The taxpayer *should* cover the costs, since the taxpayers created this out-of-control system of government-granted entitlements in the first place, and they keep reelecting politicians who protect it.
Of those taxpayers, many may well have opposed both the entitlements and the politicians responsible; they shouldn't be punished simply for being in the minority. Responsibility for the consequences lies exclusively with those whose actions contributed to the damage.
This ruling is a smackdown from the Supreme Court saying "no, you six states cannot get a judge to rewrite environmental policy for you."
It is well within the domain of the judiciary to award compensation for damages from any source, including pollution. That judicial authority is not something Congress can overrule, short of a Constitutional amendment. It does not equate to "legislating from the bench"; if anything, by claiming that following EPA regulations renders companies immune to civil suits regarding pollution, Congress is usurping the role which properly belongs to the courts.
Of course, to win such a case the states would need to show that they have actually taken damage due to these particular utilities' contributions to climate change. Needless to say, that would be a very difficult thing to prove, even under the lower standards of civil proceedings—but if they can prove it, Congress and the EPA have no business interfering.
Re:BitCoins are simply a hobby, not a currency
on
Bitcoin Price Crashes
·
· Score: 1
It's so important it's printed on the dollar bill: "this bill is legal tender for all debts, public and private". The first rule of money is, you must be able to use it to buy *anything*.
Unfortunately for your argument, that's not what that phrase means. There is no guarantee that you can buy any particular thing with Federal Reserve Notes. In the case of simple, immediate exchange there is no debt incurred, so the legal-tender part doesn't apply; merchants are free to require payment in precious metals, bitcoins, tulips, etc. However, if you owe someone money, they are legally obligated to accept payment in dollars, even if you voluntarily agreed to another medium of exchange. Obviously this is a significant injustice, and renders you a dishonorable oath-breaker, but apparently the law considers the supremacy of the dollar more important than ensuring that trade remains honorable and civilized.
The size of the network is such that even a decent-size botnet doesn't pose much of a threat. Even assuming you could put together a botnet with 50+% of the computing capacity of the network, however, there is only so much you could do with it. For example, you still couldn't spend other people's coins—for that you would need the private keys for their wallets. The main vulnerabilities are (a) you could prevent legitimate transactions from entering the block chain, a form of denial-of-service attack, and (b) you could nullify previous (recent) transactions.
The "exchanges" all seem to transfer funds in and out through even flakier services, like Liberty Reserve... and Dwolla.... Neither is registered as a money transfer agency.
Dwolla is a nationally registered payment processor. It may be small, but it's a perfectly legitimate operation. As an off-shore organization, Liberty Reserve may appear a bit shadier to some (though I haven't seen any real complaints), but nothing compels you to maintain an account there.
a crash in Bitcoins
June 1: ~$10/BTC. Now: ~$19/BTC. Where is this "crash" you speak of? Yes, there was a temporary spike to $30, followed by a slight overcorrection, but the price only dropped below the 30-day average for a few hours, total. Nothing significant unless you happen to be prone to panic—or trying to incite one.
The delay is, indeed, a problem for MtGox to solve, with cooperation from Liberty Reserve—there is nothing the MtGox can do about LR transfer policies on their own. However, you were accusing them of delaying the transfers to further a scam, which is a very serious statement for which you have offered no compelling evidence. Moreover, there are already other means of withdrawing money from MtGox accounts which do not involve Liberty Reserve or their API restrictions; in fact, the preferred withdrawal system is Dwolla, not Liberty Reserve, and direct deposit is also available to EU accounts in the SEPA zone. It is unreasonable to blame MtGox for forcing users to employ Liberty Reserve, with its attendant limitations, when there are other options available.
Dwolla is definitely associated with Foundry Co-Working, which provides mail services for a number of small businesses. As an all-electronic operation they would have little need for a unique office address.
Also, on one forum, there's the comment from a Mt. Gox staffer (?) "If we have a lot of LR activity (like, about now), withdraws will be put on hold and executed later (ie. the next day) in the order they were received." That just screams "Ponzi scheme". They're an exchange; if they're honest, they should never have a cash flow problem.
From the forum thread you cited, the reason for the delay appears to be daily API transfer limits imposed by Liberty Reserve, not cash-flow problems on the part of MtGox.
Thankfully, privacy rights don't require property rights, or else there would be no need to warrants to tap phones.
