tpudlik

I don't think people choose Python because it's virtues as a language for tooling (it's not great from that perspective, statically linked binaries would be better). They choose it because of the ecosystem of libraries. For data analysis, visualization, and scientific computing, there are no comparable offerings in any other language. And a lot of this is stuff you _really_ don't want to be reimplementing from scratch, because it's very easy to introduce bugs (around numerical stability, say) that cause your software to produce correct results 99% of the time, and plausibly-looking but totally incorrect ones 1% of the time.

Getting images of simple molecules or lattices seems tricky.

Did you mean to say it's _related_? The original "spiral" image by Ugleh is explicitly credited in the "Related Links" section.

Thank you for the prompt, that's very useful! I'm surprised GPT "knows" what Monster of the Week is. RPGs in general are somewhat obscure, and to ask for a specific one at that?

I agree with all these criticism. But I also know many human DMs who make the same stylistic decisions as GPT did here. It's definitely not good Burning Wheel, but it might be ok D&D?

On the other hand, I appreciate the GPT's willingness to follow the players' revealed interests where they lead, instead of sticking to a predetermined railroad. The comment here about engaging zombies in a multilevel marketing scheme illustrates this nicely. That's very deft DMing in my opinion. (Whether _asking_ for such a tonal shift is deft playing on part of the non-DM player is a different question.)

Yeah, maybe there's a confusion between percentage and fraction, and this was meant to be 0.001 (not 0.001%). But if so, the error is not the journalist's: I clicked through and the original paper also contains this improbable figure.

Re 2: this is not true if you're doing margin or futures trading, which was FTX's main business (and may be Binance's). This type of business is what characterizes an "exchange" (as opposed to a non-margin brokerage or a custodial bank).

Yeah! Its deepest point is only 10 m shallower than the deepest point of the Baltic Sea!

Relatively little energy is actually used for heating and cooling. In the UK, a country pretty far north, it's about a fifth of energy use (https://www.withouthotair.com/c7/page_53.shtml), and could be substantially reduced if heating was actually done using electricity (i.e., with heat pumps, rather than direct heating; heat pumps are available with "efficiency" exceeding 400%, e http://www.withouthotair.com/c21/page_140.shtml, but of course require electricity to run, while most heating today is done by directly burning fuel).

Seasons are definitely not "the biggest obstacle to solar". That would be the night and especially the weather.

There must be a little more to this, since clearly they are not a wholly owned subsidiary.

I bought and played it after reading this piece. It's a small but enjoyable game; if you're at all intrigued, I'd encourage checking it out!

Looks like the fundamental problem is that a substantial number of users are perfectly willing to sign up for providing fake likes, as long as they receive fake likes in return. A tragedy of the commons.

Interesting. I use Pigma Microns for line drawing, but never used them for writing. They are lovely!

He was the CEO of Waze, but not of Google. After the acquisition, he had to adopt Google promo processes.

He does have a passion for automation, but where's the dissonance? He recognizes it in the very opening, with the xkcd!

I believe they are talking about using Zoom to call friends/family from your work laptop.

You can export to BiBTeX (.bib).

That (nicotine carrying non-minimal health risks) is Myth 3 in the linked article.

I recommend reading it, it's very informative!

I see this as a win-win, too, but in a different way.

Early on, the US companies cut their costs: good for them, and likely their customers. In the long term, China developed the expertise needed to contribute to the development of advanced technologies: good for them, and for humanity's aggregate progress.

There's a lot of acrimony over how to divide the surplus, but this shouldn't overshadow the technological and economic development happening before our very eyes.

This is all true, but I sometimes missed the nuance of less witty but more detail-oriented histories. Harari could seem flippant at times, which mostly amused me when I agreed with him, but left me worried when I didn't. In a book this short on a period so vast, there was no space to steelman anything.

