0000

[ar9]

0000

[tutkarz]

It was mentioned few times at last. How are you going to prevent people who operates community driven exchange from running with your money and bitcoins? People are too immature still to rezist such opportunity.

[SlickMcFavorite]

I am interested -- you could fund it by selling shares (like how ASICMINER did it).

[sal002]

Are you describing Vircurex (which has shares available and is shareholder driven)?

[Phinnaeus Gage]

Let's think outside the box for a sec, with the profit aspect in mind.

What if there were a standardized trading platform that all the exchanges use, with each exchange, wherever located, limited to a maximum of 1,000 users (actual number would need to be explored).

They all operate independently, but are linked for the sole purpose of getting a weighted average. It's that dollar amount generated from said average that'll be used as the official exchange rate, no longer relying upon one entity controlling 80% of all trades, coupled with going down due to being the victum of their own success.

Anybody will be able to run an exchange as long as they pay a prevailing fee for the privilege and provide their actual vitals to TBF.

Since this would be a weighted average, each exchange, no matter how big or how small, would have their rate weighted accordingly.

A person can own more than one exchange, provided they're on completely different servers and paid the fees to operate their subsequent exchanges.

They have to had run such an exchange successfully for one year before they are allowed to purchase a license to operate a second exchange. After two years, they can apply for two more, thus growing exponentially.

Since Mt Gox has been so kind in showing us how not to run an exchange, we'll allow them to start with two such entities, whereupon next year they can apply for four more, then eight, etc. (I even let the math work better in their favor: 2-4-8 oppose to 2-2-4, as in 1-1-2-4-8 for all others, thus showing that I don't have a MAJOR red-ass toward Mt Gox)

Mark K. can easily partner with some other person to finance another exchange in their name, as long as there is full disclosure.

All owners will be required to have a business license and incorporated accordingly in their respected countries.

Although I stated 1,000 users max, the number should be such that an owner can earn a respectable income from their exchange, but not too high, for remember, they will be in position to run another exchange in the coming year(s), coupled with financing partners to operate other exchanges early on. In a sense, they'll be able to game the system, but with controls in place, so to speak.

The above was just off the top of my head from the side not damaged from the rod, thus probably needs revised, but surely a beauty can be seen in such an endeavor.

~Bruno K~

[Herodes]

How would the devs be paid (in shares) or in fiat money, or in btc ?

Where would the exchange be located ? Would the devs need to move there, or could you work remotely.

Malta may be just a great place for such a business, with all the beneficial taxation rules and easy access to mainland Europe, it is also a member of the EU and party of the Schengen Agreement. Air tickets to most european countries are cheap, and it's also close to northern africa.

Or are you located in the states perhaps ?

To run a great exchange, you would need a team of developers, sysadmins and security experts. Some of their skills might be overlapping.

The most important thing is not to find those with the most experience, but those who are thrustworthy and work with great attention to detail and do things properly, and which can be trusted.

Also, there should be at least one finance dude, one lawyer dude, and perhaps one PR dude.

+ competent customer service.

Communication with customers should be great, and responsiveness and scalability of systems should also be good.

With great funding, one could kick of a very solid exchange in a relatively short time frame, one could also outsource components of the development job to external parties, and then audit it in house before deployment. There should also be a security audit of at least two different independent professional contractors.

Also ddos protection would have to be arranged, and there would be a need for several datacenters around the world, and the matching engine should be disconnected from the web frontend, and perhaps there even could be some way of having users directed to different web-servers based on their location or profile, so that the main systems really never got congested, as if one web-frontend got swamped, the other front ends held up well. Web front ends could also be relatively dynamic, perhaps even with captchas or some sort of verification system ensuring that users are humans. And the db backend and processing engine should be on a network not directly accesible on the net, with good protection, and in the event the network connection of theirs got congested, it should be possible to route the traffic some other way.

*puh* That started to be seriously a lot of work, but I guess that's the way it's got to be if you run an exchange, or you could just start small, and build from there as problems arise. There would also really be a need for a diplomatic person (preferrably a very nice looking girl, that could dwindle bank executives around her little finger, so we could have a good relationship with the banks). I am really not kidding here. Nice looking girls really get more goodwill from many people, esp. from middle aged men.. But people are politically correct and don't admit to this. A good looking lawyer woman that knows her stuff of course.

And all this would not be cheap!

[Phinnaeus Gage]

@Herodes

Your post is probably directly toward the OP and not mine, but after reading it I see that my brainfart would need a little tweaking to be viable, yet doable, for I somehow excluded the regulations, especially here in the US, into the equation.

Using the US as an example, what is the cheapest entry amount, in terms of dollars, are we talking here? Is there a min/max dollar amount traded before such regulations are required, or if even $1 is traded, regulations are in effect?

[Herodes]

@Herodes

Your post is probably directly toward the OP and not mine, but after reading it I see that my brainfart would need a little tweaking to be viable, yet doable, for I somehow excluded the regulations, especially here in the US, into the equation.

Using the US as an example, what is the cheapest entry amount, in terms of dollars, are we talking here? Is there a min/max dollar amount traded before such regulations are required, or if even $1 is traded, regulations are in effect?

I am based in europe so I do not know a whole lot about the country in which your barn-wood business is located.  But Fincen made a statment not long ago essentially saying that all exchangers need to be regulated, how much this costs, or the details of it, I do not know about.

