Topic: P2P Exchange for bitcoin - page 6. (Read 42783 times)

pitching again my decentralized marketplace ('unbazaar') that can easily be adapted to trade cryptocurrencies. current design has all the features that you guys have been talking about, without being too difficult to make.
current design docs: http://ahdinosaur.github.io/unbazaar/docs (requires up-to-date browser), don't mind the introduction as that highlights another use case of a decentralized market, but please check out the technical design details.
also me and another person will be working full-time on unbazaar starting in around a month. of course we can always use more help.

I was thinking the same thing. Its all about their bottom line.

The system I described does use a 3rd person... But that 3rd person is chosen arbitrarily, and is not a permanent middleman in any way.

It's basically anarchy in a software. Two people needing dispute resolution will always look for a 3rd person to fairly resolve their dispute... In nations a person is always assigned to be the dispute resolver, but in anarchy they pick a non-biased, arbitrary 3rd party. This system just takes care of all of that automatically and profitably.


Of course they wouldn't... They'll refuse to look at it until it gets as big as they are...

But then they may find some use for it to free up liquidity too. In the meantime, there's nothing they can do to hurt it.

Sounds good to me. Charge on OP! Success for us all...

That is the last stage. They will not take us seriously until stage 6 is finished, when we have a working system that is stable, scalable and cannot be DDOSed or hacked they will become interested.
Some of them may join us, some may chose their own path. Our system will not destroy their business, they will benefit from this stable and secure system that they didn't have to pay anything for.
What they charge from their clients, and how they trade the markets is up to them.

That is a bold move. Do you think they will want to practically destroy their own money-makers?

We should agree on a final proposal before recruiting programmers.
Right now we are at phase 1-2, we got lot of ideas to organize into one or a few proposals that are practical in the real world.

1) Brainstorming about p2p bitcoin exchange, all ideas are welcome, even crazy ideas.
2) Organizing the ideas into 1-3 solutions  (general architecture and functional requirements)
3) Choosing one solution by voting & discussing
4) Recruiting professional and motivated programmers into the development team. Reward: honor and donations. If you know these programmers please talk to them: https://github.com/bittorrent?tab=members  https://github.com/bitcoin?tab=members
5) Architecture: Dev team makes a detailed architecture based on the final solution.
6) Developing the system
7) Invite existing exchanges MtGox, BitInstant, TradeHill et al. to use this p2p bitcoin exchange

Has anyone involved with this project tried contacting him to make him aware of it? He also said this in his blog...

"If you or anyone you know is one of the top percentile programmers and theorists in P2P protocols please contact us here at TDV."

Not sure if anyone here is a P2P programmer, but it sounds like he's interested in hearing from people that want to build something like this.

Any system that puts a seller into contact with a buyer directly without a third person is great. If it is decentralized, it makes it even better. As long as a few people are not in charge of this proposed system, that is, it is open source, it is nearly perfect. The world needs it.

Lots of great reasons.

First of all it makes it so you can't cheat bob, and bob can't cheat you. 3rd party Escrow safety.

Secondly, it adds distance. If the acceptable pingtime is 100ms, and you're 99 away from trish and trish is 99 away from bob, then you can bob can trade, despite being too far away to see each other.

Third, and my favorite, this setup gives everyone in the world an incentive to keep this client running; Free money.


Good news: Jeff Berwick just said this on his blog today:

That's some major cash, folks!

If we can get the right P2P programmer with us on this, Jeff could fund/PR it into existence easily.

Nah. Futureware...

So, Vaporware, then. Got it.

Its not made yet. That is the price to make it, its free to distribute. Plus, that was back when a bitcoin was just 60+ dollars. Sheesh...

Why is there someone in the middle? What happens if you and Bob connect directly? Why can't the quit order be sent when both parties sign something saying they are satisfied with the trade?

Free as in the BTC99 that you want to get paid? That's an interesting definition of free.

Is this free and open source? The one in my signature is.

Allow me to step in and answer some of this for poor Gollum. I know he and I don't have the exact same definition in mind, but it's pretty close; We both want a fully distributed, un-take-down-able client software exchange.


Yes.

Every client, exactly like bittorrent

i.e. If you want to trade, you gotta host trades.


Each client has a wallet inside it for each currency it can trade in. If your client has $10 USD in it, and you want to trade that for BTC, for instance, then you can't convert it locally, you have to trade others for it.

You do this by creating your order in right there in your client, which broadcasts it around the system (everyone within an acceptable pingtime range) and your order gets stored on the book of orders of all of those other clients...

As a match appears, the client most directly inbetween your client and the match's client does the trade, and collects the trading fee... So it really takes three running clients to make a trade at all.
 

All books we can make work exist on every single client, for the benefit of every client but that one.


The order books on each client are the data structure, I don't think there would be need for a blockchain.


A. Your order gets put into a short string and broadcast around, like this: "User1234BUY10BTC@120USD"

B. That string is instantly displayed on every order book within a short-enough pingtime, likely hundreds of thousands once the network grows large.

C. Bob, then creates an order to sell 10 BTC at a price @ or above $120 USD, and that gets broadcast around as well.

D. Trish's client, which happens to be running directly inbetween you and Bob, notes that your two trades are compatible and executes the trade, acts as a micro-escrow and takes her pre-set fee. (Then broadcasts a 'quit order' on both yours and bob's orders to all other books in her range.)  


Just like downloading the bitcoin wallet client, but funding the wallet with USD will be the hard part, obviously. Everything here should be doable right now except for the nationalized currencies in that wallet. -We're still thinking hard about that part.


None needed excepting of course for the brokers to fund these wallets. Think: OKPay, but hopefully better.


They'd be regional prices. Sure, you're never going to know what the price of BTC is in Japan when you're in NYC, but your trade is safe, secure, non-DDos-able, and no government can stop it either.


Clearly that wouldn't be a worry with thousands of clients in range, fully distributed.


Something else I feel I should mention here: Remember how Trish didn't lift a finger yet got paid a fee?

...That's going to be an awesome incentive for everyone and their uncle to keep these clients open and online 24/7.

Hope that helped!
 

I thought so when I wrote the "+1".

Today is a good day.

+1

That was meant as a joke