Is it time to change the undergraduate physics curriculum? | Hacker News Reader

Is it time to change the undergraduate physics curriculum?

physicstoday.scitation.org

48 pointsmiobrien30 comments

I'm a former physicist and I was expecting this article to make a different point (particularly because it's on HN). Physics education is heavily centered around--well... physics--but the majority of what most physicists do on a daily basis is really software engineering, and most physicists are woefully unprepared for that. The majority of them are talented enough to figure a lot of it out as they go, but best practices like testing and code reviews were basically unheard of when I was in the field. I wasn't in a small laboratory experiment either, I'm talking about a large-scale collaboration that cost hundreds of millions of dollars and involved thousands of people. A simple bug in someone's code could literally have a major impact on the field. Another comment asked whether universities are too focused on preparing people for academia rather than a job, but I honestly think that they're not even really focused properly on preparing them for academia.

Teaching physics is a hard problem because there's a long and rich history of building upon previous advancements. You can't teach somebody quantum chromodynamics without first teaching them quantum electrodynamics, quantum mechanics, relativity, thermodynamics, classical mechanics, and all of the mathematical methods that come along with these. It's hard to legitimately get anywhere close to the cutting edge in graduate school, let alone as an undergraduate. I hope that this has changed somewhat in the last decade, but most of the teaching methods in these subjects have historically been unchanged since the 70s. I strongly believe that it would be better preparation, for both academia and the real world, to integrate much more of an emphasis on simulation, visualization, and numerical methods into these requisite courses. These would be far more constructive for people developing skills that they simply can't get from doing problems on paper.

> but the majority of what most physicists due on a daily basis is really software engineering

Do you have a cite or explanation for this claim? Sure, particle physics is mostly software these days, but I don't think this is true for anywhere near a majority of physicists.

It's certainly also true for astronomy, quantum gravity, and most of the condensed matter physicists that I know. Which subfields don't involve at least Monte Carlo simulations or writing code to analyze data and/or control experimental apparatus?

My claim isn't that most physicist never write code, it's that most physicists spend most of their time not writing code.

Analytical work dominates in high-energy theory, quantum information, mathematical physics, cosmology theory. I'll also dispute your characterization of quantum gravity with my own anecdotal experience. If we wanted to solve this, I think we'd just open up an issue of Physical Review and count whether more than half the papers are numerical.

Outside big collaborations (e.g., particles, astronomy), my impression is that most experimental work is not writing code even though rudimentary code is of course needed for data analysis and equipment control.

We're on a site for programers, so I think the former physicists who happen to be here are unrepresentative.

> Analytical work dominates in high-energy theory, quantum information, mathematical physics, cosmology theory.

Yes, but most physicists are not theorists.

My comment doesn't assert or assume that.

All true physicists are theorists, the others are engineers. ;oP

There are way more experimental physicists than theoretical physicists, and even theorists usually at least write Mathematica code...

"Way more" is an exaggeration. I'd estimate it's a 2:1 ratio (of profs and postdocs, not techs). Second, I am not disputing that most theorists use code from time to time. Lord knows I do. The claim is that this is nothing like 80% of their job.

Optical physicist here. I spend >80% of my time writing software and agree that I was woefully unprepared for this by my education. When I was a student the prevailing attitude that skills like programming were "trivial" and that formalized programming/computer science classes were unnecessary because it could be easily learned.

My comment doesn't address whether physicist who mostly write software are prepared to do it, it addresses the fraction of physicists who are doing this.

It is interesting to see this analysis about the progress a student sees throughout their career since the situation is fairly analogous in mathematics. Although I make the more audacious claim that a "standard" pure mathematics educations (at least in the U.S.) is not even geared toward preparing students adequately for graduate school and this is somewhat intentional. Throughout the hundreds of conversations I've had with professors in my department they have routinely espoused the opinion that all of their students are lazy, stupid, and do not deserve anything resembling a quality lecture. Many of the classes taught are simply a regurgitation of proofs in some textbook on the board with absolutely no intuition, examples, etc. given. If someone asks a question in an attempt to clarify a point, or even ask for basic help, I've found extraordinary pushback from faculty members. Typically this is done through moralizing the value of hard work, but what they end up actually saying is "get the hell off of my lawn!".

What's even more sad is many of the students hold the blind belief that the school will provide them with a quality education, but in reality the professors look to cut corners at every possible step. The students who then go on to accelerate their own education are generally looked down upon as well, unless of course they act as propaganda pieces as well and make sure not to go too far so that the professors don't look bad.

It's a depressing situation which will only be remedied by figuring out how to democratize mathematics education completely. Unfortunately, I've only seen such efforts at extraordinarily high levels so that only beginning researchers who have been lucky enough to get advisors as a top 10 university will be able to readily access these materials.

