Lessons for a Young Scientist
nautil.us
nautil.us
> When you are famous it is hard to work on small problems. This is what did Shannon in. After information theory, what do you do for an encore? The great scientists often make this error. They fail to continue to plant the little acorns from which the mighty oak trees grow. They try to get the big thing right off. And that isn't the way things go.
but also
> If you do not work on an important problem, it's unlikely you'll do important work.
[0] https://www.cs.virginia.edu/~robins/YouAndYourResearch.pdf
In my experience, there's actually a neat little trick for doing this: be interdisciplinary. What one discipline considers a grand and fundamental problem may be, from the perspective of another discipline, much more narrowly-defined and tractable. Attempting to accommodate many disciplinary points of view on a problem gives you more independent experimental evidence to draw on, narrowing down the range of possible models or theories you need to investigate in depth.
The growth bit at the end is the most important one, in my opinion. In graduate school, an ecology professor of mine noted that for a field at steady state, the number of PhD students a R1 institution PI could expect to follow them on that path was 1. For a field doubling in size, it was...2.
I've approached my lab trying to be mindful of that.
Sorry if I sound dumb, but do you mind elaborating on what you meant by R1 institution PI?
R1 is the rating (top tier) research university. See a list here, https://www.brandeis.edu/diversity/dei-recruitment-hiring/ou...
1. What you mean by "steady state" really matters.
1a. There can be fields that are totally stagnant in terms of huge new intellectual advances but where the student population is exploding.
1b. There can be fields that are experiencing precipitous decline in enrollment while the intellectual and commercial field itself is in hyper-growth (CS has experienced 2 of these moments in its short history).
1c. There can be fields that are stagnant in every respect but which, for whatever reason, attract lots of funding relative to the number of occupied seats. "ML + X" for all X was like this a half decade ago or so, even for values of X in stark decline.
2. CS is growing much faster than 2x. Or at least was for the last 5-10 years.
3. Most importantly, in CS, R1 is probably the very worst type of job for a phd-holder looking to do science. You get the pay of a professor, more stress than industry, are dependent on a system that horribly exploits IC labor, and are absoutely beholden to the politics of pretty ineffectual funding agencies. R2 or teaching track CS jobs are super chill, very stable, and often pay not too much worse than R1. Industry positions come with WAY more money and often more prestige/resources.
Doing so would be painful but would force universities to be real about their priorities.
Increasing available funds without changing the system will just increase the scope of the pyramid scheme without changing anything. You have to do both.
This style of argument is commonly presented here. In discussion here very, very often, people point out "the system will be abused". Most of my friends in Science are content with where they are at. They work super hard, are super intelligent, are often making the best of smaller institutions, and they wouldn't think twice about scamming the system, abusing it, etc. Maybe the liberal mindset bias is in effect here, I'm absolutely certain there are those abusing the system. I feel like the opinions here, in HN, frequently depend so much on assuming that everyone is out to screw their neighbor, at the first chance they get. It's sad (and contrary to the guidelines at some level- assume the best, work from there).
My experience is that experiences in science can vary tremendously depending on the institution, which in itself is part of the problem. Some institutions and fields are basically as you describe, and people are basically humble, hardworking, and make do sincerely and resourcefully.
Many, many institutions, though, are dysfunctional and corrupt. Increasingly indirect funds are the primary goal in evaluating scientists and scientific contributions, period. I could tell many personal anecdotes about this, from my own experience as well as those of others (as tenured faculty), but it is not an exaggeration to say that the primary motive in many departments, at least in the biomedical sciences, is indirect funds from the federal government. This has been acknowledged as such in commentaries by formal heads of federal funding agencies.
I can explain how this distorts many things, but I have seen these pyramid schemes persist even in the presence of these discussions, without any acknowledgement by programs or colleges. They simply do not care because to do so would cost them indirects.
