Your measures are... outmoded.

Sure, cookies make things markedly easier(since data persistence is what they do, in a sort of feeble, hacky way); but there are so many more bits of information available if you want to fingerprint a user. Even better, the ones that squirm the hardest against the easy methods tend to end up with the most unusual configurations.

The astronaut is quoted as saying that it didn't taste like water from the drinking water supply. Could be that, once steeped in helmet-gunk and hair, it tastes different, could be from a coolant loop.

Neither nature nor nurture are on the side of free will(nor, indeed, is a universe that is deterministic or a universe that is stochastic).

The one plus side is that(unlike the current "Let's harass people we don't like; because they are suspicious and dangerous!") strategy, a biological marker(well chosen or not) will have the nasty-but-hilarious habit of showing up in all kinds of places. Social discrimination can be kept pointed safely downward; but if you start swabbing for DNA, anybody could end up having it, leading to entertaining collisions of policies that are tolerated only because they happen to unimportant people with people who are normally exempt from such...

"What if such a person could gain enough experience to control those impulses?"

You crazy bleeding-hearts! You'll probably be suggesting that inexperienced job applicants can be 'trained' and turned into valuable employees, rather than just being circular filed by HR, next...

Anything that reminds you of AOL Keywords can't be a bad thing, can it?

The words are pre-filled for all locations, apparently arbitrarily(with respect to their natural-language meanings), I assume that they used some cute math trick, maybe a hash of some flavor to munge (latitude, longitude) into a unique triplet selected from the dictionary, rather than assigning them fully randomly and having to store the whole collection, rather than being able to re-generate as needed; but they don't say.

Of course, for a low annual fee, you can buy a shorter 'Oneword'(tm) to advertise your place of business or whatever, so maybe somebody will give a damn about those...

"Without being able to look up the mapping from the database, the three words don't seem to be useful."

Exactly, consumer! Our awesome new system replaces those pesky, confusing, 'numbers' that hurt your little head and interoperate with basically any map, globe, or other geography system on earth, with three simple words that are meaningful only in the context of our proprietary service! Isn't it great?

Just think of the possibilities: will it be more lucrative to charge fees for service? Or maybe mine people's queries for marketable insights about their behavior?

Eh, phrenology doesn't discredit all attempts to put psychology on a physical basis, any more than phlogiston proves that chemistry is nothing but a mockery of a science...

However, even if they do find an atypically robust biological marker, I suspect that mumsy and daddy dearest are going to be disappointed:

"Violence" and "Spree killers" are really quite different things. Spree killing is a very, very, very, atypical expression of violence(there've been, what, ~20 school shooters in all of US history? Maybe low hundreds if you count all workplace and miscellanious killing sprees?), and spree killers, while they tend not to be wildly well adjusted(especially with the benefit of raging selection bias and a lot of hindsight), tend to have really banal records before their big event. Unless a test is unbelievably precise(both in terms of false positives and false negatives), the odds of finding who you are looking for, without massive false positives, are just terrible.

Violence/aggression in general are a much easier target; but it still isn't clear what use you'll make of such data(for comparison's sake, criminal records are probably a more robust predictor of future criminality than anything biological that we have, but that information hasn't exactly set the world of recidivism-prevention on fire). Unless you find the anti-violence equivalent of a statin(and even those aren't without controversy), just what are you going to do with probabilities?

We already do our best to do that by use of every other available data point...

Doesn't the phrase "There is concern that science may find biomarkers long before society can deal with its implications." carry some implication that society's ability to deal with such implications is actually improving, or at least might actually improve at some point?

On the one hand, I'm not sure why we would expect society to ever be able to 'deal with its implications'. Moral philosophy isn't exactly a progress-packed field, and people have been chewing on the issue of what moral responsibility does or doesn't mean in the absence of free will for centuries, without apparent result.

On the other hand, I'm not sure that having data delivered that we aren't ready to deal with would actually be all that unfamiliar: In the absence of good data, we don't sit, serenely withholding judgement until the facts become clear, we charge forward based on whatever scraps we have, held together with wild-ass guesses and whatever assumptions happen to flatter us.

Consider, the golden age of Eugenics, the late 19th to mid 20th century: aside from better recordkeeping, we barely had a clue beyond the vague selective tendencies that we've been using on plants and livestock for millenia. Did we let that stop us? Hardly.

Really, the biggest novelty of a (at present hypothetical) biological test is that it might defy our comfortable expectations about who the right sort of people and the wrong sort of people are. As long as it's just adding a stamp of 'objectivity' to the parol board's decision to deny some undesirable with an impressive rap sheet an early release, nobody will care; but once it shows up in little timmy from the 'burbs the hand-wringing will start.

I'm not a rocket surgeon; but I'd guess a failure somewhere in the PLLS. That component is supposed to condense and remove water vapor from the gas(as well as scrubbing CO2 and adding oxygen if needed) before reintroducing the fresh atmosphere into the astronaut's helmet.

Between a possible failure in the mechanism for removing condensation(which would cause the output to be alarmingly damp if you aren't expecting it; but would be a self-limiting problem since there just isn't that much water vapor available), there is the more serious possibility that the LCVG, or the heat exchanger that keeps the coolant water in that chilled, is leaking, which might actually end up being enough water to impair breathing(especially with surface tension but not gravity), or impair the cooling functions enough to threaten the astronaut's ability to function. If he's on a tether, they'd presumably just reel him in if he were to pass out from thermal overload; but a free walk would not be a good time to lose consciousness...

But, but, her first published letter to the editor was at age 6! And she's been involved in writing like 8 pop-health books with people who have 'MD' somewhere in their names and apparently don't have high standards! This is a serious expert we are talking about here.

Yahoo Health? Are you joking?

Unfortunately, we don't have one of those.

The UN(as its name might suggest) is representative(approximately, the details can be pretty ideosyncratic, and the Security Council is serious business) of Nations, not people. Given the revelations either connected to, or spurred by, about the spying programs various other countries(even the 'good guys', the fact that any 'bad guys' who can afford to do it are doing it has been known for ages), and other countries collaboration with the US spying program, do you feel lucky?

Heck, Mr. Secretary General himself, asserts that Snowden's 'digital misuse' has created problems.

I certainly wouldn't trust the Americans to operate internet infrastructure without spying on it; but the list of people I would so trust is Not Very Long(and none of them are in power).