speedplane

> Everyday people honestly don't understand why prices are what they are. I've had a number of conversations with people and they truly believe that prices are arbitrarily set based on how much owners want to screw customers over.

Generally, a company will charge the highest price they can get away with, irrespective of the cost it took to produce the good or service.

Prices only approach the cost of production when there is strong competition among industry players. Basic tactics like branding, consolidation, and lack of transparency in an industry help subvert competition.

> It’s like nobody understands supply and demand

Folks that use supply and demand principles to argue against UBI don't seem to understand themselves.

If everyone is given $1,000/mo, it does not necessarily follow that living costs will also increase by $1000/mo. For that to happen, the additional $1000/mo would need to actually increase everyone's demand for services, which is generally not the case. Having more money generally does not make you want something more, it just makes it easier to afford it. If more people can afford something and start purchasing it in greater numbers, then yes, the price will increase, but they are also getting something the thing of value in return.

Put another way, giving everyone $1000 will likely increase prices, but the increased prices will be less than $1000.

> 30 million adults in California x $1,000 x 12 months = $360 billion / year. 20% of adults are on Medicaid though, so if they're excluded it's $288 billion / year.

This is the real fear that progressives have about UBI... that it's a replacement for government services. Converting government healthcare into an equivalent check will effectively take away healthcare from millions of people.

PSA to folks who answered this: if you have a profitable side project, double down and try making it into something more than a side-project.

If you've built something that is actually collecting revenue, you're already ahead of a huge number of folks out there, and you can likely build it into something that can at the very least make you live comfortably, if not much more.

> New versions of Java have made great progress in the memory area.

Elastic uses huge amounts of memory (30GB is the norm). That's not a problem by itself, but Java garbage collections at this size can often take 10 seconds or more to complete on modern hardware. During this time, the server is basically down, the only solution is to increase timeouts or use replication servers.

> from a production perspective, I think security should be topic one

Two general approaches to security:

- Upgrade to a paid Elastic cluster, and use their own feature-full security suite.

- Put a reverse proxy server in front of Elastic (like nginx), and configure that to handle security.

> Great to read from someone who knows it so deeply.

I'd say it's a decent read on Elasticsearch tuning at the intermediate level, but not enough to really get high performance from Elastic.

One of the problems with Elastic tuning is that the tuning parameters depend deeply on the type of data you're indexing. Indexing text is far different than indexing numbers or dates. A mapping with a large number of fields will behave differently than with just a few. Some datasets can easily be split up into different indexes, and others are cannot be.

To really get the most from Elasticsearch, you have to know what it's doing under-the-hood, and how that maps on to your data. Elasticsearch hides so much complexity (generally a good thing), but unfortunately, it can be difficult to know where the bottlenecks are.

By now, tt seems pretty obvious that (1) there exist craft that are propelled by some physics that we are not aware of and (2) someone in the government or defense industry knows far more about it than the general public.

> You think for example the North Korean government happily tolerates the black markets there? No way.

They most certainly do. There are tons of articles based on satellite based cameras that track the rise and fall of small black markets in North Korea. Some markets lasted years, before being abruptly shut down for unknown reasons.

> “Since I left Secret I feel alive and aligned with my values and my purpose again.” > I wonder what his values and purpose are.

Few people know the founder and whether he's actually a good person or a fraud, but we do know that: he was the recipient of huge VC largess, created an app with shaky moral foundation that caused actual harm, and then shut it down after lots of bad publicity.

Not reasons to totally discount a person, but good reasons to be skeptical.

> Tell your manager. Wait four weeks for effect. You may ask for progress reports if there is no improvement. Leave.

That's a fine plan, but a simpler one is just to leave. It's exceedingly rare that someone in a toxic work environment speaks with their manager and then everything becomes okay.

The law is not behind or antiquated in this case. Bank cybersecurity has been regulated for quite some time now, and failing to adequately secure your digital assets is a compliance violation no different than failing to catch obvious fraud.

It's very likely that your bank is based and regulated by New York State, even if it isn't physically based there. Contact the NY State attorney general's office, they should take you seriously.

> There are plenty of ways to increase capacity on a railway line especially if you can build new rails, but at some point it just gets too difficult.

While trains are touted as very efficient modes of transportation, and generally beloved by nerds, I'm skeptical, largely because train tracks are so under-utilized.

Train tracks take up as much space as many multi-lane highways. On highways, dozens of cars go by every minute. But a train car only passes through at best a few times an hour. I suspect that if you took one mile of train track and one mile of a highway with the same overall footprint, far more people/goods will move through the highway-mile every hour than the train-mile.

