Aw, man. This essay gives me some perspective. I ended up hating physics by the time I got my bachelor’s in it. I remember junior year they said, we know you’re confused, you haven’t actually learned REAL physics yet, that’s what grad school is for. And most of you won’t be accepted, and of those who are, most won’t become professors even if you want to.
I figured a physics degree would allow me to get a good job, but then they said, sure, if you want to do finance, defense, or education. What?! The first two were pretty nauseating so I decided to do a year-long internship at various local public schools, and realized that I hated that too. (More the bureaucracy than the actual teaching aspect.) My school basically expected you’d double major in math or CS, so they often didn’t teach a lot of the fundamental techniques you needed, assuming you knew it already. I also learned that while my high school had done a great job teaching me English and history, the math department was not quite up to snuff comparatively, and I felt behind my classmates about “fundamentals” like linear algebra, which they had somehow been taught in high school. Sure, I knew Shakespeare a lot better than them, and even minored in literature in college (maybe biting off more than I could chew) but that clearly didn’t impress them. Why waste your time with old books when you could be making money! It definitely brought me down a notch and made me feel dumb, but I also realized college wasn’t a place for someone to learn lots of different things, but to focus on one area instead. I ended up becoming a bartender and I’m very happy with that decision. At least, once things are back to normal, hopefully within the next few months. Hopefully I didn’t just jinx it
I look back on college, and so many classes seemed like a waste of time. I understand a well rounded education, but so many classes seemed like fluff. I once took a Abnormal Psychology class. It sounded interesting being premed. For half a year we sat in a circle, and talked about feelings. I was so shy that was the last place I would open up. I used to think, If I was honest, I would admit I was still a virgin, and had a deathly fear of public speaking. I loved the teacher though. She really meant well.
If I had a kid, I would supplement his education. I would make sure he knew mathematics well. I also think physics should be taught much earlier, along with the finance, and the markets. As a kid, I was interested in woodworking, cars, engines, even electricity, but the teachers had us coloring world maps?
Counterpoint: some of the weird classes I took in high school / college ended up being useful, or at least more useful than some of my core CS classes.
I took a Psychology class in high school, and now I understand why people feel and react certain ways, because I can connect back to the concepts I learned. I took a music theory class in college, which is helping me make my own music as a hobby, and also appreciate some songs.
In contrast, I am researching programming languages in graduate school, I took an undergrad class on PL, and it barely taught me anything I didn't already know.
Bottom line is: we need a system to teach people more things that will probably end up useful and less obscure crap, but it isn't just "teach people whatever their job is and whatever they need to know as an adult, and ignore the social sciences and history because those will never be useful".
Best class I ever took was Dinosaur Natural History. A fluff class if there ever was one, but very hard to get a seat in. I really loved the class, but thought it would never really add anything to my life. Because, well, Rexy has been dead a while.
I was so wrong.
What I learned in the class was mostly the applied basics of biology, geology, and evolution. And presented in such a fun way, I gobbled it all up. Those basics later lead me to a successful career in bioengineering and a dalliance in carbon geology.
I went to a small liberal arts school (Ursinus) outside of Philly, who I only mention because it was unique in that part of getting your degree required taking CIE, or "Common Intellectual Experience" classes. Freshmen were required to take CIE I and CIE II, and the classes were scheduled such that it not only tried to mix students up between different disciplines, but they were also taught by faculty from every department, just as much a requirement for them as for the students.
The classes focused primarily on classic texts, literature, philosophy, all the big religious works, etc. The point, really, was to foster discussion and to get everyone on an even playing field their first two semesters in terms of intellectual discussion, and really picking apart what you're reading, preparing you for the world of research in academia.
A key point here is that I don't know a single person, student or faculty, who didn't feel that CIE was extremely worthwhile set of classes, and a pivotal part of our time there.
That said, after I & II, there were CIE III classes that would satisfy different electives. One of the best classes I took was CIE III - Capitalism and Democracy. It was taught by one of my math professors, a man I deeply respected and learned a ton from, and I was one of only four students in the class.
The semester was basically split up into 5 or 6 chunks, where each chunk took two countries whose basis of government encompassed (even if it was stretch), capitalistic and democratic ideas. So, we'd read papers and book excerpts from economists and political theorists (think Adam Smith, Keynes, Friedman), along with summaries of the countries, and each chunk culminated in a report comparing the two countries implementations of capitalism and democracy, trying to find the intertwining of the axioms laid down by the theorists, and how those two systems of governance and economics evolved in different political spectrums.
I was a math and comp sci double major, and am a firmware engineer today, and never dipped my toes anywhere near economics or anything politically related before or after that class, but it's really stuck with me after all these years.
> Bottom line is: we need a system to teach people more things that will probably end up useful and less obscure crap, but it isn't just "teach people whatever their job is and whatever they need to know as an adult, and ignore the social sciences and history because those will never be useful".
I agree with you here, and mostly because we all exist and make decisions in a historical, experiential, and psychological context, and you have to have some knowledge of contexts that are different from yours if you are to have any hope of being an effective leader.
As Matt Yglesias put it, the first article makes 0 sense. If they don't like what the school is teaching them, pull their kids out! Isn't that the whole point of private school? It's a choice to go there.
It's amazing how conservatives write love letters about the "free market" but despise absolutely anyone in this "free market" who do not follow conservative orthodoxy.
It's a PRIVATE SCHOOL! Send your kids elsewhere if they don't teach the kids enough Ayn Rand and Michelle Malkin!
None of these conservatives seemed to give a shit 15 years ago when rural PUBLIC schools like mine were teaching anti-science creationism, or historical revisionism in the Lost Cause/Civil War apologia. Many were shrieking their heads off if their kids weren't taught such utter bullshit.
I agree that we should teach math earlier, in elementary school the same math (fractions, percentages) were literally taught repeatedly from 3rd-6th grade. I would have loved to learn some algebra instead. The concept of doing the same thing to both sides of an equation to solve for an unknown is not terribly difficult to grasp.
As for finance, I’d rather be coloring world maps. It’s given me a great desire to travel the world. Finance makes my stomach hurt. Then again, I’m poor, so what do I know? I agree we should be taught more about it than we usually are, which is that we aren’t taught anything at all about it.
I remember learning algebra between 3rd and 6th grade. It was hidden within word problems, but 7 was exponents and everything after had a Cartesian plane
Your last statement reminds me of this rather brutal take on the US's (and really the whole world's) educational systems, reposted to HN a few days ago: https://www.cantrip.org/gatto.html.
The main idea is that if you look past the things we're told that school is meant to be for, school is pretty clearly meant to turn humans into compliant drones.
> I look back on college, and so many classes seemed like a waste of time. I understand a well rounded education, but so many classes seemed like fluff.
Even though I believe well rounded interests are important, I felt the same way as you when it came to the distribution requirements. For the most part, people took first year courses that were more about rote memorization than depth of understanding because they were easy. Some university departments even offered special courses that watered things down even further.
For the most part, it was a good experience. Things were more lively when a room had a mixture of Science and English/History majors, but that only happened because a few of us were willing to go onto second and third year courses outside of our major. On the other hand, it did feel like it was setting us back relative to those who stuck to the easy course in order to focus upon their major.
>>> My school basically expected you’d double major in math or CS, so they often didn’t teach a lot of the fundamental techniques you needed, assuming you knew it already.
Indeed, it was some decades ago, but at my college, there was substantial overlap between the requirements for the math and physics majors, to the point where just a few additional credits got you both. And it was a time when CS was still being taught in math departments at some colleges, so there were courses that you could take for either CS or math credit.
I actually came from the other direction, started as a math major and added physics, which I then studied in grad school.
In my view the affinity between physics and programming (and electronics) is no accident. Physics was computational before the computer age. We ran out of problems that could be solved in "closed" form, early in the 20th century. The Manhattan Project employed a staff of "computers" who performed calculations on mechanical calculators and punch card machines. Scientific programming predates the software industry. Today, virtually every scientific instrument is computerized, every experiment is automated. My PhD thesis experiment ran on thousands of lines of code.
Every physics student learned to program. What they didn't do is pay attention to turning us necessarily into good programmers. Since I didn't take any CS courses, I had to pick that stuff up on my own, and have continued working on improving my programming technique to this day.
> My school basically expected you’d double major in math or CS
Coding knowledge is becoming a matter of literacy for more and more professions, so it might be worth trying to pick up a bit of it before it's too late - lockdown might be an especially good time.
I'm nearly 32, I took introductory coding classes and got straight As but I can't seem to get past more complicated projects. I'm not sure if my brain is wired for it, I have a very hard time finding bugs that are sitting right in front of me, and thinking about loops for too long makes my head hurt. I've certainly tried, I'm considering getting a Mac for dev because Windows is so painful to try and learn any coding. Seriously, I'll copy and paste stuff and it just won't run or it'll give an error on Windows, every single time.
