
|
BTC Books
Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 84
Merit: 10
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 19, 2013, 04:39:59 PM |
|
That would mean you need to perform the same magic as Satoshi did with the blockchain
Ripple or colored coins. Good point, I'm pretty sure colored coins could solve that. It's a pretty complex architecture though. Yes, maybe. You're right about this though: I'm struggling to get your idea, however. How does knowledge of bank transfers inside the software help us to avoid the need for doing the wire transfers? Probably being dense here...
Putting bank info into the client software works within the boundaries of a single currency. It was me being dense, not you. It's still very effective and cheap within a currency though - SEPA is 0.9 Euro, and ACH is 25 cents - but not across that barrier. Frankly though, for new users and for general-use convenience, a single-currency P2P system would be Good Enough (TM). Cross currency gets more expensive. Not as bad as a wire transfer for sub-$2k transactions, but still high. No good at all for larger transactions. My thought was having users on a forex exchange (MtGox) to take in bitcoin and currency, and swap them on the exchange. A $1k purchase Euro > USD might cost about $8-9. Cheaper than a wire + exchange fee, but...
|
Dankedan: price seems low, time to sell I think...
|
|
|
Namealreadyexists
Newbie
Offline
Activity: 20
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 19, 2013, 05:45:30 PM |
|
From a user perspective I am only concerned about fraud. Unlike exchanges that hold both bitcoins and fiat, P2P will require one party to take the risk of the other party defaulting. How exactly is "escrow" going to work without a 3rd party arbiter?
An alternative, described earlier as "mutually assured destruction" is to have both parties put up a deposit equal to the value of the deal, and if either one of them is unhappy they both loose it, so the "bad guy" has nothing to gain. But the "good guy" has twice as much to loose, and that opens the door for another character, the "crazy guy" who just goes arround defaulting to sabotage the system.
Oh, and reputation is based on verification, and that's just another 3rd party action.
Take a look at hawala, it's based on trust. Trust is based on kinship, or one member voching for another. What we need, is just another social network, for friends with benefits.
|
|
|
|
waxwing
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 20, 2013, 06:04:35 AM Last edit: May 20, 2013, 06:26:14 AM by waxwing |
|
From a user perspective I am only concerned about fraud.
If you were only concerned about fraud, and not arbitrary actions by the government to lock accounts, then you would definitely be better off trusting large, centralized exchanges to transfer your USD to BTC, like bitfloor, Mtgox etc. Recent history and, I would say, an understanding of the politics of money, suggests that's not the right position to take. Unlike exchanges that hold both bitcoins and fiat, P2P will require one party to take the risk of the other party defaulting. How exactly is "escrow" going to work without a 3rd party arbiter? An alternative, described earlier as "mutually assured destruction" is to have both parties put up a deposit equal to the value of the deal, and if either one of them is unhappy they both loose it, so the "bad guy" has nothing to gain. But the "good guy" has twice as much to loose, and that opens the door for another character, the "crazy guy" who just goes arround defaulting to sabotage the system.
The NashX concept is only a bit different from what has previously been implemented as 2 of 2 escrow (at least that's the name I read). It's technically fairly simple. A sends the BTC into an address which can only be unlocked by A and B (B is the recipient). Take a look at the earlier discussion on this in the thread. 2 of 3 is only marginally more complex, and involves a third party, but the trust in the third party is not the kind of trust you need when storing money on an exchange or a bank account, because the third party can never take the money. Edit: https://en.bitcoin.it/wiki/User:Casascius/Escrow_scheme_draft has a good clear overview of the concept, and take a look at bitescrow.org for an implementation. It's fairly simple. Oh, and reputation is based on verification, and that's just another 3rd party action.
