throwawaymath

It drives me nuts when I see words like "undoubtedly" thrown around so confidently this way. A new book comes out about Simons and Renaissance, it enters the financial zeitgeist for a little while, and now everyone is apparently an expert on the firm's differentiating competency.

For what it's worth, what you're saying is contradicted by Nick Patterson. He did not say that Renaissance had access to clean data no one else did. What he said is that in the early days, they spent almost all their time cleaning the data. In any case, that's table stakes these days. All successful quant firms spend time sourcing exceptional data and ensuring it's as polished as possible.

The problem isn't hyperrationality per se; it's a lack of intellectual humility and self-awareness. The behavior you're talking about arises when someone looks at a problem, immediately wants to fix it and makes snap judgements to figure out how to do so. Instead, they should dispassionately consider 1) whether the problem is fixable in any meaningful sense, and 2) whether they have the requisite experience, skill and insight to solve it in five minutes of thought.

You could further distill this idea with the observation that a lot of people consider things too systematically, rather than holistically. "First principles" thinking is a powerful drug, which appears to crack open the world and solve every problem. But no problem-solving paradigm can deconstruct away unknown unknowns or inexperience.

But obviously that's not enough to reproduce everything they do, or else they wouldn't still be legendary. As it happens a lot of what makes them successful is not the sophistication of their trading algorithms and research, but also the sophistication of the execution and reliability.

As an aside, from friends there Renaissance hires researchers in three primary ways:

1. They have a small network of professors who they solicit for promising new PhDs willing to leave academia each year.

2. They watch professors and postdocs in specific specializations, and reach out to those whose research meaningfully interacts with a thesis they're interested in internally.

3. They send small groups to conferences to poach people working elsewhere in industry (particularly tech) whose work is applicable to their own.

They also do hire people who directly apply of course, but most hires are reactive. They especially like to hire people whose work or research looks like it might begin to encroach on their own, or is just notable and impressive. The math is certainly important to them, but that's just one dimension of it.

Reasonable people can disagree about whether Dorsey should be CEO. But I think the author's concerns are more varied and more cogent than your rebuttal would make them seem.

What are your thoughts on Dorsey leading two tech companies at the same time, Twitter's share performance over the past several years, and the political waters Twitter has to navigate?

My proposal attempts to activate a level of empathy and self-preservation. A tolerable amount of trauma is induced in potential drivers to get them to treat it as a necessary evil, not a rite of passage, a right or an enjoyable activity.

I think I understand what you mean when you say it sounds authoritarian, but I disagree. It's not authoritarian, it's just a more stringent set of rules on a gating function which already exists. I guess I would agree it sounds dystopian because of the VR simulations that would be involved. But then, it's a measure intended to curb tens of thousands of deaths each year, thousands of which are due to distracted driving in particular.

But it's the only way to actually get someone's attention sometimes. If you live in an urban area you should know what you're signing up for. People are going to honk, even if it's annoying to the rest of us living in the city.

To combat that, we could induce accelerated experience using VR simulation.

1. Force people seeking a driver's license to undergo classes, regardless of their prior self-education. In my opinion, a significant amount of irresponsible and incompetent driving stems from the early failures in parents teaching their kids properly. People not only don't learn everything they need to, but they don't respect driving as the dangerous activity it is.

If someone succeeds in obtaining their license, they must then re-test every five years to maintain it, regardless of their age and driving history.

2. Augment the new mandatory classes with virtual reality lessons. The VR headsets will ostensibly drill driving skills before they're practiced in a real vehicle.

But the real reason they'll be used is to force candidates to experience high fidelity simulations of hitting and killing people while in the vehicle. They will be exposed to the immediate trauma involved in hitting a family in a sedan, or killing a pedestrian who was adhering to the rules of the road. It might also be helpful to force them to watch footage of people being seriously injured and killed in vehicular collisions.

This recommendation comes from two hypotheses of mine. As the article states, people overwhelmingly know distracted driving is dangerous, but they can't help themselves. I believe this is because 1) they have no way to activate empathy for the potential danger they present to others, and 2) they do not take driving seriously enough. It is treated as an innate right with loose rules, not a dangerous activity sustained only out of necessity to keep urban society functioning.

Unfortunately, this proposal will never happen. I know parts of it are extreme, but I don't see an alternative for fixing such a systemic problem which isn't extreme.

It's materially different to have an X multiple of an NYC salary versus an X multiple of an average US salary, for all values of X.

Yes.

> We settled on an amount equivalent to how much I’d earn as a US-based engineer working 6 months, but then multiplied by X for the traction/success factor already achieved. And I also still have a stake in the project going forward.

Working where in the US? Software engineers can make wildly different amounts of money depending on where in the US they work. Even within SFBA and NYC, a subset of companies pay significantly more than others as well.

So why not just say the actual acquisition number?

Good example of this, in Python 3:

    >>> (0.1 + 0.2) + 0.3
    0.6000000000000001
    >>> 0.1 + (0.2 + 0.3)
    0.6

Most of the time I see comments like this, people don't provide hard examples of queries they found unsatisfying with one search engine compared to another.

Scalability and profitability are orthogonal. If it could scale indefinitely, you'd be right. But no trading strategy can scale indefinitely.

That doesn't say anything about whether or it "works", and it's not a reason to be suspicious of the results, in of itself. All successful trading strategies are capacity constrained.

EDIT: Looks like this paper is also already included in the list of citations for the 2019 paper :)

The historical record overlooks the people he hired who knew a thing or two about trading, while fixating on the team of NLP scientists he hired from IBM. Likewise Simons wasn't initially successful in the very, very early years. It wasn't until the late 80s that the Medallion firm really came into its own.

Does anyone else have good examples of popular scifi tropes which have formed an engineering zeitgeist, but which are realistically very suboptimal?

The person you're responding to is correct. It's an explicit design goal that a fully homomorphic encryption system would not expose any distinguishable oracle about the underlying data. Otherwise there would be no point to it whatsoever, because you'd just be performing the same computations on the data dramatically less efficiently and without any benefit.

This follows the general imperative of cryptography, which is that the outputs of cryptographically secure primitives (hash functions, pseudorandom generators, pseudorandom permutations, etc) should be computationally indistinguishable from random up to 2^n queries, for some large n (such as 128).

It's not uncommon for even undergraduate linear algebra to cover abstract spaces and notions of linearity which generalize beyond R^n. For example, the function space P_n consisting of all polynomials with degree less than or equal to n. Hoffman-Kunze, Halmos, Axler and Friedberg-Insel-Spence are all examples of undergraduate textbooks which cover this material.

This isn't just theoretical. Function spaces like P_n are useful in applied mathematics. And even if you don't use function spaces, it's very common for engineers, physicists and applied mathematicians to work in the complex space C^n rather than R^n.