http://www.newegg.com/Product/Product.aspx?Item=N82E16820461006
This is close. It will self-destruct on too many incorrect PIN entries. But you really want something with a self-destruct code. And, for a wallet, you want something that can sign Bitcoin transactions. Otherwise, you have to extract your key from it every time you use it, which would make it good only for one-time use.

Except you'd be scamming PayPal since they've explicitly made it clear that they do not agree to take this risk.

If the device received the destruct code, it would simply destroy the only copy of the key. The information then only exists in the same sense all the works of fiction man will ever create "exist" in the digits of Pi.

This requires secure hardware. Insecure hardware, or hardware of the attacker's choice, can always be "rewound" to a prior state.

The internal currency is like Bitcoin, the rest of the system (payment/credit/exchange) is not. The concept of "premined" doesn't make any sense in conjunction with a system that doesn't use proof of work.

Mining as a currency distribution system in currencies that use proof of work is necessary because otherwise there's no way to incentivize people to provide enough proof of work to keep the currency secure until transaction fees are sufficient to do that on their own. But otherwise, it has numerous disadvantages. For one thing, it consumes an enormous amount of resources for the sole purpose of making an attack impractical by requiring an attacker to exceed those resources. These resources constitute value created by the Bitcoin economy that are sucked out of it and go into the pockets of ASIC manufacturers and electric companies.

At one time, people had this idea (and I did too) that mining could be a democratic way to fairly distribute the value that crypto-currencies create. But I think we all now understand that this is economically impossible. If mining was more profitable than everything else people might do, then more people will mine, the difficulty will rise, and the profit will go away. It will always be, and necessarily so, that mining is just barely profitable if you have the best hardware and the best electric rates, because those people will increase mining until it's not profitable to keep increasing. All the money that goes into mining is value created by the crypto-currency that doesn't go into making it any better or helping it spread.

The genius of mining is that it uses the value a crypo-currency creates to provide the proof of work needed to keep the currency secure. It's a brilliant solution to a problem that Ripple doesn't have. As a way to provide the initial distribution of a currency, it's mediocre at best because it consumes much of the value the currency creates.

Right now, mining is costing the Bitcoin economy about $25 million per month. That's how much value Bitcoin, as a system, has to add just to hold the price even. Imagine if that money could go into making Bitcoin better rather than just keeping it secure.

Ripple is essentially three things in one. Ripple is a crypto-currency. Ripple is a social/community credit system. But the most important one for the short term is that Ripple is a payment system that allows people to create, hold, transfer, and exchange arbitrary assets.

No, that's not how it works. The network protection statement affected everyone's assessment of the value of their pre-orders and other people's pre-orders and it affected people's decision to buy them, sell them, request refunds, and so on. There may be people who have units in their hands today who made the decision to buy other people's pre-orders because of the network protection statement. Once you do something that change something's value, other people act based on that.

Everyone with a miner in hand made decisions knowing of the existence of that statement.

(This is a general principle. In this specific case, the network protection statement was weak and basically meaningless. I'm just pointing out that the timing argument is bogus.)

Why not just reduce the maximum to zero immediately? Ripple has no problem with a line that's over its limit. Nothing special happens except that the balance cannot increase, which seems to be exactly what you want.

A line that is at or over its limit can't be used to receive payments though. It seems odd to me that you would be willing to hold an asset but not willing to receive it in payment. If there's an actual use case for this, I can't figure out what it is. But you can get it by just reducing the limit to zero when you don't want to acquire more of the asset.

No harm done, since nobody has any gold issued by the goldsmith to give you. So you trust the goldsmith for gold he doesn't have, but because he doesn't have the gold, nobody can give it to you anyway. No problem.

Of course. You agreed to trust him. He can betray that trust, in which case you lose. That's how trust works.

Do you trust him or not? You are trying to create some case where you both trust him and don't trust him. It has to be one or the other.

This scenario doesn't make sense in the case of using a trust line to receive a payment. And if you just want a trust line so you can receive a balance from the issuer, you can drop the trust to zero immediately -- as soon as a balance is issued.

Arthur just explained why it can't. Responsible people don't default on their obligations just because other people default on obligations to them. They act as circuit breakers.

Every payment in Ripple is irreversible as well. And if you only use the internal currency in Ripple, you have that same simplicity. You're essentially arguing that giving people choices is bad because they might make bad choices that can hurt them. If you believe that, then irreversible payments are a bad thing.