Don't have data to back this up, just a hunch. But if it's true, then even though trains are far more energy efficient than cars, they're likely far less real-estate efficient.

(Also, I do suspect that this is not always true. A heavily trafficked tunnel in/out of Manhattan's Penn Station is probably more efficient real-estate wise than an equivalent tunnel. But that's a very extreme case with super high densities).

> skirting the law and abusing the commons gives one a competitive advantage. But that doesn't make it OK. The market isn't the arbiter of ethics.

Sadly, this is a bit naive, market have become the arbiters of ethics, namely: if it successful, it's good.

This isn't a traditional moral code from, e.g., Kant, Bentham, or Locke. Rather, it's an ethics based on capitalism.

Sure, wonky economics professors justify capitalism by saying it's the most efficient or provides the most utility, arguably grounded in traditional Bentham morality.

But capitalism has never simply been an economic policy, there has always been an ideological component to it (see for example, Ayn Rand). Maximizing returns on investment isn't just "efficient", it's "right".

> black and grey market activity are also I think quite an important counter-balance to the possibility of totalitarian government.

I don't really buy this: plenty of totalitarian regimes have illegal black markets, that governments often tolerate because they don't see a threat to their power.

A better way to counter a totalitarian government is via open communication and coordination of the government's subjects. Government's are far more afraid of this, which is why strict censorship and propaganda are the "go-to" tools for any totalitarian regime.

> It's a funny one because [Tax Evasion is] potentially one of the most serious crimes. You're stealing from everybody, undermining the basis of trust in the entire system in which we live and probably in the long term eroding the basis of democracy. You're creating a non-level playing field.

This is true, and in addition, it's a very unequal crime. To properly evade taxes and get away with it, you need professional help. Most standard W-2 employees can't afford it, i.e., for most, the cost to hire professional help evading taxes is larger than the taxes they could save.

However, if you're very wealthy, then the calculus changes: saving 20% taxes on $50M/yr income is a lot more attractive than saving 20% on $200k/yr income. Effectively, the wealthier you are, the easier it is to evade taxes, further exacerbating at least the perception (and quite likely the reality) of inequality.

> Problem: We make a lot of CRUD apps. > Solution: The OS should do it. > > Ehhh... Why is that the obvious solution? We can't decide on the right way to do it in user space, why does moving the problem to the OS help?

The article is indeed ridiculous. An OS should not do everything. Hardware storage resources are generally the memory, disk, and network connection (and if you're getting really deep, the cache and registers). A good OS should only provide access to those resources as efficiently as possible across a wide variety of hardware.

There is a vast myriad of ways of utilizing those resources, and it would be a fools errand to implement a one-size-fits-all approach. The better approach is to provide access to the resources, and let higher level software developers build on top of them.

A disk only database is far different than a disk database with a memory cache, is far different than a memory only database, is far different than any collection of the above coordinated via a network connection. Further, storing text is different than storing images, which is different from storing video, which is different from storing JSON or XML.

Pushing everything to the OS will often give you worse performance, locks you into a single OS vendor, and slows down innovation from third parties. Bad idea.

>> So if for example I know that my uncle is not paying all his taxes I'm a criminal too unless I report him? Or is it only for bigger frauds?

> Yes if the government can prove that you knew about it without telling, you can be charged with it.

That's not generally true. In the vast majority of cases, simply knowing about a crime and not reporting it is not a crime in itself. You only become liable when you aid-and-abet or join in a conspiracy to commit the crime.

However, it depends very much on the specific crime. Tax evasion may be entirely different than say selling drugs.

> to justify colonialist policies, Europeans engaged in motivated reasoning exaggerating the technological advantages they did in fact possess over other cultures ... and using them to justify claims of superiority not merely material but also intellectual and genetic. (A modern form of this is making fun of flat-earthers, anti-vaxxers, creationists, Apollo conspiracy theorists, Juggalos, and 9/11 truthers, and occasionally engaging in more violent rhetoric toward them.)

Nice try equating colonists with conspiracy debunkers, but the analogy does not hold up. The problem with colonists wasn't their arrogance or personal sense of superiority, it was their violent exploitation of others for their own benefit.

Sitting in an armchair smugly looking down on others isn't very nice, but it's far better than getting up from your armchair and murdering/enslaving/exploiting the people you look down upon.

People who debunk conspiracy theorists use argumentation, humiliation, and appeals to authority. Sometimes they do so improperly. But they aren't colonizing anything.

> react does not play well with non-react javascript

I'm work on a somewhat older project (circa 2015) that started without react, and have also found that to be the case.