And that's why there would need to be a competent lawyer on the team. Or at least hiring some lawyer services when needed.

[sal002]

As far as developers - I think there may be a clue here (looks like the beginning of an article about setting up an exchange) - http://www.devtome.com/doku.php?id=dvc_btc_exchange_-_episode_1_-_installing_the_software_stack

[Sage]

Another central exchange isn't the answer.  Doesn't matter who it's owned by.

The fundamental problem is centralization.  Single points of failure.  That couldn't be more the opposite of what Bitcoin is all about.

The solution is a community driven, open-source initiative to create a decentralized exchange platform that nobody owns.

[whiskers75]

Hmmm... A Node.js exchange, where the system matches up sellers and buyers automatically (while taking a small fee)...
Oh yes, and make it open source

[str4wm4n]

now this is what i call, thinking outside the gox

[ThreeJay]

I work in the IT of the finance industry for 7 years now...and what you seem to miss is that the problem is not purely a technical one. I am very certain that it is the more easy part of the task to develop a technically strong solution (scalable infrastructure, trading backend disconnected from web frontend, api for developing trading software and whatnot). I believe the hard parts are:

1. Security - You need a whole lot of infrastructure for that. Not only to secure your IT systems but also the processes around it. I mean, cold storage for bitcoins is the one thing, securing that cold storage is the other thing (good old fashioned bank robbery, anyone?). You also need to secure the company from it's employees. There is no such thing as 100% trust as long as there are millions of fiat currency involved.

2. Regulation - Everyone wants to escape regulation. But you should not. It is there for a reason. Regulation is one of the main reasons why you can actually trust a company to offer their service in a certain standard. Regulation forces you to keep your IT systems safe and to have disaster recovery plans and and and...But keep in mind that trust is never 100%.

3. Decentralization - I believe you can't. It's a wet dream, I get the notion of it. But a decentralized exchange would mean decentralized trades. And how do you fashion that? How are you going to send your dollars to someone else? By mail?

An exchange will always be a centralized, bank-like institution. And if you want to do that you should do it the 'old-fashioned' way. You should do it in a jurisidication which forces you to provide the service to a certain standard and thus will allow your customers to put some trust (not 100%!) into your company.

In terms of profit: We are talking about something which will need a lot of full-time employees. And which will need a lot of funding. How are you going to get that funding if you don't offer profit in return? shareholder-owned != community-owned.

The closest you can get to a decentralized exchange is developing a technical exchange solution and giving it away as free software. Then you can hope for many people around the world to open their own exchange. And still then, their earnings must cover their cost in the long run. Costs will be determined by regulation and thus by location. And you will probably end up with the cheaper exchanges in far-far-away-land being the most frequented and thus the market will not be much more secured than it is now.

That's just my 2 cents.
3J

[Herodes]

A decentralize exchange could've been made - but the problem is the link btc <> fiat. One way could be to have several agents in the system for each country acting as entry and exit points for fiat. But then there's always the danger of dishonest agents. But if an agent had a certain limit as to how much money they could hold, and agents were selected and appointed only by trusted members of the community, and also had to submit all their ID-papers (shit that would be centralization again...) to some hub, then they could be pursued if they went malicious.

But this is one way it could work. There could even be agents keeping stash of cash, and there could be some kind of tracking system where one kept tab of the holding of each agent, and agents could shuffle funds in between them to even out balances sometimes.

I would think a p2p exchange would be possible, like the Bitcoin-QT klient, only as an exchange,and when someone wanted to deposit money into the system, they had to do it through an agent. That would function much like bitcoin-otc today, but the difference would be that bitcoins could be traded against fiat money on the p2p exchange. And people could be doing daytrayding like they do it on the exchanges now. But some people would eventually run with money, and how much money could an agent store before it became too much ? But if there were fees collected from the service, these could be accumulated and offset any risk of an agent making a run, but again the collection of a fee would require some centralized structure ? Unless of course fees collected could be kept with extra trusted agents as insurance against some agents making a run.

Careful selection of agents would make the system more thrustworthy, I think.

Could it possibly work ?

[ThreeJay]

Well...that's an interesting suggestion you make, Herodes.

But, I don't think it will work:

The identity of agents would have to be more or less public to make transactions of fiat money to them. That would make them vulnerable to crime. And we are talking non-virtual crime here: Kidnapping, blackmail, robbery, murder. Sort of ugly things...

You have to bear in mind not only the risk such agents would pose them selves but also the risk such agents would be exposed to. I for one, wouldn't want to be an agent.

[Praxis]

Is this : https://github.com/p2p/bitcoin-exchange  relevant?

[johnyj]

You can not move fiat money ELECTRONICALLY without any centralized organization

I think localbitcoins.com is one of the choice, this is essentially already a very distributed exchange system. Buyer and sellers meet personally and carry the trade using cash, which has nothing to do with existing banking system

But this system have lots of limitations, you can't trade fast, you can't carry a case of dollars on the street, and you really don't want to reveal your idendity if huge amount of money are involved

These just showed how much we are used to dependant on a centralized authority when it comes to trust

The trust model in ripple is worth studying, but then you are dependant on your social network, which again have some limitation and privacy concern