P2P Exchange for bitcoin

[gollum]

http://irdial.com/blogdial/?p=3505

The question is, who is going to put it together and release it?

We will do it open source - Im sure there are many talented people here at bitcointalk with competency in the field of programming and finance.

https://github.com/p2p/bitcoin-exchange

[1714964600]

1714964600

[1714964600]

1714964600

[1714964600]

1714964600

[1714964600]

1714964600

[Rassah]

http://irdial.com/blogdial/?p=3505

The question is, who is going to put it together and release it?

We will do it open source - Im sure there are many talented people here at bitcointalk with competency in the field of programming and finance.

All if this sounds too much like, "I have this awesome idea, which is WORDS. I don't actually how how it will work, technically, of it it's even possible, and I don't feel like actually executing it myself, because my idea is so good that someone else will like it, figure it out and code it for me."

[gollum]

http://irdial.com/blogdial/?p=3505

The question is, who is going to put it together and release it?

We will do it open source - Im sure there are many talented people here at bitcointalk with competency in the field of programming and finance.

All if this sounds too much like, "I have this awesome idea, which is WORDS. I don't actually how how it will work, technically, of it it's even possible, and I don't feel like actually executing it myself, because my idea is so good that someone else will like it, figure it out and code it for me."

I can code, but Im not a programming guru like some of you guys. But people like me can at least try to contribute with ideas and architecture.

[Rassah]

http://irdial.com/blogdial/?p=3505

The question is, who is going to put it together and release it?

We will do it open source - Im sure there are many talented people here at bitcointalk with competency in the field of programming and finance.

All if this sounds too much like, "I have this awesome idea, which is WORDS. I don't actually how how it will work, technically, or if it's even possible, and I don't feel like actually executing it myself, because my idea is so good that someone else will like it, figure it out and code it for me."

I can code, but Im not a programming guru like some of you guys. But people like me can at least try to contribute with ideas and architecture.

Sorry, I know ideas are needed, but I'm probably just upset that we've been having these ideas, mostly the same ones, for almost two years, with no one being able to come up with anything new, or how to actually come up with the solution to the fiat storing/transmitting problem (this token thing sounds barely promising). In a way, the fact that we still haven't solved the P2P fiat problem is a testament to the superiority of Bitcoin to cash.

[juice]

yes he is ok !

what about : http://bitcoin-otc.com/

looks intresting: http://bitcoin-otc.com/vieworderbook.php

easy web frontend  isnt it

[gollum]

yes he is ok !

what about : http://bitcoin-otc.com/

looks intresting: http://bitcoin-otc.com/vieworderbook.php

easy web frontend  isnt it

Yes Bitcoin-OTC is a great solution! If we can make it decentralized and make it easy for brokers to participate this could be our official exchange.

[mobile4ever]

Sorry, I know ideas are needed, but I'm probably just upset that we've been having these ideas, mostly the same ones, for almost two years, with no one being able to come up with anything new,

This is one I am fairly sure you have not seen:


https://bitcointalk.org/index.php?topic=145389.0


Its a PHP plugin for Wordpress.


Many here have plenty of good ideas. Can we mix a few of them? One thing about software; you only have to make it work one time and then it does the rest.

[gollum]

http://irdial.com/blogdial/?p=3505

The question is, who is going to put it together and release it?

We will do it open source - Im sure there are many talented people here at bitcointalk with competency in the field of programming and finance.

All if this sounds too much like, "I have this awesome idea, which is WORDS. I don't actually how how it will work, technically, or if it's even possible, and I don't feel like actually executing it myself, because my idea is so good that someone else will like it, figure it out and code it for me."

I can code, but Im not a programming guru like some of you guys. But people like me can at least try to contribute with ideas and architecture.

Sorry, I know ideas are needed, but I'm probably just upset that we've been having these ideas, mostly the same ones, for almost two years, with no one being able to come up with anything new, or how to actually come up with the solution to the fiat storing/transmitting problem (this token thing sounds barely promising). In a way, the fact that we still haven't solved the P2P fiat problem is a testament to the superiority of Bitcoin to cash.

This time it is about the future of bitcoin. If we are so lazy that we let MtGox's failure destroy the reputation of bitcoin, then its a shame Satoshi wasted his time creating bitcoin.

[mobile4ever]

This time it is about the future of bitcoin. If we are so lazy that we let MtGox's failure destroy the reputation of bitcoin, then its a shame Satoshi wasted his time creating bitcoin.