I started off as a Physics major in the 80s before switching to Mechanical Engineering. I was pretty far along in the Physics curriculum before making the switch and one of the most interesting things to me was the different approaches to the respective upper division Mechanics courses.

The Physics version was the hardest course I experienced, spending hours coming up with closed form solutions. Rarely were any actual numbers used when solving these problems. On the M.E. side, once a sufficient set of equations were derived to model the problem we would use a computer to come up with a numerical solution that was close enough.

The Physics course had no tests and consisted of roughly 30 problems we had our whole college career to complete. The M.E. course had tests where we were under time constraints to present a numerical answer.

While the M.E. course was more "practical", the the I remember about the topic in general are retained from the Physics course.

All that to say that I think there is a ton of value to a classic Physics curriculum but most of that value is indirect.

"30 problems we had our whole college career to complete"

That sounds like a really interesting way of evaluating students - where was that?

This was at Cal. State Chico.

However, it was not a department thing. It was a particular prof that ran the course this way.

Other profs taught the same course in a more traditional manner.

Please let's not reduce the amount of fundamental physics in a bachelor's degree. I have to work with people who have graduated relatively recently (recent to my graduation from Exeter Uni. in 1977 that is) and they often have a woefully inadequate grasp of the basics. This has a strong impact on their abilities as engineers later, in my field at least (transformers). They can learn to code after they have mastered basic EM, QM, and Newtonian mechanics.

> Teaching physics is a hard problem because there's a long and rich history of building upon previous advancements. You can't teach somebody quantum chromodynamics without first teaching them quantum electrodynamics, quantum mechanics, relativity, thermodynamics, classical mechanics, and all of the mathematical methods that come along with these.

I feel I was taught these things very inefficiently, though. Teaching tended to recapitulate the historical sequence in which the formalisms for each subject were developed, rather than developing the modern perspective from the outset. For example, Hamiltonians are typically taught in a final-year classical mechanics course (if they appear at all), but if they were taught early then statistical mechanics, electrodynamics and quantum mechanics would all be much easier. This even gives you a nice entry point to techniques for simulations via perturbative methods.

I'm not a physicist, though, so maybe there is some deep underlying reason for the way the curriculum is set up.

There is a weird tension between universities preparing people for academia or preparing preparing people for a job. On the one hand, universities are academic places. They are not job training, they are where you learn the scientific basis of a subject.

On the other hand, cold hard statistics says that most people don't stay in academia (and considering reports, academia is already saturated with willing recruits). So therefore it makes sense that universities prepare people for a job at a company.

This issue is exaggerated in the Netherlands (where I live) because we have a formal difference between 'University' and 'HBO' (Translates to Higher Vocational Training). On the face of it, the former is academic and the latter is focused on getting a job. However, University has higher acces requirements, and is therefore more prestigious. This means that University will probably land you a better job than HBO even though HBO probably prepares you better for doing your job. Thus, Universities here are focusing on soft-skills and preparing people for the working life. It seems they are forgetting they are academic because people who should be going to HBO are going to university for 'Virtue signaling'. But I digress.

In general, there seems to be an assumption that college will prepare you for life afterwards. It is thought unreasonable that someone with only a college education isn't suited for a job. The response to this is to focus on including such practical matters in a college education; as opposed to encouraging students to get extra-curricular experience.

/rant

> ...universities preparing people for academia...

I don't think there are many university programs explicitly designed to prepare people for academia. Where it happens, it's mostly just a case of people teaching what they know.

> So therefore it makes sense that universities prepare people for a job at a company.

This, of course, ignores the possibility that "learn[ing] the scientific basis of a subject" is not useful in an industrial career.

This article is more of a compare and contrast between the US and UK education systems as a whole and very little if not zero on the undergraduate physics curriculum itself.

I majored in EECS, an engineering major, and I only took 3 semesters of physics. And so to disagree with a point in the article, I loved my out of major and out of engineering classes. At Berkeley you’re limited to 4 years and summer school. There were classes I wanted to take in my major that I couldn’t. But I took a lot and after 4 years, I was wiped and more would have been torture.

I prefer the US system. It’s more flexible.

Having been through the track a few years ago at a reasonably large university, it was somewhat limiting that the teaching style basically pushed the narrative that "these people thought this stuff a while ago but they were wrong and here's why, so shut up about it lest you be made a fool." It totally kills the inquisitive and investigatory nature that is the field of physics. A side-effect to this, too, was that when students aren't testing the boundaries and running into all these limiting cases with their own inquisitions, they don't get the kinds of insights that really make understanding physical systems intuitive. For instance, did you know Hamiltonian systems must have an even number of dimensions? It makes sense now that I think about it wrt Newton's 2nd, but I learned this last week in a graduate-level math course.