Yes and no. You can sign a message with the private key of one of your bitcoin addresses to verify identity. That's the beauty of bitcoin - it actually creates cryptographically very solid identities, but these identities are totally disassociated from government sanctioned identities. Reputation systems can be built on top of this bitcoin identity. I personally am not in favour of overemphasising the reputation aspect in this, though, because it seems that this would tend to a centralization of the money flows, which is really what (I think) we should be trying to avoid. Take a look at hawala, it's based on trust. Trust is based on kinship, or one member voching for another. What we need, is just another social network, for friends with benefits.
I'm certainly well aware of hawala as a concept, I guess most of us here are. It's an entirely different approach to Bitcoin, involving trust in third parties. The two are not mutually exclusive, but they are completely different systems.
|
PGP fingerprint 2B6FC204D9BF332D062B 461A141001A1AF77F20B (use email to contact)
|
|
|
aceking
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 20, 2013, 07:20:07 AM |
|
how you will wire money in a p2p exchange ? , i don't see it possible.
|
|
|
|
themusicgod1
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 20, 2013, 08:48:04 AM |
|
I'm talking about a secure protocol that is P2P and totally decentralized. I understand Ripple accomplishes this but it is controlled by a single entity and regulatable
Ripple will be completely decentralized. It is not designed to be a 'single entity', and though there will be nodes that will be regulated -- there will be those which will not. The final project is however not yet complete, though they must be getting close by now. Unfortunately I don't understand how ripple accomplishes this, it requires a network of friends to develop a trust relationship, so for those who have never used it before and don't have friends that use it, its worthless?
Or am I just missing something.
The 'something' you are missing is Gateways, which obviate the reason for these difficulties. Obviously, it's better if you have the network of friends and trusted contacts, however.
|
|
|
|
|
Namealreadyexists
Newbie
Offline
Activity: 20
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 20, 2013, 06:34:27 PM |
|
because the third party can never take the money It is not enough. There should be a mechanism for the 3rd party to make the right decision, and I don't see that being addressed by the escrow system. For example me and you agree to trade my bitcoins for your USD. I put the bitcoins in escrow, you wire me the money, now it's turn for the 3rd party to release my bitcoins to you. Based on what? If I don't admit I received your USD, you'll have to initiate a dispute, provide your bank statements, etc and this is becoming just another legal battle. How is the 3rd party supposed to get perfect information to make the right decision?
|
|
|
|
charleshoskinson (OP)
Legendary
Offline
Activity: 1134
Merit: 1008
CEO of IOHK
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 20, 2013, 06:38:43 PM |
|
ipple will be completely decentralized. It is not designed to be a 'single entity', and though there will be nodes that will be regulated -- there will be those which will not. The final project is however not yet complete, though they must be getting close by now. I've asked the Ripple community several time to record a lecture for the Bitcoin Education Project on how ripple works and explain how to conduct commerce. They have never replied back. I really hate being ignored. Yet, until I get a detailed explanation on Ripple, I honestly don't know if what you are saying is completely correct.
|
The revolution begins with the mind and ends with the heart. Knowledge for all, accessible to all and shared by all
|
|
|
solex
Legendary
Offline
Activity: 1078
Merit: 1006
100 satoshis -> ISO code
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 20, 2013, 09:54:06 PM |
|
ipple will be completely decentralized. It is not designed to be a 'single entity', and though there will be nodes that will be regulated -- there will be those which will not. The final project is however not yet complete, though they must be getting close by now. I've asked the Ripple community several time to record a lecture for the Bitcoin Education Project on how ripple works and explain how to conduct commerce. They have never replied back. I really hate being ignored. Yet, until I get a detailed explanation on Ripple, I honestly don't know if what you are saying is completely correct. If this project is to be permanently derailed then the way to do it is by trying to get it done via Ripple. This is the best outline solution thus far.
|
|
|
|
usscfounder
Member
![*](https://bitcointalk.org/Themes/custom1/images/star.gif)
Offline
Activity: 84
Merit: 10
|
![](https://bitcointalk.org/Themes/custom1/images/post/smiley.gif) |
May 20, 2013, 10:22:44 PM |
|
I am designing a P2P cryptocurrency exchange that can plug into sites like mtgox and btce: https://bitcointalk.org/index.php?topic=209269.0USSC Litecoin-P2P-Server - A Decentralized P2P Client-Server Application & Exchange For Fast Transactions That Utilize Any Cryptocurrency
Application Features:
1. Directly and internally handles wallet.dat files and the Bitcoin/Litecoin protocol.
2. Completely separate database for user accounts with decentralized replication of virtual-server to other P2P servers.
3. Utilizes banks of wallet.dat files that are not mapped to any specific user account. Evenly distributes coins to internal banks of wallet.dat files.
