1e-9

I would expect them to use statistical models to estimate probability of kill given the state of the target, drone, and environment at the time of a simulated weapon firing. Certainly not perfect, but you probably won't find many test pilots willing to have a live missile fired at them. Instructors are able to judge pilot effectiveness from simulated firings, so it should be good enough to evaluate a drone.

It seems like an entirely valid statement. AI can certainly make faster decisions than a human and it can utilize accurate 6DoF calculations and multi-sensor fusion that humans can only roughly approximate in an intuitive way during battle. I'm sure there will still be areas where humans will be better, such as analyzing complex visual information or predicting high-level reasoning of an enemy pilot, but I see human-versus-drone encounters as a clear loser for human pilots in the relatively near term.

A Tomahawk is a drone, just not the kind discussed in this article. A Tomahawk is subsonic missile designed to attack land and sea targets. The drone they are talking about here would be able to attack land, sea, and air targets. It would be vastly more maneuverable, faster, reusable, have AI specific to air-to-air combat, and I assume it would have greater versatility in selection of its munitions during an engagement.

Hits can be simulated.

> Based on his analysis, the single biggest effect on fertility is the education of women.

Although it's not surprising that controlling for education of women would provide a better model fit, it seems unlikely to me that this is the main causal relationship. I think logic and evidence suggest that the main driver of fertility is child mortality.

Humans appear to adjust fertility to maximize the probability that their offspring will successfully reproduce. Higher child mortality in a population tends to result in higher fertility as insurance against loss of a child [1]. Lower child mortality tends to result in lower fertility.

The human species would not be around today if we lacked a natural tendency to place utmost priority on offspring reproduction success. Parents must trade off the number of children they have versus the quality of care per child. Our current rate of child mortality reduction is shifting the optimal strategy to fewer children and higher quality of care per child. I see improved education of women (and men) as largely a consequence of this trend. I expect this will continue until disrupted by high-mortality events such as natural disasters, disease, or major wars.

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4411230/

Singing also appears to reduce snoring and sleep apnea by strengthening the pharyngeal muscles.[1]

[1] https://file.scirp.org/Html/5-2460085_31850.htm

You might find the Meanderful blog informative.

https://meanderful.blogspot.com/

That is a common viewpoint, but one I strongly disagree with. I exclude illegal disruptive orders, since those are addressed by market reg disciplinary action. An aggressing order that provides new information to the market is not harmful. On the contrary, it makes the market more efficient by definition. If a market maker has to widen their spreads or lighten their liquidity, it is a sign that they are doing a worse job than someone else of determining market price (again, barring illegal activity). If a market maker is doing a good job, then they should be thrilled to have anyone take their passive orders all day long, because it will result in greater profit for the market maker.

The markets need aggressive orders as well as passive orders. There is nothing inherently bad about liquidity taking. It is generally just as beneficial as liquidity providing. If anything, the misuse of passive orders is probably more common than the misuse of aggressive orders because passive orders can be used to influence a market without trading.

Even the taking of liquidity in the anticipation of a price move is not inherently bad. It is only bad if it degrades market efficiency (for example, by adding instability). An aggressive order that anticipates a price move due to an existing market inefficiency (for example, a sudden over-reaction to a report) benefits the market by speeding price discovery and adding stability.

As to your question of how an aggressive HFT strategy could benefit Vanguard, consider that if Vanguard places a passive order, it is because Vanguard hopes to attract an aggressive matching order. If they are unable to attract one within their time requirement, they will likely need to change their order to a less-profitable price. Their time requirement could be quite short during times of great volatility. An HFT strategy able to more-quickly recognize that Vanguard's original price was attractive would save Vanguard the cost of a price change. This benefit can be particularly large during significant market stress when participants are forced to liquidate assets in order to manage risk.

> It seems to me that the field contributes nothing to society,...

Better markets allow our economy to prioritize resources better and convert assets more efficiently. This increases prosperity.

Good HFT improves markets.

Of course, there is bad HFT, but such actors tend to not last long due to natural market dynamics and regulatory enforcement.

Yes[1].

[1] https://www.nature.com/articles/s41598-017-09388-7

Previous discussion: https://news.ycombinator.com/item?id=22809199

It appears that the lack of violet light contributes to myopia[1]. There is abundant violet light outdoors.

[1] https://www.nature.com/articles/s41598-017-09388-7

There's evidence it may also reduce snoring and sleep apnea by strengthening the pharyngeal muscles[1].

[1] https://file.scirp.org/Html/5-2460085_31850.htm

It probably confers survival benefit. Those who liked to hear dystopian tales were more likely to survive bad times and reproduce successfully. Thinking through how someone else handled a tough situation prepares one for similar situations in the future.

Slow market making is still market making, although rarely a good idea in my view. There are instruments with market makers operating at millisecond time scales. Not nearly as many as a few years ago, but they do still exist.

> The boss says to build a thing, you build the thing.

Sounds like a typical entry-level job to me.

> They're not making the world a better place; they're not providing a valuable service; they're just exploiting and extracting and skimming some dollars out of the market.

Firms providing useful services through their trading allow our economies to prioritize resources more efficiently. Better resource prioritization increases prosperity, which makes the world a better place in a multitude of ways. The useful services provided (for example, increased market stability) go beyond the seemingly-simple benefits of "increased liquidity" and "tighter bid/ask spreads" that are often mentioned. This becomes easier to appreciate once you delve into the enormous number and diversity of traded instruments and their interrelationships. It's far more than just cash equities.

Firms pursing parasitic trading are harmful to markets. Their behavior tends to backfire in the long run due to a combination of regulatory action and natural market dynamics.