I'd love to see more of a seminar-type setup for courses where every week an investigation is done and yhe students try to figure it out, while the teacher provides a gentle nudge in the right direction when mistakes are made along with the lessons learned from history. IMO this would make for a much more solid understanding of the scientific method and progress re: "standing on the shoulders of giants."

I'm a current undergraduate majoring in physics at a major US university, and I strongly disagree with the assumptions and the conclusions that comprise this article.

Firstly, the article lists two anecdotes as definitive proof that everyone hates their gen-ed classes. This is almost invariably wrong: because students have enormous flexibility in choosing their gen-ed classes (these literally be any number of classes out of thousands at a liberal arts college) they are much more inclined to pursue a legitimate area of interest outside their major. In fact, many of my friends actually change their major _because_ of a gen-ed class they were required to take.

This leads to my next point about not having a set major: upwards of 50% of undergraduates at my university change the major that they applied to the school for, and over 90% of graduating students conclude that they ended up in the correct major. To reduce all flexibility whatsoever in majoring, I think, is a fatal flaw that does much more harm than good. In fact, the little "good" the 3-year-major approach has – i.e. less money/time spent at college, end up more advanced in respective field – is far less valuable than the tangential skills one may pick up from gen-ed requirements.

That point about scheduling being so complicated "that at least one company, Hobson’s, makes money selling software" to navigate the process is just ridiculous: companies make money delivering burritos; does the burrito selling process need to be rethought? My high school had software that they paid for to schedule classes, and so do many.

Finally, the author's conclusions are not only contradictory but paint a worrisome future for undergraduates. We should "give students more freedom to pursue their interests," but also create "less choice" for classes? In addition, his last point about teaching students "communication, facility with widely used software, and teamwork" is literally the purpose of gen-ed requirements.

Undergraduates should be treated as adults: many (if not all) colleges have time-honored "honor code" traditions, wherein students are expected to act honorably and maturely in examination requirements. If we are to instead prescribe each course each student should take, remove all flexibility to study unrelated subject areas, and create a uniform assembly line through which every undergraduate passes without deviation, not only would students lose the most valuable parts of a university education, but they would also be woefully unprepared for the variety of challenges that life faces.

I frankly hoped to read something along e.g. the line of Scott Aaronson mentioning that now we understand quantum mechanics sufficiently better to approach teaching it differently. Then he starts it with a concept of probability more broad than the classical one.

We might have a good approach to classical physics - though I wouldn't bet on that, may be we really should teach classical mechanics closer to SICM, by Sussman, approach. But quantum physics leaves doubts even with undergraduates majoring in physics.

I like constraint based physics, and phase-space/Lagrangian/Hamiltonian methods are very useful both for solving difficult mechanical problems, as well as, seeing connections with QM, optics, etc. However, for most freshmen, the math is simply beyond them in their first year. Unless you can delay until multivariate, partial differentitals, and at least ODE... it won't stick. On casual examination SICM looks more like a substitute for Goldstien's Classical Mechanics, but perhaps it is better. Nominally, I'd recommend that in the late 2nd or 3rd undergraduate year.

One thing that I'm quite happy with in my physics education is that the faculty went to quite some effort to make sure that the math and physics tracked each other so that you would see the theory and practice twice and intertwined.

One thing that was missing (as other's have mentioned) was numerical methods and programming techniques, which I think could also be covered in a parallel track.

> However, for most freshmen, the math is simply beyond them in their first year.

You can introduce Lagrangian and Hamiltonian mechanics very early on if you're okay with hand-waving over some of the analytical details and presenting it mostly as a problem solving tool. E.g. David Morin's Introduction to Classical Mechanics does a great job with minimal formal prerequisites (freshman calculus is adequate for the majority of the book) though its exercises and problems, which are the book's highlight, will be too challenging for most freshmen.

I think a low-brow problem solving centric approach is ideal as an introduction to Lagrangian mechanics in particular (not so much Hamiltonian mechanics) since the advantages are so stark and obvious compared to the Newtonian formalism. You don't have to worry about how velocities and forces transform to write down the equations of motion. You just write down the Lagrangian and take derivatives. That's something every student will immediately appreciate.

A bit of a side note, but you caught my eye with the phase-space/Lagrangian/Hamiltonian methods. I always loved those methods (including Fourier Transforms).. but that seems to make me an oddity for having both struggled to pass Calculus 2 (as in with a B) while aceing the Classical Mechanics math for Hamiltonian and Lagrange methods and all those phase space problem domains. Makes me suspect that people tend to comprehend certain subsets of math better than others. Still those mathematical methods are going to beyond most first year undergrads.

A thread in the physics education I received was "we used to have (old model), but then that experiment caused us to think (new model)" ad infinitum. Maybe there's just a problem with my own brain, but I experience significant interference between the old models and newer models. In particular, mass is super confusing to me in a way that, say, electric charge is not.

Reply on news.ycombinator.com