4. End-user client software accounts are assigned to specific virtual-server allowing transactions speeds comparable to a centralized system like Visa, MasterCard, and Pay-Pal.
5. End-user accounts and wallet.dat banks are replicated to other P2P servers.
6. P2P server network monitors each other for online or offline status. Virtualization server replication to replicate virtual servers across network in the event of physical server seizure or DDOS attack [virtual servers house wallet.dat banks].
7. Double spend attacks are mitigated by denying end-users access to wallet.dat files or banks. [Wallet.dat files are internally encrypted].
8. Uses Bitcoin/Litecoin protocol as a lower level protocol like web-browser applications use the transport layer in the OSI model stack.
9. Uses Bitcoin/Litecoin client and wallet.dat files as lower level application.
10. Eliminates need for confirmations by end-user clients and web-server or web based services.
11. No need for end-user to keep wallet.dat files on local computer or phone. Account is in the P2P cloud.
12. Account cannot be frozen or seized by anyone - even the server operator.
13. No changes to the existing Bitcoin/Litecoin network or protocol. Utilizes existing Litecoin network and protocol.
14. End-user application plug-in to existing Bitcoin/Litecoin client.
15. Government cannot track nor trace nor freeze your funds.
16. Funds cannot be traced.
17. Anyone can run a Bitcoin/Litecoin P2P Server (it's decentralized). If the server goes offline or is seized, the virtual server, accounts, and wallet.dat files are still safe and reallocated to other P2P servers.
19. Can be used as an peer-to-peer (P2P) exchange. Can utilize other exchange sites such as btc-e and mtgox for fiat conversions.
20. Completely Open Source.
This is how to best use cryptocurrency protocols
Bitcoin and other altcoins are best suited to operate as lower level applications and should not be designed for direct end-user control or use. Like web browsers use tcp/ip for lower level operations, the Bitcoin application and protocol should operate much the same way. End-user applications and services need to be built on top of the Bitcoin protocol and application.
There is no need to write code from scratch. all of the above can do coded by combining the bittorrent protocol [for virtual server and wallet.dat bank replication] and onion router protocols [for P2P Server communication] with the existing Bitcoin/Litecoin protocol [This would make a hybrid bittorrent/onion router/Litecoin application].
I will post on here design specs but I will not code it for you.
You have my permission to use the above designs. I will not make a patent or copyright claim against you so long as you keep it open source. I do not own any patents on the above system. It came from meditation and thought.
You don't necessarily have to use the Litecoin protocol, Bitcoin or Worldcoin will work as well. But I recommend the Litecoin Protocol [fast and secure].
I have been an Active Directory admin, and MS Exchange admin, Banyan Vines Street Talk Admin, And have supported everything from X.25 to X.400 and Protocols from RIP, OSFP, BGP, etc...
If you have any questions on how to design the application, I will post answers on here. You can PM me or just post your design questions here. I will help you. If you code an application for this then please give me credit for some of the design. That is all that I ask and maybe a little donations in Litecoin [PM me for address].
Get to Coding.
Thank you.
USSCFounder
Here are some of the first design specs to help you code the above system.
The system should be coded to run on an LAMP server using PHP and MySQL only. Perl can be used to facilitate server side scripts and systems commands as well. The end user client or plug in can be coded in C if desired.