Sure there would be. To tap your phone directly (without your permission) would violate your property rights, obviously, but tapping a line at the phone company without their permission would just as obviously be a violation of their property rights. If you're smart, your service contract with them will ensure that they can't grant anyone permission to listen in to your conversations voluntarily, in which case the police would still require a warrant. Alternatively, you could simply take steps to ensure your own privacy, such as encrypting your communications.
The only reason to create fake transactions is to keep this pipeline full. If it's already full of real transactions then the bot miners will mine those.
To what end? You don't need fake transactions to mine bitcoins. A block with no transactions apart from the 50 BTC reward is perfectly valid.
But the proceeds go to the bot masters not the people paying for the computing power. It does no good to dismiss this with a wave saying tough luck to those who got themselves rooted.
I'm not simply dismissing the problem, but it's hardly specific to bitcoins. Any profitable and CPU-limited task is a candidate for botnet activity. If those bots couldn't mine bitcoins they would spend their illicit CPU time on other problems. Compared to most of the alternatives, mining bitcoins is positively benign.
Your ability to disrupt the network scales exponentially with the amount of computing power and BTC you have.
I won't argue with that—it's a very high barrier, but it is well-known if you can come up with enough computing power (a majority of the network) you can cause significant problems for other users. You still can't spend other people's coins, but you can trivially delay their transactions and/or attempt to enlarge the database. (Note that it isn't actually necessary for every client to store a copy of the whole database, so long as all the blocks remain available on request, so the database size isn't really a problem in the long term.)
However, you should keep in mind that no one else is compelled to accept your new blocks, and enforcing a smaller block size limit (for all blocks or just new ones) is a matter of changing a few lines of code in the client. For that matter, the community could even disable your existing accounts by refusing to recognize transfers from them as valid, or introduce a web-of-trust system for miners as an additional requirement for block generation. The protocol is not fixed in stone; there are plenty of ways to deal with deliberate disruptive influences should they ever become a problem.
You're right, of course. I suppose I should have read the entire comment before responding.:) Different block criteria would indeed create permanently divergent block chains.
So basically, I can get a smallish botnet of 10000 nodes or so, buy/mine a ~1000 BTC, then proceed to make as many transactions between the nodes in my botnet as I can. If I keep at it long enough the size of the database will become so large that noone could possibly store all that data conveniently - and you need to store all transactions ever made to verify transactions - rendering the whole system useless.
The rate at which blocks are generated is fixed at about one block every ten minutes, and there is an upper limit to the size of each block. Any transactions which don't fit into the current block are left for the next one, with priority being assigned to transactions which offer the block generator a fee. Ergo, the size of the block database would increase at a predictable rate regardless of attempts to flood the system with small transactions.
And then I haven't even mentioned the insanity that the whole system relies on the fact that there are more honest nodes than dishonest - so just get more dishonest nodes than honest and you can get all bitcoins in existence.
Not true. The specific property which relies on most nodes being honest is simply the protection against double-spending. If you could guarantee the ability to outpace the rest of the network and create a longer block chain then you could spend some of your coins, invalidate those transactions, and spend them again. (Until your reputation gets around, anyway.) You can't take control over anyone else's coins, since you don't have the private keys for those wallets.
What exactly would happen in the long term of this situation is an open question. Both sides of the fork could continue indefinitely as independent currencies or one of them could die out as its userbase dropped in favour of the other.
That's only true if the two forks remain exactly the same length(*). The official bitcoin client software resolves forks by adopting the longer one, so which ever side ended up with the most computing power would eventually be accepted as the official version. At that point anyone who submitted transactions in the alternate fork could re-submit them to the winning block chain. (The risk is that transactions might not be resubmitted, which would be bad for anyone who accepted them while following the other chain.)
(*) Length is determined by cumulative difficulty, not simply the number of blocks, so you can't disrupt the system by running a private network and generating a large number of fake blocks at low difficulty.
Even if CPU mining in JavaScript isn't efficient enough to actually be cost-effective, it could prove to be an interesting and unobtrusive alternative to embedded advertising as a way of supporting web sites. The upcoming addition of WebGL and OpenCL via JavaScript should help as well.
There seem to be some misunderstandings here about how bitcoins work:
There are two problems with the highspeed wash transactions. First it generates large numbers of events that have to be propagated across the whole network. (bitcoin requires all nodes know about every transaction).
