I dunno. I think there probably needs to be a kind of emotional literacy. I always told my kids when they were growing up, "Listen to your feelings, but don't believe them." The reason is you need to be aware of how your feelings are coloring your judgment before you can even apply any critical thinking skills you have.

"I am a high status individual because of my superior genes."

I never said it wasn't rambling nonsense. But it works for him.

I don't think papers with bullshit *results* are common. I think what's common is papers that aren't very interesting where that fact is obscured with language you could describe as bullshit, because it's intended to shape your attitude toward a boring result.

An interesting result is one that challenges expectations or common sense assumptions, but any program that pursues such results will *inevitably* produce mostly boring results. Still, boring is not the same as useless; it's important for someone to check things. It's far better to be able to cite a boring result than it is to rely on an common sense assumption, because sometimes common sense was wrong.

This is actually effective communication. You just have to see what is being communicated.

We were taught to write as if we were destined to write legal briefs or user manuals for some complicated product. As students we were judged on our ability to relate facts and concepts clearly and to arrange them into a memorable or persuasive structure.

A human doing this is like a dog walking on its hind legs. You can train any dog to do that, but it's not natural. Our paleolithic hunter-gatherer ancestors evolved a set of cognitive skills that *could* construct or critique a persuasive essay, but that's not something they ever did or had use for. I think human communication started with simple stuff like this: be afraid of that animal; the tribe on the other side of the hill is bad; I am one of you; I am a strong leader; don't mess with me. In other words things that communicate emotions and attitudes towards things. Stuff that goes right from the language centers of the brain into the reptilian brain without critical processing.

In other words, I believe bullshit is the original form of human communication. It remains far more prevalent than we are aware of, because we seldom *study* it; it's a perfectly natural behavior for us to produce and consume it. And it remains incredibly powerful. Even something that's complete gibberish when translated into a fabric logical propositions can do a potent job of painting an emotional picture of who is in the tribe and who is out; who is on top of the tribe and who's at the bottom.

"Leveraging" is corporate speak.

But then you come the occasional landmark paper, and suddenly the writing gets a lot better.

I've puzzled over this: is it cause or effect? Does a landmark paper get cited because it's better written, or does it get better written because the author expects people to read it? I think it's effect. When an author knows he's got a winner, he dispenses with the frippery. With pressure to publish to pad your CV, most papers are half-baked or nothing burgers, and the lack of substance is routinely disguised with inflated language. Clear writing comes across as cheeky.

I think it's become a kind of fashion. My company was submitting a research proposal in conjunction with a well-known lab at Harvard. I took it upon myself to edit the final proposal draft, so I streamlined the overblown academic prose from the Harvard group. The Harvard PI told me to revert it back. He acknowledged that my version was better writing, but the proposal reviewers were used to bad academic style and if we didn't write the science parts that way we'd look inexperienced. The technological part of the proposal on the other hand had to be well-written because it was stuff the reviewers wouldn't be familiar with.

Well, yes, the 2015 Nobel Prize in medicine was awarded for the discovery of ivermectin *to treat worm infections*. Ivermectin is absolutely a great drug for a helminth infection. That doesn't mean it's sensible to jump to the conclusion it's good for viruses.

Anti-parasitic drugs like ivermectin *are* perennial candidates for anti-viral use because they have potent physiological effects and often shut down viral replication in a petri dish. But that's a very low bar to pass; in the human body the dynamics of the drug don't necessarily allow you to achieve those effects without dosing so high it would harm patients more than it helps. That's an absolutely normal result when testing a drug that shows promise in vitro.

It's not ivermectin fear-mongering to say that it's not a cure-all. Research into ivermectin for viral diseases is ongoing, and perhaps it will work for some of them -- dengue is apparently a promising application.

If you're a spy, people who have access to a broad array of sensitive information are the most attractive targets. This is either a senior non-technical officials, or they technically sophisticated junior workers -- your Chelsea Mannings and Edward Snowdens.

Each of these types of people represent a qualitatively different vulnerability that requires a different approach to securing. You don't have to tell a low level tech with access to lots of highly sensitive information to use end-to-end encryption, he probably understands that better than you do. What you have to look out for is a growing dissatisfaction with working for what to a clever young person will tend to look like a brain-dead, hypocritical bureaucracy. Show me a young person who has climbed to a position of trust in a system based on his technical talents and there's a good chance he doesn't think much of the system.

For the high level apparatchiks, they *are* the system, and while you can't always take loyalty for granted, the biggest vulnerability is lack of technical sophistication. They absolutely are unable to judge what is a safe way of communicating, you need to give them a whitelist of approved methods.

That's kind of the point -- not having to talk to other people.

This product seems designed to take some of the effort out of organizing and analyzing information, but that effort is important to engaging your brain with topic. Just firehosing documents into an AI and letting it do the work is certainly easier, but it really makes you less relevant. The same goes with interacting with human beings. It's a pain in the ass, but it's necessary if you're going to engage with other people.

Like Ireland, Germany and Italy in the 1800s and early 1900s. The KKK in the 1920s was bigger north of the Mason Dixon line and was mainly aimed Irish, Italians and French immigrants. But people really underestimate the power and attraction of becoming American, which was what all those people ended up becoming. We're still America, but with pizza and spaghetti and corned beef on St Patrick's day.

Subject matter expertise is, however, quite a difficult point. While I think most of us would agree that critical thinking is independent of subject matter expertise, it doesn't mean that I can blythely take my universally applicable critical thinking skills and confidently overrule the judgment of subject matter experts.

So if the dermatologist say the thing on Aunt Betty's arm is a harmless skin tag, and I, armed with Google and my critical thinking skills say its cancer, who should Aunt Betty believe? The doctor, or her idiot nephew Googling symptoms?

But here's the problem: sometimes you *can* be more right than someone with more expertise than you. I've actually been the idiot telling the doctor he was wrong, and ended up being right. The attending physician thought a family member was having a psychiatric episode, and I, a non-doctor, thought it was encephalitis. And in fact it was my critical thinking skills that allowed me to spot the errors she was making and challenge them. To her credit, when she couldn't answer my objections she brought in a neurologist who immediately confirmed my suspicions.

Logic really isn't the problem. In real life, sure, sometimes people affirm the consequent or something like that, but the big problem is evidence -- gathering it, evaluating it, and weighing it. Real life situations are full of red herrings and bad data and it's always necessary to pick some data to focus on and not others. The flip side of this is that it's always possible to obstinately challenge experts even when they're *right*. But they're not always right, or even justified.

So really the argument you need to have isn't about logic, it's about evidence is included and what is excluded, understanding that including and excluding is a necessary process.

The supply is fixed, therefore the value is unstable. Under varying economic conditions this can lead to either deflation or inflation, which is why Bitcoin sucks as a currency. Denominating a contract in Bitcoin adds an element of risk and uncertainty you don't get with fiat currencies, which can be managed to have stable or at least predictable future value.

Bitcoin is a shitty currency, because people treat it as an asset. They're *hoping* for the value of Bitcoin to change in a substantial way, which is what props up its value. If people bought Bitcoin for spending, its value would be far less.

Criminology and geography too.

Sociology has been in something of an upheaval; ironically though conservatives have embraced the strong programme of sociology of science because they reject what they see as the privileged position of scientific consensus.

You really have no idea what you're talking about.