Virtual Servers Virtual servers are just database tables that give a type of centralization to P2P servers facilitating a single point of transaction for a specific user account; thus allowing for speedy transactions that could not be otherwise obtained by conventional P2P cryptocurrency networks. Virtual servers can be configured to be reassigned to other physical servers in a few minutes in the case of ddos attacks or physical server seizure by authorities.
virtual-server-001.user.table user-id-key | coins.litecoin | coins.bitcoin | coins.namecoin (bob) XXXuser-id-key-001XXX | 100 | 0 | 20 (alice) XXXuser-id-key-002XXX | 0 | 100 | 20
The coins can then be evenly distributed so that no one account can be linked to any specific wallet:
virtual-server-001.litecoin.wallet-bank-001.table wallet-id | coin-amount wallet.vs-001.bank-001.litecoin-01.dat | 25 wallet.vs-001.bank-001.litecoin-02.dat | 25 wallet.vs-001.bank-001.litecoin-03.dat | 25 wallet.vs-001.bank-001.litecoin-04.dat | 25
virtual-server-001.bitcoin.wallet-bank-001.table wallet-id | coin-amount wallet.vs-001.bank-001.bitcoin-01.dat | 25 wallet.vs-001.bank-001.bitcoin-02.dat | 25 wallet.vs-001.bank-001.bitcoin-03.dat | 25 wallet.vs-001.bank-001.bitcoin-04.dat | 25
virtual-server-001.namecoin.wallet-bank-001.table wallet-id | coin-amount wallet.vs-001.bank-001.namecoin-01.dat | 10 wallet.vs-001.bank-001.namecoin-02.dat | 10 wallet.vs-001.bank-001.namecoin-03.dat | 10 wallet.vs-001.bank-001.namecoin-04.dat | 10
(MORE TO COME)
Additional design specs:
virtual-server files can then be propagated to other P2P servers simply by sharing the files via bittorent to the other P2P servers:
virtual-server-001 files to be propagated to P2P network:
virtual-server-001.user.table virtual-server-001.litecoin.wallet-bank-001.table virtual-server-001.bitcoin.wallet-bank-001.table virtual-server-001.namecoin.wallet-bank-001.table
wallet.vs-001.bank-001.litecoin-01.dat wallet.vs-001.bank-001.litecoin-02.dat wallet.vs-001.bank-001.litecoin-03.dat wallet.vs-001.bank-001.litecoin-04.dat
wallet.vs-001.bank-001.bitcoin-01.dat wallet.vs-001.bank-001.bitcoin-02.dat wallet.vs-001.bank-001.bitcoin-03.dat wallet.vs-001.bank-001.bitcoin-04.dat
wallet.vs-001.bank-001.namecoin-01.dat wallet.vs-001.bank-001.namecoin-02.dat wallet.vs-001.bank-001.namecoin-03.dat wallet.vs-001.bank-001.namecoin-04.dat
It is that simple to make a virtual-server. Each P2P server would have local copies of the virtual-servers on the network but only a few would actually be online. Every virtual-server would exist in an online state only on one of the P2P servers on the network. That means if virtual-server-001 was online in a New York P2P server, an exact copy of it on the Moscow server would be offline. In the event that the New York server was seized or ddos'ed and went offline, virtual-server-001 would then come online in Moscow.
Its a layered P2P application; one P2P application layered on top of another P2P application.
(MORE TO COME)
|
|
|
|
toffoo
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 21, 2013, 03:52:29 AM |
|
I can envision some clever developer coming up with an open-source distributed exchange client that could perhaps cryptographically escrow your BTC and LTC holdings, maintain a distributed order book, and then instantly swap the BTC for LTC when a trade has been agreed, without counterparty or clearing risk.
Something like that would probably catch on.
I am designing a P2P cryptocurrency exchange that can plug into sites like mtgox and btce: ...
Well that was quick!
|
|
|
|
areebmajeed
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
May 21, 2013, 06:55:46 AM |
|
I know how to make a p2p Exchange >> I'll come to your house and I will give you a dollar.
|
|
|
|
|