Quantum computers don't just turn exponential time into linear time. There are certain problems where they're known to be able to do that, but there are a lot of problems where that's not known. Finding a suitable nonce is a lot like reverse phonebook search, and Grover's algorithm operates in order sqrt(N) guesses, instead of the classical N guesses. Assuming this is in fact the best way to use a quantum computer for mining, this has a curious effect. It means that if the difficulty quadrupled, it would take only twice as long to find a suitable nonce – which is effectively twice the hashpower, but still finding blocks slower. The effective hashpower of a quantum computer increases with increasing difficulty, but it still falls behind. The difficulty would still be able to increase to keep up with hashpower, so there's no existential threat to Bitcoin mining.

You should definitely only "trust" entities which you actually trust to honor their IOUs, but additionally trusting only one entity per currency is probably a good idea. Trusting more than one increases default risk substantially, because if either defaults, you probably lose all of your money.

XRP is not debt. Once Ripple leaves OpenCoin's control, they will not have the ability to change the number of XRP in the system. They may still own a large portion of the XRP, which will give them significant leverage on the price if they choose to use it, but this is very different from having an unlimited ability to print money. For example, you know at any moment what fraction of the XRP you control, and that fraction can never be diluted without your permission.

Simple, just base it on the logarithm of the price.

I've noticed something funny about the way bitcoin.clarkmoody.com colors the candles... it seems that the candle is colored not based on whether the open is greater or less than the close, but whether the close is greater or less than the previous close. Typically the open of one period will be close to the close of the previous period, but it can be different. There's an example of a strangely colored candle now: the 2013-5-22 22:15:00 (Chicago time, if it matters) candle for the M15 view is colored red even though the open is 124.40000 and the close is 124.40002. (The previous close is 124.43900.) Is this intentional behavior? It seems counterintuitive.

The only thing you exchange on Mt Gox is IOUs for USD and BTC. You have to trust that Mt Gox will let you withdraw your balance at the end of day. (The law might help you if Mt Gox defaults, but a gateway can attach legal meaning to its Ripple IOUs as well.) Ripple isn't replacing Bitcoin, and you shouldn't treat gateway IOUs as any more secure than fiat or bitcoin on an exchange. However, Ripple does potentially provide a very interesting replacement for the exchange component of the Bitcoin world.

As for whether XRP will compete with Bitcoin: I agree with the skeptics in this thread that the degree of control OpenCoin has is a good reason to shy away from using XRP as a currency and a store of value. Until such a point where OpenCoin's control has become negligible, XRP won't be as "free" a currency as bitcoin, and I think this is a compelling reason to not use it. This isn't really a criticism of Ripple, because using XRP as a currency isn't the point of Ripple. I have about 200 XRP total, and I'm going to sell all the ones I get in the giveaway immediately. Using or speculating in XRP is fine if you want to do it, but I don't. This doesn't make me less excited about Ripple.

You only need to trust them if you hold XRP. Who cares if the XRP market crashes when you're using Ripple as a distributed cryptocurrency exchange?

TradeFortress's Ripple scam is the "I'll give XYZ if you give me your private key" of the Ripple world. He's shamelessly duping people who don't understand Ripple. That's not a flaw in the underlying protocol, but it is a "social flaw" surrounding Ripple.

I seriously don't understand why people are paying so much for XRP, though. It's a system with a lot of promise and some flaws (including social flaws), but the market cap for XRP is *insane*. It's a totally unproven system. We don't even know if its fundamental innovation (the consensus system) really works in the real world.

Perfect. Couldn't have said it any better.

Both statements are more or less false. I assume the "almost no value" is so that it's easy for people to create new accounts and process transactions, and the "substantial value" is in order to act as DDOS prevention.

The various reserve requirements are adjustable, so they can remain low no matter what value XRP have. And the expected behavior of transaction fees is the following: typically they will have a very low XRP cost (transactions should be cheap, assuming the network can support it), but during a DDOS they will automatically increase in order to make the DDOS uneconomical. In order to continue the DDOS, they need to keep the network at maximum load as the fees compound – to do this for any extend period would require paying truly extraordinary fees. In the mean time, anybody wanting to use the network will just need to prove that their transactions are more important than the DDOSer's transactions by paying a higher fee than them on *one* transaction. (Rather than the many transaction fees per second the DDOSer is paying.)

