That's not really the problem. Any increase in the effective tax rate is going to drive jobs out of the country, that's just how it works. It's supply and demand. Make operating a company in the US more expensive and demand for real estate and employee labor moves to alternative suppliers of tax jurisdictions that have lower "prices."

The underlying problem is that the entire concept of "corporate profit" is accounting fiction. Corporations only exist on paper, and you can make the paper say whatever you want.

In particular, the problem is that the existence of profit violates the notion that assets have a single "market price" that can be identified for the purposes of transfer pricing. Walmart buys a widget for $.79 and sells it for $1. That's how they make their money. The fact that a subsidiary does the same thing (and happens to be a low tax jurisdiction) is not something you can legitimately complain about; of course they're paying less for things than they're selling them for, that's where profit comes from.

Which is the source of the problem. How much is it worth to Microsoft's UK subsidiary to have a license to sell Windows? How much is it worth to Google Australia to be able to use the Google trademark? There is no objective way to determine exactly what these things are worth. So lo and behold, the things that subsidiaries in high tax jurisdictions buy from subsidiaries in low tax jurisdictions are on the high side of the acceptable range, and the things they sell to the same subsidiaries are on the low side of the acceptable range. And it doesn't take much of that to eat all of a company's profits -- making profit is hard even when you're trying to do it.

The fact is, corporate income tax is inherently broken. Corporations are a legal fiction, which means corporate income is a legal fiction, as is its location and tax jurisdiction. Trying to tax something fictional is a great way to get a thousand tax accountants to assert that it doesn't exist, but not a particularly good way to raise money. If you want to raise money, tax something that has a physical location: Sales, labor, real estate, etc. Microsoft can easily arrange for its local subsidiary to be unprofitable; they can't avoid collecting sales tax on Windows while continuing to sell it to customers in your jurisdiction.

The problem is not who is getting wiretapped, the problem is who and what is being obligated to support it. The original CALEA applied to AT&T. AT&T can figure out how to navigate a federal statute.

But now they're wanting to impose it on software. The last thing this country needs is laws that end up throwing J. Random Hacker at some university graduate program or tech startup in federal prison for publishing a new VOIP protocol without consulting a team of attorneys.

On top of that, the traditional phone network has crap for security. Any jackass with a lineman's handset can stand in front of your building and listen to your POTS telephone calls. Implementing wiretaps for that is easy because the phone company already has the cleartext, and it doesn't really make the security any worse than its current level of non-existence. By contrast, the way VOIP should be implemented is with end-to-end encryption -- but then the VOIP provider can't wiretap you, because they don't (by design) have access to the cleartext. Which is the only way to make it so that if the VOIP provider gets hacked, the infiltrators can't intercept your phone conversations.

Enshrining insecure designs into the law that allow foreign governments to conduct industrial espionage against U.S. companies is a bad idea.

Look, you're obviously trolling. I'm not a lawyer, but you said, "You're an accomplice in any illegal activity if you fail to take any steps to prevent it." So I'm an accomplice if a bank robbery is occurring 3000 miles from me and I'm watching it on television, not taking any steps to prevent it? You don't have to be a lawyer to realize how blatantly wrong that is.

As for your other examples, the majority have no intervening malicious third party. The owner's negligence is the sole cause of the harm, clearly distinguishing them from the wifi case. And I question whether the remaining examples are even accurate, especially as general principles. For example, "Man leaves baseball bat laying on street corner, kid comes by, picks it up. It's later used to rob a convenience store." You propose that should give rise to liability? I don't think so. And the fact that you seem to think it's a good idea to impose criminal liability on a convenience store clerk with the audacity to get robbed while taking a piss shows the clear failure of your position. Talk about blaming the victim.

But notice something else: Cars, fire suppression systems, oil platforms, cigarettes, firearms, animals. These are all things for which mishandling has a material propensity to result in corpses. Nobody has ever died as a direct result of open wifi.

More than that, it is possible to generate an enormous list of counterexamples. Man commits robbery in city park; city not guilty of robbery. Man bums a cigarette from another man, later dies of cancer; second man not liable for death. Walmart sells a firearm to a man who then uses it to commit a murder; Walmart not guilty of murder. Here's a good one for you: AT&T customer commits wire fraud using telephone; AT&T not liable for wire fraud. Man sends bomb threat in the mail; US Postal Service not liable for making bomb threat. Man borrows DVD from local library, takes it home and copies it; library not liable for copyright infringement or DMCA circumvention. I could go on.