That  wont happen. Too many lives are at stake, plus freedom is at stake as well.


http://lfb.org/today/what-bitcoin-is-teaching-us/

At the end he says:

....the prospects for freedom will become bright again.

Bitcoin is still in beta, so the only way is up.

[marcus_of_augustus]

But I trust a P2P open order book more than I can trust the orderbook of a server that easily can be hacked or ddosed.

So, effectively, the brokers publish promises as their order book? 

Does this get linked to a particular customer?  If you place a sell order, you then see it on the public list?  How are cancels handled?

In theory, you could have the exchanges actually act as banks.  For example, they could issue chaum digital cash.  Mt. Gox could give you 10 Mt. Gox dollar for $10 and buy 10 of them back for $9.90 (or whatever spread made them profitable).

These "dollars" could then be traded using a alt-chain. 

In fact, you could exploit the main chain if you wanted to.  You could use the coloured coin idea with 1 satoshi coins = $1.  Basically, coins generated from a particular source would be alt-coins.  For example, if the coin came from a particular address.  However, an alt chain might be better.

However, it is done, mints convert fiat to/from digital tokens and then they can be handled digitally.

A Federation of OpenTransactions signing servers could have this system up and running in 2-6 months (maybe less if more resources) ... the "exchange servers" are not so much exchanges as Notaries that sign off on any trades/transfers that pass through them ... i.e. it is questionable whether they need to worry about MSB BS because they do not actually transfer anything, just sign off on it.

[loanexpress]

A Federation of OpenTransactions signing servers could have this system up and running in 2-6 months (maybe less if more resources) ... the "exchange servers" are not so much exchanges as Notaries that sign off on any trades/transfers that pass through them ... i.e. it is questionable whether they need to worry about MSB BS because they do not actually transfer anything, just sign off on it.

If someone knows how to do this, please share. Last time I looked at OT it was all very abstract. Yes there is code and everyone agrees it's elegant and amazing, but HOW DO I USE IT?

I'm happy to throw a server or 2 at it - but there needs to be some sort of map on how to get there.

[Luckybit]

Two things I want to stress. This exchange has to be even more simple than mtGox. It should connect to people on Skype or AIM or ICQ etc and let them order through that interface or web or any other.

The API should allow anything to plug into it so it can be ubiquitous.

It must be modular, so that it can be easily pluggable, extendable, etc.

Basically I think what is needed is a protocol and an API, written in a simple language like Python so anyone can easily read it and make use of it. Ruby could work too for it's worth. If it must be written in C or C++ then write it so that high level scripting languages can make use of all it's functions and API.

As far as how the exchange will work that is up for debate but definitely make the code well written, modular, easy to read, easy to upgrade, and let as many people plug into it as possible. So if the core of it is C++ or C then let as many from the community write code in their own languages for their own plugins as possible.

The goal should be to write code which can be reused a lot by a lot of people and code which can be read and improved up by a lot of people. I think it might be useful to start completely from scratch here, but design is the most important part and that is the part which in my opinion needs the most debate and by design I mean what is the best possible design from a computer scientist or engeering phd perspective?

Maybe it's time to look up some of the academic journals and see whats out there or find some really well written code and see which examples we can use for case studies which have been known to scale up. This code has to scale up without losing flexibility and remain decentralized (not easy).

[Sage]

It needs to be done without servers period!

Techies, is this a potential option:

https://bitcointalk.org/index.php?topic=174326.0

[Luckybit]

http://irdial.com/blogdial/?p=3505

The question is, who is going to put it together and release it?

We will do it open source - Im sure there are many talented people here at bitcointalk with competency in the field of programming and finance.

All if this sounds too much like, "I have this awesome idea, which is WORDS. I don't actually how how it will work, technically, or if it's even possible, and I don't feel like actually executing it myself, because my idea is so good that someone else will like it, figure it out and code it for me."

I can code, but Im not a programming guru like some of you guys. But people like me can at least try to contribute with ideas and architecture.

Sorry, I know ideas are needed, but I'm probably just upset that we've been having these ideas, mostly the same ones, for almost two years, with no one being able to come up with anything new, or how to actually come up with the solution to the fiat storing/transmitting problem (this token thing sounds barely promising). In a way, the fact that we still haven't solved the P2P fiat problem is a testament to the superiority of Bitcoin to cash.