One thing that's upsetting to me is that the people who are well-situated of course are not harmed. They benefit, and are happy to continue with the status quo because they have thrived under it, and don't have to deal with the consequences in terms of misplaced shame, lost years and salaries, and lost opportunities. It also distorts scientific process, leading to endless discussions of p-values versus Bayesian CIs that don't matter because the problem is hype chasing and monetary incentives.
The thing is, it's not even necessarily about the scientists, many of whom may be very well-meaning as you describe. It's at a higher level, in terms of program and hiring strategy, departmental organization, etc. The people doing well may not even be aware of the fact that for every group of theirs that is doing well, getting along, etc. there are many times more people who weren't allowed in the group, who got in with a bad crowd, who were manipulated or taken advantage of, or so forth.
I often get this sense that the counterarguments go something like this: "well, science is still making a lot of progress" and "good people are getting along fine" which has this Machiavellian flavor to it. The question isn't whether or not progress is being made, it's how much better could things be if they were different?
I'm also not saying funding shouldn't be increased, I'm just saying that there have to be serious structural changes to go along with it as well, otherwise we're going to end up with the same problems at a bigger scale.
Since the 1950s US institutions have been producing many more grad students than there could ever possibly be positions for. This simply increases the competition and favors anti-competitive approaches to science (block alternative theories funding, buddy up with a few others to control what papers get accepted to conferences etc. )
The issue is that there really is less and less work on the bottom (farming etc). There is actually less work in even blue collar due to how things are made (thrown out). So where do people work?
Why wouldnt 2022 have like a population of 60% scientists? Would you rather them fill bullshit jobs?
I think we need a more stable funding model for science that doesn't incentivize gaming the system. I'd love it if 60% of the population had stable employement as scientists while robots did everything else for us, but spending 100x on our current science infrastructure would probably mean less output unless we allow greater independence and stability.
As for the fields finishing decades ago. Yes some of them have become quite well defined (Physics) but there still is plenty of room for applied science and engineering in these areas.
I assume you mean $ for infrastructure as in physical buildings or something, not $ for research? Either way, I can imagine no world in which 100x $ will mean less output in Science. I'd entertain less output/$ arguments, but flat out adding money meaning less Science being done?!? If you do mean this then I supect you haven't written grants, nor (formally) reviewed them on a board, nor struggled for reagents, used surplus computers, or pined for a better microscope, or figured out what field work to cut etc. etc... if this is your take.
Ruined, really? Like living on the street ruined? Anyone with a PhD in Physics has opportunities. They might not be able to get a tenure track job in a Physics department, but there are many, many other opportunities, even in academia. For example, several of the best oceanographers I've worked with are Physics PhDs.
> Ruined, really?
The "I" in that sentence is doing a lot of work, and Katz is a sort of generally a foolish know-it-all on everything outside of his field. See, e.g., his defense of the Iraq war. Sounds like he knows everything, but the blind spots are astounding (and now, two decades later, finally obvious to even the least informed layman.) I remember reading it and shaking my head at the time -- 'this is what dunning-kruger looks like!'. The piece aged about as well as I would've predicted at the time.
From what I've heard about Katz and his advising... style... the selection bias of this "I" is particularly perverse.
That said...
> Like living on the street ruined?
Yes, and this is a MUCH more serious problem than you're giving it credit for.
PhD stipends are NOT normal salaries/wages! AFAIK it's literally the only full-time position where the employer does not have to pay FICA taxes. That means no unemployment benefits. That means homelessness if you don't find a job immediately after graduation.
Why might you not find a job immediately after completing a physics phd? Perhaps you graduate during a multi-year hiring slow-down caused by a global pandemic or a financial crisis. I.e., you're one of roughly a third of recent cohorts.
I've known many people who failed to find a post-doc and ended up homeless for a period of time because they didn't realize that they couldn't even rely on federal unemployment as a stop-gap. One of my wife's former peers slept on our couch for a couple months.
I've know many, MANY more people who would've been homeless for a period of time in their late 20s after half a decade of working full-time had it not been for family support. Truly astounding numbers. Perhaps 1/2 to 2/3rds of my wife's graduating cohort fell into that category.
