Whether the implant records a person's thoughts is not material. A diary is not protected under the 5th even though it is a recording of private thoughts, so why should an implant that records thoughts be treated differently?

I think rgmoore has a point, the key is not that a device records thought, but when does device data become indistinguishable from private thought. If a person had brain damage and there was an implant that would effectively take the place of that damaged part of the brain, say memory, then I think it would become a legal issue.

There is also the issue of how easily the data could be retrieved. The court weighs the potential probative value of the evidence against the rights of the individual and that generally precludes procedures which endanger one's health, cause severe pain, or lasting trauma.

When looking for the WaPo article I linked to I noticed they had added a correction since I first read it:

CORRECTION: An earlier version of this story mischaracterized the prescription drug overdose death rates. In aggregate, states with medical marijuana laws had higher death rates than those without, though the authors’ statistical analysis did find that the laws were in fact associated with overall decreases in overdose deaths.

Which means that the idea that legalized states had lower death rates is not true.

Also, the author of the study wrote a piece for the NY Times in which he said:

Using death certificates compiled by the Centers for Disease Control and Prevention, we found that the rate of prescription painkiller overdose deaths increased in all states from 1999 to 2010. But we also found that implementation of a medical marijuana law was associated with a 25 percent lower yearly rate of opioid painkiller overdose deaths, on average. In absolute terms, we estimated that states with a medical marijuana law had a total of about 1,700 fewer opioid painkiller overdose deaths in 2010 than would be expected based on trends before the laws were passed.

Since the author states that all states had increases in OD death rates, the summary's claim that legalization led to "a dramatic reduction in overdose fatalities" is also not true. The "reduction" in OD deaths was due to the difference between the statistically expected death rate in legalized states using the non-legalized state death rate trends compared to the observed death rate. In other words, and being very rough with the math, the non-legalized states had their OD death rates increase 4x, but the legalized states only had a 3x increase, therefore the legalized group had 25% fewer deaths than would be expected if they were following the non-legalized trend.

I meant to say states moved from the illegal to legalized group.

This study has been misreported nearly everywhere. The study didn't find states with legalized medical marijuana had fewer deaths than non-legal states. Legalized states continually had more deaths per capita, and both groups had dramatic increased in opiate OD deaths over the period covered by the study. The researchers found OD death rates in legalized states increased ~25% less than expected.

I don't have access to the full study, but this chart included in this Washington Post article shows both groups OD death rate increase dramatically over time. It's interesting to note the change from 2009-2010, which significantly narrowed the gap between the groups. Prior to that year both groups seemed to be on similar trend lines. That said, groups moved from the illegal to legalized group over the course of the study and I'm not sure if or how the chart was adjusted for those changes.

On the other hand if you're a Tor developer interested in disrupting the NSA unit assigned to hack your system why not just say you receive regular leaks from the NSA unit assigned to hack your system.

Really all 40k of them are "new students". This is a different group of kids, presumably taking a different test, perhaps with different prep methods, so it's not so easy to isolate the effects of the extra 8K students. If we were able to isolate the "new students" from the core 32K that would have taken the test anyway, and assume the core group would have gotten the same 68% pass rate we could estimate that the new group had around a 33% pass rate.

33% is a pretty poor pass rate, but on the other hand that represents 2600 students who passed that may not have taken the test at all. And taking the test is usually a no risk proposition as reporting the score is generally voluntary for college applications and many will be taking them after they have already been accepted to a school.

That said there is some risk to failing. The first is the hard to quantify effect on motivation to continue in the field. This kind of discouragement can negatively affect some, though it also can help guide students into a field that may be a better fit for them. Also, some high schools record AP scores in the student's transcript, and if a student does poorly on the exam it can be made exponentially worse if the admission board notices they did well in the AP class itself since it calls into question how rigorous the school's course work/grading system is.

So you are saying that Amazon has somehow found a way to actually ship items for free, to both the user and itself? Otherwise, whether they call it "free" or "covered" the cost of shipping is covered by the product purchase price.

Actually, you are incorrect.

The 5th absolutely protects the "contents of one's mind". The Supreme often uses those exact words in many cases affirming the notion. In fact, one common analogy the court uses in deciding these issues is that the government can force a suspect to produce the key to a safe, but it cannot force him/her to produce a combination for a safe.

The 5th does protect against coerced testimony, and that includes passwords or knowledge of physical evidence. However, once the government knows that physical evidence exists and is in the possession or under the control of the suspect they can compel the suspect to produce the evidence, whether that evidence be business records, personal papers or computer data. So while the suspect cannot be compelled to reveal his password, if the government knows an encrypted drive contains incriminating evidence it can compel him/her to produce the unencrypted data. That is why the suspect are not ordered to reveal the password, but to unlock the data for the police.

One key distinction here is that the government must have ‘reasonably particularity’ concerning the evidence it requests. They cannot merely suspect the evidence may exist. Cases where defendants are compelled to produce unencrypted drives usually have other factors such as U.S. v. Friscou where the government had recordings of the defendant talking about the incriminating contents of the encrypted data. Or, in re Boucher where police saw the unencrypted content of a laptop that was turned on, but then lost access when it was powered down.

In the safe analogy the police cannot compel a suspect to produce a combination, but if they have sufficient reason to believe the safe contains a incriminating accounting ledger, they can compel the suspect to open it and produce the ledger.

In the most likely scenario that the trunk is never damaged in an accident then at some point you are left with a very expensive carbon trunk lid with an embedded battery that no longer holds a charge.

To generate page views, of course.

"When critics of passenger-rail subsidies, such as Randal O’Toole of the Cato Institute, suggest that the highway costs are mostly covered by the gas tax, Lind counters with figures from a 2008 Federal Highway Administration paper: the FHA reports that highway user fees, including gas taxes, only cover 51 percent of costs. By contrast, Amtrak in 2010 covered 67 percent of its operating costs from ticket fares and other revenue."

That's BS. First, Amtrak's "Operating costs" are all inclusive, while the FHA's, or indeed all government highway spending, represent only a small fraction of the operational cost of the highway system. Add in a large part of the US auto industry (sales, parts and service, insurance, fuel, etc) and you will get a more accurate picture of the "operational costs" of the highway system.

Or simply compare the federal costs vs revenue by mode of transportation per passenger. Which is something the Bureau of Transportation Statistics did in 2004: http://www.bts.gov/publications/federal_subsidies_to_passenger_transportation/pdf/entire.pdf

They found:

"On average, highway users paid $1.91 per thousand passenger-miles to the federal government over their highway allocated cost during 1990-2002"

"Passenger rail received the largest subsidy per thousand passenger-miles, averaging $186.35 per thousand passenger-miles during."

In other words, if a highway user and a rail user both travel 1000 miles, the feds earn $2 from the highway user while they spend $186 on the rail user.