Here is a way to code it. Start by using a language a lot of people know to make a basic open exchange protocol. This could be done in C because C is the most widely used language there is. The basic functions can be written in C and then an API can be created. This API should allow someone such as myself who prefers high level languages like Python to code certain aspects. Basically the protocol should be language agnostic so that as many different programmers as possible can join into the project.

The project has to put a specific focus on countering trading algorithms and on preventing attacks which can game the system. This would require an engineer with PhD level math and programming skills beyond most of ours to come up with some ideas. The first step for any of us who are grad or Phd students is to use our resources to research all the journals on this issue to see whether or not any novel ideas have been found.

For example game theory may be part of the solution just based on my hunch but once again you'd have to look it up and read a journal or two to figure it out. There may be all kinds of solutions but the main thing is research. One way to facilitate research is by offering a Bitcoin bounty to anyone who does the research and discovers the solution.

I don't know how the bounty could be set up but I think this is the best way to encourage the masters or Phd level students in here to actually research the theoretical solution. Once a theoretical solution is proposed then it should be voted on and if it wins the contest then the winner should receive the Bitcoin bounty.

At some point start a Kickstarter or something like this and have the community officially back the project once a theoretical design solution is found. It's not enough to look at already existing projects because the design we'd need for this one would have to prevent all kinds of manipulations, and being distributed doesn't mean it would be completely immune to attack. Also I see this as a chance to strengthen Bitcoin and cryptocurrencies in general. So why not do it right? Bitcoin was done right, so gather up the best brains and start the contests to see which ideas and proposals are the best.

[wumpus]

Guys, the forum is overflowing with topics that initiate a p2p bitcoin exchange. I'm not kidding. Yesterday I investigated them all, and I chose help developing bitcoinx (2nd place was darkexchange, at a distance). You can do your research and you'd probably end up joining bitcoinx as well . See user killerstorm, project armoryx, project bitcoinx.

Interesting, thanks for the links! I wasn't aware of any projects actually in development. It has gotten lost in all the theorizing and arguing noise.

Overview of current projects:

- bitcoinx: http://www.bitcoinx.org/  

Protocol description: https://bitcoil.co.il/BitcoinX.pdf
Other code, as well as links to the forum can be found here: http://www.bitcoinx.org/resources/  
Repositories here on github: https://github.com/bitcoinx
Wiki: http://wiki.bitcoinx.org/index.php/Main_Page

- darkxchange "Dark Exchange is a distributed p2p exchange for bitcoin.". Uses the I2P network.

Source available at https://github.com/macourtney/Dark-Exchange  
Described on bitcoin wiki: https://en.bitcoin.it/wiki/Dark_Exchange .
Also has a forum thread: https://bitcointalk.org/index.php?topic=26063.0;

Not really automated P2P exchanges, but P2P nevertheless:

- #bitcoin-otc (peer to peer trading with PGP web of trust and reputation system)

- Running an exchange in Google Docs: https://bitcointalk.org/index.php?topic=174117.0

Any others?

[Luckybit]

It needs to be done without servers period!

Techies, is this a potential option:

https://bitcointalk.org/index.php?topic=174326.0

You know, it could be done via sneakernet even but the point is I think we need to start with a Wiki of some sort. We need to have a design contest.

Something like design the best open distributed exchange protocol and win X amount in Bitcoin.
If there is no current way to even do these sorts of contests then we need a way to reliably reward people in Bitcoin who come up with brilliant designs and solutions. The way to avoid having to redesign or avoid embarrassment is to do it right the first try and the way to do that is to hold contests and have peer review.

So let's start with a big contest which lasts perhaps a week or a couple weeks and then see who wins and the winner of that contest should be the person who designs the protocol.

[Luckybit]

Guys, the forum is overflowing with topics that initiate a p2p bitcoin exchange. I'm not kidding. Yesterday I investigated them all, and I chose help developing bitcoinx (2nd place was darkexchange, at a distance). You can do your research and you'd probably end up joining bitcoinx as well . See user killerstorm, project armoryx, project bitcoinx.

Interesting, thanks for the links! I wasn't aware of any projects actually in development. It has gotten lost in all the theorizing and arguing noise.

Links:

- bitcoinx: http://www.bitcoinx.org/  ... has protocol description and even a download, but where is the source code? When I click join, it says "Register with your Google account". Uh, no.

- darkxchange: https://github.com/macourtney/Dark-Exchange   "Dark Exchange is a distributed p2p exchange for bitcoin.". Uses the I2P network, https://en.bitcoin.it/wiki/Dark_Exchange .