End-all-be-all web frameworks seem to be the rage now, but I'd prefer to see a solid library of micro-frameworks that can be stitched together (like jquery plugins, but better).

Also, the idea of a web server transferring over a huge javascript library and app just to load a single page seems wrong. There has to be a better way, a better amalgamation between server and client.

> I’m wondering if Wilbur Ross is cheering this virus on. This is like a gift from the Heavens to the Trump Administration

The coronavirus will definitely hurt China, but I don't see how it will help the U.S. in trade policy. If anything, it'll just delay any real trade agreement the U.S. and China could make as China devotes it's political attention to the virus.

> This is an industry founded upon the principles of innovation and invention, is it not? Perhaps tech can invent a union that can overcome issues found in the unions of the past.

Throughout human history, unions are one of the few ways that ordinary citizens can actually stand up against concentrated power (democracy, revolutions are some other ways). Rather than trying to invent something new, maybe study the past a bit more.

> Do we always live for ourselves? Do we never try to live up to expectations of our family or friends? Or those who believe in us, being on our side in life?

People can of course change over time and be influenced by people around them. However, entering a relationship with another person based on the presumption that you can "change" them is a recipe for relationship disaster.

There are billions of people on this planet. If you're looking for a relationship, you're better off finding one that suits you rather than someone who doesn't and trying to change them into something else.

> How do you know I’m not your neighbor? I’m middle class in the tech world and built an airplane.

My dad was a pilot and engineer, and we built many things together in my childhood (go-carts, computer video games, an automotive speedometer, more).

But there was one thing that I always wanted to do with him: build an airplane. As a 13 year old, I found different plans from vendors, created a budget, put together a basic timeline, but could never convince him to do it.

In fairness, building a plane is a huge multi-year commitment, and almost certainly more expensive than buying a used Cessna. But the price is not completely out of reach for many folks (<$100k). The real draw is the amount you learn during the process and seeing your handiwork.

I'm still sad we never did it, but it was a big ask of a dad with a full-time job from a twerp son. As a consolation however, he did buy a small wing-section kit and we got to rivet a few panels together over a weekend. First time I ever used rivet fasteners!

> Facebook and other large successful companies try to only commit crimes when they know the profit from doing so exceeds the risk of being caught and punished.

While this statement is often true (big companies make these cost-benefit decisions), I don't think it really captures what is happening to Facebook. Doing a cost-benefit analysis like this requires having data on potential costs and benefits. In most cases, Facebook didn't have this data.

When an individual, company, or institution is faced with a decision, but lacks data to properly evaluate it, they fallback on other decision making tools, including instinct and principles.

Facebook almost certainly didn't know the consequences of the many decisions they made, they didn't know the potential costs or benefits. But they did have some underlying principles, namely: maximizing platform engagement, convincing people to share more personal data on their platform, and exploiting user data for better advertising.

With these foundational principles, you can end up making some really bad decisions even without a proper cost-benefit analysis.

> Serious question, why is this a metric? Changing your mind should be a good thing. ... why is consistency such a highly regarded metric?

It's not about consistency, it's about being right consistently. Unfortunately, most decisions whether made by an individual, CEO, or politician have to be made with imperfect information. The data just isn't there to do a full analysis. In data's absence, when you need to make a decision, you rely on instinct and foundational principles.

Of course, when the data becomes available you can adjust, as anyone should. But when you have good enough instincts or principles, you can operate effectively without good data, and that makes for a good leader.

> Just because you cannot detect something does not make you wrong nor does it mean it doesn't exist. What a ridiculous statement.

If you cannot detect something, it's can be equally ridiculous to say it exists.

> There's a difference between lying about one position and making entire linkedin profiles that you set up. One's mischievous and one's deception.

As a human, you can only bring a few things to the table: inherited wealth, demand for your talents, and credibility. The first is a gamble, the second is impossible to control given the market, your credibility is the only thing you can really guide.

Be honest, be trustworthy. It takes forever to build, but in the end, it's how people make judgment calls about everything else.

> The marginal cost of deprecating something or breaking something is higher than the incremental cost of workarounds.

So why are we in this python 2 to 3 mess?

> No amount of async/await or goroutines can solve this problem. A non-blocking web server is a godsend for I/O-bound tasks

In the past, we were told that threads were cheap and to use them heavily, especially to achieve parallelism. Now with the advent of async models, we're being told that threads are expensive, and often that a single processor/thread async model is better than a multi-threaded blocking one.

I'm not a luddite, I do agree that async is often better. But I wonder how we got tricked into thinking more and more threads were the answer and how we avoid such trickery again.