The terms "closed platform" and "walled garden" have a very specific meaning, and it doesn't apply to Windows. From Wikipedia (my emphasis):

A closed platform, walled garden or closed ecosystem is a software system where the carrier or service provider has control over applications, content, and media and restricts convenient access to non-approved applications or content. This is in contrast to an open platform, where consumers have unrestricted access to applications and content.

It's obvious that Microsoft has absolutely no control over what software can be run on Windows. Compare that to Apple's iPad, where you can't install anything that's not approved by Apple (unless you jailbreak it first). That makes iOS a "walled garden".

Now, maybe we agree that it was foolish for Microsoft to enable any kind of "autorun" feature. The point is that in an "open platform" (that is, one where the user has complete control over what can be run on it), the user must also have enough power to do dumb things like running an unknown program from a pendrive that was just plugged in. How easy it should be for the user to do that is another discussion.

Actually, to "believe" in science, the only thing that's strictly required is that you believe that the universe is knowable. Even the ways you use to know more stuff (the "scientific method") are not "a priori", that is, if you can think of a better way to discover stuff about the universe, then it will become part of the scientific method.

Science doesn't a priori reject the possibility of a creator (God), just as it doesn't a priori require that the universe came into existence of its own accord. The Big Bang theory is just the best answer we have so far when asking questions about the start of the universe. Science doesn't give definitive answers, since it's always possible that you'll find out later that you didn't know everything there was to know about something.

I have to agree with this one. "Son of a submariner!" is the best line I've ever seen in a videogame.

That's very silly. It's a convention to write "God" when you're talking about the the supreme being (like the Christian God, as opposed to other gods like Zeus, where "god" is usually not capitalized), even if you don't believe in it.

See for instance Wikipedia: in "God"

But in "Zeus"

That pretty much confirms the point he was making; clearly the iPad had a price set in dollars, not bitcoins. Just because you didn't have to convert your bitcoins to dollars, it doesn't mean the conversion didn't happen.

People here are stuck at the (inflammatory) headline and are missing the point entirely. Of course prices (i.e., the value of money) vary, but when the "currency" you're using fluctuates wildly -- to the point where it's possible to make or lose a lot of money just investing in it, and not using it, like a lot of people have been doing with bitcoin -- that's not a healthy currency.

Now, clearly the article is a bit (a lot, actually) on the FUD side ("We don’t really know how this coin is created. [...] Thankfully its plunge will be a salutary caution to most folks."), but it's good to least try to understand and respond to it, and not some imaginary argument you think he's making based on the headline.

That's a nice idea, but it doesn't really solve the underlying problem. Imagine that you're convinced that TeliaSonera is friendly to governments in Central Asia (as the story seems to imply). So it would make sense to trust them (a lot) to attest government-friendly identities in that region. But it would be silly to trust them (at all) for anything else.

In the end, trust in a CA has context. It's not enough to simply assign a number to convey how much you trust a particular CA; what you're really interested is how much you trust a particular CA to attest a particular identity.

Dalvik bytecode is almost always generated by compiling Java source to Java bytecode (.class) and then using "dx" to convert one or more .class file to a .dex (sometimes the conversion is done by converting a .jar containing .class files to an .apk or a .jar containing .dex files).

I'm not sure where is the "official" documentation for dx, but you can find its source code in "com.android.dx" in the Android SDK source (here you can see its "Main" class, for example). Or just google "android dx tool".

Much as I dislike Oracle, I don't feel Google played fair on this, they took someone else's technology and effectively stole it.

How exactly did any technology got stolen?

1) Dalvik is completely different from the usual Java VMs (Dalvik is register-based, while Java VMs are stack-based). That's how they got away with a VM that interprets (modified) Java bytecode without infringing any patents related to Java (as was confirmed by the "Oracle vs Google" case).

2) Android uses the exact same interfaces than Java, but that has been standard practice for decades (Unix, for example). No code between Android and Sun/Oracle's Java is the same, except for the stuff that must be the same to implement the same interfaces. The "Oracle vs Google" confirmed that's OK (as the summary says).

In the end, what happened is that Google didn't want to pay Sun/Oracle for a license to use Java mobile, so they implemented their own Java-compatible system (or rather, bought it from other people). That's how technology evolves. As someone else said here a few posts back: imagine if anyone wanting to do a Unix-like OS had to pay a license to someone. Where would we be now?

I agree with your main point, but this example doesn't fit in it very well. It's true that Newtonian physics has been superseded by relativity and quantum mechanics, but it's still "settled science" -- it's taught in every physics course and it's widely used. Just because we know some model doesn't fit perfectly every possible situation, it doesn't necessarily mean we stop using it.

Newtonian physics (like classical electromagnetism, thermodynamics, and a lot of other theories in Physics) is still useful when the system you're studying is guaranteed to be within some range of parameters -- not moving too fast, not too dense, not too cold or too hot, not too small, etc.

That's not to say there aren't many examples of "settled science" that have been completely abandoned. If you want a nice and clean example to illustrate your point, you might want to use the phlogiston theory. It was dominant for a large portion of the 18th century, but we now know it's completely useless.

This is one of those topics where there are a lot of conflicting studies on the exact numbers (on how much energy you get compared to what you put in), but it seems that everyone agrees that corn ethanol is particularly bad: http://en.wikipedia.org/wiki/Ethanol_fuel_energy_balance.