Chris Lee's impressive overview of quantum cryptography. Arstechnica, to my mind, does the best science reporting on quantum mechanics.
A real world example of the future of secure computation is found in the Danish sugar beet industry. All the sugar beets in Denmark are purchased by a single company. The farmers buy the rights to sell a certain volume of sugar beets to the monopolist. However, individual farmers must buy these rights based on estimates of their own production. As a result they often find themselves in the market to buy or sell rights.
Now, the last thing they want to do is buy and sell these rights through auctions run by the monopolist—monopoly power must be limited after all—because the monopolist would have the production records of every farmer and be able to use that to bid the price of the rights down. The market is setup by sellers simply listing the number of units they are willing to sell at each price. Buyers make a similar list of how many units they are willing buy at each price. The results are put together into two graphs of price and volume. Business is conducted at the intersection of these two curves.
Key to the success of this market is that no one knows who is in the market, or how much any particular farmer wants to buy or sell. A more open auction would reveal too much information, leaving farmers vulnerable to market manipulation by the monopolist and each other. So, secure computation could enable a way to create markets that are less open to manipulation, creating a fair economic mechanism.