arghbleargh

Sounds intriguing, but after reading through the website for 5 minutes, I still have no idea how this is supposed to work.

From the pictures, it looks like they've set up some kind of system to translate certain physical movements into changes in the projected images. It seems like a non-technical user can only make superficial changes within a limited set of pre-defined behaviors. But from the description it sounds like there's a lot more to it than that...

I find a lot to agree with in this article, and it explains it pretty well, but the part about the Federal Reserve is a bit too simplistic. The value of money is determined by a complex interplay of factors, including government spending/taxation and international trade. The Federal Reserve has only limited influence on it via manipulation of interest rates. Nevertheless, the main point stands: the strength of fiat currency lies in the implied social contract that measures will be taken to ensure your savings today are not worthless tomorrow.

Looks very interesting! I don't completely understand how to use it (and I guess not everything works yet?), but the UI looks nice, and it looks like it has some really cool features.

I have a project along these lines as well: https://github.com/zhaizhai/turnbase . That's a somewhat cleaned-up version that I put on github recently; I actually started it almost five years ago. You can see an example implementation of Battleship here: https://github.com/zhaizhai/turnbase/blob/master/games/battl...

I once had visions of a "code-blocks" interface like you're doing, but you've certainly gotten a lot farther than me on that front. I'd love to find out more about what you're doing and if any of my code can help (will send you an email). I don't actually have time to work on this project seriously, so I'm happy to share whatever experience I've gained so that it doesn't go to waste :).

Crucial question whose answer I didn't catch while reading the articles: how do you determine who is considered a "scientist"?

Not to say that the thing about men vs. women isn't problematic, but that quote is taken out of context. The suffrage part is a completely separate bullet point and isn't saying that women shouldn't have the right to vote as your quote suggests.

Also, the main point preceding the quote that men and women may have different natural dispositions is a reasonable one. He only oversteps by implying men are better leaders, a claim that may be defensible under some very specific definition of "good leadership" but is mostly just inflammatory in this case.

This seems like a tricky case because if the users and drivers are being completely rational, it's not clear that Uber's deception should have affected their behavior. Sure, the drivers expected to make X% of fares, but realistically they probably estimated what that would come out to be not from looking at customer fare data (in which case the deception would have materially harmed them) but rather by tracking what they or other drivers were actually getting paid.

Still, if the allegations are true, a culture where it is fine to engage in such active deception is not a good sign.

50,000 miles/year * 20 years = 1,000,000 miles does not seem reasonable...

Isn't it simple enough to make three bounds checks to essentially do the same thing you did with the zero padding? You can even make it a function

    is_in_bounds(x, y, z)

It makes sense to control pedestrian traffic in some intersections, but it just seems like they're not even trying to optimize pedestrian throughput. What makes the lack of feedback for crossing buttons worse is the fact that many of them are so conservative about letting you cross that you question if they are working. I encounter a lot that will refuse to let you cross if the light is already green, even if there is still plenty of time left. Unfortunately, lowering pedestrian accidents looks great, while saving pedestrian time is hard to measure.

Wait, so will Sage still be open source?

Yes, I think this has to be a major reason why people distrust science, and frankly I am sympathetic to it. Too many studies amount to a 2% statistical difference (with an analysis rife with biases) with no causal mechanism (or a highly speculative one). Relying on this type of evidence, while not worse than pseudoscience, still drastically weakens your position when trying to argue against claims like "vaccines cause autism". (Remember when fat was bad?) Also, even for the examples given in the piece, I think the truly "scientific" view has to be more nuanced:

- vaccines cause autism: OK, we have no reason to believe that they do, but we don't know for sure that nothing in them can cause autism. But even if they did, we believe the effect must be quite small.

- people are safer owning a gun: The science can't possibly say whether a particular person is safer or not owning a gun. But it does suggest that you may be underestimating the dangers of owning a gun.

- genetically modified crops are harmful: So far they have not really been harmful, but we have only a vague idea of the likely consequences down the road.

- climate change is not happening: Well, it's obviously happening, but a lot of similar questions are not clear. There's really no way to prove for sure that it's caused by humans or predict its consequences. We should still worry about it because of the potential consequences, but at the end of the day, we are still mostly working with educated guesses.

At the end of the day, even when science is wrong, I think it tends to give answers that are much closer to the truth than e.g. random intuition. So that's why some people arguing against pseudoscience try to paint the actual science as absolute truth. They're using the ends to justify the means. But their arguments are sometimes over-reaching, and that can ultimately make them lose credibility.

I think this highly depends on what you mean by self-taught. Probably a majority of good mathematicians are "self-taught" in the sense that they learned most of the things they know from reading books or thinking on their own rather than taking classes. What you won't see commonly is someone who has a day job but becomes really good at mathematics on the side. This is probably because of the sheer amount of training it takes to become what is considered "professional-level" in math. It takes quite a bit of background before you can even hope to do something that hasn't been done before. The same is not true for programming.

There is a linear programming formulation of the shortest path problem: see https://en.m.wikipedia.org/wiki/Shortest_path_problem#Linear.... Indeed the vertices of the simplex method would correspond to paths, but I think running the simplex method on the dual formulation is more akin to Dijkstra's algorithm. Actually, I think the simplex method on the dual and Dijkstra's are equivalent, although I did not work through the details carefully.

As someone who has done a fair bit of both math and programming, I will agree that having some math knowledge has definitely helped at various times when programming, but I don't think these specific instances (e.g. the 10 examples given in the article) constitute the primary reason it's a good idea to learn math. For the purposes of handling those specific situations, I think you would do fine to just learn the math as you encounter the need for it (and 95+% of the time you don't need anything beyond high school).