> Seriously, I'll copy and paste stuff and it just won't run or it'll give an error on Windows, every single time.
I've had this as well, in part it's because I don't know PowerShell / CMD that well. I'd go for Mac, I did it and never looked back. Buy the new M1 Mac Mini even, if you have the spare money.
> I'm not sure if my brain is wired for it,
I've thought about this question as my brain partially isn't and learned a thing or two on how to overcome it.
My issues:
1. Being detail oriented isn't my strong suit.
2. For a long time I wasn't able to grok it.
3. I have hard time visualizing code.
My solutions:
1. Use all kinds of tools that will kill the detail. Simple things like syntax highlighers and linters are a given. The most bang for buck was when I learned how to use a debugger, that was almost 10 years ago.
2. Follow a computer systems course like NAND2Tetris and after that hack computers at hackthebox.eu (especially the x86-64 assembly challenges). By understanding the internals of how computers work at a systems level, you essentially are capable of grokking a lot more. Take a compilers course while you're at it, I failed mine, but learned a lot from it anyway that helped to grok more stuff.
3. Debuggers help here as well, practicing pure math helps (I'm not good at that either) and simply doing data structure and algorithm exercises help.
---
I've learned over time that making a statement like "my brain isn't wired for it" isn't helping. In part this is because it isn't true, what I was lacking was a certain perspective that comes quicker to some people. For me, it took some time, but now I have it :)
You're right about the last bit, I'm really frustrated by people who say "Math isn't for me!" because for me, I love math, as long as I have a good teacher. If I have a bad teacher, though, which isn't uncommon, it's a nightmare. I can't learn higher-level math when it's just a series of proof, corollary, lemma, whatever, without any motivation. I have trouble picking what things I want to study with CS, because it's all interesting. Too much breadth, not enough depth. I still haven't figured what's best for me. I probably have ADHD but haven't taken any medicine for it; maybe that'd help. I trip over the details and I know I'll get it if I keep at it but I have to keep looking up syntax and it drives me crazy. I've liked Python because of the simplicity. But then I want to learn the nitty gritty of architecture and systems programming, and people tell me not to waste my time and focus on the higher-level stuff.
I disagree here. Being able to use gdb makes things easier. I had 2 programming courses under my belt after taking a computer systems course.
People make it more difficult in their minds than it is. The key tricks being: one step at a time, use a debugger and google everyday assembly instruction that you encounter.
I’ll give you this: studying computer systems made me a better programmer.
Late reply, but gdb wouldn’t be used in nand2tetris. Their online grading system accepts only a couple of high level languages. I used Python to write the assembler and compilers.
What I was trying to say in my previous comment was that one would not want to be a raw beginner when starting nand2tetris. One would need to be thoroughly familiar with basic Python data structures. I used lists, dicts, and a class I wrote to handle the AST nodes.
I’d recommend nand2tetris before taking a full compiler course.
We’re agreed that low level understanding is very helpful. And, it’s interesting.
I'm more than twice your age, and I can say that Windows is the worst coding environment there is.
If you just want to learn to do software, but don't want to dig into computers per se, then a cheap Mac is a good choice.
If you do want to learn software and computers, then use Linux. You can install it on most any old PC or laptop you have or can scrounge. Learn how to admin the box. It's an important skill for experimental physics and engineering physics these days.
Can I ask what languages/tools you've tried to learn? (Hopefully nothing complicated like C++?) If you have a link to something you followed that confused you in the
past, I'd be happy to see if I can pinpoint the issue and suggest an alternative. A lot of tutorials, courses, etc. definitely do assume Linux/Mac, so they're painful on Windows. And conversely if you find a Windows-based one, you'll find it painful on other platforms. It's really important to pick something with less friction so you can focus on the concepts initially.
That's the other thing. Pretty much every tutorial I find is front-end web development, which I understand is most of what we interact with on a computer, but not especially appealing to me (fiddling with CSS doesn't feel like programming). The intro classes I took were Java, IDL (computational astrophysics, learning to solve ODEs) and also Javascript via HTML5 Canvas (a really fun class on chaos theory and fractal art). I got all As and thought it was easy. Beyond that, learning on my own, it got really hard really fast and I didn't know what to do.
I'm interested in iOS app development but that requires a Mac. I'm probably gonna grab an M1 Mac Mini.
I've tried CS50, got most of the way through it but it's tough to be motivated by self-guided courses when I'm not forced to do it. I was able to do the first week or so of Advent of Code before it got way too tough. I love self learning mostly, but for programming it did help to have lectures, grades, and office hours. I have choice paralysis where I just dabble in all sorts of things and don't know what to commit to.
If there's any programming-adjacent technology that is practically guaranteed to make you want to pull your hair out, it's CSS :-) I don't find web dev appealing in general either, so lots of sympathies there.
It's nice that you've tried Java and computational stuff and found those more appealing. Re: CS50, my understanding is that while the course is well-structured, they do start with C, which is better suited for people who really like learning from the "bottom up", rather than those who learn better "top-down". You might be in the latter group.
If all that sounds about accurate, it's probably better to focus on courses that start with higher-level technologies like Python, C#, or Java in the beginning (but stay away from web dev). Python is the most likely to feel "natural", so that's what I recommend to people just starting out. You could check out Berkeley's CS 61A for example (though I think MIT/Harvard have Python-based courses too). If you stayed on Windows, I'd recommend Visual Studio 2019 (not to be confused with Visual Studio Code), since it has everything well-integrated together. But courses focus on UNIX-based tools, so you'd indeed want a Mac if you're following anything that uses the terminal. And if iOS app development is your goal then you definitely don't want to go with Windows, though I don't have any experience with that so I can't give you pointers (hah) there.
CS in general isn't about just coding though; the algorithms and data structures are extremely important, and they aren't necessarily easy to pick up, so that's normal. And self-guided courses in general definitely need a lot of motivation, and that's easier said than done for most people, so don't feel like it's just you. You probably want to pick something where you have a good dose of interest & curiosity to drive yourself instead of just relying on the course, regardless of what you're doing.
Not sure how helpful this is but best of luck trying it with a Mac!
> I've certainly tried, I'm considering getting a Mac for dev because Windows is so painful to try and learn any coding
As a happy Mac user for many years, I agree it is worth it. However, in case you aren't ready to spend the money just yet, you can start by dual booting your PC to switch between Windows and Ubuntu or other Linux distro.
Yep, you can learn programming in any OS. (OS is mostly orthogonal to programming itself really.) Especially if you are using something popular like Python that is available for almost any platform.
My whole childhood and young life I wanted to go into physics academia. I wanted to have influence - to push the boundaries of science. I had an idyllic view of this purely meritocratic system where people were paid to explore. My first year of undergrad utterly destroyed this notion. I left as soon as I had my bachelors degree. I knew what would await me in grad and post grad work and I'd seen post docs cry after endless work for grants. I wanted financial freedom and happiness, and that eventually won over pushing the boundaries of science. So sorry young me - I quit too.
> I'd seen post docs cry after endless work for grants.
People knock the recent Alien films but you know the scene where the android is leading the team through a theater of charred bodies and you think "holy shit, this is a HINT, how do you not see it?"
Well, when I got home, I realized that the disgruntled postdocs around me -- some of whom were smarter and worked harder than me and still got thrown out like so much trash because their thesis didn't pan out -- were also a HINT. How did I not see it before?
Thanks, Prometheus. You weren't a great movie, but you had a very important impact on my life.
I have felt this way about work. I notice that people above me in the org chart always seem stressed, constantly in meetings, working late and arriving early cause their normal day is taken over with meetings, fighting with other teams, politics, dealing with uncomfortable situations, etc. When my manager asks me about my promotion plans I just smile and nod.
I think much of the hate towards Prometheus was unwarranted and overblown although it was flawed. You could easily point to plot holes and dumb decisions by characters in Star Wars or any other classic sci fi.
And it was still better than 95% of films released that year or since.
I finished my physics degree in the early 90's, and went straight into industry. At the time, we all knew about the problem of producing more PhDs than could possibly all get academic jobs. We called it "the lie" and "the birth control problem." My dad said it was the same in the 1950s when he got his degree.
We pursue the degree for deep personal reasons that are hard to discourage, even in the face of harsh economic reality.
I did a back of envelope calculation based on the number of grad students, the number of post-docs, their duration, etc., and concluded that I would need to be a superstar to ever get tenure. And I was not a superstar. On the other hand I've had a good career in industry, with no regrets. I live in a nice town, and rarely have to work long hours.
