A Bitcoin without Mt.Gox

[BTCLuke]

Now that we're all agreed that Mt.Gox must not be the Single Point of (massive) Failure in Bitcoin, we've got to have a serious discussion.

Sadly, I don't see any exchanges on the horizon that can actually handle the volume that bitcoin ALREADY has... Even after we lose the tire-kickers today, the runner-ups like BTC-E and Vircurex are spotty and never even had a real-time API for a decent trading platform to track.

...And if one of them or a new exchange did step up to fill Mt.Gox's shoes, then we'd be in the same boat again... One point of failure that'll be DDossed into similar problems.

It's really pathetic if you think about it. Bitcoin's future is going to look extremely dim to professional analysts after today.

So how do we SURVIVE this re-occurring problem?

I only see a few options: (Please add more if you can think of one!)

• 1. Invent a decentralized Exchange. (The perfect option, but it would be extremely challenging; likely a challenge for Satoshi himself!)

• 2. Find a way to somehow equally split the use of exchanges. (Decentralize membership. It's cheating the free market, but hey, it might work!)

• 3. Go back to being a sub-$1 bitcoin that only the most freedom-minded people would want to own. (And somehow keep it down there artificially.)

• 4. Approach a Wall Street exchange and convince them somehow to take us on. (Pandora's box 2.0)

Options 3 & 4 sounds scary as hell to me, so it appears that all the best options require that we find a Decentralized answer of some sort; and it's likely going to take another coding genius to do it.

Until that occurs, Bitcoin will always be way too volatile for most people to take seriously as a currency, because today's events could happen at any time, and even Bitpay can't save anyone from serious financial pain when this occurs.

Help us Satoshisan Kenobi, you're our only hope...

[empoweoqwj]

That's very pessimistic ... especially if you mention ***holes like btc-e.com as an alternative.

There is a vacuum to be filled, it will be filled, and it will be a lot better than MtGox. Fast forward 12 months and we will have completely different (and competing) players. Until then, enjoy the bumpy ride 

[qxzn]

How about an MPEX-style solution? A well-capitalized company opens a premiere exchange, and rents a limited number of "seats" at the exchange, say 100ish seats, for some market-determined rate.

This company focuses on a few core competencies: being fast, being reliable, and being secure. Managing these three is a good bit easier with only 100 clients. The bulk of the AML, dealing with customers, etc. then rests on the shoulders of brokers, which are the companies renting the seats at the exchange, and trading on behalf of their customers.

Perhaps this is the niche Tradehill is aiming at.

[muyuu]

Decentralised exchanges already work, but they cannot play in the same league for trading speed.

You cannot decentralise bank transfers without friction. There is P2P trust and other considerations.

Slower trading as standard would probably benefit BTC at this point, but as long as there's faster and more efficient trading, it will cannibalise the market.

My prediction:
MtGox will improve their performance and for the time being things will continue "as usual". I don't think this is going to change any time soon.

[BTCLuke]

There is a vacuum to be filled, it will be filled, and it will be a lot better than MtGox. Fast forward 12 months and we will have completely different (and competing) players. Until then, enjoy the bumpy ride 

Of course new companies will attempt to step up, but the biggest problem is that everyone who wants to trade will head straight for the most popular one, because the most volume is there to trade with, and voila, today happening again is inevitable.

Without major action, this cycle will continue forever... And as long as it's happening, Bitcoin cannot be a Currency... Too volatile.

How about an MPEX-style solution? A well-capitalized company opens a premiere exchange, and rents a limited number of "seats" at the exchange, say 100ish seats, for some market-determined rate.

Hmm, I'm having trouble imagining MPEx with so much organization and money behind it. Could you elaborate? Where is the capitol coming from? How will it be organized? Where will the real-time trading come in? Etc...

Decentralised exchanges already work, but they cannot play in the same league for trading speed.

Could you point us to one that works?

There's got to be a way to add some real-time trading to the mix... Even if it's just some sort of a session that pops into existence for a region...

You cannot decentralise bank transfers without friction. There is P2P trust and other considerations.

I hear you, but if we just solve some of the problems we may be able to hybrid the system between other, existing exchanges and have no central point of failure... Worth a mental excercise, at least.

