Why Bitcoin crashed, and how Ripple might avoid the same fate

To the surprise of very few people, Bitcoin has crashed. That’s not to say it’s a goner, but even those who bought into it just before the craziness of the last two weeks are now looking at losses: from a U.S. dollar exchange rate high of $ 266 two days ago, the crypto-currency is – at the time of writing – trading at around $ 70.

What went wrong? Apart from being a bubble (albeit a bubble of a kind we’ve never quite seen before), it looks like Bitcoin fell victim to a single point of failure. But wait, you say, it’s a decentralized currency – how can that happen?

Facepalm, face palm That single point of failure is the most popular Bitcoin currency exchange, MtGox. There are other exchanges, but the bulk of Bitcoin trading happens there. MtGox claims to have been hit over the last couple weeks’ mania by the twin ills of denial-of-service attacks and sudden, excessive popularity, both of which amount to the same thing: MtGox’s systems falling over. The operation (which is based in Japan) has also shut down its own service at least once in an attempt to “cool down” the market.

And every time that has happened, a panic sell-off has been the result. That’s not surprising: MtGox’s status as the best-known exchange has led it to become the main data source for most of the Bitcoin rate visualizations out there, so when Mt.Gox goes down it affects visibility for a lot of people. And when people can’t see what’s going on, they panic, find another exchange and sell, sell, sell. Same goes for the biggest exchange unilaterally deciding to cool down the market – hardly a sign of viability.

(Some people have theorized about more sinister motives, too.)

As I write, MtGox is conducting an AMA session on Reddit, in which it is explaining what it’s doing to stem the problems:

“Upgrading computer systems means ordering more servers (two weeks timeframe), setting up (one day), load testing (two weeks) and deployment (one day). It’s a process that can take up to one month in total… We are now enforcing new rules for people placing large amounts of trades in order to reduce risks of lag.”

Hardly ideal. But what – apart from boring old state-issued currency – is the alternative?

Introducing Ripple

The ideal alternative for Bitcoin as an ecosystem is to try to even the load between different exchanges, or at least settle on one that doesn’t fall over when people get keen on the currency. However, there is also an emerging rival of sorts called Ripple (not to be confused with the charitable donation tool of the same name).

Ripple Although its use is also pseudonymous, Ripple isn’t quite the grassroots effort that Bitcoin is: on Thursday its sponsor, OpenCoin, picked up a round of funding from Andreessen Horowitz, FF Angel IV, Lightspeed Venture Partners, Vast Ventures and Bitcoin Opportunity Fund, “an investment vehicle for Bitcoins and Bitcoin-related companies.” OpenCoin’s development chief is Jed McCaleb, the guy who founded MtGox in 2010.

However, while it doesn’t have the same Stick-It-To-The-Man vibe as Bitcoin does, Ripple does have a few advantages over its better-known rival. Chief among those is the fact that it doesn’t need currency exchanges: in fact, it is its own distributed currency exchange.

Ripple can be used to convert dollars into rupees, or for that matter Bitcoins, and send them across the world for the nominal fee of one “ripple” – this fee is only charged to stop people from swamping the system with millions of transactions. OpenCoin says a ripple is worth around a thousandth of a cent, and the company will put 100 billion of them into circulation — three quarters of which it will give away and a quarter of which it will keep for itself, in the hope that the value goes up. No more ripples will ever be created.

So, ripples are both in-service tokens for the mechanism of sending and exchanging real money, and a virtual currency in their own right. Of course, Ripple users will need to get their hard cash into the system somehow, so the system employs what it calls “gateways.” Anyone will be able to act as a gateway, even individuals and convenience stores – although online services will probably be the most convenient.

A faster self-regulating network

Ripple is a bit like Bitcoin, in that the network verifies transactions – this is essential if you’re removing the “trusted third party” role that banks usually fill, because someone needs to ensure that people aren’t double-spending their virtual money. However, there’s a big difference in how this happens.

hand shake With Bitcoin, nodes on the network called “miners” compete with each other to verify each block of transactions every 10 minutes. To verify a block, the miner has to complete a complex computational puzzle faster than its rivals do. In return for the electricity spent in doing so, the miner gets a certain number of freshly minted Bitcoins – this is how the system keeps working, and how new Bitcoins get brought into the system.

With Ripple, the nodes on the network also maintain a shared ledger – the equivalent of Bitcoin’s blockchain – but they don’t compete with each other to do so. Instead, the system uses a complex consensus mechanism to make sure transactions get verified and added to the ledger.

As this process has nothing to do with mining the virtual currency, there is no need to control the timing of the verification, meaning transactions can happen within seconds rather than in 10 minutes or more it takes with Bitcoin. This is clearly a big advantage, and there are others, such as the ability to create a chain of IOUs, either through people they personally know and trust, or by using ripples.

But…

Thus Ripple solves some of Bitcoin’s problems: transactions can take place more quickly, there’s no need for shaky third-party exchanges, and the whole shebang doesn’t need to waste a bunch of electricity on solving computational problems. However, I suspect Ripple will have its own problems.

The first is to do with money-laundering. This is also a big potential problem for Bitcoin – although good luck to anyone who tried that this week – but I’m a bit confused about how well-controlled these “gateways” will be. Ripple’s own explainer states that “your neighbor or corner grocery could be a gateway,” but OpenCoin told me that “we believe these gateways should be properly licensed and regulated in the same way as other financial institutions,” Does. Not. Compute.

The second problem is OpenCoin’s role in Ripple. The company maintains that it’s “just here to pay to develop and promote the network,” and doesn’t control Ripple, but at the same time it’s a for-profit company (hence this week’s investment) that has a vested interest in seeing the value of ripples increase. At the very least, there may be an inherent problem of perception here, particularly for those subscribing to Bitcoin’s core ethos.

Ripple will also need to find enough “validating nodes” to ensure an above-board network consensus process. Unlike Bitcoin’s miners, these nodes don’t get anything for their efforts other than seeing the system stay up and running. Granted, they also don’t need to expend as much electricity in doing their job, but their participation is still not a sure thing.

Finally – and perhaps most importantly – Ripple is really hard to understand, compared with traditional “fiat” currency. A lot of pieces need to be in place for it to work, and a lot of education needs to take place too. I would even go so far as to say that Ripple is more complex (even on a conceptual level) than Bitcoin, and that’s saying something.

Still, Ripple is interesting, and perhaps having an official sponsor will make it more viable than Bitcoin. Whatever happens, it’s another step towards the post-experimental use of digital currencies.