I opted out of academia with TT offers in hand because I don't understand how people can run a group staffed by grad students and look themselves in the mirror in the morning.
> PhD stipends are NOT normal salaries/wages!
I always saw it as a really good deal to be getting paid to go to school!
> I don't understand how people can run a group staffed by grad students and look themselves in the mirror in the morning.
The several that I know do this by helping them find jobs --after-- before they are done. And if the jobs are outside of academia, they are usually well paying (engineering, or software engineering etc, in the defense industry).
Ugh.
You and I both know that PhD students spend very little time in coursework, and that the quantity and quality of that coursework in nowhere near the differential in their earnings. You later mention Software Engineering. Those jobs can be had with a BA/BS and no graduate coursework. So the earning differential is what? Lowest end 30K/yr and up to 100K/yr? Over 5 years, that's hundreds of thousands of dollars. That's WAAAY more graduate coursework that the typical PhD student takes during their program.
And that's assuming the employee pays for coursework.
Do you know what I pay my "junior employees who just finished undergraduate"? At least six figures. AND our company pays tuition if they decide they want to take some graduate coursework. Which isn't exactly a rare benefit. I'm also highly confident I do a better job at mentoring than a huge portion of R1 faculty. Many of my direct reports have more first author papers than the average PhD candidate.
Pay your junior employees what they're worth. Or, if you can't, at least pay into their federal disability/uneumployment insurance so that if they get hit by a bus and can't work they aren't completely fucked! Or, if you can't do either, at least be honest about how horridly exploitative the situation is, both with them and with yourself.
Again, if R1 professors are not exploiting these poor kids, then why are R1 professors the ONLY mid-level managers in the entire country whose direct reports don't receive unemployment insurance? Seeking out the only exemption in the entire country to paying your direct report's Social Security, Disability, and Unemployment is absolutely absurd and horridly unethical if you stop and think about it for even a second.
I don't know how people who work in the academic industry are able to convince themselves that what they are dong isn't reprehensible. I guess if you're smart enough it's easy to talk yourself into anything.
> The several that I know do this by helping them find jobs --after-- before they are done.
That's totally irrelevant. It's the years of absurdly underpaid labor that I would feel guilty about, not the job prospects at graduation.
> And if the jobs are outside of academia, they are usually well paying (engineering, or software engineering etc, in the defense industry).
I find that most academics who say this are full of shit and did very little to actually help their students find employment.
I once had a conversation with a math professor who told me that he tells his phd grads to just "go into finance" if they can't find an academic job. This was in 2009.
That aside, all of those jobs are available by simply dropping out after year two with a masters degree. Or, often, right after undergraduate.
Again, I don't get how academics look themselves in the mirror in the morning. I'm glad I turned down the R1 TT.
[1] The Art of Doing Science and Engineering:
https://en.m.wikipedia.org/wiki/The_Art_of_Doing_Science_and...
It would be a mistake for somebody in the current age to think of it as a meritocracy.
So, if you're a young scientist, decide if it's worth the squeeze and plan accordingly.
There's close to zero of that left in academia.
just that it is far less than what most people expect of science
It has always been that way. Science was the domain of wealthy white men who came from thoroughbred backgrounds. It's more of a meritocracy than ever, in some ways, in that you can pay to attend a university and you can shoehorn your way into the system. I'm a scientist, I grew up in a single parent family with zero academic background. However, this is not the norm, and it's a rough road full of petty bullshit and gatekeeping... choose accordingly.
I don't know what it is with HN and this fetishism for the past and its superior ethical and technical state. It's, as a general rule, entirely false. Any time you read "things aren't like they used to be" it should trigger intense skepticism. In fact, things used to be awful. It was just accepted as the norm and poorly documented in their reasons for being awful and now we have forums where the public can easily and openly discuss such issues.
The merit can be thought in two ways
- Can people that want to and deserve to be scientists become one?
- Once you are scientists, will hard work and ability reward you with a satisfying career?
I agree that in the former we made great progress and the situation is better than ever.