That isn't strictly true. Every full client does need to know about every transaction which has been accepted and integrated into the block chain. (There is a relaxed form of verification which can work without the full block chain, but at least some of the clients need to have the full list.) However, only the miners really need to know about unconfirmed transactions, and priority for both confirmation and propagation is given to transactions which offer a fee—which these high-speed transactions can't do without breaking the bank. Ergo, most of these high-speed transactions would probably be ignored.
Second, every transaction causes a work event that generates new bitcoins.
Blocks, which award bitcoins to the miners which generate them, are created about every ten minutes. This interval is regulated by periodically adjusting the difficulty threshold by consensus, and is independent of the number of pending transactions. When a block is generated, the miner that finds it decides which transactions it includes; if there is insufficient room for all the known transactions, they will typically be prioritized based on the fee offered (which goes to the miner). Real transactions will thus crowd out the "fake" high-speed ones which offer no fee.
...but the harm is that the cost of processing the transaction is borne by the owner of the computer not the botmaster.
Well, true, but the answer to that is to secure your own computer. Would you prefer that the botnet's CPU time be spent cracking people's bank account passwords instead? At least processing transactions is a useful and perfectly legitimate service, even if the cost is externalized.
In a free-banking system, without a central bank or an equivalent of the FDIC, reserve requirements are dictated by the risk of a run on the bank and competition between banks. No bank can easily afford to be proven bankrupt, unable to meet its immediate obligations to its account holders; moreover, a bank which operates with excessively low reserves runs the risk of being deliberately bankrupted by its competitors.
Far from increasing the reserve requirements, typical reserve levels were higher before the Federal Reserve and the FDIC were created to spread out the risk.
This is orthogonal to the bitcoin system, which has the advantage of not needing any banks just to hold bitcoins securely or transfer them over the Internet. If someone were to put their money in a bitcoin bank—for the interest, presumably—I should hope they would recognize that their interest-bearing account is an investment, and carries a corresponding investment risk, as with any other interest-bearing loan.
Sure, these sorts of things have a limit, somewhere.
The limit is that speculative hoarding only makes sense if you anticipate someone actually buying the commodity in the future. If that should ever become a concern, speculators will attempt to sell while they can rather than holding on to their assets, which would increase the supply and drive down the price. The same mechanics which limit the prices of regular commodities subject to speculation also apply to bitcoins.
In short, present speculation is both driven and limited by estimated future (non-speculative) demand. It's not an unbounded feedback loop.
Adding a small transaction fee, which goes to the miner who confirms the transaction by adding it to the block chain, generally grants the transaction priority over all transactions without fees (or with lower fees for the same amount of transaction data). Paid transactions are also more likely to propagate through the network, which again makes them more likely to be confirmed quickly. The normal fee is 0.01 BTC, although there is a recent proposal to reduce the propagation threshold in the official client to reflect the recent upsurge in value.
That's not the total number of transactions, just the number which have not yet been confirmed (integrated into the block chain). Note that all of the transactions which include an offer of a transaction fee (typically 0.01 BTC) to the miner which incorporates the transaction into a block occurred within the last 20 minutes (one block ago, on average), and all but one of the remainder occurred within the last hour.
Over the past 30 days, over 4.2 million USD and 1,066,094.21 BTC were exchanged via https://mtgox.com/, one of the main trading sites. While still small compared to the currency exchange market as a whole, that's hardly reminiscent of anything you'll find in a high school.
There's plenty of profit in peace. It's just spread throughout society rather than focused on governments and military contractors.
This is one of democracy's (or perhaps society's) greatest flaws: a small risk or opportunity to each of millions of individuals will almost always lose out to a large risk or opportunity to a small number of individuals, even when the former is much more significant in the aggregate. Few would get worked up over personally losing $1, but a half-dozen well-connected lobbyists will make a major effort to secure $25 million for themselves by taking $1 each from 300 million others (even though half of the total is wasted). Even ignoring the cost of redistribution, society loses 150 million dollars' worth of productivity in the process—but, unlike the prospective recipients, no one has a personal stake in opposing the measure.
In more concrete terms, it's no surprise that the broad benefits which peace brings to an entire society lose out to the much narrower profitability of warmongering to military contractors, military employees, and certain politicians.