My prediction:
MtGox will improve their performance and for the time being things will continue "as usual". I don't think this is going to change any time soon.

Not without some real answers.

...And again, until they appear, bitcoin will never be a currency that the masses can accept.

[BitCoinLoft]

That's very pessimistic ... especially if you mention ***holes like btc-e.com as an alternative.

There is a vacuum to be filled, it will be filled, and it will be a lot better than MtGox. Fast forward 12 months and we will have completely different (and competing) players. Until then, enjoy the bumpy ride 

Bitcoin follows natural evolution. Didn't you expect shakes and turbulence? No need to panic. Bitcoin will finds its way. It may be bumpy, yes, but there's no other way. Going to Wall Street and stuff like that is not realisitc, nor desired for Bitcoin.

On the other hand, MTGOX can easily fix several problems in the short run. For example, limit trading API, introduce captchas, impose some limits on the trades so that people don't put hell lotta 0.001 trades and so on. I am sure there are working on these items.

But there is a good side of all this. Increase of awareness, PR, lot of noise. This was the price to be paid for making Bitcoin more popular.

[MarKusRomanus]

If Bitcoin trading quickly recovers from this fiasco and finds a new somewhat stable level for a while, it will have a positive impact on its media coverage.  The fact that it bounced back (if it does.. and it looks like it wants to be around $100 usd now) and continued on at a fairly high price will add more legitimacy to an argument that its just a currency battling to emerge on an infant trading system.

[BitCoinLoft]

If Bitcoin trading quickly recovers from this fiasco and finds a new somewhat stable level for a while, it will have a positive impact on its media coverage.  The fact that it bounced back (if it does.. and it looks like it wants to be around $100 usd now) and continued on at a fairly high price will add more legitimacy to an argument that its just a currency battling to emerge on an infant trading system.

Yes, you're absolutely correct. Currently, the biggest problem is the grade of trading platforms. None of the currently available trading floors are good. Bitcoin needs high grade trading platforms to become a viable currency. Yes, it's recovering pretty fast and let's hope this trend continues.

[BTCLuke]

New Exchange idea:

I've just thought through a new way to do exchanges themselves... Very decentralized:

This really should be done as another Peer-to-peer client, like bitcoin itself. You wanna trade? Download the client.

To run this client, your network connection would have to be fast enough, but I think speed is the only hardware requirement.

Anyone passing that test can install this client, which is somewhat like a bitcoin wallet. You run it, you can trade on it, but most importantly, if you are running it, then other people "near" you will have more connectivity to trade with those around them too.

Anyone, anywhere can log into the system, get routed to their local network of clients running ATM, and trade amongst them... The further away the person you are trading with, the longer the trade will take to execute; some trades will be fast, others will be slow, and real-time charts will not likely encompass the whole planet... Just those near enough to you to not lag the system too much.

It's different, I admit... But it's not as bad as what we have now, is it?

For the trouble of running the client, you also get a tiny take from the trades executed on your machine. (Fees go to your BTC address, but of course they'd be smaller than Mt.Gox's present take.)

Yes, It's a major program, but not as major as bitcoin itself... I don't even think this would need to SHA anything... And it has it's own built-in incentive for people to run it.

What do you guys think? Doable?

[muyuu]

Decentralised exchanges already work, but they cannot play in the same league for trading speed.

Could you point us to one that works?

There's got to be a way to add some real-time trading to the mix... Even if it's just some sort of a session that pops into existence for a region...

You cannot decentralise bank transfers without friction. There is P2P trust and other considerations.

I hear you, but if we just solve some of the problems we may be able to hybrid the system between other, existing exchanges and have no central point of failure... Worth a mental excercise, at least.

Last time I checked they were a work in progress but since I'm not a short term trader I haven't really needed them so far. I plan to get back there when I have more free time.

Check  Dark Exchange, p2px, bitcoinx. Dark Exchange is open source but I think it's kind of abandoned (lack of interest).

For these things to catch up I think you'd need traditional exchanges to completely stop working or be so unreliable people would need to hedge elsewhere.

Despite all the doom and gloom, this is not the case just yet.

