lewis500

I write little academic papers for a living as a professor. I think this is good advice on how to write like cormac mccarthy. More important is to pick an academic writer whose journal articles you love and copy them. One of my favorites is Paul Krugman. He’s not in my field but apparently he’s good enough that people want to read him. He uses lots of semicolons and punctuation. I understand where this advice is coming from, though. Lots of academics are truly horrible writers and believe complication itself makes their writing erudite.

Me too. I think that because of the meta programming capabilities we’re going to see a lot more development in that direction. Recently I started using Turing.jl, which uses a sort of declarative syntax that was easy to write and understand.

I just thought it might be nice if you did a breakdown of the physics of traffic and maybe how you coded that part. As for teaching, I think it could be used to teach about like signal timing etc in a fun way.

Do you have a writeup of the traffic simulation rules that you're using? I peeked in the code and saw the intelligent driver model is being used for acceleration. I suppose you also have a lane-changing model and such, too.

Your architecture could be very useful in teaching.

I agree I would recommend industry to almost everyone. In fact, engineers ask me about grad school several times per year and I almost always discourage them. Too many people think of grad school as sort of a vague "next step" that will "open doors" or else a way of "leveling up" as though life is an RPG and a grad degree is a special badge or skin you can get. I encourage them rather to think concretely about what academia specifically entails: reading abstruse papers, writing papers with little chance of being appreciated, debugging software, writing grants, giving presentations, etc. Academia is only appropriate for a rare type of person, like being a classical musician.

I don’t think they’re excuses. They’re explanations of how things happen. The asymmetrical relationship especially explains how professors can get away with treating students badly, but it isn’t a justification. If you read Bad blood, for instance, E Holmes and sunny treat everyone horribly, but ultimately lots of people quit because they have other options.

The pressure on the professor and the selection for obsessiveness explain motives.

I’m a professor. I have a few thoughts about this.

First, the grad student/professor relationship is inherently asymmetrical, much more so than most employer/employee relationships. The degree has to last years or else you get nothing, and there are no hard and fast terms of the relationship like there is in many jobs. The professors recommendation might be the most important thing you get, and that is always running in the background. So many professors treat their students like robots.

Second, professors are essentially super students. We became professors because we have extreme standards. Your engineering manager became a manager because he was an engineer and got promoted, but having a professor as a boss is like if a famous open source developer was your boss. Nothing about becoming a professor selects for kindness or empathy, unlike a manager at an ordinary company who might be trained and evaluated—-especially in a large company—-on the happiness of his workers. So a lot of professors have essentially obsessive attitudes and no empathy. Add to this the fact that many students are escaping countries where they have little future. For example, the past few years, the Iranian economic crisis has caused a surge in Iranian applicants.

Third, the professors themselves are under extreme pressure. Everybody knows this.

Fourth, the work a professor outsources to students tends to be the stupidest and worst tasks, because the students usually aren’t good enough to make big contributions. This is depressing for the students: they’re smart people who signed up to work with a famous professor on work they cared about, but their day to day is to do things anyone could do for almost no money.

Finally, the professors are under pressure to create these huge labs. We don’t make profits, so things like funding and number of PhD students are taken to be metrics for evaluating. The US news ranking absurdly includes phd’s per tenure track professor as one of their metrics. So a lot of people are accepted who really shouldn’t be there, and the research is designed to have tons of busy work.

If supply and demand matters then it would seem employment does not exist because of societal wish for people to justify their existence with madeup jobs, but rather because of many independent decisions between workers and employers. This was my point.

“We keep inventing jobs because of this false idea that everybody has to be employed at some kind of drudgery…. He must justify his right to exist.”

This is a framing designed to get attention but is obviously empty. "We" don't "invent" jobs. There isn't some big council of job inventors we all sit on with the purpose of giving everyone a right to exist. People and organizations decide to hire people for various reasons, and people sign up to be hired---usually so they can make money. Jobs are agreements between people. This is true even of government jobs in a democratic society: people are hired by various government agencies which, with a couple of jobfare exceptions, hire people for reasons of their own---not because of a mandate to make up jobs for people. This is a plea to an imaginary decision-maker.

Why isn't it a flying three-wheeler?

This is the type of decision-making the authors of the Green New Deal need to grapple with. E.g., the Green New Deal calls for high speed rail everywhere. But when you actually try to build HSR people in affected communities file lawsuits and force decisions that increase the cost, slow down trains and delay construction. For the Green New Deal to actually happen, the federal government would need to override local decision-making.

The weird thing is that, even though a lot of policy is oriented toward making you a conscious consumer (high deductibles, coinsurance) in many cases you just can’t do it,at least not very well.