BitcoinExchange Services (forum thread; no independent website for now) offers various USD-denominated store-brand gift cards, pre-paid debit cards, and VISA gift cards in exchange for bitcoins (at a small premium). It looks like he/she/they may also accept bitcoins in exchange for USD via PayPal, although they stopped selling them via PayPal due to fear of frozen accounts. I haven't used this service myself, but the thread contains many positive reviews.
It's what's known in economic circles as "consuming capital". Yes, we can afford these things now, but only by sacrificing our ability to produce as much of these goods in the future for a given amount of labor and/or raw material.
Capital (tools—the means of production) enormously increase our ability to produce goods compared to gathering and producing goods by hand. However, those means of production wear out over time, and need to be maintained and/or replaced if that increase in productive capacity is to continue. If fewer resources are allocated toward capital investment than are required for basic maintenance, there is more available for immediate consumption, but the capacity for production of new goods is impaired.
Consumption is an essential part of economic life; after all, the entire point in producing capital or consumer goods is to use them. However, the balance between investment and consumption must be considered if the standard of living is to be maintained, or possibly even increased. Over-consumption in the present decreases the future standard of living, both at a personal level and for an entire society. At the moment there is quite a bit of both.
Open the gates and let bogus, contradictory records spread across the network.
The bogus records wouldn't be accepted. In the event a client receives blocks from contradictory branches of the block chain, the chain with the greatest length (cumulative difficulty) wins. Any shorter chains will be rejected. That means you would need to outpace the entire rest of the network for quite a while to effect any significant changes.
You could, of course, do as you described and trick a local group of clients into accepting your modified history by cutting them off from the outside world for a time. However, those transactions would not infect the rest of the network when the isolation ended. The clients would revert to the longer block chain and resubmit any (valid) transactions processed during the isolation period.
Wages could fall much faster than the prices of things like food.
They could, but they won't. Historically, wages have always trailed goods. In an inflationary economy, prices of goods rise faster than wages (decreasing purchasing power). In a deflationary economy, prices of goods fall faster than wages (increasing purchasing power).
Preventing the sort of meddling you just described, which only serves to propagate the misallocation of resources which led to the problem in the first place, is a deliberate design feature, not a flaw.
Slashdot Mirror
The taxpayer *should* cover the costs, since the taxpayers created this out-of-control system of government-granted entitlements in the first place, and they keep reelecting politicians who protect it.
Of those taxpayers, many may well have opposed both the entitlements and the politicians responsible; they shouldn't be punished simply for being in the minority. Responsibility for the consequences lies exclusively with those whose actions contributed to the damage.
This ruling is a smackdown from the Supreme Court saying "no, you six states cannot get a judge to rewrite environmental policy for you."
It is well within the domain of the judiciary to award compensation for damages from any source, including pollution. That judicial authority is not something Congress can overrule, short of a Constitutional amendment. It does not equate to "legislating from the bench"; if anything, by claiming that following EPA regulations renders companies immune to civil suits regarding pollution, Congress is usurping the role which properly belongs to the courts.
Of course, to win such a case the states would need to show that they have actually taken damage due to these particular utilities' contributions to climate change. Needless to say, that would be a very difficult thing to prove, even under the lower standards of civil proceedings—but if they can prove it, Congress and the EPA have no business interfering.
It's so important it's printed on the dollar bill: "this bill is legal tender for all debts, public and private". The first rule of money is, you must be able to use it to buy *anything*.
Unfortunately for your argument, that's not what that phrase means. There is no guarantee that you can buy any particular thing with Federal Reserve Notes. In the case of simple, immediate exchange there is no debt incurred, so the legal-tender part doesn't apply; merchants are free to require payment in precious metals, bitcoins, tulips, etc. However, if you owe someone money, they are legally obligated to accept payment in dollars, even if you voluntarily agreed to another medium of exchange. Obviously this is a significant injustice, and renders you a dishonorable oath-breaker, but apparently the law considers the supremacy of the dollar more important than ensuring that trade remains honorable and civilized.
The size of the network is such that even a decent-size botnet doesn't pose much of a threat. Even assuming you could put together a botnet with 50+% of the computing capacity of the network, however, there is only so much you could do with it. For example, you still couldn't spend other people's coins—for that you would need the private keys for their wallets. The main vulnerabilities are (a) you could prevent legitimate transactions from entering the block chain, a form of denial-of-service attack, and (b) you could nullify previous (recent) transactions.
