> Developing useful software robots is a whole lot cheaper and easier than building and programming meatspace robots

Google also bought Boston Dynamics ( http://www.theverge.com/2013/1... ), who make very meatspace robots - in fact their robots are modeled on animals.

> There is no way for anyone to gain anything from those coins without a publicly viewable transaction appearing in the blockchain.

This is incorrect. Each of the early blocks that were mined went to a different bitcoin address, so they hold 50 BTC each. Each address has a different secret key associated with it. The original miner can exchange the secret key for cash, without making a bitcoin transaction. Of course, since the original miner also knows the secret key, they could still spend those coins, so doing a private sale like that would involve trust.

On a larger scale, the original miner could enter into a partnership, with adequate legal protections. They would sell a part ownership of all the coins in return for some money, and again not have to move the coins.

Just hold it in front of the anti-aircraft radar for 30 seconds.

There's a reason the first microwave ovens were called "Radaranges" (http://www.radar58.com/radarange/radar.html), and were made by the Amana division of Raytheon, a military radar manufacturer.

That limit is set by the finite size of a transaction (~ 250 bytes), and the hard limit of 1 MB per block in the block chain. Thus you can fit 4,000 transactions/block. Blocks are generated every 10 minutes (600 seconds) on average, thus ~7 per second.

The block size limit is intended to not overwhelm average PC's running a full bitcoin client (i.e. a node on the bitcoin network). There are several ways to deal with this limit. One is simply to gradually increase it, and migrate from user PC's to a distributed network of servers with more processing capacity. Another is "off chain transactions". For example, Coinbase.com has both 940,000 consumer wallets and 23,000 merchant accounts. So if a Coinbase user shops at a Coinbase merchant, the transfer is internal to their books, and does not need to hit the network. Eventually other aggregators can bundle up multiple user transactions and send it on the public block chain as a single large transaction to another aggregator. The details of who gets what amount can travel as a separate data file between them.

That's pretty much what happens in the traditional banking system. Banks settle up with each other once a day at a clearing house (usually the district Federal Reserve Bank). They add up all the day's checks going between a pair of banks, and then one of them pays the other the net difference. The actual payment goes across a private payment network (FEDwire) that only financial institutions have access to. In the old days, they had to swap piles of physical checks at the clearing house. With modern debit cards and electronic payments, it goes through an "Automated Clearing House" (ACH) which tallies up the amounts, but it is the same idea - lots of small transactions aggregated into one big daily clearing of the net balance between banks.

Do you think the idea of chained-hash record-keeping with proof-of-work is retarded? That's what allows bitcoin to keep a history of transactions which cannot be fiddled with, by anyone. Bitcoin happens to be the first implementation, but it can be used for any kind of digital data whatsoever. With this method, you can prove the data has not been tampered with or accidentally been changed (i.e. software or hardware errors). I'd say that is the opposite of retarded, it's very advanced and useful.

Do you think that money that only exists as data in a computer is retarded? Did you know that 88% of US dollars are that way? And that the vast majority of them are backed by other digital assets? (Mostly loans and Treasury bonds).

If not these, then which part do you find retarded?

The Hongkong and Shanghai Banking Corporation (HSBC) was founded by a Scotsman in 1865 to take advantage of the opening of trade with China, including the opium trade. They have been laundering money for drug pushers from the start, back when the British were the drug pushers.

You seem to have missed the 400 MW Ivanpah solar thermal plant just west of Las Vegas: http://ivanpahsolar.com/

Actually, they were $13.50 at the start of 2013.

> The question is, will they continue to go up, or will they crash and burn.

Bitcoins are useless without the payment network of which they are a part. The network moves money from one person to another. The value of one bitcoin unit is then driven by demand to use the network. A year ago, BitPay, a merchant processor, had 2400 merchants signed up. Now they have over 20,000. If that kind of growth rate is sustained, the coin will also go up.

About half a million people hold a significant amount of bitcoins, where "significant" means > US $80, an amount of cash a person might keep in a physical wallet:

http://bitcoinrichlist.com/cha...

Over time the early adopters will spend some of their coins, otherwise what's the point? You can't eat bitcoins. So the distribution will tend to approach that of any other asset in the world. The exact same thing happens to the founders of any successful tech company - the early employees end up with a big share of the company.

> because by 10:40AM they will get caught.

This is a complete misunderstanding of how the bitcoin network works. Individual transactions go out from the sender across the P2P network. Each node checks the transaction sending address against past transactions to see if it has enough balance. If the first one to arrive depletes the balance, the later ones get rejected and not passed along to other nodes. Each node can do this because it has a complete history of bitcoin transactions in the block chain + a memory pool of recently arrived transactions not yet in a block.

