That means road subsidies benefit employers who don't have to pay higher wages just so their employees can afford to get to work.

track circuits...can't tell you which train [is on a section of track], nor distinguish between maintenance vehicles and trains, nor can it tell you how fast or long a train is.

RFID can tell you those things.

Problem is, you'd end up screwing over the poor - that is, all the people who cannot afford a Prius or similar hybrid/electric vehicle.

Don't the poor usually walk, ride bikes, and take mass transit? Did you know that the poor love tolls more than other income classes because tolls displace taxes the poor would otherwise have to pay?

It would also jack up the price of nearly anything that is transported over the roads...

Actually, what jacks up the price is when we don't charge users full price for use of the roads, leading to a distorted, inefficient market for transportation.

(Don't give me any garbage about how everyone should live in cities -- what a drab, sad world that would be.)

Yes, what a drab, sad world it would be if people lived where they weren't an economic burden on others.

And building a website without reliance on JavaScript can be really tricky and limiting.

A website should always be navigable without JavaScript. One that isn't, is a sign of laziness, pure and simple. The use of JavaScript should be limited to eye candy, and in many cases CSS will do what you want.

The Chinese don't really know very much about rice. That's why the purpose of other dishes in Chinese cuisine is to help the rice go down.

The Japanese are the real rice connoisseurs. In Japanese cuisine, rice is a thing of worship, probably as an artifact of the Shinto religion where everything has a soul. A Japanese person will tell you that their domestic rice is the best, followed by California rice. But to the Chinese, rice is rice.

Buffalo eggs.

Unless it can do it so much better than doing it some other way and it's used so often that it's worth the space. For example, a rice cooker.

(Before someone says "well, they're just hard tests", the 6th grade tests had college level reading material on them.

To play the devil's advocate, if you get a perfect score on a test, you don't really know how good you are, you only know that you were good enough to pass the test. Getting an answer wrong reveals an upper limit of your ability, if knowing that is worth anything.

But how to you discern whether that 10% improvement is due to the work of the teacher or due to the work of a paid tutor outside the class?

If one student improves by 10% due to the addition of a paid tutor, and another student regresses by 10% by dropping a paid tutor, they cancel each other out. If all students got paid tutors, then their percentiles would not change.

The poor kid might be stuck at home by himself watching tv and eating a McDonalds value meal while his single mom is working her 2nd shift job...

And chances are that these circumstances were reflected in the test scores at the beginning of the year. I already explained this.

And evaluating based on the difference between beginning and end year testing doesn't work either because (at least theoretically) both kids are being exposed to new material throughout the year, so the disparity only increases.

Do you understand how percentiles work?

If a student starts the year at the 30th percentile and ends the year at the 40th percentile, then the teacher was probably pretty effective, even though the student is still under-performing.

If students from wealthy families score better than students from poor families, then that will be reflected in the evaluation at the beginning of the year, so this "value-added" methodology corrects for family backgrounds.

So we can glean some useful information about teacher effectiveness from student test scores.

That $68 billion is in year-of-expenditure dollars, not 2008 dollars. In 2013 dollars, that $68 billion (actually, $67.6 billion) is $54.9 billion according to the 2014 business plan.

It's unfortunate that even journalists don't understand inflation.

By the way, the central valley segment is only $27.8 billion in 2013 dollars. It's called the IOS, the Initial Operating Section. It's cheap because the land is flat and sparse. The bookends will be expensive.

The original cost estimate was $33 billion in 2008 dollars. The current estimate is the equivalent of $51.1 billion in 2008 dollars. If you'll do the math, you'll notice that $51.1 billion is not double $33 billion.

Very good. Now let's discuss how "opportunity costs" mean this should be completed as soon as possible and why it doesn't make sense financially to have airports close to downtowns.