My wife got an allergy test at a local allergy clinic, but she had 20% coinsurance so we tried to figure out how much it would cost. Everyone there treated us like garbage. Even when we called the billing department in advance they told us there was no way to know in advance how much the basic allergy test would cost until she saw the doctor. So we went to the clinic (stupidly), and the nurses just acted like we were crazy. Finally one did tell us what the bill would be:$12k. I had researched these tests online and was floored; that’s way more expensive than it should be, and the doctor who gave her a referral had told us it would be a reasonable price. So we were in the awkward position of just leaving the appointment. The doctor then basically talked to us somberly like we were the poorest people on earth, and said “listen, most of the test is for allergens that don’t exist in northern California.” If I just give u the local ones it’ll be like $1800.” I started laughing. Later on I read an NPR story about a lady who had a$48k allergy test from there. Just a complete scam, much worse than a car dealer because at least there they will tell you a price and they will not sell you ten extra useless cars evenif you don’t explicitly say “don’t give me ten superfluous cars for no reason.” Anyway, we got the limited test, and at the end the doctor said “listen I’m not supposed to do this, but given the financial hardship I’ll make an exception. If I don’t put the results in the system you’ll only be charged for the visit.” And he just wrote the results on a yellow notepad paper and tore it out for us.

Tldr we were treated like annoying busybodies for asking the price at an in-network facility, and then treated like indigents when we balked at paying way more than anywhere else charges, for totally bs procedures.

Normally when markets have so much unsold inventory, prices fall to clear it out. But prices can be "sticky" and won't go down. All this unsold inventory makes you wonder why the prices are so sticky.

Here you can see one reason: "Shanghai homebuyers came out in droves to protest a developer's decision to cut prices in an apartment complex."

I'm sure there are others. Like if you're holding a loan for a giant apartment complex, and the owner cuts the price, it becomes immediately obvious that the loan is not going to be paid back in full. But say it's a 30-year loan: if you get the owner to keep the prices high and the units unsold a little long, maybe you can unload the loan onto some other bank or investor before the reckoning comes. Or maybe you're a bureaucrat who will be fired if the prices go down.

I have also heard that in China it can be difficult for person to rent out a vacant property. So people are holding onto empty units just as a store of value, like bars of gold.

It’s inaccurate to say that we “can’t” escape it. The federal reserve has been raising interest rates to slow inflation and growth in employment, and for years they have paid interest on bank reserves at the fed and constantly talked up plans to unwind their portfolio (selling off assets). If they had done less of all that we would have higher inflation.

Same is true of some other central banks. I always am frustrated when I hear about japan not being able to raise inflation rates much. If their activities have no impact on inflation at all, why not buy up all the securities and foreign exchange in the world by printing yen? The Japanese could live entirely off the dividends because of this magical property of their central bank.

I think this mainly makes a difference with regard to reading the code. I'm looking at a struct in someone's code and ask myself "How does this thing get used?" With classes I'll be able to see its methods right away, and if it had inherited then I could see right away which parent to go check.

But the problem is not that big. Maybe the methods are right next to the struct, or if I search the codebase for the struct I can find which methods use it. Still, would be cool if Juno had some feature like "show which methods use this."

> Do colleges exist to promote education and intellectual curiosity or to train workers?

I don't really think the teaching part at most universities oriented toward training workers. If they were oriented toward training workers, then the administration would be worried about what skills are in demand by employers...probably training students on particular software, computer languages, jargon and techniques that are common. There would be more emphasis on getting various certifications while still in undergrad. I can't speak to promoting "intellectual curiosity" because it seems hard to define and if you did I bet it would depend mainly on the nature of the student...probably his/her genes, peer group and childhood role models...not on the actions of someone the student sees three hours per week during adulthood.

As it is, a professor can (very often, though not always) be totally incomprehensible and irrelevant to work, as long as they are not offensive, without any repercussions nor even any gentle communication that he/she explain things clearly. Humans tend to craft stories favorable to themselves, so even if the student evaluations are negative, the professor can just blame e.g., short attention spans in the era of marijuana and fortnite...or the professor just might not look at the student evaluations if they don't want to. Over time, some professors even develop a sort of perverse joy at confusing students...their confusion being the sign that the professor is very smart and that he is "challenging" them, when in all likelihood he is just asking them things he never explained clearly in the first place. In my undergrad, an EE professor bragged that the class average was a 23%; he also used his TA as a translator because he could not speak english.

It is expected most everywhere that students will learn work skills on the job and in internships...not at school. Professors at research universities spend all their time thinking about hard frontiers of a topic, and that frontier might have little to do with what you do on the job.

wow this is brutal. thanks for sharing. I had mused on the idea of pitching them an opinion piece about my field, traffic, but not anymore.

Why does it say "homes" when the first sentence says "urban housing stock"?

Why would a country want to buy up only shares of companies in that country? You could have a large sovereign wealth fund, but most countries are very small and it would not make sense to only buy domestic stocks.