What this means is that you will usually only need a very small amount of XRP to use Ripple (leading to a possibly low XRP price overall), but a determined attacker can pay an extraordinary cost in order to make the network temporarily more expensive to use. Remember that if a DDOSer makes it so that getting a transaction through costs a dollar, they're paying for many transactions per second at the same cost, or tens of thousands (hundreds of thousands?) of dollars per hour. If such an event happened over a long period of time, the price of XRP would probably spike as people realized they needed it to send transactions – this is where the apparent paradox of almost no value and substantial value is resolved. However, given the built in cost and lackluster effects of running a DDOS on Ripple, it's hard to imagine that anybody would actually try to do this, and the price of XRP would remain low. The mere presence of this variable XRP transaction cost acts as an effective anti-spam device.

Of course, this has been explained to you before, so I don't expect that you'll be convinced by this repetition of the argument...

The key here is that Ripple creates economic incentives for people to mediate this kind of transaction. Person C can hold WeExchange and Bitstamp IOUs, and let Ripple trade them for a small fee. Person A would only use WeExchange, Person B would only use Bitstamp. If Person A wanted to send USD to Person B, Ripple would give Person C their WeExchange IOUs and give Person C's Bitstamp IOUs to Person B. There's a market for Person C's fee, so it will tend to approximate the actual cost of moving money between gateways (in bulk, after netting.)

If nobody's providing this service between two gateways, there's a huge incentive to step in and make it possible, because you can set whatever fees you want. If the fees are too high, there's an incentive to step in and compete by trading IOUs with lower fees. It's more or less inevitable that if there are two gateways in use for a certain currency, a market for liquidity between their IOUs will spring up. Currently I understand these markets are rather sparse and illiquid, in part because the client doesn't have support for setting those fees yet, and in part because there's only one major gateway right now.

Ripple does a number of very nice things for Bitcoin. My money says that if Ripple takes off, it gives Bitcoin a significant boost. Why? First, it can in principle make bitcoins easier to buy, and limit the number of points of failure of this infrastructure by acting as a distributed exchange. If the social lending catches on (unlikely in the near term, imho) then you don't even need to depend on gateways in order to buy bitcoins, it would just facilitate trades through peer networks.

More importantly, it hopes to be a currency-indifferent payment system. This means that anybody who accepts any kind of Ripple payment, also accepts Bitcoins. It also means that if you are selling a product, if you accept Bitcoin payments through Ripple, people can pay you in USD and it will be converted on the fly. Basically, it's a distributed, omnidirectional BitPay-like service.

There are more subtle reasons why Ripple would be a boon for Bitcoin: one is that the economics of providing liquidity means that the cost of Ripple payments between gateways in the same currency is related to the underlying costs of moving that currency around. This means that even on Ripple, payments in BTC will tend to be cheaper than payments in USD, so if people use Ripple, they may want to start using bitcoins. (Or they may want to start using XRP for the same reason.)

All of this is assuming that Ripple delivers: that OpenCoin delivers on its promise to open source rippled, and that the consensus mechanism turns out to be a good and scalable and secure way to operate something like Ripple, among others.

The role XRP play in this is uncertain – it does compete more or less directly with Bitcoin, but people may or may not end up using it as a currency if Ripple takes off. It has advantages and disadvantages over Bitcoin as a currency – faster confirmations and cheaper transactions, but (more) limited privacy and a troubling relationship with OpenCoin. Also, everybody who accepts XRP also accepts Bitcoin (as mentioned above), but the reverse is not true.

The bid depth down to 89.99 is greater than the entire ask depth.

A bitcoin for only a few loaves of bread? Those are really cheap bitcoins!

Depth warning: <44k BTC to $1 trillion.

LAST CHANCE TO BUY UNDER THIRTEEN DIGITS!

Received, thanks!

Hm. I sent 4 XRP a bit ago and haven't gotten any response. The transaction (8ADBA83054C44031F823583FF13DA512CACE00DD1BD20AB21D590C3DE5EF55D6) isn't showing up on ripple.com/graph either.

I wonder if Bitstamp's involvement with Ripple has put them on more people's radar as well.

Very interesting, thanks for the detailed reply!

I don't see an obstacle to such an attacker, for example, zeroing their debt with all their creditors, or causing somebody to trust them and owe them money (although there's no reason to believe that person would pay that debt), or transferring arbitrarily large amounts of XRP from other accounts, or even creating XRP (although if they broke the 100 billion XRP cap, I don't know if validators would consider it valid). As far as I can tell, they can rewrite the ledger at will, so long as it appears valid afterwards.