Characterizing all (or even most) HFT firms as harmful is inaccurate in my experience.

I went through a short beat-the-casino phase for kicks. You can increase your edge significantly beyond standard card counting by targeting flaws in casino randomization methods. Randomization is hard. For example, card shuffles aren't random, not even mechanical ones, and particularly not human ones.

It is possible to carefully observe and model a variety of randomization processes at different casinos, find a weakness in a particular game at a particular casino, and construct a profitable strategy that is nearly impossible for the casino to anticipate. Even so, they will eventually notice someone who is a consistent winner, assume you are cheating, and force you to play a stressful game of cat-and-mouse that will lower your expected profit, increase your variance, and lead to a less-than-desirable lifestyle, in my opinion. A team can improve these things at the cost of adding management and human resource headaches.

If you have the talent to succeed at this, my experience is that the same talent can be applied in much more productive endeavors that pay far better and yield a much higher quality of life.

Yes, the E-mini S&P 500 Futures March contract (ESH0) did hit a Velocity Logic circuit breaker at 9:13:12 PM ET, soon after Trump announced the travel restrictions. The halt lasted for 10 seconds and then trading resumed. It's down 4.4% now and is getting close to the low price limit for the day, which could halt trading completely until the CME lowers the limit at 9:30 AM ET.

Fat and air float. Muscle and bone sink. Your overall buoyancy depends on your particular combination of lung capacity, fat, muscle, and bone. Also, fresh water is not as dense as salt water, so it is more difficult to float in.

Decomposition produces a variety of gases that increase buoyancy.

We are far past time for a successor. The x86 family has been an architectural mess since the 8086. It's time for someone to provide an alternative that motivates the market to dump the baggage and start with a clean slate focused on security and efficiency, as opposed to compatibility. Unfortunately, I see nothing on the horizon. Neither ARM nor RISC-V qualify in my view. We need something more radical that has dramatic advantages.

The researchers discovered that two particular proteins appear to degrade health in worms and mice by harming mitochondria. Worms, mice, and humans all produce these proteins in increasing amounts as they age. The researchers improved the health of aging worms and mice by suppressing the proteins. They found reason to believe this might also be possible for humans.

Sure, there's an imperfect competitive framework in place. It can be improved. Regardless, AMD failed to compete effectively in that framework over the last few years. The fact that AMD now has a superior offering attracting significant orders implies that competition is possible in spite of the unfairness. AMD probably would have done much better previously if they had produced better products.

I whole-heartedly welcome AMD's new offerings and have completely switched to buying servers with AMD CPUs, but I have to disagree with the characterization of "ruthless milking" by intel. There was no instituted monopoly here. Others were free to compete, but they failed. Intel's high prices motivated AMD to create something better and now they have. If anything, the high pricing was a screwup on intel's part and good for the consumer. Intel could have better protected their lead by charging less (but not as less as now) to discourage competition. I'm glad they didn't, even though it cost me for the last few years. Now that there is a valid competitor again, maybe we will see better conformance to Moore's Law.

Edit: To clarify, I will not argue that intel has a clean record of competitiveness. My point is that they didn't "ruthlessly milk" the consumer with high prices. Instead, they stuck a knife in their own back by charging excessively high prices without sufficiently innovating. This created an opportunity for AMD and motivated them to create something better.

Perhaps it was due to the ability of cotton to absorb up to 27 times its weight in water. It seems like water-saturated cotton would provide significant evaporative cooling, which might outperform wool's protective properties for short periods of time.

It seems that sleep pattern variation is an evolutionary consequence of the survival benefit of having someone awake when others are sleeping [1]. A good combination of night owls, insomniacs, and early risers in camp probably made it far less likely a sleeper got hauled off by the hyenas. This also seems to be the reason that anxiety (sense of danger) reduces our sleep. There's a part of our brain that doesn't differentiate well between today's worries and predation risk. For this reason, I think that addressing anxiety is often the best way to improve sleep.

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4972941

Harder problem => Increased productivity variance

For extremely hard problems, it's not difficult to measure greater than 10X differences between highly-talented senior developers and the average productivity of those capable of meeting requirements.

Most negotiation strategies attempt to optimize deal terms. This article makes the point that when post-deal relationships are important, it is better to optimize a combination of deal terms and quality of the post-deal relationship. It concludes that a softer approach better achieves this broader goal by yielding a much better relationship even though it might be at the cost of worse deal terms.

I largely agree with this, although the researchers omit the important tactic of developing an acceptable alternative deal with a counter-party's competitor and then focusing negotiation on comparisons to it. Done well, you can still drive a hard bargain in a soft way, while maintaining the relationship by diverting the counter-party's resulting negative feelings away from you and towards the competitor.

You are describing a recursive Bayesian approach, which can have significant computational and storage advantages for filtering (for example, Kalman filters). For this to work well, the prior must be able to adequately represent the learning of the posterior, which may be practical with a self-conjugate prior or a Monte Carlo approximation such as what particle filters use. In practice, for nontrivial machine learning applications, self-conjugate distributions rarely model the problem well and good approximations of the posterior into a concise prior are rarely practical.

> We should ask, “what evidence are you providing that your priors are any good?”

This is valid. Anyone pursuing a Bayesian approach should be asking themselves this question about every prior they use. To fully benefit from a Bayesian framework, one needs to construct models with understandable parameters for which there is some sound theoretical or practical insight that can be embedded with priors and that is not well-represented by the training data. Doing this can help your solution avoid the kind of wildly unpredictable and costly mistakes you might get if you used a completely blackbox approach. For critical applications, this can be highly useful. If you can't come up with priors that are clearly beneficial, then you are likely better off using a non-Bayesian approach.