I was extremely lucky that both of my parents are industrial scientists, so I knew there was an honorable alternative to being a professor. The only hitch is that my own professors in physics could encourage my change of plan, but had no idea how to help me pursue it. I guessed correctly that computer programming would be important. ;-)
While I was not a research superstar, I was very good at making complicated machines work for other people. That's what I do today.
This is something that has fundamentally bothered me about our current society. I consider all the positions of people who do things that, when all has been said and done, do not provide value to humans. Then I consider all the positions that will advance humanity. The mismatch is jarring and telling. Something is wrong.
I imagine a society where all the funding into ad tech goes into cutting edge research, where we might have to wait a decade or two but before too long we reap the rewards. Even in small cases that that don't matter, even something like creating a video game, how much of the budget goes into the game and how much goes into advertising? Across the gaming industry, how much of the budgets for creating games goes into games that expand the mind, introduce players to new views, and garner interest in new areas of life and how much goes into reskinned slot machines?
This pattern seems to appear in many places. Short term optimization over long term gratitude as the social level leading to and caused by the same as the personal level. While it isn't the worse humanity has done, and it isn't anything worse than seen in other species, I wonder if this is a more pervasive and passive Great Filter that'll cling to us even after we have overcome others like the threat of nuclear war.
From the point of view of wider society, people who don't produce tangible or comprehensible benefit are indistinguishable from parasites.
All the people around them are supporting them, how do you know they're really worth it? If a physicist spent 40 years producing useless stuff, would anyone notice? Does that matter? How many string theorists should we support as a society, anyway? And why say, researchers? Why not artists, interpretive dancers or historians? Critical race theorists? What does all this say about our values?
The plumber is in the physicist's apartment, knee deep in crap, fixing the toilet. Suppose the plumber has interests, things she would do if she didn't have to wade through crap all the time. Suppose she's actually a fairly talented artist. Why does the physicist get a "free ride", intellectual freedom and financial security, while the plumber deals with the crap?
Of course in our society you _do_ get a "free ride" by owning enough stuff. That's a separate, but related conversation.
You don't know anyone is really worth it. That's why it's called "research." You pay smart people to explore being smart, because even though most of the effort and money are wasted every so often something wonderful falls out.
Try building GPS without relativity. Try building modern chip designs, lasers, or many medical applications without quantum theory.
It's no different to funding start ups. Most fail. A few win big.
But... a room sharing startup is still utterly trivial compared to finding life on other planets or inventing a new kind of computing. Or game changers like EM theory, quantum theory, relativity, Shannon's information theory, or Turing/Church.
And a room sharing startup is still vastly more useful than the weaponised parasitism of most of the financial sector.
Academics are not the problem. The problem is poor allocation of resources, especially creativity and intelligence - all of which have been badly warped by the peculiar value system propagated by contemporary economics in ways which have left them far less productive than they have been in the past.
Would you think the same about speculative social science? Some nerds find theoretical physics especially cool for its own sake, as opposed to say pure philosophy, but this is not most people's position. They think even space exploration, something much more concrete, is a waste of money. Your value judgment is right, and theirs wrong? They're the one having to clean the literal toilets. The thing also is that quantum mechanics was developed with a fraction of today's money being spent on physics.
On the one hand I completely agree with everything you wrote.
On the other hand, it sounds like what you are proposing is that there should be socialism for sufficiently talented people (I am not actually averse to this, although it seems fairer if it could somehow be extended to everyone).
Private capital will only fund ideas which a) have quantifiable benefit which b) can be captured. This aligns well with patentable pharmaceuticals or adtech, not so well with theories of quantum gravity. Stuff which benefits everyone conventionally has to be funded by government (although a patronage model does not seem impossible).
In the ideal case, how would we work out who are the sufficiently talented (they get socialism, let's call them the eloi), and who will who make their shoes and clean their toilets (it's capitalism for them, let's call them the morlocks)? What fields would be eligible for this kind of support and who would choose that?
It sounds like your complaint is that the existing systems for doing this (grants, tenure, research impact, prestige) are mis-allocating. I agree that existing systems aren't working well. Is it just that there is not enough funding, or does there need to be a rethink somewhere?
It seems fundamentally to be an information problem. That is, it's hard to figure out who and what are worth funding. The information problem is more difficult if the research has a longer time horizon.
Yet our current system shows we do have tolerance for dead weight, be they geniuses who happened to have bad luck in the theory they chase or be they frauds. One way to rephrase the issue in light of this is to ask if we have optimized the dead weight for the benefit. If we double the scientists and accept that we will increase our percentage of dead weight, won't we still all benefit from the increase in non-dead weight. Have we really reached the point of diminishing returns?
Yet another thing that bothers me is how benefit is viewed. Ability to defraud others, addict them, create fear and then provide the safety for that fear are all seen as benefits where as a scientist who pursued the wrong theory is not. Yet in terms of how better off society is, isn't the scientist winning even if they net contribute nothing?
I don't claim to have answers. The more I ponder the nastier the problem becomes. Some of the answers I've seen other gives, things that could be summarized as variants of socialism or related, fail to account for other flaws in humans.
It somewhat reminds me of trying to figure out how to run a government. Like someone pointing out all the problems with democracy, yet seeing that there isn't a better solution. Well, incremental improvements in the specific variant of democracy aside.
> I would need to be a superstar to ever get tenure.
The real issue is that tenure is not given to highly competent people. It's given to "superstars" and that is more of a political title than anything else.
I actually think that issue is overblown. I observed some of my friends (and relatives) progress through the academic track. Today I live near a major research university, and a lot of my friends and acquaintances are professors.
My grad school friend who went the furthest, the quickest, is in fact a brilliant physicist. Another friend has had a long career teaching at a regional college, a job that he was really well suited for. Those are just a couple of examples, and of course they're anecdotes.
In my case, no amount of politics could have saved me from the fact that I didn't have a good idea for my own research program.
I can't tell you what happens at the absolute top tier, since I don't know any of those people, or at the very least, haven't asked them.
What I do suspect is that nobody's career is untainted by incidents that involved politics if you were to examine carefully enough.
"I had an idyllic view of this purely meritocratic system where people were paid to explore."
I feel like our society and schooling really pushes the idyllic visions of things as well as the idea of strict meritocracy. I feel like this has left many of us jaded when we realize much of what we were told as we grew up were only half-truths concealing a bleak reality from us, and in some cases dooming us to self-inflicted issues like the one's described in the article.
School is just so much nonsense. I used to roll my eyes when I heard people say this, thinking it was a trope for rebellious teens - but, there's a huge amount of stuff you learn and do that has no real value and a huge amount you could learn that would be valuable that people don't teach.
Even obvious things like a course on how to cook or do your taxes or any of the things that people actually do in life aren't taught at all. At least they weren't for me. What's the relative utility of any K-12 lesson series compared to a series showing people how to budget or how to prepare affordable and healthy foods?
In reality we kind of expect people to learn everything they need to know outside of school and many things they don't need to know inside school.
I used to think that School was much nonsense too, so much that I spent 1 year or 2 mostly skipping it. But:
>In reality we kind of expect people to learn everything they need to know outside of school and many things they don't need to know inside school.
Yea and isn't school system aware of this? After all, if you learn how-to cook & do your taxes anyway (out of necessity, which is a stronger learning signal nevertheless) why teach it with public funding?
The opposite of your complaint is: "Ugh why are school teaching such obvious stuff that you will sooner or later learn anyway? Complete waste of taxpayer money!"
I see school as a incomplete(lazy) attempt to teach a broader populace "context" behind everything we have learned as a society so far. But that is too wide, and too deep to cover fully. Hence we cut of at some width and at certain depth, and focus on things you wouldn't learn or even know in a everyday life, the "unknown unknown" so to speak, for the young, energetic, mind.
My diastase for school mostly boggle down to how utterly incompetent they were even in this regard, in practice.
But maybe you have a more informed view on things that I missed in the short text.
> After all, if you learn how-to cook & do your taxes anyway (out of necessity, which is a stronger learning signal nevertheless) why teach it with public funding?
Plenty of people eat poorly because they can't cook, get their taxes wrong in a way that doesn't plausibly benefit anyone (unless this meant to be is a job creation program for the IRS), spend years not realising that they could start a business, have court proceedings go badly in a way that just wastes everyone's time...
School isn't going to fix the broken justice system. I've seen a trooper knowingly hold a false charge resulting in loss of liberty and then lie to the court about why he was amending it - his supervisor didn't care, IAD didn't care, the magistrate wouldn't even let us present our argument. The stuff they would teach in school about the system would be a joke because it's basically an oligarchy.
I can understand the pain of leaving that dream, but in a way, a tainted soil is rarely a good dream anyway.
I so wish I could do beautiful coding work but societies are not aligned with beauty (even at zero cost) so i'm bailing toward financial independance and peace of mind. Hopefully you and I can make time to join club or do thinking on our own terms.