The relay time per node averages about 1 second. If there is as much as 5 seconds variance among the scammers, only one transaction will get relayed across most of the network. Miners receive transactions like everyone else on the P2P network. When they find a valid hash for a block of transactions, they send it back out to everyone else. Each node then verifies the hash is legitimate, and if so adds it to the growing "block chain", and subtracts the included transactions from the memory pool. But the node had the transaction within seconds of it being first sent. A block just means the contents of the transaction are now backed by a lot of computation, therefore it will be hard to ever change in the future. The hashes are chained by including each hash as part of the data for the next block. Thus as more blocks get added, a given transaction would require repeating all the computation that has happened since to change it, which gets increasingly unlikely. But double-spending is mostly prevented within seconds, because nodes won't forward a transaction that doesn't have enough funds to cover it.

Checking for double spending happens with each node in the bitcoin network as it relays the transaction. This takes seconds. Each node compares a given transaction as it arrives to the past transactions for that sending address. If there is not enough balance, it dumps the transaction. Nodes can do that because they have a complete history of past transactions (the Block Chain) and a memory pool of recently arrived transactions not yet in a block. Since transactions typically go through ~5 nodes from sender to everyone on the network, transactions are checked multiple times. Thus the incidence of *attempted* double-spends are less than 1 in 10,000. Successful ones are much less frequent.

When a hash for a new block is discovered by miners, they send it out over the same P2P network that relays individual transactions (that is how they get the transactions to put in a block in the first place). Each node then verifies the hash is correct, and adds it to their copy of the Block Chain. Then they delete the transactions in their memory pool that are now in the block.

Your statement "It's simple to double spend" is incorrect for a number of reasons. Someone has to hack their wallet software to allow it, then relay a transaction by different paths, because only the first arrival at a given node is allowed. Inevitably miners will accept only one of the transactions into a block, and whichever arrives first is the one they work with. Unless the sender manages to balance the double spends evenly across different parts of the network, most likely only one of them will reach the majority of miners, and thus get put in a block. The other one will get filtered by nodes before it even gets to the miners. Doing a double-spend while in line at Target is difficult because you are using a portable device which links to the internet through a single path. Even if you had a hacked app that sends your money to Target *and* another of your own addresses simultaneously, your nearest nodes that you relay the transaction through will have variable delays, and one of them will get one of the transactions out faster. If it's not the one to Target, they won't see the payment arrive at all, and tell you to try again.

If you are selling a car or a house, it would be wise to wait for a number of confirmations, and in addition check that the balance in the sending address is "mature" (over an hour old). But for small scale store checkout, zero confirmations are quite enough. The risk is much lower than "shrinkage" (theft by store employees mostly), and the fees, fraud rates, and charge-backs are way lower than for bank card transactions.

> "I'll take bitcoins, but you only get your bread once the transaction clears"

You don't need to wait that long. For a transaction to reach the seller's computer, it has to be relayed through the network. Each node compares the transaction against the balance for that address, as recorded in the block chain plus recent transactions not yet in a block. If there isn't enough balance, the transaction isn't relayed further, it's dropped. Thus the incidence of "double spending" (trying to spend a balance you don't have) is less than 1 attempt per 10,000 transactions.

For buying a loaf of bread, that risk is low enough for the shop owner to ignore. "Confirmations" happen when the transaction is included in a block, and then added blocks are chained after it. Since each block takes a rather large amount of computation to find a valid hash, and the hashes are chained, undoing a transaction takes more and more work over time, becoming exponentially less likely. If you were buying a car or a house you would be wise to wait for some number of confirmations, but not for a loaf of bread, or other small transaction.

Ty Warner, the inventor of Beanie Babies, became a billionaire. Obviously they were worth a lot. But Beanie Babies can't be sent to the opposite side of the world in minutes for pennies in fees, so they are not as useful in a payment network as bitcoins are. Bitcoin derives its value purely from the fact that Western Union, PayPal, and bank wire transfers suck. Which would you choose, to pay tens of dollars in fees to send money somewhere, or to pay 8 cents?

The bitcoin code is open source. All you need to do is spin up an exact copy, let's call it bitcoin2.0, with a whole new set of coins. Aside from having a fresh block chain, it's otherwise identical, and user software can be made to handle both interchangeably. This has already somewhat happened with the collection of altcoins. They are mostly similar and use different names, but they add to the digital coin universe:

http://coinmarketcap.com/

We just go full metric. 1 Satoshi = 10 nanocoins = 10,000 picocoins