▲ 209 points
back
104 comments
Quoting the last two paragraphs of the blog post (cuz I think that's where the gist of this post lies):
> Research should never be driven by success, because the pursuit of it either disheartens, or beguiles, even the greatest of men. An ideal Research should truly ever be ignited by curiosity. It should be fueled by the desire to know. And it should be done for the pursuit of truth and truth alone, no matter what it is. Feynman investigated the curious wobbles only for the pure joy of satiation. And the rest, as they say, is history:
> "I went on to work out equations of wobbles. Then I thought about how electron orbits start to move in relativity. Then there’s the Dirac Equation in electrodynamics. And then quantum electrodynamics. And before I knew it (it was a very short time) I was “playing”–working, really –with the same old problem that I loved so much, that I had stopped working on when I went to Los Alamos: my thesis-type problems; all those old-fashioned, wonderful things. It was effortless. It was easy to play with these things. It was like uncorking a bottle: Everything flowed out effortlessly. I almost tried to resist it! There was no importance to what I was doing, but ultimately there was. The diagrams and the whole business that I got the Nobel Prize for came from that piddling around with the wobbling plate."
The implication of the quote is that anything "ignited by curiosity" is worthwhile. This is all fine and good, as noble ambitions go, but it does not answer the key questions for academia: (1) Which papers to accept for publication?, and (2) Which projects to provide with funding?. The context of course is that there is only limited room for papers in publication venue (journal and/or conference) and limited funding for science.
How do we capture external value from letting people stay in an environment where the ey can stay curious and work on whatever they want at the time?
Does it always escape?
Could everyone work like this? (Surely not)
How can people that want to work like this do so while others sort of harvest the value that may not be harvested by those who are consistently curious and researching?
This was a memorable passage from the post for me:
"This is because all efforts trivially lead to some results. It’s an entirely different matter whether it is an expected one or not. But such is the world we find ourselves in that anything short of a ground-breaking discovery is treated as an unacceptable embarrassment. And thus, every marginal increment is touted as ground-breaking, and every deviation is repackaged to look like an intended result (which is again touted as ground-breaking)."
It's even worse than this in my opinion, as the reasons ground-breaking discoveries occur are generally not why the field thinks (or asserts?) they occur. So it's not just the results that are misrepresented, it's the whole chain of things leading to them, and what comes afterward: how they came about, who was involved in what capacity, who considers them ground-breaking or not and why, and so forth and so on.
The second (quoted) paragraph is from the book which is definitely worth reading.
We complained about greedy publishers making millions, and touted open access as the solution. I used to get a couple of review requests a month, with maybe 6 weeks of deadline. Now I get 10 review requests a month with two weeks (max) deadline for open access journals, who are making even more millions than the greedy publishers by publishing basically any crap with a ridiculously high fee. So, now I deny 90% of these requests because it's just absolutely meaningless, and so the requests go to other people less critical (and who write quicker reviews) which means the papers are even easier to publish. In the end, the whole thing is just a joke. If I wanted to really keep up with my research area I'd have to read 10 papers a day.
The right model should be like the Oscars. Write your paper, put it up in arXiv for free, and let people read, review, and react to that. At the end of a certain period (months or year depending on it's a journal or a conference), the committee chooses a bunch of them and publishes them officially, somewhere. And leave all the rest of the crap out.
That's my general sense of how things should be, although increasingly I see issues with that in terms of basic "visibility" dynamics in terms of the search engine problem and so forth.
For example, Google seems to be downweighting search results older than a certain amount, and I've noticed that if you go that route, it can be difficult to find certain unpublished/preprint-type papers of a certain age unless they appear in press. I think this might lead to an exacerbation of fad dynamics, because then papers that people know and remember are those that get a lot of buzz -- this happens now of course too, but I think it could be magnified in a different system.
You could introduce new search engines and so forth, but that adds a layer of complexity to it, and I think journals always have a way of drawing attention to certain new papers and maintaining their real-world availability.
I'm not sure I have a solution to this, but I've sort of become convinced that the approach you're describing, even though it's something I'm sympathetic to, has its own sets of problems that aren't necessarily better. I also wonder if journals will just end up being reinvented because there seems to be a logical train of thought that goes "well pay attention to those you trust to find papers" -> article curators -> fickleness about what's curated -> curation by committee/group -> reinvention of journals.
