By "in the middle of" you mean "not yet in control of and had no power over," right?

So the timeline is:

1) Motorola brings up possible patent issues with Microsoft in Germany. For the moment we'll set aside if they're valid/invalid or if this is real or patent trolling. But Motorola said, "Hey, Microsoft, that's ours."

2) Possible phase of negotiations in Germany.

3) Microsoft doesn't like how the negotiations are going, so they sue Motorola in US court.

4) Motorola says "I don't think so" and sues Microsoft in German court for violation of German patent law.

5) US judge says "well Microsoft sued here first, so this is where international jurisdiction should reside."

So the lesson here is: regardless of the country of the dispute, and regardless of the merit of the patents, and regardless of how negotiations are going, and regardless of whether you're the plaintiff or defendant overseas, as long as you sue in US court first you can get a US judge to slam the brakes on any patent issues.

How is that reasonable?

Between 70 and 80%. That's a HUGE difference. That means that compared to the other computerized systems out there you're either totally awesome or really suck.

That's like saying, "I did a lap in a Formula 1 car, and I'm either 15 seconds ahead of last year's world champion, or I'm a minute behind the field."

You haven't done this before, have you?

[ Disclaimer: I work for Google. ]

I've been here for several years and I've never ever seen an interview schedule with 10 people. Are you including the recruiters, all the people you spoke to over the phone, all your interviewers, and the person who took you to lunch as "interviewers"? Did you interview here multiple times? Because nothing I've seen -- and I've done over 100 interviews here -- matches your description.

Nah, I haven't been arrested yet.

Okay. I believe it went

• I pointed out about six fallacies in a parent post.
• You called me a fuckface and threw in some new numbers on 1 of the 6 items.
• I took your numbers and recalculated.
• Then you called me a fuckface again and dug up some more numbers.

Does that bring everyone up to speed? Good. Let's continue.

The right side with the 9.08 and 65.15 is an attempt to introduce some fuzzy math about "social costs" ("Includes pavement, congestion, crash and noise. Excludes pollution." [Emphasis mine]) I reject this column since it's nearly impossible to accurately determine how they arrived at these numbers. Instead we'll focus on actual pavement costs. In which case the cost ratios are somewhere between 21 (small trucks vs cars) and 819 (big trucks vs cars); Given that these only go to 1 decimal place, it's possible that car costs are in the range [0.05,0.14] and still round to 0.1 cents per mile.

Given that the average driver in the US drives about 13,000 miles their costs are around $300/year and not $400. (Note that I'm making my own case worse). So the average truck pays 13900 / 300 ~= 47 times what a car does.

I was citing the diagram for weight distribution between the axles of a semi, and nothing else. I'm sorry you got confused by that.

So IN CONCLUSION, what you've shown is that if we look at only "urban interstate" traffic, ignore "rural interstate" traffic, and assume that an average semi weighs no more than about 25 tons (as opposed to the 40-ton maximum that shipping companies aim for) then and only then do cars and trucks break even. In all other cases, trucks pay less (proportionally) than cars do. Unless you want to venture from the confines of actual road costs and try to include fuzzy concepts like "societal costs". But I'm sure a guy of such rigorous devotion to hard data wouldn't want to do that.

So now that you've dug up all this data to help me prove my point, what's next?

Greetings and salutations to you, too. I see you're bringing a dizzying intellect to the table. Taking in account your FIRST TWO FACTORS and the per-axle weight distribution here (6T in front, and CARGO/4 per drives and trailer axle), the 50,000lb truck from my earlier example does a trifling 2,396 times the damage to the road as an average 2-ton car. A fully-loaded truck would do 4,308 times the damage as a single car.

Indeed. Now that we've done it your way and shown that the truck merely does 2,400 - 4,300 times the damage as a single car, what's next? Are you going to argue that the taxes and fees paid by the working-class Joe on his car are proportional to the taxes and fees paid by the shipping company? I pay about $75/year in federal gas tax (10K miles / 25mpg * $0.184/gal), plus another $125 in state vehicle registration, taxes, and fees. Does each semi truck pay $180K (less weight, gas only) - $860K (fully loaded, gas+fees) per year for road upkeep? (This is using your math, remember?) And please show your math.

Wrong. Backwards. State income taxes are a line item deduction from your federal income taxes. Increased state taxes result in less federal income. (See Schedule A).

Comparing anything to California is invalid because California has so many Constitutionally-mandated spending requirements and Constitutionally-prohibited tax sources that it's basically a given they're going to be broke year in and year out. What about Nevada? They have no income tax at all and are currently facing a $1.8 billion dollar deficit on a $3.6B budget; that's even worse than the federal government, as a percentage of money spent.

Okay, let's look at that. That truck bringing groceries to your store can weigh (legally) up to 40 tons, but let's conservatively say it weighs 25 tons. That's 12.5 what a good-sized car weighs. Taking into account that road wear is proportional to the fourth power of weight, and one semi bringing groceries to the store causes as much wear and tear as 24,414 cars. Do you think that semi pay 24,000 times as much in taxes and fees on a per-mile basis as you do? If not, then business owners are getting a lot more out of their road and fuel taxes than you are.

If you honestly think that anyone well-to-do keeps more than $250,000 in a single savings account then you'd make the world's worst financial advisor. Even the moderately wealthy have their money tied up in investments (not FDIC-protected) and their savings spread across multiple financial institutions in order to minimize risk. That's not even taking into account that the FDIC is broke, and the institutions where the rich keep their investments just get a direct federal bailout when they go under. So in summary:

• You were wrong about the tax deductions
• You mislead about the efficacy of income taxes versus sales taxes
• You used a misleading metric for "benefit" in a few cases, and
• You have no idea how to invest, and when banks go bust the working-class get screwed while the investing-class and upper-class get a bailout.

Would you like to be wrong about anything else today?

It's kind of difficult to do that when the Venn Diagram circle representing the GOP reflexively recoils when the Democratic circle reaches out to overlap it.

Don't worry. If you've got an electronic version stored somewhere, they've already grabbed a copy.

While this is a useful data point, it's not conclusive. If the root cause is some electronics error whose symptoms are a sudden acceleration and (according to two victims) no response to the brake, it's not surprising that the black box -- presumably using the exact same input controlling the engine -- would claim that the accelerator was fully pressed and the brake was untouched.

I whole-heartedly support your philosophy of turning what works in your house into a law that applies to everyone. For example, in my house the mother of the kid has sex with me on a regular basis. Clearly there should be a law stating that in these situations, women who have children should sleep with me.

Completely true.

In 2005 the state of Virginia wanted me to fork over AT LEAST $3,000 (!) to get per-precinct turnout figures. Not per-precinct results; those were free at the State Board of Elections website. But if you wanted to know actually how many voters showed up at each precinct they said it would take 4-6 weeks and "reproduction costs" would be between $3,000 - $5,000 for them to send me a CD with the PDFs. They had per-county/city turnout results (also on the website). But apparently getting it down to the precinct level was going to be a massive undertaking, requiring one staffer to spend three to five weeks full-time collecting this data and then another week to digitize it and send it out.

My lawn.

Off it.

Now.

P.S. To the ~0.8% of you to whom this does not apply: sorry.