I think for this it might be better to start from scratch. Sure if portions of their code is well done and useful then use it, but where is their code so we can peer review it?


And even with their code it doesn't mean their designs are going to solve all the problems. We need the perfect design first, then ANYONE can write the code. The protocol design is the key, and the underlying algorithms are what make a program efficient. And that is something I admit I'm not very good at, and few people actually are. Satoshi happened to be one of the people who is good at that, how do we find the next Satoshi? I suggest a design contest.

[wumpus]

And even with their code it doesn't mean their designs are going to solve all the problems. We need the perfect design first, then ANYONE can write the code.

Years of experience with open source has made my way of working is 100% opposed to yours. Sure, only fools rush in, but I usually like to figure out the basics then have something working first, then in time make it better. You will never have a "perfect design". What do you think would have happened when Satoshi first argued for years about a perfect design? There would still be nothing but an academic project. It's a myth that ANYONE can (or will) write the code.

And even when starting completely from scratch you need to learn about the previous projects first, even if only to know why they failed. But let's not start with a NIH attitude, everyone going their own way is what kills projects not what makes them.

Edit: I updated my post above to reflect all the github repositories etc, should be more complete now.

[Sage]

Guys, the forum is overflowing with topics that initiate a p2p bitcoin exchange. I'm not kidding. Yesterday I investigated them all, and I chose help developing bitcoinx (2nd place was darkexchange, at a distance). You can do your research and you'd probably end up joining bitcoinx as well . See user killerstorm, project armoryx, project bitcoinx.

Interesting, thanks for the links! I wasn't aware of any projects actually in development. It has gotten lost in all the theorizing and arguing noise.

Links:

- bitcoinx: http://www.bitcoinx.org/  ... has protocol description and even a download, but where is the source code? When I click join, it says "Register with your Google account". Uh, no.

- darkxchange: https://github.com/macourtney/Dark-Exchange   "Dark Exchange is a distributed p2p exchange for bitcoin.". Uses the I2P network, https://en.bitcoin.it/wiki/Dark_Exchange .

I think for this it might be better to start from scratch. Sure if portions of their code is well done and useful then use it, but where is their code so we can peer review it?


And even with their code it doesn't mean their designs are going to solve all the problems. We need the perfect design first, then ANYONE can write the code. The protocol design is the key, and the underlying algorithms are what make a program efficient. And that is something I admit I'm not very good at, and few people actually are. Satoshi happened to be one of the people who is good at that, how do we find the next Satoshi? I suggest a design contest.

Big plus++++

Design contest is the next logical step.

Lets get that step going today. Seriously, today.

What would it take?  I'm not a techie.  What's the best collaborative platform to use here?

[Luckybit]

And even with their code it doesn't mean their designs are going to solve all the problems. We need the perfect design first, then ANYONE can write the code.

Years of experience with open source has made my way of working is 100% opposed to yours. Sure, only fools rush in, but I usually like to figure out the basics then have something working first, then in time make it better. You will never have a "perfect design". What do you think would have happened when Satoshi first argued for years about a perfect design? There would still be nothing but an academic project. It's a myth that ANYONE can (or will) write the code.

And even when starting completely from scratch you need to learn about the previous projects first, even if only to know why they failed.

Edit: I updated my post above to reflect all the github repositories etc, should be more complete now.

mtGox is broken right now because they didn't properly design their code base. They relied on a flawed design because they didn't actually re-engineer or rethink everything.

Academic projects are useful for coming up with basic prototype designs. The design is extremely important when you want something to scale up without breaking down for instance. MtGox was not designed to scale up which is why they have lag. Just look at Napster, that was the initial design. They built it because it could be built at the time and it solved the problem but it didn't solve the problem in the best way so it got shut down, then we had to go through Kazaa, Bearshare, Limewire, and a bunch of other petty rip off protocols before we actually got to the well designed stuff like Bittorrent, Tor, I2P. All the while Freenet was in the background being worked on since 2000 around the same time as Napster but it didn't stop people from copying Napsters weak design over and over and all of them getting shut down and all their users having to switch from service to service.

That is what I want to have avoided. If it takes a design contest which lasts a week to avoid years of that then so be it. If you don't want to go the academic route you don't have to, but why should we believe your design is better than any other if it's not going to be peer reviewed either on this forum or at some academic journal? If you're afraid of peer review then build it yourself but don't expect people to join in if we don't know the mathematics are sound, that the algorithms will work, that the solutions are scaleable, that it's actually decentralized and not easily attacked etc.