The "exchanges" all seem to transfer funds in and out through even flakier services, like Liberty Reserve... and Dwolla.... Neither is registered as a money transfer agency.
Dwolla is a nationally registered payment processor. It may be small, but it's a perfectly legitimate operation. As an off-shore organization, Liberty Reserve may appear a bit shadier to some (though I haven't seen any real complaints), but nothing compels you to maintain an account there.
a crash in Bitcoins
June 1: ~$10/BTC. Now: ~$19/BTC. Where is this "crash" you speak of? Yes, there was a temporary spike to $30, followed by a slight overcorrection, but the price only dropped below the 30-day average for a few hours, total. Nothing significant unless you happen to be prone to panic—or trying to incite one.
The delay is, indeed, a problem for MtGox to solve, with cooperation from Liberty Reserve—there is nothing the MtGox can do about LR transfer policies on their own. However, you were accusing them of delaying the transfers to further a scam, which is a very serious statement for which you have offered no compelling evidence. Moreover, there are already other means of withdrawing money from MtGox accounts which do not involve Liberty Reserve or their API restrictions; in fact, the preferred withdrawal system is Dwolla, not Liberty Reserve, and direct deposit is also available to EU accounts in the SEPA zone. It is unreasonable to blame MtGox for forcing users to employ Liberty Reserve, with its attendant limitations, when there are other options available.
Dwolla is definitely associated with Foundry Co-Working, which provides mail services for a number of small businesses. As an all-electronic operation they would have little need for a unique office address.
Dwolla is a nationally registered payment processor, with backing from The Members Group, a subsidiary of the Iowa Credit Union League. This isn't some shady fly-by-night operation.
Also, on one forum, there's the comment from a Mt. Gox staffer (?) "If we have a lot of LR activity (like, about now), withdraws will be put on hold and executed later (ie. the next day) in the order they were received." That just screams "Ponzi scheme". They're an exchange; if they're honest, they should never have a cash flow problem.
From the forum thread you cited, the reason for the delay appears to be daily API transfer limits imposed by Liberty Reserve, not cash-flow problems on the part of MtGox.
Thankfully, privacy rights don't require property rights, or else there would be no need to warrants to tap phones.
Sure there would be. To tap your phone directly (without your permission) would violate your property rights, obviously, but tapping a line at the phone company without their permission would just as obviously be a violation of their property rights. If you're smart, your service contract with them will ensure that they can't grant anyone permission to listen in to your conversations voluntarily, in which case the police would still require a warrant. Alternatively, you could simply take steps to ensure your own privacy, such as encrypting your communications.
The only reason to create fake transactions is to keep this pipeline full. If it's already full of real transactions then the bot miners will mine those.
To what end? You don't need fake transactions to mine bitcoins. A block with no transactions apart from the 50 BTC reward is perfectly valid.
But the proceeds go to the bot masters not the people paying for the computing power. It does no good to dismiss this with a wave saying tough luck to those who got themselves rooted.
I'm not simply dismissing the problem, but it's hardly specific to bitcoins. Any profitable and CPU-limited task is a candidate for botnet activity. If those bots couldn't mine bitcoins they would spend their illicit CPU time on other problems. Compared to most of the alternatives, mining bitcoins is positively benign.
Your ability to disrupt the network scales exponentially with the amount of computing power and BTC you have.
I won't argue with that—it's a very high barrier, but it is well-known if you can come up with enough computing power (a majority of the network) you can cause significant problems for other users. You still can't spend other people's coins, but you can trivially delay their transactions and/or attempt to enlarge the database. (Note that it isn't actually necessary for every client to store a copy of the whole database, so long as all the blocks remain available on request, so the database size isn't really a problem in the long term.)
However, you should keep in mind that no one else is compelled to accept your new blocks, and enforcing a smaller block size limit (for all blocks or just new ones) is a matter of changing a few lines of code in the client. For that matter, the community could even disable your existing accounts by refusing to recognize transfers from them as valid, or introduce a web-of-trust system for miners as an additional requirement for block generation. The protocol is not fixed in stone; there are plenty of ways to deal with deliberate disruptive influences should they ever become a problem.
You're right, of course. I suppose I should have read the entire comment before responding. :) Different block criteria would indeed create permanently divergent block chains.