> Hopefully you and I can make time to join club or do thinking on our own terms.
Reminds me of a thought that's been sitting in my quotes file for a while.
"What one wants is to be able to talk with a diverse club of smart people, arrange to do short one-off research projects and simulations, publish papers or capture intellectual property quickly and easily, and move on to another conversation. Quickly. Easily. For a living. Can’t do that in industry. Can’t do that in the Academy. Yet in my experience, scientists and engineers all want it. Maybe even a few mathematicians and social scientists do, too."
-- Bill Tozier, Diverse themes observed at GECCO 2006
> Erdős would show up—often unannounced—on the doorstep of a fellow mathematician, declare “My brain is open!” and stay as long as his colleague served up interesting mathematical challenges.
I can also generalize that to human existence. I've struggled with work life a lot, I've seen good people going robbers, depressed.. but 90% of them came in the morning with the best intentions.. somehow in this system, naturally a sad state of chaos emerges that makes most people very frustrated if not more.
The immediate context is the article to which I linked. It's a stand-alone observation by someone attending some conference in 2006.
How did I encounter that quote? I have no idea. I thought I found it on HN, but going by HN search, I can only find myself posting it, twice before in the last 5 years. I must have found it elsewhere.
I saved it, because it strongly resonated with me - it describes what I would love to do for a living, and what I also think the society needs to improve both research and engineering.
I quit 3 years into my physics/math degree (my university had a joint 5-year masters program) with basically the same insight. Took a masters in CS instead and wrapped up the necessary credits for a bachelors in physics so I'd at least have a piece of paper to show for the effort I put in.
My thought at the time was; If I want to do physics full time, I better make it rich first. I saw that the tenure track academic route seemed very far from my naive teenage dreams of what being a physicist would mean, and it was unrealistic that I'd even get accepted into grad school not being fully straight As. Currently enjoying a pretty decent career developing software for banks, so I can't say I regret my decision, esp. after reading articles like this.
I think this is a problem for any "deep" occupation which can't readily be measured by immediate financial benefit.
If I look around me, I can identify only a very few who are paid a living wage to do a) a thing that they enjoy that b) doesn't have a direct, immediate benefit to others.
It's not just physicists, it's theorists of all kinds, the arts, history, and yes, underwater basket weaving too.
It's worse than that, those areas are taken over by know-nothings who are great at politics and terrible in their field.
The people deciding tenure and grants are the best physicists, they are the best politicians who happen to have had a physics degree, or physics adjacent at some point in their education.
Still my favorite essay I've ever read on the experience of grad school, five years after publication. Of course I've never been to grad school, although if I were to go it'd be in physics (try reading Anathem without wanting to drop everything and get a PhD in physics afterward, I challenge you).
I've spent the past decade or so living frugally & working in big tech. I now have enough saved up that I can conceivably work on a couple contracts a year then spend the rest of the time working on whatever I want. I think I know enough to do useful, interesting things in computer science without going to grad school. So that's one path I can take. The other would be to go to grad school for physics, but able to maintain greater material comfort than the author of this piece - and having already made peace with the end of a burning desire to shake the world. I wonder whether those two things would make a difference to my experience.
Raptor conservancy. It's often cold and wet (an outdoor job) and often menial (someone has to clean the projectile poop) and I'm a newbie where the young people are my mentors. And I don't get paid (the previous 30 years are funding this).
But the experience, on a summer evening, in the rolling countryside, of a Peregrine Falcon diving at 100mph and hurtling right past you is unbeatable.
I know, some people have it good and "never work a day in their life". I would probably write software when I retire, but it would be personal projects.
Unless a degree in physics would be useful on your current path in Big Tech, I'd recommend staying in your current career and keeping physics as a passionate hobby.
It is not at all necessary to go to graduate school to immerse yourself in physics. A Deepdyve subscription will expose most of the literature.
Do as they do in England, "read" physics. It is a never ending joy for me.
Having to do physics to earn a living from it diminishes the joy, because you have to do what others want you to do and not follow your own course.
"Still my favorite essay I've ever read on the experience of grad school, five years after publication."
Not all grad school is like this. It varies greatly by subjects and also by degree level. For example, a Master's degree can be easily attainable through night school while working. Education PhDs have their own challenges, but it usually doesn't resemble the types of questions and struggles the author describes about physics.
This resonated with me pretty hard. Despite starting as a nuke e, I figured very quickly that the situation was rather grim, so I got my BS in physics, with a ton of credits in related math, computer science, and some other hard sciences for giggles. All the time, though, I kept my eyes peeled.
What I saw was some very intelligent people whose hard work, creativity, and genius I admired making very little money, scavenging for grants like jackals, their spouses working second jobs, their clothing a little worn, their cars breaking down even as they labored over Big Issues. That is the payoff?
I did not pursue it any further. Yes, the degree itself impresses people (sometimes more than really necessary, is my feeling), and it certainly primes you to be able to switch gears for entirely different frameworks, but it has opened doors.
Meanwhile, I occasionally look at the odd paper, sniff around. For strings (oh, not strings, superstrings) (oh, not superstrings, M-theory) it's all out of my league, but not so much so I cannot feel the shape of things and just how far they are from being more than just very abstract mathematical constructs. "Fiber bundles," exactly how many strings and shiny hag-stones are we out on that particular Glass Bead Game?
I accept the possibility that the universe may simply be too difficult, on a mathematical level, for mere humans to understand. It could happen. On the other hand, I suspect that the development of mathematical models of physical phenomena without frequent touching of the "real world," fumbling for guideposts, why, you could very well wander off into a paper map of your own creation, unrelated to the world you wanted to explain.
> too difficult, on a mathematical level, for mere humans to understand.
to me the current situation looks similar to when we were building more and more complex combinations of epicycles in order to more and more precisely model the "Sun and the planets rotating around the Earth". Similarly we have right now a couple of dogmas, like say Copenhagen interpretation and the recently emerged dark matter, which while not confirmed by experiments yet can't be questioned (classic sign of dogma) and as a result the science resources are available only for the complex epicycle constructs based on those dogmas.
Well, right now we have some perfectly valid mathematical constructs that nonetheless have nothing to do with reality. Gabriel's Horn is a good starter example, but you can wave it away with giving the paint "atoms" a finite size, but then there are more subtle constructs which lie in wait, like the Banach–Tarski paradox.
This will doubtlessly irritate some, but math for me is a tool and if it is not used somehow in reality, I start wondering what it is for, then. Occasionally utility will pop up later, and good, because you don't know what will end up being handy eventually, but some branches seem so esoteric I am at a loss. Again, probably my own ignorance, but I worry that those brighter and better educated than I might be led astray by castles in the sky, just from the beauty and the sensation of satisfaction you get from finding your way into the castle, if that makes sense.
>math for me is a tool and if it is not used somehow in reality, I start wondering what it is for, [...] some branches seem so esoteric I am at a loss.
i'm of a kind opposite opinion when it comes to math - myself a math PhD dropout i welcome more and more complex and esoteric constructions in math without any regard for any relation or lack of any such to physical reality. When it comes to physics though i think an opposite rule should be applied - any math constructions brought in should be checked against the physical reality.
> any math constructions brought in should be checked against the physical reality
But that's not how things work in practice. Parts of the mathematical framework used for describing a particular physical phenomenon are always a kind of mental scaffolding and do not have a direct representation in reality. This is how math always works in science. Think of the complex numbers, for example: they are used extensively in quantum physics and electrical engineering, but that does not mean that you can see them, touch them or measure them.
>Think of complex numbers, for example: they are used extensively in quantum physics and electrical engineering, but that does not mean that you can see them, touch them or measure them.
yes, we use complex numbers there only because we have a very consistent mental map between their properties/behaviour in the math model and the physical reality. The physical reality doesn't obey the math model, it is a successful and useful math model which seems to consistently describe and predict the reality in some range of conditions. So when a new model is brought in you'd want it to have at least some partial consistent mapping to reality, and with refining of a model the mapping is expected to improve. If refining of a model turns instead into rounds and rounds of complexity growths without matching massive improve in the consistency of mapping to physical reality, you'd naturally start to question the model, and may be think, kind of VC style, about spending say a few percent of resources going into that massive complexity growth of the currently dominating, yet still far from working that good, model into exploring some alternative models.
"One of the oldest tactile illusions is the Aristotle illusion. It is easy to perform. Cross your fingers, then touch a small spherical object such as a dried pea, and it feels like you are touching two peas. This also works if you touch your nose."
Math is its own reality. Well many, depending upon which axioms you start with. Computer Science is the same, the theoretical computation stuff that is. In reality, we can only build pseudo-Turing machines with limited tapes. In the end these are all equivalent to FSAs. Albeit very large ones.