Lengthening the second is unlikely: for every physical phenomenon with a duration (from the rotational period of other planets to the frequency of atomic transitions), you would forever after have to wonder, is that value quoted in"new seconds" or "old seconds"?

Redefining units leads to ultimate backwards compatibility problems.

Abstract of a talk with the same title by one of the authors (Ju Sun; coming up at Duke University on February 5th, 2018):

Many problems arising from scientific and engineering applications can be naturally formulated as optimization problems, most of which are nonconvex. For nonconvex problems, obtaining a local minimizer is computationally hard in theory, never mind the global minimizer. In practice, however, simple numerical methods often work surprisingly well in finding high-quality solutions for specific problems at hand.

In this talk, I will describe our recent effort in bridging the mysterious theory-practice gap for nonconvex optimization. I will highlight a family of nonconvex problems that can be solved to global optimality using simple numerical methods, independent of initialization. This family has the characteristic global structure that (1) all local minimizers are global, and (2) all saddle points have directional negative curvatures. Problems lying in this family cover various applications across machine learning, signal processing, scientific imaging, and more. I will focus on two examples we worked out: learning sparsifying bases for massive data and recovery of complex signals from phaseless measurements. In both examples, the benign global structure allows us to derive geometric insights and computational results that are inaccessible from previous methods. In contrast, alternative approaches to solving nonconvex problems often entail either expensive convex relaxation (e.g., solving large-scale semidefinite programs) or delicate problem-specific initializations.

Completing and enriching this framework is an active research endeavor that is being undertaken by several research communities. At the end of the talk, I will discuss open problems to be tackled to move forward.

There is much more redundancy in terrestrial cables (because they are vastly cheaper to lay), so they're much less of a single point of failure.

Not exactly the same concept. Need-based admission requires need-blind aid, but not the other way around: especially for foreign students, many good US colleges need-based aid if you get an offer, but are less likely to admit you.

The idea of truth-seeking is laughable. You are required to switch sides each time.

I'm with you, except for this bit. The side-switching is the primary way for truth-seeking to enter debate. If you only look for arguments for one side, you're a demagogue. Truth-seeking consists of searching for the strongest arguments for both sides. Most people struggle to argue at their best for both sides simultaneously, so debate sequences them.

This depends on what the alternative is. For people who can't afford an American sticker price for a car, the choice is between a less safe (but cheaper) car and the near-suicidal experience of riding a motorbike.

Selling "unsafe" cars may actually save lives.

Please note that the study defines any office space shared by more than 6 people as "open plan". Also, although there's a (statistically) significant difference between 1 person offices and "open plan", there is no significant difference between 2 person offices and"open plan".

I think both you and the original commenter agree that the "intangibles" make a significant (perhaps decisive) contribution to the heritability on outcomes. But that's where you both disagree with the article, which argues for the importance of "tangible" factors (growing up in a neighborhood with exclusionary zoning; receiving an expensive college education; etc).

Like the OC, I've seen the dramatic impact of the "intangibles" in my own homeland, a former Communist state. Many of the leading businessmen and professionals today are descended from pre-communist elites, even though these elites were not only deprived of all their material wealth but actively discriminated upon during the five decades of communism (for example, by receiving "class background" penalties for university admissions). The idea that "money breeds money" is intuitively plausible, but turns out to be largely an illusion.

It sounds like you agree not just with the overall premise, but with the article's specific recommendation, which is to "start with a serious cost benefit analysis and then require localities to refrain from any new regulations without first performing cost-benefit analyses of their own".

Very few self-taught people contributed to research in the 20th century. I can't speak with authority about mathematics, but in my own field (physics) the only self-taught researchers I can think of were people trained in allied fields (math, chemistry) who ran into physics problems in the course of work in their own discipline.

The root cause, I suppose, is that it takes much less time and effort to just get a PhD than to make an original research contribution, so most people get credentialed along the way. Academic programs also immerse you in current research work, making it possible to figure out where you could make contributions in the first place.