Programming ability (similar to other STEM abilities) are power-law distributed, rather than Gaussian [1]. From this, it follows that training has very high probability of producing competent programmers but very low probability of producing exception programmers.
Not necessarily. Surely there are people out there who have the potential to be exceptional, but never reach it, because they never got started, never realized they'd be exceptional, etc. Converting the potential exceptionals into actual exceptionals, is a very big win.
I've been using ownCloud for over a year and it works great. The only issue is that it is not as convenient as Dropbox. It needs a server to host it, which is not free; and the server requires maintenance, security patches (see: heartbleed), upgrades, etc.
From the article: "what these two sets of massive reductions show is that companies like Amazon were probably charging far higher for their services than it cost to build and run them"
It absolutely does not show that. What it shows is that these are commodity services with little differentiation, and the only lever is price.
The question is: do you need to be someone with the size/capabilities of Google to compete in this market, or there are many companies that can make that "commodity" ?
Because if the answer is the first one, this looks like a good place to be in.
It's quite possible that Amazon and/or Google is prepared to operate their cloud (or some parts of their cloud) at a loss, or at least at zero margins. Look at Amazon's financial statements and you'll find they operate almost all their business on zero margin.
The reasons for that differ - Amazon optimises for growth and market domination over profits, while Google doesn't want to see Amazon control the entire cloud market.
You're making an assumption that they aren't now operating at a loss. I don't think they are but I wouldn't assume it. It wouldn't be the first time they've operated at a loss in an attempt to corner the market.
State and municipality bonds are sort of like U.S. Treasury bonds, except they're issued by a state or local government. They pay a fixed interest rate, but your interest is usually tax-free.
Of course, they pay a lower interest rate than taxable bonds. Thus, the only reason these tax-free bonds would be better than taxable bonds is if you are in a high enough marginal tax bracket to compensate for the difference in interest between these tax-free bonds and taxable bonds.
An example should make this all clear. Assume you have a marginal tax rate of 40%. (You earn a lot of money.) If you invest $100,000 in a typical taxable bond that pays 10%, each year you will get $10,000. But subtract out the 40% tax, and you are left with only $6,000 after-tax.
Compare this with what would happen if you invested that same $100,000 in a tax-free state bond. This bond only pays 8%, being tax-free and all. Each year you get $8,000. But you pay no tax, so you are in fact better off than if you had bought the taxable bonds!
Now, lets change your marginal tax bracket to 10%--you don't earn much money, so your marginal rate is low. You invest that same $100,000 in the taxable bond that pays 10%, you will get $10,000 pre-tax but only pay your marginal rate, 10%, on that sum, so you are left with $9,000.
With the tax-free state bond at 8%, you only get $8,000, though you don't pay any tax on it. So at your low marginal rate, you would have been better off buying the taxable bond.
My point is, tax-free bonds are probably only useful if you earn a lot of income each year to put you in a high marginal rate.
Source: I'm a law student and we learned about tax-exempt state/municipal bonds today in Tax!
Municipal bonds, loans to your local government are free of taxes (both state and fed). I have excess cash that would normally be in certificate of deposit in muni bonds currently making 3% tax free. There are risks, best to consult with a money manager if you are uncertain on how to gauge the associated risks.
[1] http://www.sciencedirect.com/science/article/pii/S1877042813...