Don’t get me wrong. I think the actual activities parents get their kids to do can be wonderful experiences, especially if you undertake them on your own. What I’m skeptical of is that these experiences hydraulically translate into success as adults, which is what the parents are often trying to accomplish.

Latex lets you write equations quickly. Whether that makes it worthwhile to learn depends on how many equations you write and how quickly you want to be able to change them.

I also like being able to write my papers in vscode, with all the shortcuts and vim settings that I’ve gotten used to from coding. If Vim is second nature to you, it’s frustrating to have to switch over to ms word. Vscode has a good latex plugin.

That said, it did take some getting used to. Anything hard and potentially useful can become a “fetish” whereby we fool ourselves it’s the usefulness that motivates us but really it’s the hardness. I recall a time when I tried to use matplotlib to format all my plots, with labels and such. Horrible experience but I was really proud of these ugly graphs I’d make entirely from code. Then one day i just started saving them and opening them in illustrator for all the bells and whistles. Haven’t looked back.

Looking back on my own childhood, I'm skeptical how much anything alleged to "enrich" the children actually made a difference. I recall most of all piano lessons. When I was a kid, every kid had to learn piano. Today, no adult I know ever plays piano; they forgot how because they don't like it and no one wants to hear it because it's boring. I know one serious professional musician. He taught himself guitar for our high school punk band and got super interested in guitar technique.

Now, you could say they learned "discipline" from the piano lessons. But how do we know? I don't see any relationship between who took the most piano and who became a doctor, etc. My view of piano lessons, and other enrichement, is that they are sort of homeopathy for parents' anxiety about their kids' future. Even when there is nothing to do for someone's medical condition, lots of people will turn to quack treatments out of anxiety, and these quack treatments involve a host of immeasurable concepts like "toxins" proposed to imitate the kind of causation we know of for better-known treatments. Same with piano lessons: your parents are too anxious about you to sit on their hands, so they make you practice piano and tell themselves, and each other, that you're learning "focus," some magic property that will stay in your brain one day even when they're not around. I don't buy it. Nobody made their kids take piano because they read a bunch of studies proving it makes your kid choose to go to med school or whatever; they clung to it automatically because they wanted to believe in it.

I don't understand why Musk is so invested in the idea of cars in tunnels. Does he really have a substantially better tunnel boring machine/tunnel boring workflow? If so, there are lots of things to use it for other than car tunnels. It's kind of like if he started making electric cars and then said, "I am launching the world's first all-electric pizza delivery service."

The O'Hare airport connector in Chicago is a better example of what cheap, fast tunneling could be used for than the car skate mechanism. This article points out how remarkable it would be if things go according to plan: https://www.citylab.com/transportation/2018/06/for-1-billion...

There is a network effect. Imagine if you made a ridesharing app with only one driver. No one would use the app due to high wait times. That's an edge case but it gets the point across: too few drivers and no one will use your app, and if no one uses the app no drivers will enlist in your service.

I’m obviously joking about wishing they got bad results.

I keep trying to get my semi-retired parents to change their lifestyle, but unfortunately they keep getting blood tests which say their cholesterol and blood pressure and such are perfect. It is all the more frustrating because my dad smoked for about thirty-five years, and according to his veins it's as though he never smoked at all. I am thinking of asking the doctor if he can fudge the results.

In all seriousness, it's probably genes, but I've also started to suspect it's due to coffee. They each drink about twelve cups of coffee per day. In the past few years, there have been more and more studies suggesting coffee is surprisingly good for you.

Another advantage is that they both worked for themselves; they never had bosses making them do excruciatingly stupid things for years on end.

The subsidy ratio here is incredible. Alabama has used subsidies to lure automakers and other factories, but our incentives are peanuts compared to this.

For example, we are getting a $1.7 billion Toyota-Mazda plant for $350 million in incentives, creating about 4k jobs. The state has a 4-5% income tax and 9-10% sales taxes, so I'd bet that, even if the 4k number is exaggerated, then the state and local govts will earn much of that money back within a few years (provided it doesn't create some pollution disaster).

Would've been good if, in spite of just being text and a few buttons, it had been built with react, redux, redux-saga and had some contorted webpack build resulting in five JS bundles.

"This revolutionary theory – his Liquidity Preference Theory – is still considered too radical to be acceptable today."

What? Liquidity preference was taught in my undergrad macro class using the textbook by Greg Mankiw. It's the basis of the IS/LM model. And central banks are constantly worrying about interest rates and liquidity. To whom is it too radical? Probably some people disagree with it, but I wouldn't call it very "radical" today. The author is fighting straw men.

Congestion pricing is certainly do-able, and it would have the significant advantage of charging all vehicles rather than just TNC's. But it is technologically more complex to install readers everywhere than to tell TNC's to vary the tax they already pay. In other cities downtown congestion pricing has cost a couple hundred million to set up, and it takes some time.