Things are completely broken imho (in academics in general, not just journals) but I'm not sure what to do about a lot of it. Maybe the problems in academics have always been social weaknesses that get brushed under the rug, but now with so many people involved there's too much to brush.
Over the last 50 years the value / norm system of mediocre minds have won. I think it’s as easy as that, and you see it everywhere.
Sadly I think it has to do with external threat: When there is one (like during the Manhattan project) the mediocre can accept a standard of excellence, because deep down they know what the alternative is. But when there isn’t one the cultural focus shifts towards mediocrity, and “politics” becomes the dominant game. And the less socially focused curious minds are easily outmaneuvered.
The financial and attention industries (and crypto, lets call that fin too) then gobble up a lot of the
geniuses (and plenty of smart but not genius people)
I would say that we failed to establish an (educational) system that would produce higher average of so-called "mediocre" minds. When you have a educational system that is, in it's core, unchanged for more than 100 years, and where the only goal is to enable everyone to pass and not to learn, then it is hard to see any other outcome.
Speaking of publication pressure, motivation etc.:
I spent my post-doc trying to tackle a rather fundamental challenge (in the architecture of analytical DBMSes). The result of my efforts took a long time to materialize, and their form was not a short article of the kind you put in conferences, but a long tract. I believe it is very useful for my (now former) field; but I could simply not get it published. Not because I was rejected for lack of novelty or anything - it was simply not in the "least publishable unit" form, and not easily (or at all) reducible to that form. The large chunk of preliminaries required motivation in use, and the motivation required the preliminaries etc. But in my field, people don't publish new research as monographs. Bottom line - put the monograph on arxiv, very few people noticed it, and life goes on without it having left a mark.
Plus, by the time this was done, my post-doc had ended, and my weird former career and lack of "publishable paper generation skills" meant I didn't have much chance at another position.
So, in a sense, my personal experience is a bit of the flip-side to that of Feynman, described in the post. Still, the sentiment definitely resonated with me.
The skill here is to find some ways to chunk the final output into smaller sizes. That usually means compromises and sometimes publishing some stuff that you want to reference later in lower level (or only tangentially related) conferences just to get it citable for later work. It's usually not fun and sometimes a huge compromise so I fully understand (and applaud) that you didn't go that route.
If you're still interested in picking up an academic career I noticed you spent some time in Europe (if the link under this post is your puplication). I'm not sure about the Dutch system but in Germany there are universities of applied sciences. Requirement for tenure as a professor is usually PhD+3 years work experience in the field outside academia. It's very teaching focused and you have to slowly ease into research again because you'll start doing it as a one person show but it's a decent option for an academically inclined person with a PhD to "come back" into the system from odd/unfitting career paths. Not sure if this construct exists in other countries but my guess would be the anser is yes. These days there's also plenty of positions that are international programs so language might not be an issue.
> it was simply not in the "least publishable unit" form, and not easily (or at all) reducible to that form
I know a couple of people from grad school who worked on massive infrastructure projects ranging from designing the ISA to verilog implementation to compilers, simulators and everything in between. I think they faced similar challenges in that it's a lot of important work but not easily publishable. It's unfortunate that we don't have a good way of recognizing this type of work.
A computational model for analytic column stores
Try treating the conference paper as a 15-page long abstract. You can still try to publish now, you don't need to be a postdoc for that!
A (tenured) professor told me he presents these ideas as a 2x2 for new untenured faculty: One axis is to work on what you think will be fun, or work on stuff you're not interested in, but that you think will get you tenure. On the other axis, you get tenure or you don't. Evaluate the likelihood of happiness in each box.
He thought it broke down roughly like this:
fun + tenure = Awesome!
fun + no tenure = Not bad, you still got paid for six years to work on fun stuff, which probably more than you'll get anywhere else.
no fun + tenure = Okay, but was six years of no fun really worth it?
no fun + no tenure = What am I even doing?
He thought it was clear that working on fun interesting stuff before tenure maximized happiness. Your milage may vary.
While I love 2x2 matrices, there are two incredibly important things missing. The first is the career of the PhD students you (as the faculty) are mentoring and supervising. Sure, you might do fun things, but if it ends up not helping their career, then you as the faculty have really failed them.