The only way this can happen is if mining suddenly becomes less profitable than it is now through some mechanism other than a difficulty increase. Otherwise, anyone who finds mining profitable today will still find mining profitable and will continue mining. There are really only two ways this can happen. Either the price of Bitcoin drops or the price of power increases. Either way, if you're mining you're just as screwed as the other miners. In other words, any change that makes other people want to stop mining will also make you want to stop mining.

The reality is this simple -- if mining isn't profitable, you don't want to be mining. If mining is profitable, you still don't want to be mining because profitable mining means difficulty increases. The only time you want to be mining is if for some reason you can mine more efficiently than most other people. This could be because you have already bought mining equipment as a sunk cost or it could be because you have an unusually low price on power.

Otherwise, it's near certain to be outperformed by just buying Bitcoins. This is so for two reasons. First, buying Bitcoins has the same upside benefit and downside risk as mining -- you profit from an increase in price and suffer from a decrease just the same. Second, buying Bitcoins allows you to time your buying and selling more flexibly -- you can buy Bitcoins when the price drops and sell earlier if the price rises.

There is, however, one big upside to mining -- the "bigger fool" principle. You may be able to buy a miner, mine with it until it costs more to power than it can make, and then *still* find someone to buy it from you at a high price.

If it were profitable, everyone would be doing it until it wasn't profitable. It is not possible, so far as we know, to design a scheme such that mining is profitable for people who don't have optimized setups (most efficient hardware, lowest price on power, and so on).

If the output voltage is correct and the current is low, it's not the VRM's fault. Check your VRM's for any physical damage. Check to make sure the plastic cover was removed from your heat sinks before they were mounted to the ASIC. What baffles me about your numbers is that your output voltages are so high, and evenly so across the board. They should be around 0.75V or so.

Because the ASICs are built on similar technology to CPUs and GPUs, it's reasonable to assume the range of permissible temperatures is about the same. The leaves two unknowns:

1) What temperature is the miner reporting exactly? Is it die temperature? Heat spreader temperature? Board temperature?

2) Should we worry about the VRMs overheating even if the temperature reported by the miner is in the safe range?

This assumes that everyone who wishes to mine would have to buy a miner to do so. There are people for whom miners are a sunk cost -- for example, every one of KNC's customers.

You can compute the 24 hour return from the hash rate on any mining profitability calculator. I agree that seeing an estimate would be handy.

Yes, because it depends on the pool's luck.

Eligius already pays you the maximum amount that is fair as soon as possible, with the exception of avoiding very small payments because they just make spending expensive. You cannot make it faster because it is already as fast as it can be. Revenue from blocks found is paid out as soon as it is earned by the pool, typically at the precise same instant.

I like having a mining pool that's completely free, charges no fees, and works. If that means they have to ask for donations on every web page to keep the doors open, I'm fine with that.

Yes, the payout schedule is complicated. The complexity is needed to ensure the pool is fair to everyone. My advice -- don't worry about it. Just watch your hash rate to make sure your equipment is working. If the payouts were somehow cheating people, lots of other people who love to do the math and check them would be complaining quite loudly.

Two main reasons. The first is that increased demand for the currency makes the practical applications more expensive. The second is that decreased demand for the practical applications reduces the value of the currency. Or, to put it more generally, a change in the demand on one side (practical applications or means of exchange) affects the price on the other side.

These aren't necessarily a big deal and certainly aren't fatal flaws. Gold has this flaw and it was still quite successful as a currency historically. But there's no reason a crypto-currency has to have this flaw, so one would expect crypto-currencies that don't to be superior unless the tying of the currency to the practical application provides advantages that outweigh these disadvantages.

To be clear, I am *not* saying that Namecoin is broken because of this tying. I'm just saying that it's a technical difference between Namecoin and Bitcoin that could cause some rational people to prefer Bitcoin over Namecoin, even if they were both brand new and introduced today.

Mostly because it's extremely illogical to price goods in terms of domain registrations. The more practical applications something has natively, the less useful it is as a currency. NMC's name registration costs are fixed by the protocol.

I think it can be salvaged. A particular block, well in the future, will have to be chosen. After that block, every name will have to be recomputed to determine its correct ownership under the new, correct, rules. At that time, all names will revert to their correct owners, and from then on, the exploit will be impossible. It's just going to take the Namecoin community to come together, agree on the new rules and block number, get the work done on time, and so on.