But at the same time I also think the latter, which is closer to what the word "meritocracy" means, is in worse shape than ever
However, this is primarily tied to the incentives in place for judging science. Extremely important work often has no value in the current structure, due to all of the publish-or-perish and University hierarchies that have been written about ad-nauseum but that have yet to significantly change. I have nothing meaningful to add to that conversation in fact. We know what the problems are. We just don't fix them.
It seems that it has always been this way. See "How Innovation Works" by Matt Ridley for many examples - and fight on.
The problem with that is that it could always be treated as a subjective measurement. And who are you going to appoint for that? The people who are already in the system. Surprise, you have created (again) the perfect conditions for nepotism to arise.
1) Before the 19th century science was the domain of the landed gentry or those who could find sponsorship from someone of means. Things have been different since then, but positions have always depended on academic pedigree. It has always been (and likely will continue to be) easier to get a permanent position at a great science University if your academic training ran through a great science University. This makes sense in some ways, but it also reflects residual bias (i.e. the halo effect of great Universities).
2) If anything, the system of recruiting for faculty positions has become MORE objective and meritocratic over the past few decades. Here is a true story that one of my older colleagues tells about getting his first academic job in 1964: a senior Professor called my colleague's Harvard PhD advisor and said, "We need someone in your area. Do you have any good students at the moment?" Six months later my senior colleague had a job at UC. That can't really happen any more (though the system is still far from perfect).
3) One thing that has changed is the ratio of applicants to positions. This has gotten a lot worse over the years. As a result, training has stretched out quite a bit. Postdocs are common/expected in many areas of science now, and they are often fairly long. But many of the most capable and persistent manage to find a way.
4) Always remember, science is ultimately about people. Therefore, it is inevitably political. Heck, if it weren't for the fact that, in the long run, science ultimately has to explain and predict things in the real world, it would be just as dysfunctional as Congress...
I don't think one can reach that conclusion (or the opposite one) yet; there's plenty of people who have found new ways to game the system in order to secure their future, so that counteracts many of the systematic measures that exist to create a fair platform for everyone.
Same here, I know the struggle. At some point I just left because of a terrible incident of abuse I suffered that led me to evaluate my options and move to the industry where pay is 10-20x and where I've found a much healthier environment to thrive.
I really dislike people who are unfamiliar with academia that try to push this fetishized/romanticized version of what THEY think academia is, with little to no factual knowledge of what it truly is.
No, it's not. I could write a book with all the extreme situations of abuse I've witnessed first-hand while in academia.
None of that would fly in a private company, simply because there is at least, like it or not, a premise of making the company profitable that still somehow favors merit over other things. You don't have that in academia, there are many instances in there where there is literally NO incentive to act in good faith.
Look at what happened at Uber, for instance, it took a while but the CEO got ousted and they took a hit on their public image. There are many (MANY) instances of sexual abuse and coercion that have been going on for DECADES at some of the world's most recognized institutions, and nothing happens. Some people are just untouchable in their little academic universes.
Do not speak about something you don't know in depth. The image that the regular folk has romanticized about science is false and needs to disappear, because it only favors those who are abusing the system to their benefit.
Academia is by no means a paradise and there are lots of problems (largely caused by how academia is funded and the resulting incentive structures). Problems also include abuse etc., but the picture you paint is neither accurate nor warranted. It seems like your view is clouded by some bad experiences you've had, but you can't generalize those.
And I do know what I'm talking about.
Read what Scientific American did to Wilson mere three days after his death:
https://razib.substack.com/p/setting-the-record-straight-ope...
A meritless attack by an untrained hack, with no evidence and proof whatsoever:
> seemed to have little to do with Wilson, and everything to do with opportunistically shoehorning particular views about social justice into the practice of hard science.
and
> ... attacks on science are legion, genuine fear of the social-media mob is rampant in academia and whole careers are “canceled” on a specious basis ...
That is a much better lesson for a young scientists of where science is headed.
The point of science is the search of truth, it's impossible to get closer to the truth of the world if the only answers the researcher accepts are those that matches their political views.