This issue of students is something often missing in discussions of the modern research enterprise. Speaking as a computer science faculty, if a PhD student wants to compete for an academic job, they need about 1 publication per year. Sure, back in the 1970s and 1980s you could get a CS faculty job with just 1 good publication, but that's not the case anymore. Of course, everyone is doing this, which then leads to more publications overall.
The second is the overall funding of science. There's an unstated compact that the public is funding science to help solve important problems and lead to breakthroughs that can advance the economy. This is the framework that has been in place in the USA since WWII thanks to Vannevar Bush, and is why NSF, NIH, and other funding agencies exist. Sure, as researcher, I strongly agree that we should be doing things we view as fun. But those fun things should have some angle as to why others should care. Another way of framing this is Pasteur's Quadrant, which is good science and good applications.
The more general advice I give to all students (undergrad, MS, and PhD) when looking for jobs: do something that is (1) fun (because there are many job opportunities in our field and life is short), (2) something that will keep pushing your boundaries (because you're smart and will be bored otherwise), and (3) good for humanity.
As someone who has had tenure, this is exactly correct. The catch is that that process doesn't end with tenure.
> But such is the world we find ourselves in that anything short of a ground-breaking discovery is treated as an unacceptable embarrassment. And thus, every marginal increment is touted as ground-breaking, and every deviation is repackaged to look like an intended result (which is again touted as ground-breaking)
This is a very nice quote.
Richard Feynman was a math genius, this is why he could derive groundbreaking results "effortlessly". There's this narrative around Feynman that he derived new physics results just for the fun and out of intuition that I believe is pretty wrong. If you read Feynman's work you'll see it's full of heavy maths and it's not intuitive at all (at least if you're not a genius like him). I don't think that the way he worked can be applied to everyone, because when you're not a genius and you do things for the fun of it it's very probable you'll end up wasting your time. This is why usually you have advisors that help you to find important problems that need to be solved.
> X was a Y genius, this is why they could Z.
This is a non-explanation. Chalking something up to genius is just a specific kind of chalking it up to magic. "Here's this person, how does their mind work, what goes inside on it? Magic! They're a genius!"
It may be so that the way Feynman worked followed from the way he thought. If so, that doesn't mean we just throw up our hands and recommend it to people who fit our genius-pattern. Instead, we need to explain genius, which involves treating it like any other object of scientific study. Once we do that, then we can figure out how to cultivate it. That's what this article should inspire.
The entire point of Feynman's worldview is that he is not "great", he is just an individual, and plays to his individual interests. Those interests become strengths because of the natural interest.
It is not about being the best. It is about contributing something that you actually care about.
> when you're not a genius and you do things for the fun of it it's very probable you'll end up wasting your time. This is why usually you have advisors that help you to find important problems that need to be solved.
Also add if there are not great examples to learn from. Most radical ideas require mastery of the topic. Incremental work can be done copying and simply adding one on top. Advisors have gone through all of this experience. Most people underestimate power of good mentors.
Feynman didn't think he was a genius, he said that he was just an ordinary person who worked very hard. The story in the article actually points this out; he was just curious which led to very beneficial working methods. The idea of geniuses is more of an excuse than an explanation.
I've always been a big fan of the "Feynman method of Problem solving":
Write down the problem. Think very hard. Write down the solution.
Me too. But I guess it does border on that famous meme:
1. Draw some circles. 2. Draw the rest of the f*ing owl.
Personal anecdotes from a family member of mine who is just breaking into genetic research at an Ivy:
- Power tripping principal investigators
- Cutthroat competition leading to actual sabotage (researcher deliberately social engineered their way into a former lab and unplugged a sample freezer)
- Gossip that makes Mean Girls look tame
- Abysmal, poverty tier wages for people who have PhDs
That's the tip of the iceberg really. Pretending the iceberg doesn't exist and doesn't hit ships is also part of the problem.
But you still should trust the science
I agree that unaligned incentives have lead to the replication crisis, but I fail to see how research driven by “curiosity” and not by “success” will align those incentives. The author admits that poor incentives have determined the course of academia/research, but I think rather than getting away with incentives altogether and just letting people be “creative”, we should find better ways to align incentives, and encourage a system of reliable peer review to restore the trust in research.
Talented people have intrinsic motivation and don't need artificial incentive systems.
Extrinsic rewards or punishments are at best poor substitutes and at worst destroy intrinsic motivation.