>Gabriel's Horn is a good starter example, but you can wave it away with giving the paint "atoms" a finite size, but then there are more subtle constructs which lie in wait, like the Banach–Tarski paradox.
What about e or pi? Any transcendental number suffers the same problem as Gabriel's Horn. What point is there to knowing pi beyond a million digits? Recently saw a video that we know almost nothing about a power tower of pi size 4. Not even if it is or isn't an integer. Side note, e^ln(2)=2 is the trivial case of raising one transcendental number to another to get a natural number. This despite both numbers not being related to reality even less so than something like sqrt(2).
But is the research important? Or is it just impact factors and grants? There's so many researchers. It feels that some kind of optimism rather than say mathematical capability predicts who does that. Certainly I think it's very different than 20 years ago. How has the amount of publications changed over time? There just can't be enough original critical thinking for that amount.
Why three dimensions? There's so much deep math and physics and even philosophy around the abstractions of what's a measurement, what's an observation, what is information, what is communication, what is the direction of time, etc. I keep thinking that could eventually lead to a completion of an understanding of reality when proved deep enough. But why three dimensions? It seems so arbitrary in the context of everything else. But there has to be a reason.
Well, that's four, actually. This is what is needed for electrodynamics to work, and we being "electric creatures" only care about these. It's like different animals that are adapted to seeing different colors. (I don't know, some of them probably think they live in one dimension, why not?)
You can have as high of an ideal as you want about physics, quantum gravity, Einstein, Feynman, and hard work, but you should at least be aware when you start to abuse yourself in the name of this ideal. Fifteen hour days in the library? Reading paper after paper while ignoring the feeling you don't have a direction? These are moments where self-empathy can go a long way. And the more you dig into your ideals the less of a pedestal you put them on, and you can see them more clearly as they really are, and as human. Einstein was very close-minded about physics in his late career. Feynman was brilliant but also said a lot of other things other than "Shut up and calculate", which, by the way, the context of which is "eventually you just have to shut up an calculate" (from one of his books I believe). Quantum gravity isn't the only deep mystery in physics. And why does glory always seem to the main motivation for pursuing quantum gravity?
> "But in the end, I never knew what I was looking for and I didn’t find it."
To me it felt like you were looking for meaning and purpose in quantum gravity and its (mis-)characterization as the Grail, or at least in the process of working on quantum gravity. But it seemed like it was difficult to admit to yourself that it wasn't working, which ultimately makes the going even tougher.
I'm in a physics phd right now and I see this disconnect in myself and in other academics a lot.
I think what the author is missing here is how does he define "succeeding" in the field. It sounds like for himself anything short of "theory of everything" is failure, but for others, finishing a PhD is success. So on his terms for himself, terms even his advisor is way short of that goal. Heck, even Einstein was. But then he sees others just completing an arbitrary thesis and thinks of them as the successful ones. The author has just failed to reconcile his expectations and hopes of himself with what he's seeing others do.
In other words, I think even if the author had "succeeded" to the point of being top in his class and exceeding the research output of his advisor, he'd still be unhappy, still consider himself unsuccessful. Opting out was probably the best decision he could have made, and his life is probably more productive thereby. And heck, maybe he even still has a better chance of finding some theory of everything via some analogy he comes across in his day to day work than by a life of research in "fiber bundles".
I last read this article before I started grad school (in computer science). I'm reading it again, now, more than two years after I left.
My experience was the polar opposite, if the poles were bent around like a horseshoe magnet. Rather than suffocating under thick expectations, I was gasping for air without direction.
My first advisor recognized that his lab's interests had shifted between my application and my arrival, so he gave me free reign to figure out where my interests lay for my first year. (The nexus of programming languages, concurrency, and logic, if you're wondering.) He moved on to another university, and as an introvert who'd barely been out of state without my family, much less on my own in another country, I chose not to go with him.
I met my second advisor in a seminar course on static analysis and formal methods -- his first on joining the university. It was completely up my alley, and we got along really well. Given the situation with my first advisor, we were on the same wavelength about becoming his first official student. But he thought he was getting a grad student who had done real research for a year, and in reality I still had no idea how to approach research. He had to lower his expectations a few times, and since he started from an expectation that I wouldn't need much guidance, I never really got the amount of attention I asked for. We ended up parting ways with bad feelings.
I can't really blame my advisors personally. I still don't know enough about the day-to-day of the academic system to judge. And I'm sure there are places where (and advisors of whom) you can expect more support.
But my takeaway, from my experience and from this article, is that if you go in without knowing what you're getting into, nobody's going to be there to tell you what to do. You either luck into finding a sustainable approach, or you don't.
I still want to go back to grad school. I think my experience will help me know better what my needs are and where those can be provided, and thus make me a less risky proposition for a university to take on. But I regret not taking a gap year to really dig in and understand the system I was going to put myself through.
(EDIT: My advisors are good people, and I wish them well. I hope that's clear from what I've written. Things just didn't work out.)
The author was sort of like me. When I entered my Ph.D., I had a dream of finding something really big. But I was disappointed to find that those low hanging fruits of big questions had been answered. Those left were impossible to answer with current technology. People were just guessing the answers by some hints, rather than evidences. I decided to leave to do a very ordinary job after getting my Ph.D.. I do not regret it. What I have learned is if you really like a subject, you should be fine with solving those trivial problems, while enjoy the fact that the real questions are there to be answered, but you may not see them answered in your lifetime. Otherwise, don't do a Ph.D. in theory directions.
Drawing yet a bittersweet story. There's one word written all over this experience - "burn-out".
Science, just as any other domain of this human life does need these youthful, dreaming, some may call "naive", yet daring believers.
This energy is precious, just as it is also short-lived and fragile. If it sparks at a right moment and could burn for the needed time, it may bring into view those hard to find solutions to well known hard to crack problems.
However, should it glow at not so right time and for too long, that youthful source just burns out.
The best a wise advisor/mentor could do is to instill some burning discipline. Or the very least to recognize when the burning may become unsustainable. And intervene, channel it, help to recharge, lead.
Perhaps, developing this sort of discipline should be parents' job. Still, a lot of talented and strong minds simply overheat and shut off in one way or another just for self-preservation.
No blame for quitting. I just wonder if there a way to design these young minds for stronger more efficient TDP, so to speak?
The section on being ill-prepared and out of his depth compared to fellow (mostly international) grad students with stronger math backgrounds stretching back to elementary school was almost identical to my experience in a top 5 operations research PhD program. A typical American AP Calc HS curriculum and a CS undergrad (even one with theory/ML heavy upperclass work) just don't cover the math that's relevant for top level research work in most applied math domains (theoretical CS included). I and every other US student in my cohort (roughly a quarter of the class) wound up dropping out - none of the international students did. It was undoubtedly for the best, but there is still a twinge of sadness to have closed the door on my dreams of pushing the frontiers of knowledge (foolhardy and grandiose as they may have been).
The biggest failure in graduate preparedness I found was that you learn the 'how' (eg: how to calculate a derivative or integral by hand or how to compute a determinant) without a deep understanding of the 'why', which means you don't truly Know those subjects. Even if such information is presented, it's typically not the focus of evaluation and is therefore not retained. Proof-based math coursework from first principles is the only way to really progress meaningfully in understanding these topics in a way that translates beyond just applying the mechanics you've learned. Real analysis is the first baby step most will take along that path, but the greater your depth/breadth of math knowledge the bigger your toolkit to solve problems in your chosen domain will be. It's hard to suggest subsequent steps without knowing your interest (eg: for OR, measure theory is critical but other topics might pay higher dividends for other fields).
> When you say calc-heavy, you are speaking about just cranking the wheel or number crunching?
There's a clear disconnect here: in my Swiss, CS undergrad calculus classes (1st and 2nd year), nothing was remotely as mechanical as those two choices. That (practice in hard math problem solving, not mechanical application of techniques) is probably what the grandparent is missing.
I feel like during graduate school I was somewhat like the author, but this line really is the answer to this:
> Rajeev, I believe, just liked doing the math. The thing was a puzzle he could solve, so he solved it. For him that was enough.
I started my grad school career also wanting to go into particle theory, I ended up doing plasma physics honestly because it was the opportunity I could find. I didn't really dislike math or theory or anything. I also sort of wanted to chase the grail or whatever, although I moved on after being disillusioned by HEP in general because of some of the things the author mentioned. That said, I think it wasn't until my the first year out of grad school in first post-doc (probably current, even though the grant names have changed it's the same group even though it's been a few years) that I realized no, I love doing, that is, the actual doing. I love getting on a computer, running simulations, writing code, analyzing data, closing my eyes and imagining electrons in a plasma. I mean I care about the "grail" which I'd assume would be fusion ultimately, and understanding some of the plasma physics more immediately, but the reason I'm in it isn't really the problem, I love the work itself.