I'd be curious to hear which law school you attended that taught you that, considering how wrong it is.

You neglect that your examples are the exception rather than the rule. Cars are large, fast, dangerous and expensive. They have their own rules. Almost nothing works that way, nor should it. The person responsible is the person responsible, not random innocent bystanders who happens to be in the general vicinity or who lent general purpose tools to someone not knowing they would be used for nefarious purposes.

It's a little more complicated than that. For one thing, if you tax based on number of employees, the employers who can will stop hiring people in jurisdictions with higher taxes. Which is the opposite of what those jurisdictions want, so they won't want to tax based on that.

But more than that, it doesn't really work: How do you define a corporation? If a holding company in Bermuda owns the US R&D outfit, and you tax based on the profits of the US R&D subsidiary, the executives will arrange for the subsidiary to make minimal profits and the holding company or another subsidiary to make more profits. Conversely, if you tax based on the entire conglomerate, they stick some labor-intensive manufacturing company into the conglomerate which employees a hundred thousand people in some third world country, which dilutes the 5000 US employees doing R&D into nothing.

The problem is that there is almost no industry where profit margins are so massive that they can't be completely consumed by systematically but only ever so slightly overpaying for everything, and buying it from subsidiaries in other jurisdictions who end up reporting profits there, where taxes are lower.

There is really only one way to prevent this. You tax based on revenue rather than profit, so that "transfer pricing" doesn't matter. But revenue taxes are very silly unless it's a VAT, since otherwise you get tax paid on tax and it compounds based on the (extremely arbitrary) number of transactions that occur, which highly favors extremely vertically integrated companies.

So you're really left with a VAT instead of an income tax. Which is probably not a bad idea, if you combine it with a yearly refund in a fixed amount per taxpayer to make it progressive.

This is true as far as it goes, but the behavior when overclocked is telling for more than how well you can overclock: At the risk of stating the obvious, the chips the overclockers are having heat issues with are the ones Intel is manufacturing. That means Intel isn't going to be able to ramp the clock speed very easily for the same reasons that the overclockers are running into trouble, unless there is some significant and avoidable flaw in the chip or the process that they can remove in future revisions.

On the plus side, this gives AMD a little breathing room to try to catch up a little.

Certainly. When you pass laws that are vague and non-compliance results in outrageously expensive litigation, you can easily predict that the response will be inefficient prophylactic measures. Ignoring that and passing the legislation anyway makes you responsible for causing the response the legislation induces.

I want to clarify what I'm advocating here. I don't really see a problem with a law that prohibits you from discriminating in hiring on the basis of race, as long as the law doesn't allow the conclusion that you have done so based on a collage of cherry picked circumstantial evidence. But some of them do. (Especially the ones at the state and local level which are often carelessly drafted and passed with minimal debate.)

The problem is that you're not taking into account how businesses react to laws like that. They don't just say "we're not racists, so no problem, we'll just keep doing what we've been doing," instead they put together a huge bureaucratic compliance process. In many cases they establish quotas because in the past people have argued that underrepresentation of minorities is evidence of racism, leading to hiring of unqualified applicants in order to satisfy the quotas. Even without quotas, there is often internal pressure to hire questionable minority applicants whenever they are nominally qualified, to the exclusion of those who seem to be a better fit but are the "wrong" color.

Racism is not an effective method of fighting racism.

The Microsoft case was a very different set of facts: Microsoft did agree to a license from Sun, and did call their version Java, but then violated the license contract by making their version incompatible. Moreover, Sun sued Microsoft for antitrust violations, and Microsoft settled the case rather than Sun winning in court.

The Google case is about copyright and patents. Oracle is claiming a copyright over the API (which is ridiculous), and for a collection of patents the majority of which have now been rejected by the Patent Office and the remainder of which appear to be non-essential performance optimizations that in the worst case could probably be worked around without breaking anything serious (and it has yet to be decided whether even those claims will hold up in the end).