>it is not uncommon for many to inflate marginal results or engage in dubious practices, leading to a noise that greatly burdens the reviewers of these journals/conferences ... It is no surprise then, that many low-quality (or downright incorrect) works escape the sieve.
Eh, I blame the reviewers for this in large part. Every reviewer craves novelty like a drug addict. They'll reject papers because "it's not novel, just a bunch of engineering/implementation", or reject them because a mere 1.5x improvement does not meet the bar (Is it even science if it's not an order of magnitude ?). Then people will cry about dubious papers. The academic community deserves this.
In my experience you find a lot of spectrum. Keep in mind that reviewers today are the authors tomorrow. Some people in academia are incredibly arrogant and view themselves as the next Feynman, and dismiss marginal advances or small results. Anything under perfection and lots of work behind isn't worth publishing. But you have also the honest ones that can recognize value in the small advances.
But in general you are right: we deserve this because the amount of arrogance is too high, and more often than not you have to deal with someone who firmly believes they deserve the Nobel price.
Would someone qualified mind explaining, in layperson's terms, how the dynamics of wobbling plates led to QED, keeping in mind I barely passed partial diff-EQs all those years ago? A cursory search online only leads to commentary on Feynman's memoir.
I'm not sure there's any strong or direct connection in the academic sense. My recollection from reading the rest of the Feynman story that was quoted is that the wobbling plates problem lifted him out of a self-perceived slump or depression. So what's important is the impact that working on a fun and unimportant problem had on his state of mind and emotions, more than any particular insight or inspiration that problem had on QED stuff.
It's not that dynamics of plates gave him insights about QED. Working on a simple problem have helped him to get back into working rhythm of his phd years.
I don’t have full explanation of the connection that would be satisfying. However, I think it mainly comes down to symmetry group of the spinors and their 720 degrees symmetry. It is called plate trick for a reason after all: https://en.m.wikipedia.org/wiki/Plate_trick
This quote says to me imposter syndrome can tap on your shoulder even if you’re as good as Feynman, so no need for anyone else to take it personally.
And then I thought to myself, “You know, what they think of you is so fantastic, it’s impossible to live up to it. You have no responsibility to live up to it!”… I realized that it was also true of all the other places, including my own university. I am what I am, and if they expected me to be good and they’re offering me some money for it, it’s their hard luck.
It is one thing to pursue something out of curiosity, and another to spend hours upon hours every day for weeks on the same thing.
I think your curiosity might give way to other obligations at some point which is why you have to be somewhat goal oriented.
Worth noting that I have spent hours pursuing a hobby as a kid: programming. And I think that is probably true of many of the readers here.
I think the computer lends itself to this in a weird -- perhaps unique -- manner. The act of programming, or writing, is like a video game in that you get this reward just for the act. And it is easier to spend hours for days for weeks programming (or writing) than say mulling over other matters.
One trick for turning a task you cannot spend time on into a task you can spend time on is to turn it into an activity. This is why mathematicians have to write out their equations and diagrams when they're mulling plates.
There is some strange disconnect between citizen and worker - a citizen has tenure unless they violate fairly high bars, but a worker has no tenure - yet we expect better results? maybe tenure is something we should consider at business?
Hah. As for the wobbling plate, it reminds me of the super secret revelation of what happens when you juggle nuts and wrenches in spaaace!
https://medium.com/illumination/why-a-tennis-racket-led-russ...
this is a lengthy diatribe borne out of frustration of having to constantly produce a peice of work, not out of will or pleasure but from pressure to show that one is doing somework.
i am ignorant when it comes to politics in academia; however, isn't this the reality of academia? one has to show work irrespective of quality if one is to maintain one's position?
Most academics are struggling to get tenure. They're not Feynman who has the good fortune to not have the threat of perishing.
Speaking of Feynman, does anyone know that one poem he either wrote himself or stole and edited a bit?
All I remember is it was the same sentence, repeated, with a different word overemphasized each time to show how you can change the meaning of a sentence with a bit of… emphasis.
A lot of Feynman's stories come down to being dishonest for personal gain. His charm and charisma are great at hiding what would be pretty controversial stuff from a more stuffy/boring person.
can't help but shake my head as I read about Feynman turning down a fancy job to remain in academia, at humble Cornell (and later, Caltech I guess).
That's all very well if you're Feynman...