I think it's valuable to be chase after lofty goals and such but really doing science, with all its warts (which are legion), you have to enjoy actually doing the work, whether it's in the lab, doing calculations, or running simulations like myself. Money and stability is another issue, but if I could get paid to do this with no hicups, I would literally stay here my entire life.
"... stay the stressful course of wandering in the dark and guessing ..."
I feel like that's a highly accurate description of my career path (I'm a dev, not in physics). The part about dreading going back into work and the feelings of being inadequate - that the best effort is only treading water - is basically how I feel too. I wonder how common that is.
Ultimately this guy should have had the mind to shop around more for other professors who motivate their work better. Maybe do some experiments? You're not stuck with one advisor, and it isn't unreasonable to think that it's your job and not a convenience, or someone else's job, to find good motivation for your work.
He didn't want to do experiments. He wanted to chase the holy grail in theory. He was full of himself. He didn't want to be Hamilton. He wanted to be Newton. Hamilton was brilliant and amazing and changed the world. I would happily settle for being Hamilton.
When I finished the experimental work for my degree and was working on the thesis, the lab with its 10 KGauss magnet and other equipment was assigned to the next graduate student.
One afternoon we were hanging out in one of the grad student offices, when one of the guys had a problem - his desk calculator stopped working. After doing a bit of troubleshooting, it turned out that the wall plug strip was dead, the breaker had been tripped. We went searching for the problem and discovered that the new experimental physics PhD candidate, while hooking up a power supply, had wired a switch directly across the ac line and turned it on.
Back in the office, with the breaker reset, we were having a good laugh about this. The fellow's supervisor wandered in and leaned against the wall, listening to us.
After a while, he said, "You know, I had to stop him from doing it two more times."
I heard later that he moved into theoretical physics.
> So I did strike out on my own. A subject called random matrix theory was hot in physics at the time, applied to condensed matter phenomena and to the quantum behavior of classically chaotic systems. The relevant math was familiar to me from engineering. I figured I’d just dive in, learn the subject, and come out the other end with a publishable discovery.
Did he find something interesting on random matrices? Years ago I was researching random cycles and found a phenomenon of emergent order, that seems related to the universality of random matrices. But I'm just a programmer playing around with numbers.
I was also lost for a while working on my physics Ph.D. The course work ended with the state of physics in 1930 (this was the late 80's). I was also trying to do Lattice Gauge theory which involves many different areas: statistical mechanics, high energy, ... I really struggled trying to get up to speed on these.
My wife was finishing her finance doctorate so I had a deadline. I had already worked on some statistical mechanics projects and had the idea of downsizing and just expand on them. After this decision, I was done in a year (and before my wife was finished).
As the essay shows, it is very easy to get lost working on a physics Ph.D. After getting my degree, I ending up teaching CS for ten years before getting into software development. While it was stressful at times, I don't regret my physics education. I learned a lot (sometimes by being thrown into the deep end).
This article made me subscribe to the magazine hoping to find more amazing articles, but I don’t think there was ever an article that had this kind of emotional impact as this one in nautilus.
Still this single article was worth the subscription.
I have to say that if you have difficulty with the definitions and basic intuition of fiber bundles, you probably don't have a great future in high-energy theoretical physics. Fiber bundles are the foundation of modern approaches to gauge theory, which is ubiquitous in high-energy theory. Any junior or senior math major should be able to grok them with little or no difficulty (speaking from experience: I learned them from a physics book as a junior). I can appreciate that an engineering grad might not have much experience with comparably abstract math, though.
Life is too short for these shenanigans. But, I think that you should still go for it if and only if you understand what you’re getting yourself into. There ain't no such thing as a free lunch.
I think this is the most important comment. Most people don't realize until much later, that everything is a trade. This for that. You get an advanced degree at the cost of: higher income? free time? finding the right partner? That's what life is about. Everything has a price, and not knowing the cost is the biggest mistake.
>> you should still go for it if and only if you understand what you’re getting yourself into
> Most people don't realize until much later, that everything is a trade. This for that.
> Everything has a price, and not knowing the cost is the biggest mistake
I agree. Yet people often -- perhaps nearly always -- choose to pursue things based on limited information. Some people decide at a young age that they want to be a physicist, or a ballerina, or an astronaut, or a doctor, or a professional basketballer. Perhaps the decision is based on a dream or naive external understanding of what the occupation involves that isn't based on the reality of what the training and doing the work actually entails. Let alone the realities of the labour market: is there an oversupply of people willing to do the work compared to opportunities? What base rate of intelligent, hard working aspirants actually succeed at turning that job into a career? What compromises -- in ethics, in freedom of action and thought, in terms of making the work pay a living wage -- need to be made in order to have the career succeed?
Another lens: universities are in the business of selling education. If the young customer wants to buy an education in some specialised niche based on a naive misunderstanding of the true costs in doing so, who is the university to disabuse the customer of their naivety and turn away their custom?
Similarly for people wishing to pursue a career in academia: if we think of the system as some kind of industrial process, what inputs does it require? Funding from grants is helpful, a few blackboards, access to libraries and journals. But we also need a feedstock of smart, well-trained, hard-working low-cost graduate students to feed into the machine.
"In short, some young people think that science is a good career for the same reason that they think being a musician or actor is a good career: "I can't decide if I want to be a scientist like James Watson, a musician like Britney Spears, or an actor like Harrison Ford."[0]
>choose to pursue things based on limited information.
I've pondered over this for a while and one workaround I've found is to: rather than ask people "what do you want to be when you grow up?", ask people " where do you want to be when you grow up?". The former question almost always returns a wish/pursue based on limited information and idealization. The latter often gives bland or blank answers, but can be answered from your current (less) limited information about your own preferences, rather than your own wish and idealization of others/other things
This article resonates quite strongly with me. I also did a PhD in physics which I grew more and more disillusioned with as it progressed.
I did physics in undergrad, then worked for a few years as an engineer. I was the only one on my team who didn’t have a PhD, and saw the limits of my own knowledge, so I went back to school to get my
PhD in experimental condensed matter physics.
Originally I thought I’d struggle through the classes, then do some kickass research, and come back to being a full-time engineer.
Turned out to be quite the opposite. My first semester transitioning back to being a full-time student was rough, but after that I really enjoyed the classes, and for the next couple years I wanted to do tenure track research in academia.
I then signed on with a faculty member athe department hired as my advisor, to study nanoscale physics (eg superconducting nanowires). Fast forward a few more years and I started to hate the daily grind.
Firstly, my advisor was hired in my third year, so we had to build up the lab from scratch. Every little bit of cryogenic equipment, data measuring apparatus, data acquisition and analysis code. Took a couple years before we could do “real” research of interest.
Also we had no senior grad students or post docs to learn from. I felt way behind all my colleagues who were in functioning labs and could start taking publication-quality data after just a few weeks of joining their group.
Most of my day-to-day wasn’t even physics but annoying laboratory maintenance. Eg finding cold leaks in the cryogenic equipment (these only leak at temperatures of 4K or less, so it took lot of trial and error to find). Or tracking down ground loops on my circuits that were blowing up my fragile nanowires. Or fighting the sputtering chambers we used for deposition (and realising they had been contaminated with magnetic materials making it impossible to grow clean superconducting films).
I saw my friends at other schools write paper after paper in top journals working in their well-established labs with huge-budget clean rooms and facilities with full-time maintenance and support staff. Meanwhile in my lab we had limited equipment, no maintenance staff, and we had to fix stuff in our ‘spare’ time, etc.
I became jealous of my theorist friends who could work anywhere with just their notebooks and a pencil.
I became jealous of my astrophysics friends who were assigned observing time on Hubble and other telescopes, where teams of technicians would make sure their observations would go smoothly and provide them with perfect raw data.
I saw that without any seminal publications, my postdoc opportunities were slim. I thought about changing research areas, but came to the realisation that I’m stuck doing the same narrow brand of research that I had been doing the past few years. Because no faculty member would pay for a postdoc who wasn’t ready to hit the ground running as a domain expert.
I saw some of my successful friends with lots of amazing papers struggle to find a postdoc appointment. Or if they found one struggle to find a second, or a good faculty position. And if it was a struggle for them, my lack of publications would make this nearly impossible for myself.
I was working 12+ hour days making barely enough to live on. And not enjoying it at all, and feeling depressed, with difficulty getting out of bed to come into the lab.
Then one day pretty much on a whim I applied for an internship as a quant on Wall Street. I was invited to an interview where I was grilled over two days in 11 separate one-on-one sessions. In between sessions I would look out the window of the 30 floor of a skyscraper in NYC and wonder if I was betraying myself and my fellow physicists, as if academia was a cult.