So basically, I can get a smallish botnet of 10000 nodes or so, buy/mine a ~1000 BTC, then proceed to make as many transactions between the nodes in my botnet as I can. If I keep at it long enough the size of the database will become so large that noone could possibly store all that data conveniently - and you need to store all transactions ever made to verify transactions - rendering the whole system useless.
The rate at which blocks are generated is fixed at about one block every ten minutes, and there is an upper limit to the size of each block. Any transactions which don't fit into the current block are left for the next one, with priority being assigned to transactions which offer the block generator a fee. Ergo, the size of the block database would increase at a predictable rate regardless of attempts to flood the system with small transactions.
And then I haven't even mentioned the insanity that the whole system relies on the fact that there are more honest nodes than dishonest - so just get more dishonest nodes than honest and you can get all bitcoins in existence.
Not true. The specific property which relies on most nodes being honest is simply the protection against double-spending. If you could guarantee the ability to outpace the rest of the network and create a longer block chain then you could spend some of your coins, invalidate those transactions, and spend them again. (Until your reputation gets around, anyway.) You can't take control over anyone else's coins, since you don't have the private keys for those wallets.
What exactly would happen in the long term of this situation is an open question. Both sides of the fork could continue indefinitely as independent currencies or one of them could die out as its userbase dropped in favour of the other.
That's only true if the two forks remain exactly the same length(*). The official bitcoin client software resolves forks by adopting the longer one, so which ever side ended up with the most computing power would eventually be accepted as the official version. At that point anyone who submitted transactions in the alternate fork could re-submit them to the winning block chain. (The risk is that transactions might not be resubmitted, which would be bad for anyone who accepted them while following the other chain.)
(*) Length is determined by cumulative difficulty, not simply the number of blocks, so you can't disrupt the system by running a private network and generating a large number of fake blocks at low difficulty.
Even if CPU mining in JavaScript isn't efficient enough to actually be cost-effective, it could prove to be an interesting and unobtrusive alternative to embedded advertising as a way of supporting web sites. The upcoming addition of WebGL and OpenCL via JavaScript should help as well.
There seem to be some misunderstandings here about how bitcoins work:
There are two problems with the highspeed wash transactions. First it generates large numbers of events that have to be propagated across the whole network. (bitcoin requires all nodes know about every transaction).
That isn't strictly true. Every full client does need to know about every transaction which has been accepted and integrated into the block chain. (There is a relaxed form of verification which can work without the full block chain, but at least some of the clients need to have the full list.) However, only the miners really need to know about unconfirmed transactions, and priority for both confirmation and propagation is given to transactions which offer a fee—which these high-speed transactions can't do without breaking the bank. Ergo, most of these high-speed transactions would probably be ignored.
Second, every transaction causes a work event that generates new bitcoins.
Blocks, which award bitcoins to the miners which generate them, are created about every ten minutes. This interval is regulated by periodically adjusting the difficulty threshold by consensus, and is independent of the number of pending transactions. When a block is generated, the miner that finds it decides which transactions it includes; if there is insufficient room for all the known transactions, they will typically be prioritized based on the fee offered (which goes to the miner). Real transactions will thus crowd out the "fake" high-speed ones which offer no fee.
...but the harm is that the cost of processing the transaction is borne by the owner of the computer not the botmaster.
Well, true, but the answer to that is to secure your own computer. Would you prefer that the botnet's CPU time be spent cracking people's bank account passwords instead? At least processing transactions is a useful and perfectly legitimate service, even if the cost is externalized.
In a free-banking system, without a central bank or an equivalent of the FDIC, reserve requirements are dictated by the risk of a run on the bank and competition between banks. No bank can easily afford to be proven bankrupt, unable to meet its immediate obligations to its account holders; moreover, a bank which operates with excessively low reserves runs the risk of being deliberately bankrupted by its competitors.
Far from increasing the reserve requirements, typical reserve levels were higher before the Federal Reserve and the FDIC were created to spread out the risk.
This is orthogonal to the bitcoin system, which has the advantage of not needing any banks just to hold bitcoins securely or transfer them over the Internet. If someone were to put their money in a bitcoin bank—for the interest, presumably—I should hope they would recognize that their interest-bearing account is an investment, and carries a corresponding investment risk, as with any other interest-bearing loan.
Sure, these sorts of things have a limit, somewhere.