To me, the primary value of math for programming is that it is a pure form of exercising many of the skills that help you write good code. For example:

- fluency in logical reasoning

- turning vague intuitions into precise statements (translating business requirements into code)

- formalizing proofs (covering all cases, establishing invariants)

- developing abstractions to succinctly describe relationships

- solving a problem systematically

You can also learn these skills by programming, of course, but probably not as quickly, because you will be distracted with other tasks such as debugging, setting up your dev environment, waiting for your program to run, etc. So in my opinion, what's important is not so much the immutability of the math itself, but rather the process of doing the math. I would be interested to know if others have had similar experiences.

While you are correct in pointing out that these contests only test a relatively narrow subset of the skills needed for good programming, I don't think it would be correct to say it's based on "rote learning". There can be some amount of pattern-matching, but it's not as if you can just read the solutions to 100 problems and expect to do well.

Also, I know for a fact that many (but not all) people who were good at these contests are also good all-around programmers. (In fact, you probably do better at these contests by writing more readable code, even accounting for the time cost. But some people can get by with messy code as well.)

The notation is indeed a bit non-standard (we would generally use intersections nowadays), but it's not completely ad hoc. If you think of P_n as the indicator function of the polygon (i.e. P_n(x) = 1 if x is in the polygon and 0 otherwise), then it makes sense. (There is still a little bit of sloppiness from using P_n to denote the area of a polygon as well as the polygon itself.)

It should be noted that from the original paper, the asymptotic formula that Oliver and Soundararajan conjecture still says that each possibility for the last digits of consecutive primes should occur about the same number of times in the limit. It's just that the amount by which the frequencies vary is more than you would expect from the most naive model of primes as being "random".

Providing opportunities for kids to learn coding is great, but it's possible to go overboard with it (teaching pre-readers?). For example, I don't think it should be considered one of the core subjects that gets taught throughout K-12 (math, science, English, history).

Let's face it--being able to program anything interesting actually takes a tremendous amount of hard and often tedious work, and most kids won't be able to get over the hump. Think playing an instrument: can be very rewarding if you're good, but most kids don't want to grind out all those hours of practice. If it's problem-solving skills that we're trying to teach, math or even puzzle games engage those modes of thinking more directly, more generally, and with a smoother learning curve.

Does anyone have any insight into how Skyscanner differentiates from the many other flight booking websites (Kayak, Expedia, Travelocity, etc.)?

It definitely seems true, in my experience, that one-on-one tutoring achieves educational outcomes far superior to traditional classroom learning. However, I am wary about the ability of technology to scale this up. One-on-one tutoring is more than just providing a customized curriculum and the expertise to answer any questions the student has.

A huge part of learning effectively is developing the right habits. The startups in the article are not really focusing on that, probably because it's the type of thing that people usually learn indirectly through social interactions (e.g. a child emulating their parent). I also believe much of the value of a tutor lies in the social connection with the student, providing them with someone to model their learning habits after as well as someone to keep them focused and motivated.

I think this social aspect of teaching is very hard to capture with technology, sort of like how it's hard to learn a language without living in a country where everyone speaks it. Technology can help, but I feel it will be more like a 0.2 sigma improvement rather than 2 sigma.

When this kind of device is being used outside, does it actually have a noticeable impact on the air quality? Purifying enough air to fill Madison Square Garden sounds impressive, but it's nothing compared to the volume of air that cycles through the atmosphere.

"He just wants to invent good things for mankind,” said Ahmed’s father, Mohamed Elhassan Mohamed, who immigrated from Sudan and occasionally returns there to run for president.

Wait... run for president?

Interesting that engagement seems to be waning from a lot of Evernote users (based on this anecdatum and the article). Is it because the product has gotten actively worse, or do you find less need to use it these days (e.g. you realized it was more hassle than it was worth)?

So... can somebody explain what the difference between "essential" and "non-essential" Flash content is, and how Chrome can tell the difference?

I think the article agrees with you, but another point it makes implicitly is that programmer A is also on the hook for creating a brittle abstraction. The article is specifically warning against the temptation to deduplicate things for its own sake, without having in mind a good abstraction that describes the shared functionality implemented by the duplicated code.

It’s worth noting that none of these approaches fundamentally change the N^2 nature of the work to be done, but do substantially reduce work at reasonable levels of contention.

I don't think that statement from the article is true if I understood correctly... the performance gains are precisely because you're reducing from N^2 to N log N.

An interesting theoretical question would be to identify the optimal backoff policy. I think FullJitter should be close to optimal, but maybe you can squeeze out a little more with something more sophisticated.

Edit: I just realized the DecorrelatedJitter (not sure why it's called that?) makes a lot of sense as a minor optimization because if you already waited a long time and still failed, that suggests there is a lot of contention, and you should wait even longer.

I played with Legos a ton as a kid, and I have a completely different view on what made Legos good. Building Legos with only basic bricks can be fun, but there's only so much depth to it. It was actually the special pieces that enabled you to make really interesting things, and I didn't feel like they stifled creativity at all. Not having the special pieces would be like cooking with only flour, butter, salt, and water. You can probably make a few tasty things, but I don't think you would want to eat that every day.

Here's an interesting small extension of the problem: prove that it is not possible to guarantee survival if 64 is replaced by some number that is not a power of 2.

In the same vein, an easy way to get the tests to pass is to add a bunch of "if" statements to handle each test case. Of course, that is counterproductive in terms of the usefulness of the software.

However, because so much weight is put on testing, that is essentially what happens in school a lot of the time. For some reason, it seems more obviously absurd in the software context.

What about Sandbox doesn't suit your needs?