Comment by jyunwai | Hacker News Reader

Comment by jyunwai | Hacker News Reader

Chemistry has been uniquely difficult as a subject to study in my view. You need to be good at all of memorization, problem-solving with theory, and laboratory experiment: it's difficult to succeed with a weakness in any one of these. Though one could technically try to be a computational chemist for people who didn't enjoy the experimental work, you'd still have to go through a lot of labwork as a student. I have a special respect for people who've succeeded as chemists as a result.

Also, as a side note of personal curiosity that's on a bit of a tangent in case anyone with experience in chemistry is reading this: I heard an anecdote about improving results in teaching labs once that I've thought about from time to time, but I'm not sure if it's credible.

A chemistry undergraduate was running experiments at a teaching lab, when he decided to go the extra mile and look up how to calibrate his equipment. He then calibrated his equipment carefully before running the rest of his experiments each time that semester. His experimental results were so consistently close to the theoretical results, that he was accused of falsifying data by his instructor (though ultimately, he didn't receive any penalties). Would this be plausible, or a tall tale? I was always wondering since then, if taking the time to calibrate all of one's tools before an experiment could have such a big impact for experimenters—or alternatively, if the student actually was manipulating the data. But then again, maybe this was some kind of statistical fluke, and he attributed the randomness to his calibrations.

You are completely right. I am a chemist and this isn’t a self indulgent rant but there are those who “get” chemistry and those who don’t. We can teach and train a chemist to work in a lab - but one who groks it? Difficult to create. Sadly, there are scant opportunities for glory in chemistry. Salaries are usually low, issues with mental health are rampant, and it’s generally a career of high suffering. (For a white collar role) Many of us regret our choice, because we all feel like Walter White, funnily enough. Talented, but little to show for it. Most of us don’t start cooking though

Yeah, that’s it.

I didn’t "get" chemistry back in my school days and it was a real frustration because, well, of course I had bad grades but what saddened me were that I still found the topic to be very interesting and I loved physics, which I was pretty good at and I could see how the two were magically interconnected. But I failed to grasp the "logic" behind the system.

That’s truly one of my regrets because chemistry is probably one of the most important fundamental science for humanity’s future.

But that’s ok, I love computer science too and I did "get" it (mostly). Thanks for the awesome semiconductors, chemists !

The way chemistry is taught in schools isn't very logic in a lot of ways – it's based on heuristics. It's mostly some empirical rules, but if feels like you have more exceptions from those rules than real applications. The reasons are that 1) each molecule is a complex and complicated quantum mechanical system and 2) each observable unit of chemistry (one or more substances and their transitions in reactions or change of state) is a thermodynamic system with complex statistics. Highschoolers - not unlike alchemists - lack the math to describe these problems, so they are taught heuristics that are useful in understand a lot (but not all) everyday-chemistry.

Your phrase “logic behind the system” resonates with me. I loved and thrived in organic chemistry but it took hours and hours of sitting working through syntheses over and over-pulling books off library shelves looking for more examples. When things clicked it was beautiful. Many of my pre-med classmates were determined to memorize their way out of it, and I pleaded with them that it was impossible. It’s not list of parameters it is a way of thinking and more complex pattern recognition. They were likely busy with other pre-med stuff and had to allocate time elsewhere. I ended up pursuing microbiology/molecular biology as I didn’t have the financial runway to switch or expand majors at that point in my life, no regrets but I loved the logic of chemistry and probably could have led to some cool things had I done both.

> We can teach and train a chemist to work in a lab - but one who groks it?

This happens a lot in mature fields. Mechanics generally can't design cars. Doctors generally can't come up with drugs. IT staff can't generally write software. Pilots can't generally design aeroplanes. Homeowners can't generally build houses. The operator/builder split is real!

Aren't your examples also true about immature fields?

Very few doctors have every come up with drugs.

Few of the early pilots after the Wrights designed their own airplanes, but airplanes were certainly not mature by 1912.

When did home building change from an immature field to a mature one? I struggle to think of when most homeowners built houses.

Saying "IT staff can't generally write software" sounds like saying "sailors can't generally pilot a large vessel" - both are specialized abilities in a larger field.

> Saying "IT staff can't generally write software" sounds like saying "sailors can't generally pilot a large vessel" - both are specialized abilities in a larger field.

I'm not saying they aren't. That's why I said that there's a gap between build and operate.

It still has nothing to do with the maturity of the field.

How many astronauts could design a rocket that could get to space? I believe the answer is zero, irrespective of maturity.

> Salaries are usually low, issues with mental health are rampant, and it’s generally a career of high suffering. (For a white collar role)

Are you specifically referring to grad school and work in academia, or is this very location specific for people who've started their career? Because I know tons of chemists who went into industry after their Ph.D and they earn on the high side of overall STEM degrees.

Pharma, polymer producers, chemical bulk goods, petro, ... They all pay 6-figure salaries before mid-career. Of course it's not FAANG, but it's very comfortable.

So either my chemistry friends purposely got skills in grad school that transfer well into industry, or the German language region has unusually strong pharma/chemical industry.

I would say your last point. Chemical field is of extraordinarily high status in Germanic countries. Its like math for Francophones last century. (People learning the language for the field) Merkel? Angewandte Chemie? BASF? Not sure if that will last thanks to a lot of it based on fossil fuels.

I think many people are faced with a choice of studying physics or chemistry at a graduate level and, since up until then chemistry is generally much harder work than physics if you are an above average student, gravitate towards the subject where their ability so far has made life easy for them.

I still remember my introductory chemistry lecture at uni where the instructor said that a whole bunch of us had coasted up until now, and that in his experience those that had had to work hardest to get there, even if they had lower grades, would be the ones to succeed (he was right!)

I'm a little confused by your anecdote. High ability students who coast along gravitate towards physics, which is easier for them. But your professor said that the chemists had coasted?

This matches my experience studying chemistry after graduating from highschool with 0 effort. Studying chemistry is hard work in a lot of different fields: brute force memorization, understanding the intricacies of arcane heuristics with multiple (historically inspired) levels of "reasoning", applying lots of basic math and some higher math, applying some higher math in the context of more modern and somewhat less confusing models, and doing manual work in the lab with all kinds of equipment. It was a lot of fun, but I wasn't ready for the insane amount of work that I needed to put in. For comparisson, I've also done some courses in economics and business administration – 0 effort, just like basic school.

This heavily depends on the kind of experiments, especially the possible sources of error. Calibration can reduce most errors in physical chemistry where the measurements themselves are the usually the main part in the experiment. From my point of view, calibration of the devices should be part of the teaching lab – but equipment is often very old, so you don't calibrate the equipment itself, but rather try to measure some kind of error and clean up you data (theoretically, this is a calibration as well). If you the teaching lab was about synthesis, calibration of pipettes, scales etc. will reduce the error as well, but having a more precise stop watch doesn't make you a better runner: the result can still be influenced by non-calibratable things like the cleanliness of your flasks, the consistency of your heating device, the purity of your reactants, the speed with which you can perform certain working steps etc. – also, synthesis can involve quite some luck.

Calibration very often makes the most difference in analytical results.

This sounds like a pH meter scenario.

Where the instrument is subject to more drift because it is a community device, and undergrads doing identical experiments had never gotten very ideal results. The variation was accepted for "as good as it gets" and the powers that be were not prepared for an outlier that goes the extra mile for quality.

Sometimes the most important lesson is that readings which are assumed to be accurate enough to calculate consistency (but are not fully verified) can often have more inherent variation than dissimilar bench techniques.

Reply on news.ycombinator.com