The limit is that speculative hoarding only makes sense if you anticipate someone actually buying the commodity in the future. If that should ever become a concern, speculators will attempt to sell while they can rather than holding on to their assets, which would increase the supply and drive down the price. The same mechanics which limit the prices of regular commodities subject to speculation also apply to bitcoins.
In short, present speculation is both driven and limited by estimated future (non-speculative) demand. It's not an unbounded feedback loop.
Adding a small transaction fee, which goes to the miner who confirms the transaction by adding it to the block chain, generally grants the transaction priority over all transactions without fees (or with lower fees for the same amount of transaction data). Paid transactions are also more likely to propagate through the network, which again makes them more likely to be confirmed quickly. The normal fee is 0.01 BTC, although there is a recent proposal to reduce the propagation threshold in the official client to reflect the recent upsurge in value.
That's not the total number of transactions, just the number which have not yet been confirmed (integrated into the block chain). Note that all of the transactions which include an offer of a transaction fee (typically 0.01 BTC) to the miner which incorporates the transaction into a block occurred within the last 20 minutes (one block ago, on average), and all but one of the remainder occurred within the last hour.
Over the past 30 days, over 4.2 million USD and 1,066,094.21 BTC were exchanged via https://mtgox.com/, one of the main trading sites. While still small compared to the currency exchange market as a whole, that's hardly reminiscent of anything you'll find in a high school.
There's plenty of profit in peace. It's just spread throughout society rather than focused on governments and military contractors.
This is one of democracy's (or perhaps society's) greatest flaws: a small risk or opportunity to each of millions of individuals will almost always lose out to a large risk or opportunity to a small number of individuals, even when the former is much more significant in the aggregate. Few would get worked up over personally losing $1, but a half-dozen well-connected lobbyists will make a major effort to secure $25 million for themselves by taking $1 each from 300 million others (even though half of the total is wasted). Even ignoring the cost of redistribution, society loses 150 million dollars' worth of productivity in the process—but, unlike the prospective recipients, no one has a personal stake in opposing the measure.
In more concrete terms, it's no surprise that the broad benefits which peace brings to an entire society lose out to the much narrower profitability of warmongering to military contractors, military employees, and certain politicians.
BitcoinExchange Services (forum thread; no independent website for now) offers various USD-denominated store-brand gift cards, pre-paid debit cards, and VISA gift cards in exchange for bitcoins (at a small premium). It looks like he/she/they may also accept bitcoins in exchange for USD via PayPal, although they stopped selling them via PayPal due to fear of frozen accounts. I haven't used this service myself, but the thread contains many positive reviews.
It's what's known in economic circles as "consuming capital". Yes, we can afford these things now, but only by sacrificing our ability to produce as much of these goods in the future for a given amount of labor and/or raw material.
Capital (tools—the means of production) enormously increase our ability to produce goods compared to gathering and producing goods by hand. However, those means of production wear out over time, and need to be maintained and/or replaced if that increase in productive capacity is to continue. If fewer resources are allocated toward capital investment than are required for basic maintenance, there is more available for immediate consumption, but the capacity for production of new goods is impaired.
Consumption is an essential part of economic life; after all, the entire point in producing capital or consumer goods is to use them. However, the balance between investment and consumption must be considered if the standard of living is to be maintained, or possibly even increased. Over-consumption in the present decreases the future standard of living, both at a personal level and for an entire society. At the moment there is quite a bit of both.
Open the gates and let bogus, contradictory records spread across the network.
The bogus records wouldn't be accepted. In the event a client receives blocks from contradictory branches of the block chain, the chain with the greatest length (cumulative difficulty) wins. Any shorter chains will be rejected. That means you would need to outpace the entire rest of the network for quite a while to effect any significant changes.
You could, of course, do as you described and trick a local group of clients into accepting your modified history by cutting them off from the outside world for a time. However, those transactions would not infect the rest of the network when the isolation ended. The clients would revert to the longer block chain and resubmit any (valid) transactions processed during the isolation period.
Wages could fall much faster than the prices of things like food.
They could, but they won't. Historically, wages have always trailed goods. In an inflationary economy, prices of goods rise faster than wages (decreasing purchasing power). In a deflationary economy, prices of goods fall faster than wages (increasing purchasing power).
Preventing the sort of meddling you just described, which only serves to propagate the misallocation of resources which led to the problem in the first place, is a deliberate design feature, not a flaw.