The whole thing feels like Oracle just being extremely petty and vindictive because they overplayed their hand in the license negotiations, Google called their bluff and now Oracle is butthurt and acting irrationally. (I mean lets face it, the invalidation of the majority of the patents they've tried to assert has already hurt Oracle more than the lawsuit is likely to hurt Google, and that had to have been a predictable outcome of all of this if Oracle had been thinking clearly.)

The open source implementations of Java are licensed under actual open source licenses. There is not one word in the GPL that says you can't use it on a mobile device. The way Sun/Oracle maintains control over "Java" notwithstanding that they released their own implementation under the GPL is by licensing the trademark. If you want to call it Java(TM) then you need a trademark license. Sun has said as much. But Google isn't calling it Java(TM). And given that, it seems like they could pretty easily have done away with the copyright claims going forward just by licensing one of the open source implementations which inherited from Sun's GPL'd implementation, if they wanted to. Apparently instead they've decided that the copyright claims are silly -- it is an independent implementation -- so they'd rather fight it. (I can't say I blame them. Oracle is a bit of a jerk. "We want $6 Billion dollars durr hurr.")

Google doesn't have a license from Sun for Java (or Java(TM) or whatever), the negotiations broke down and Google decided to do their own implementation instead. What do Sun's license terms have anything to do with anything if Google hasn't agreed to them?

The copyright question is very much whether they "need" a license to do what they did, i.e. creating an independent implementation that implements the same API. (I think there is also another question about a few lines of a range check function that is line-for-line identical between Sun's OpenJDK and Google's version... but it seems to have come out that the reason for that is that the code was originally written by a Google engineer and then accepted into OpenJDK. Way to reward your contributors, Oracle.)

Are you seriously suggesting that they bought off the judge? Give me a break.

Do your friends not have their own money? Also, play Starcraft. It has spawn copies.

It's 2012, is this really a thing? You can get an AMD A-series laptop which is faster than any existing console for like $400. What kind of Pentium 4-era dumpster bait are your friends using?

So let them use your other computer. Most people with computers have multiple computers: Either you have a desktop and a laptop, or a new computer and an old computer, or your computer and your brother's computer, etc.

And don't be such a pedant. An alternative doesn't have to be 100% better at every individual thing in order for it to be better on net.

Have them bring their laptops?

Your argument generalizes to the idea that any product for which the seller pays for shelf space has negative value. This occurs pretty much anywhere that distributor intermediaries have sufficient market power to abuse it, specifically where customers choose the distributor intermediary based on more considerations than solely which third party service (e.g. search engine) they offer.

Kindly troll elsewhere.

I'm not saying you're necessarily wrong about Steam, but it's worth keeping mind that today's minor inconvenience can easily turn into tomorrow's accursed blood-boiling permanent loss of every game you own.

Let me give you an example: Suppose this app store model catches on in games, and Microsoft, Apple or Google creates a Steam competitor that really storms the market, and Valve and Steam go the way of the dinosaurs. Really, even regardless of the reason, this is bound to happen eventually: No company lasts forever. Maybe it goes the other way and DRM dies out entirely, and they ultimately decide they aren't even going to support the legacy ones anymore.

One way or another there is going to come a day when they turn those servers off, and a day after that when you buy a new computer and can't install a single one of your games on it because the Steam servers are dead and gone. You have to make the decision today as to whether you want to have to tear your hair out on that future day or not.

As for me, I won't be buying any game with DRM. Ever.

That'll show them.

Voting is necessary but not sufficient. The vote comes too late: If the incumbent wins, he thinks he's got a mandate to keep doing what he's been doing. If the incumbent loses, the new guy thinks he's got a mandate to do whatever he wants. In no event does voting send the right message, unless your vote actually changes the outcome of the election -- and good luck with that.

You need to make sure they understand what you need them to do before the election, that way they can actually do it. We didn't stop SOPA by voting anybody out of office -- we stopped it with millions of letters and phone calls, threatening to vote them out. If it actually comes down to you having to make good on the threat, you can certainly vote them out if you have the numbers, but by then it's already too late.

You can't just make a credible threat and then execute it. You're missing the most important part, which is to make sure they understand that you're going to kick them to the curb if they don't do what you want. So get off your ass and write a letter to your congress critters.