I was offered an internship, and really liked it. This was my ticket out.
I went back to school at the end of the summer to write up my thesis.
Many of my physics friends were curious about my internship and were considering doing this themselves. All of them told me not to tell anyone else that we had spoken.
I wrote a few drafts for papers, one was published, defended my thesis and got myself outta there.
It would be interesting to do some kind of writing style analysis on posted articles and the comments about them. I wonder if you could show that the style rubs off on the comments.
Like with this:
> I was invited to an interview where I was grilled over two days in 11 separate one-on-one sessions. In between sessions I would look out the window of the 30 floor of a skyscraper in NYC and wonder if I was betraying myself and my fellow physicists, as if academia was a cult.
You could have pulled that straight from the OP. The tone, the cadence, and the way it interrupts a high-level summary to insert a pretty image.
Interesting suggestion. I read the OP's article a few years ago and it totally resonated with me. And I'm glad to have seen it here again, and wrote out my own experience above. Maybe one day I'll publish a longer version of it somewhere.
The thing with the wall street interview that you mentioned was just how diametrically opposite it felt compared to my life as a grad student. And I wanted to point this out.
I was working in a basement lab, doing long hours often by myself, in a dark drab dreary dungeon of sorts, feeling cut off from the world.
And then for the interview I walk onto a trading floor in a skyscraper in NYC with hundreds of people shouting into telephones or at colleagues across the way, TV's broadcasting news scattered around the floor. Most traders with crazy screen setups (eg 2 rows of 4 screens) full of blinking charts and tables of numbers updating in real time. Wide windows surround the floor with amazing views across Manhattan and beyond. I was put into a little windowed office for the interviews.
In the first day of interviews, the window offered a great view of the Statue of Liberty and Ellis Island. And I thought back to my great great grandparents who came to this country through Ellis Island for opportunity, and now I was given an opportunity and I hoped not to blow it.
And then into the second day the window faced uptown, showing an endless expanse of skyscrapers. I started questioning myself, feeling like a fish out of water, totally outside my element.
The summer internship itself was great, finance was nothing at all what I thought it was. There are all sorts of risk and pricing models, and it was full of people like myself who grew disillusioned with a career in academia but were still able to apply math/coding/modelling/analytical skills to something interesting and exciting.
> All of my classmates had taken up with advisors who were, like most physicists, experimentalists, the researchers who do the hands-on work of, say, smashing particles together at accelerators to see what comes out. Theorists like Rajeev, or for that matter Einstein and Feynman, who instead do the noodling necessary to explain the results of experiments with math are fewer and further between.
Wow, I got so used to the idea that almost all physicists are theoreticians that this looks quite foreign :) .
I love physics. But going back to university sounds like a nightmare. Those halls squash large swaths of neuro-atypicals and swallow you with debt. Luckily we have the internet now and we can learn by ourselves.
Bell Labs. Many great ideas outta there. When they hired Jansky and he accidentally discovered radio astronomy. And nothing happened.
But that led to the adventures of Grote Reber[0] (an inspiring figure). In 1937 the 26-year-old built a radio telescope dish in his backyard. A ham radio operator with a EE degree, he completed the first VHF survey of the sky - in his spare time. In 1939 he discovers Cygnus A. But his work 'fell between fields'. [1]
During WW2 he works with John 'Big Ear' Kraus. In 1954 Reber moves to Tasmania (less radio interference, less Big Bang). Inspired by his results, in the 50s Ryle and Hewitt get busy at Cambridge. In 1974 they win a Nobel Prize. Down under, Reber just kept doing what he loved.
P.S. Interesting coincidence: his mother had been a schoolteacher; one of her students: Edwin Hubble.
It's a beautifully written essay but he misses the obvious. What was the point of spending all this time in what he admits but doesn't make explicit was a dead end. He would have ended up spending his life pretending to make important discoveries and doing something worthwile, in reality he would have been making "toy models".
He definitely made the right decision. I don't understand the "I quit" angst, it's sunk cost fallacy, no he didn't quit, he made a rational decision. It was a waste of his time and most likely everyone else who he mention's time.
He did quit, there should not be shame in quitting, but he did quit. Others have the same difficulties as him, they don't quit, nor they achieved anything, that is the risk. Now he makes money in Wall Street. By his standards, he hasn't achieve anything yet, nor will ever, other than procuring for his family, which is not a small thing! But his dreams were about "The Grail", if they still are, then he quit. He quit a path that would most likely left him in the dust, but that it was the only path that could have possibly taken him to "The Grail", the current path won't take him anywhere near it. So unless he is able to change the narrative of his own life, he will forever be a failure, independently of how much money he can make in the trading floor.
I agree 100% but his path to the grail was stifled where he was.
He made the right decision to escape but maybe he would have been better off living like Grigori Perelman and devoting himself mind body and soul to the grail. That’s what it takes, he should have left as soon as he felt the misalignment with the grail and the toy models he was working on as he was involved in fruitless make work.
I don't necessarily see it as angst. He didn't want it enough to keep at it. We're maybe taught to see that as a failure, but it can be a liberating realization.
I went to art school. I would not recommend most arts to anyone who can bear doing something different. You will likely sacrifice most of what people consider a normal life to be an artist. If you don't want it so bad you have no choice, you can have a very nice life. "Wanting it" can be a burden.
Presumably he did not exactly cure World Hunger at Wall Street either, though he almost certainly did not have to worry about killing cockroaches at night.
I think the crucial insight in this essay is glossed over; the author allots only a few lines to an explanation of how the other graduate students didn't need to work as hard as he did to understand the fundamental concepts relevant to his project. This, I believe, is the heart of his issue. He comes to the conclusion that "he quit" or "he didn't know what to work on", but it appears to me that he simply wasn't as high-caliber as the other students, and only persevered for as long as he did largely due to stubbornness and his tremendous work ethic. (Also, he mentions the other students' more-rigorous mathematical backgrounds, but he does not mention their IQ [1].)
Immigrant outperformance of native-English-speaking students is a complicated topic; the way I choose to look at it is in terms of population competition. India has 1.3B, China has 1.4B, the U.S. has 0.33B. Assuming that the IQ distributions are exactly equal (ehh), for every one 130 IQ American, there are 4 130 IQ Indians, and 4 130 IQ Chinese. If they all compete for one spot at an American university, the Indians and Chinese will have to work much, much harder to win the "top spot from <nation>" than the American will. A look at the questions and scores for entrance exams in undergraduate physics and math for the India Institute of Technology vs. a look at the questions and scores for entrance exams in physics in math for Stanford should confirm this vast gap in curriculum and competition. Any cursory examination of formal education in terms of course complexity, workload, and exam importance of Korea, China, Japan, etc. vs. the U.S. or other western countries should also lay bare the incredible gap in preparation and competition. As I like to explain to those who support a "global meritocracy" for college admissions, the implementation of their plan would result in a majority Asian student body at every single university in the U.S.
The author's work ethic is tremendous- I have never worked anywhere near as hard as he did for that sustained period of time. But the fact of the matter is that a higher-IQ, more-prepared student would not have needed to struggle as hard as he did, and from his description of the foreign students that surpassed him, they didn't. I suspect he would have been happier if he'd accepted his lot much earlier and leaned into a higher-paying, less-hours career and studied physics in his free time.
Another question- did he refuse to quit because he loved physics, or did he refuse to quit because his vision of his life was overly focused on holding the identity of "guy who works on theoretical physics"?
[1] Many people dislike using IQ to measure anything, and its most stringent detractors say that it is a meaningless metric. It is worth pointing out that while IQ is a completely made-up metric, it is a specific measure of memory and abstract reasoning as applied to western-style logic puzzles and patterns. This makes it a useful metric when trying to evaluate one's ability to study and perform western-style math or physics, which are giant collections of western-style logic puzzles and patterns. Many people who believe that IQ does not predict anything have never had the opportunity to spend a lot of time interacting with someone who is so much higher-IQ than they are that it shows in regular interactions. I am regularly shocked at how much more quickly my extremely smart friends can pick up and memorize concepts than I can.
>Immigrant outperformance of native-English-speaking students is a complicated topic
I wouldn't even say 'outperformance' is an accurate description. I have learned from, taught, and been colleagues with many scientists from across the globe and each education system has its merits and drawbacks. East and south asian education systems highly emphasize quantitative skills. Western education emphasizes critical thinking.
The hurdle for many western students is late introduction to higher mathematics, but you are ignoring the hurdle for many eastern students, flexibility and creativity of thinking. Eastern education systems focus heavily on rote memorization, stringent quantitative evaluation, and strict mentor student hierarchies that discourage questioning. I have encountered many colleagues from more structured education systems really struggle with developing their ability to formulate novel lines of inquiry.