My prediction:
MtGox will improve their performance and for the time being things will continue "as usual". I don't think this is going to change any time soon.

Not without some real answers.

...And again, until they appear, bitcoin will never be a currency that the masses can accept.

I'm ambivalent about it being a currency for the masses. I think an scenario where bitcoin is an alternative for a few million smart, conscious users is more than acceptable to me.

In any case it's definitely not a currency for the masses now and it definitely wasn't while it sky-rocketed vs the US$. You don't want a currency moving so fast in perceived value, regardless of the direction of this movement.

[empoweoqwj]

That's very pessimistic ... especially if you mention ***holes like btc-e.com as an alternative.

There is a vacuum to be filled, it will be filled, and it will be a lot better than MtGox. Fast forward 12 months and we will have completely different (and competing) players. Until then, enjoy the bumpy ride 

Bitcoin follows natural evolution. Didn't you expect shakes and turbulence? No need to panic. Bitcoin will finds its way. It may be bumpy, yes, but there's no other way. Going to Wall Street and stuff like that is not realisitc, nor desired for Bitcoin.

On the other hand, MTGOX can easily fix several problems in the short run. For example, limit trading API, introduce captchas, impose some limits on the trades so that people don't put hell lotta 0.001 trades and so on. I am sure there are working on these items.

But there is a good side of all this. Increase of awareness, PR, lot of noise. This was the price to be paid for making Bitcoin more popular.

Yes I did expect the bumps. The evolution was exactly my point. Bad exchanges / thieves will disappear and get replaced with better alternatives.

As for fixing problems at MtGox, you are assuming MtGox are competent. Everyone could see these problems multiplying every day for the last few weeks, so why didn't the management (presuming they have one) act before the cataclysm? The horse has bolted now.

The good side to me is:
1) There is nothing wrong with bitcoin itself
2) Everyone now understands one dominant trading platform, as with any virtual (sic) monopoly is a terrible idea
3) The complete lack of competence shown by MtGox can only hasten the competition to get their act together, and come to market

I really don't believe this idea that we are stuck with MtGox for good just because it currently high trading volume. Its a very amateur operation, as demonstrated by that press release they put out the day after the latest crash saying how "impressive" the stats were. Employ a PR person guys, don't have a geek write your fb press releases.

[BTCLuke]

Check  Dark Exchange, p2px, bitcoinx. Dark Exchange is open source but I think it's kind of abandoned (lack of interest).

Just looked into your suggestions, p2px is vaporware, Dark Exchange holds promise as a simple p2p version of #bitcoin-otc or bitcoinary... -But not intended to do real time trading.

Bitcoinx appears to be the whole enchilada, and it is an exchange for anything of value at all, but I don't know if it has real-time trading either. I'll talk to them about the specifics.

For these things to catch up I think you'd need traditional exchanges to completely stop working or be so unreliable people would need to hedge elsewhere.

Actually the way I've got it in mind, it can actually compete with the traditional exchanges, and additionally incentivize their own existence too, much like how bitcoin mining works.

I'm ambivalent about it being a currency for the masses. I think an scenario where bitcoin is an alternative for a few million smart, conscious users is more than acceptable to me.

I figure that bitcoin was designed to fall right inbetween legal tender and local scrips:



However that is a worldwide projection, so we're certainly not talking about a few nerds here... That's mostly everyone on the internet. Bitcoin will always be the most desirable money for people online.

It will never achieve this vision with such horrible volatility of course, so I'm really thinking hard about how to solve this weak link to the puzzle.

Thanks again!
-Luke

[muyuu]

Check  Dark Exchange, p2px, bitcoinx. Dark Exchange is open source but I think it's kind of abandoned (lack of interest).

Just looked into your suggestions, p2px is vaporware, Dark Exchange holds promise as a simple p2p version of #bitcoin-otc or bitcoinary... -But not intended to do real time trading.

Dark Exchange is the truly P2P way and it cannot be any other way without introducing banking in the system, which then makes it less P2P.

Bitcoinx appears to be the whole enchilada, and it is an exchange for anything of value at all, but I don't know if it has real-time trading either. I'll talk to them about the specifics.

For these things to catch up I think you'd need traditional exchanges to completely stop working or be so unreliable people would need to hedge elsewhere.