As this article describes, the difficulty in science isn't just the math, it's about knowing if question you are trying to solve is valid, novel, and adds to the field. You need both to be successful. Unfortunately the author of the article seemed to fail at both, resulting in his exit from science altogether.
Those seem like they could be orthogonal issues. The US system could have more rigor and be more engaging to bright students. The creativity issue could be more a function of wider society than the education system itself. Rajeev came off as a very creative person who somehow filtered out of that kind of a system.
Your argument using IQ is silly, and not because of political correctness, but because IQ is not the best metric to use if trying to measure success at physics.
When trying to compare IQs between countries, there are systematic differences and systematic measurement errors.
When we start to talk about extremely skilled people with high IQs, it gets even harder to make sweeping generalisations.
The vacuity of your argument shines when you think that by decreasing the IQ cutoff by 4 to 126 will more than quadruple the number of applicants that pass, yet a population IQ difference of 4 is hardly likely to lead to much differentiation in output (because other factors like persistence are wayyyy more important - and limited immigration filters for people - people that tend to be high persistence achievers that tend to succeed in other areas of their lives, like physics!).
Just stick to the population numbers - which is all your argument washes down to.
Also generally beware of IQ arguments, because at an individual level, the standard error of measurement for IQ is 7: “Therefore, if a person had an IQ of 130, his or her score would be between 123 and 137 points 68% of the time (i.e., there is 68% confidence that the person’s IQ is somewhere between 123 and 137) and between 116 and 144 points 95% of the time.” as per https://www.questarai.com/resources-docs/briefs/Assessment-B...
The specific numerical rating of IQ doesn't matter. I picked 130 to grab a arbitrary instance of "equal performance on an IQ test". "120 vs. 140 IQ" doesn't give a tremendous amount of insight; what gives a tremendous amount of insight is "performs very highly on all forms of IQ tests" vs. "performs poorly on the same IQ tests".
Most people that are good at physics tend to gravitate towards the same things- strategic games like Go and Chess, computer programming, puzzles, riddles, etc. Those people also tend to score highly on IQ tests, because for whatever reason their spacial reasoning, short- and long-term memory, and ability to track multiple chains of logical steps and substitutions are superior when compared to other humans.
This article was quite literally about how a "person that tended to be a high persistence achiever" failed at physics relative to other students who apparently achieved superior results with less effort. The author made allusion to failing to track the same mental tinkertoy model that his advisor had no trouble constructing. What I am alleging is that people who score highly on IQ tests have less trouble following tinkertoy models in physics. That, combined with early preparation in formal math, is how the foreign students easily outcompeted the author. Not only did they beat him on physics topics, they'd beat him on arbitrary tests of memory, logic, and spacial reasoning as well.
It doesn't matter how great my tenacity is, how hard I knuckle down and hit the books- I will never pick up on concepts and remember logical chains like Steven Wolfram. It's not a matter of practice and application. He and others are running better hardware than I am. I focus on my own strengths and am very happy. The author did not, and he is not happy per his own admission.
Late exposure to higher mathematics sounds perfectly plausible. Someone getting into sports at 15 is rarely going to compete with those that started at 10, assuming equal innate talent. The same principle applies to academics.
Would be nice to see some envelope math on the respective importance of IQ and socioeconomic background/nepotism in academia (many PhDs have parents with PhDs etc).
Feynman had an IQ of 125 which is just moderately above average, and if we were admitting people into top PhD programs based on IQ he likely would have never become a physicist of his stature.
IQ is a deeply flawed measure of intelligence.
Merely the idea that intelligence can be measured by a single test with a single output seems quite crazy to me.
Feynman's IQ was reported by himself after he took a school test. I have zero doubt that if a representative sample of the population took a scientifically controlled IQ test (and everyone participated honestly, to the fullest extent of their abilities) that Feynman would have placed near the top if not at the top of the distribution. His incredible memory, ability (and desire) to track lengthy chains of logic and patterns shine through all of his own writings and the writings about him.
"IQ score" is meaningless; "aggregate performance on tests concerning memory, logic, and patterns" is not. Do you really think that Feynman had a middling ability to perform the feats of mental acuity typically represented on an IQ test?
Perhaps I should make myself clear; rather than "IQ", I'll say "performance relative to other humans on arbitrary tests of memory, logic, and patterns". The raw scores on these tests don't mean much, and high performance on these tests does not guarantee success in physics; but prior success in physics almost certainly guarantees high performance on these tests.
Did you read the article? It is quite literally the story of a highly-motivated, creative, perseverant individual who did not have the innate ability to deal with the high-concept logical models presented in his PhD program that his advisor and other students did. Who cares about an "IQ score"? If you subjected him and the other students to a battery of mental processing tests, you'd find that his performance was worse relative to theirs.
" I have zero doubt that if a representative sample of the population took a scientifically controlled IQ test (and everyone participated honestly, to the fullest extent of their abilities) that Feynman would have placed near the top if not at the top of the distribution."
...He did take an IQ test and this did not happen.
You're just basically admitting here that no amount of evidence will ever cause you to rethink your worship of the IQ test.
so...ok then!
I have a theory that the slavish obedience many people have for the IQ test as THE ONE test of intelligence is because these people once did well on the IQ test so OF COURSE it proves how smart you are...
You are clearly too excited about bashing the concept of the IQ test to actually read any of my comments, or the article itself.
1. Feynman self-reported that he took an IQ test once at school and received a score of 125. He also often lambasted such types of tests and didn't care about them, indicating that he likely didn't try his hardest on these tests, nowhere near as hard as he tried thinking about physics. If he tried to the fullest extent of his abilities on a test that was not self-reported but instead administered in a monitored situation, and a representative sample of the world also took the same test under the same conditions, it is undoubtedly true that he would score near the very high end of the other people taking the test. Feynman's intellectual ability to recognize and memorize patterns was clearly top-notch. I say again: do you really think that Feynman had a middling ability to remember patterns, deduce and track logical models, and memorize things, which are the skills tested by the IQ test?
2. I said over and over again in the comment that you're responding to that "the IQ test" is essentially a stand-in for tests of puzzles, logic, patterns, and memory. It doesn't matter what the test is; some people will outperform everyone else on these tests, and those people are also going to be overrepresented in the high ends of physics.
3. I do not have a particularly high IQ and do not perform well on IQ tests. Compared to my cohort in college, I struggled to focus and understand complex mathematical and programming topics. I instead leaned into other skills like writing and creative thinking to complement my technical degree, and now am very happy and successful. My ability to pattern-match, remember things, and track long chains of logic is weak compared to some family members and more brilliant friends; what sets me apart from them, and apparently from you, is my willingness to consider the possible worth of subjective metrics like the IQ test.
The desperate grasping of straws trying to explain away Feynman's unimpressive IQ score is pretty entertaining here.
" I say again: do you really think that Feynman had a middling ability to remember patterns, deduce and track logical models, and memorize things, which are the skills tested by the IQ test?"
See, now I'm confused because for all your effusive (and IMO misplaced)praise of the IQ test you don't seem to grasp how the scores work.
A 125 on the test is in the top ~5% or so of all test takers.
In no world is that a "middling" score. It's not far from two standard deviations above the mean, but below what many would say is "genius" level and below the Mensa cutoff for whatever that's worth.
It's a really great score, nowhere even close to middling!
I read it again - and enjoyed it again (as much as one can enjoy a cautionary tale). Really I'm just curious how such old articles come back to the front page.
Aside for some controversial subjects, which seem to get an extra ding from dang, it appears to me that articles get boosted to the front page by getting a good number of votes quickly.
There's enough turnover here that many would not have seen (or remembered :) the earlier articles.
Because some people may have already read it and the title may not be enough for them to realize it. It is not shaming information, it is just another piece of data.
I figured a physics degree would allow me to get a good job, but then they said, sure, if you want to do finance, defense, or education. What?! The first two were pretty nauseating so I decided to do a year-long internship at various local public schools, and realized that I hated that too. (More the bureaucracy than the actual teaching aspect.) My school basically expected you’d double major in math or CS, so they often didn’t teach a lot of the fundamental techniques you needed, assuming you knew it already. I also learned that while my high school had done a great job teaching me English and history, the math department was not quite up to snuff comparatively, and I felt behind my classmates about “fundamentals” like linear algebra, which they had somehow been taught in high school. Sure, I knew Shakespeare a lot better than them, and even minored in literature in college (maybe biting off more than I could chew) but that clearly didn’t impress them. Why waste your time with old books when you could be making money! It definitely brought me down a notch and made me feel dumb, but I also realized college wasn’t a place for someone to learn lots of different things, but to focus on one area instead. I ended up becoming a bartender and I’m very happy with that decision. At least, once things are back to normal, hopefully within the next few months. Hopefully I didn’t just jinx it