Actually the way I've got it in mind, it can actually compete with the traditional exchanges, and additionally incentivize their own existence too, much like how bitcoin mining works.

I commend your efforts, but there is simply no way P2P will be competitive vs a centralised + liquid exchange. People will arbitrage the hell out of it until it makes no sense for 90%+ of the people to use anything else than MtGox or whatever centralised exchange is there.

For it to truly catch up massively, you need all popular exchanges to fail and fail very badly.

However it will still be useful as a minor alternative.

[BTCLuke]

Dark Exchange is the truly P2P way and it cannot be any other way without introducing banking in the system, which then makes it less P2P.

Don't be too sure about that yet... I think I've got this part cracked.

Over in this thread, I've laid it out better, but not too completely yet.

It would be regionalized, based upon Pingtimes. Each client would log on, note all other clients within X milliseconds from them, and create a board (like BTC-E's main trade engine) for that region.

Every client would create it's own board, and only those within Xms would access it. When that client logs out, that board ceases to exist.

[muyuu]

Dark Exchange is the truly P2P way and it cannot be any other way without introducing banking in the system, which then makes it less P2P.

Don't be too sure about that yet... I think I've got this part cracked.

Over in this thread, I've laid it out better, but not too completely yet.

It would be regionalized, based upon Pingtimes. Each client would log on, note all other clients within X milliseconds from them, and create a board (like BTC-E's main trade engine) for that region.

Every client would create it's own board, and only those within Xms would access it. When that client logs out, that board ceases to exist.

Can't see that possibly being competitive. For starters, the lack of standard, curated and supported ways of transfering FIAT would massively hinder volume.

All these things are academically very interesting, but sadly simple beats complex most of the time. As long as "normal" exchanges continue to work, this isn't going to catch up.

I hope you prove me wrong.

[BTCLuke]

Can't see that possibly being competitive. For starters, the lack of standard, curated and supported ways of transfering FIAT would massively hinder volume.

Competition: What if every client could set their own desired fee? That's some serious Competition!

Lack of Standards: This would obviously need a core development team; I'd suggest modeling it exactly after bitcoin itself... In fact I'd love for the bitcoin developers to work on it closely.

Transferring Fiat: Volume would be whatever trades are within "quick enough" pingtimes. You're thinking about a central exchange here again... So once you use a broker to get your fiat up to your client in one form or another, the client just has to securely hold onto the credit there (I guess it would serve as your wallet too much like how mt.gox has a wallet for your bitcoins there) and trading is agreed between the buyer and the seller on a central client between them, not their own. (The client with the shortest pingtimes to both. This client's owner would get the fee.)

simple beats complex most of the time.

Because people were more stupid than they are being forced to become in this increasingly decentralized world.

P2P tech is the solution for every worthy problem known to mankind. World peace, economics, hunger, space research, every problem you can imagine is a problem because someone has power that is using it incorrectly. (A central point of failure.) Decentralizing takes power away from that point in every case and returns power to the masses on a distributed scale.

AnCaps like you and I will benefit from this trend the most, I'd think.

As long as "normal" exchanges continue to work, this isn't going to catch up.

Don't worry, I'm a free market kind of guy... I recognize that a newer, better exchange or three will come along to replace MtGox soon... And it's true that these exchanges will offer a trading environment with more volume than any decentralized solution could ever do...

...But there would be a strong set of reasons for the decentralized environment to exist, if not dominate. Most obvious among them:

1. Everyone is incentivized to run these clients in order to collect the fees. (Direct Income)

2. The rates on such a service cannot be affected by speculators nor DDos attacks, so it'll always be safer, and if nothing else a safehaven in times of stormy weather. (Such as when the centralized exchanges are being Ddossed again.)

[muyuu]

Lots of wishful thinking I'm afraid.

I will benefit from any advance in this direction but you are fooling yourself if you believe it will overtake traditional exchanges any time soon (or within our lifetimes). Unless traditional exchanges fail miserably, in which case we will just have these solutions but bitcoin adoption will remain very low.

[dutt]

Just hold on, old money (Jews) have taken notice of Bitcoin by now and we'll see a good legit exchange soon