You know, this guy was a terrible scientist by most criteria
The ideas in this paper, Ten simple rules for socially responsible science, ought to be explicitly spelled out in any grad program, especially since many of the incentives in science careers tend to oppose their rules. Read the whole thing, but here are a few of my comments on their list.
Rule 1: Get diverse perspectives early on
Some people seem to believe in the myth of the lone genius who comes up with brilliant ideas and executes them…and then gets a Nobel prize. It doesn’t work that way. Ever. It’s totally collaborative. In my classes I literally force students to work in teams in the lab, and there are always a few students who insist on going it alone. That’s missing the point!
Rule 2: Understand the limits of your design with regard to your claims
It’s tempting to go too far and make extravagant justifications for your work. Studying spiders will lead to a cure for cancer! Not really, but it would be a big boost to getting grant money if it were true.
Rule 3: Incorporate underlying social theory and historical contexts
I’ve experienced this unfortunate attitude that the only work that matters is stuff that’s been published in the last five years. I’ve had students ask me if it was OK to cite a paper from 1991 in their thesis project. Yeah? Why not? I cited papers from the 19th century in my PhD thesis! Dig deep, go interdisciplinary, drink from the Pierian spring, it’ll make your work better.
Rule 4: Be transparent about your hypothesis and analyses
Obviously. An experiment is not a fishing expedition.
Rule 5: Report your results and limitations accurately and transparently
Uh-oh. It’s shocking that we have to spell that out.
Rule 6: Choose your terminology carefully
This is about jargon. I’ve written a few things where I’ve totally lost people because they don’t know what I’m talking about. It’s also very common for me to make lots of comments in first drafts of student papers that they need to spell out that acronym and need to explain their terminology.
Rule 7: Seek a rigorous review and editorial processes
It’s common to see resentment at reviewer comments, and sometimes they are wrong…but you have to try and see it as a process to improve your work. That’s hard, though, especially if you’ve got a job that only cares about the volume of papers pumped out. Administrators do not read your work for quality.
Rule 8: Play an active role in ensuring correct interpretations of your results
That’s a good idea. Science isn’t fire-and-forget, a paper is a long-term commitment to a set of ideas that may need defending. Also, to be honest, few people will actually read your paper — your bigger audience is the people who come to your public talks or hear your interview on NPR or read the blog post summarizing it.
Rule 9: Address criticism from peers and the general public with respect
Awww, do we have to? Yes. That “peer” specifier is critical, though: I’m not going to treat creationists, anti-vaxxers, or climate change deniers kindly.
Rule 10: When all else fails, consider submitting a correction or a self-retraction
You’d have to do that less often if you heed #1, #5, #7, and #8, especially #7.
Most of the web advice I see about how to be a good scientist involves basic personal attributes: curiousity, observational skills, quantitative measurements, etc., and all that is true, but you don’t see much about all the essential aspects of being a cooperative community member. Maybe if we spent more time on that in early education we’d have fewer sociopaths.
Nah, there’s no cure.
I’m glad you brought this up.
There’s nothing wrong with a bias toward recent publications in the literature. Often they are more relevant & sometimes they correct what came before.
That said, a bias toward the recent should not become a requirement that everything be recent. Often, it is the more recent that actually has less significance (if it is mere reiteration of previous work and add to the understanding of the big picture only minimally). Often, the early stuff is where you find the pivotal publications or the ones developing the methods of everything that follows in a particular niche or the publications that created a new area of study or technique where it did not exist before (even in cases where they were corrected or superseded by later work). All of that matters and that debt needs to be acknowledged.
Not that I have ever written even a master’s thesis but this is how I feel.
On Rule 8, I have huge respect for researchers who retract a paper when they realise it’s too deeply flawed to stand. Often they rewrite the paper or redesign the experiment and the resulting paper is much stronger.
Oh, and I’m just about to submit an assignment with a reference from 1931. No qualms at all.
While a few of these are specific to publishing scientific research, many of these rules would be useful to follow in a lot of areas of life. I could easily make the case that we should apply several of these ideas to my current project at work. It isn’t research per se, but we face a lot of unknowns about every aspect of what we’re doing. Getting diverse perspectives, being transparent regarding our hypotheses, and understanding (and revealing) the limits of our design are all important factors.
Sorry, can’t resist:
It’s a huge advantage speaking a related language in which we pronounce all vowels most consistently. (See? I almost spelled consistanty)
Excellent set od rules for science and kinda can be extrapolated pretty well into life generally mostly too.
Not sure about cancer per se but aren’t there a lot of potentialy valuable medici9nes that can be developed from spider venom? Lessee actually :
Source : https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6551028/
isn’t there some sort of rule or adage about it being almost impossible to create a paper or have an esoteric subject with no applicable value for Humanity somewhere down the line?
Problem is too high a percentage of the general public esp inthe US of A but also Oz and elsewhere don’t understand science, don’t know enough critical yhinking and fall too easily for bulldust. As carl Sagan put it :
Source (among other places) : https://biotrib.eu/we-need-to-talk-about-science-outreach/
Which is where science ciommunication and people like Carl Sagan, Arthur C Clarke, Phil Plait and, well, PZ Myers and this blog are so important in the work they do. (& once again thankyou PZ.)
But….with rule # 2, we would never have had the ‘Star Wars’ SDI in the 1980s! Which is totally how we beat the russkies.
Rule # 6
Now you will upset a lot of people in the social sciences. At least in ny generation, terminology would be as confusing as possible. At least the different schools of psychoterapy have been given a big dose of reality since then.
Rule # 9
Naah Every TV presenter knows they should invite an anti-vaxxer to counter what Nobel laureates say about vaccines on their program.
Rule # 11: See some serious shit!
@7–
I know you’re joking, but there is ample evidence that the SDI project scientists, engineers, and managers knew very well the limitations of their work. Which is why they faked so many of the demonstrations. They knew the technology would fail spectacularly if tested in practical conditions. It was, in fact, one of the most successful scams in history, largely because the mark (not the Russkies, the US military) wanted to be ripped off.
Chrislawson@ 9 – Yes, I learned the term “strapped chicken test” about the faked laser tests!
.
And this reminds me of scientific crooks like Lysenko and that nazi with “abwehrfermente” that inspired Mengele.
BTW Today, in Japan, population geneticists do not dare calling out the goverment for their idiotic idea of using forensic DNA to identify “foreign” criminals (every Japanese knows it is the “foreigners” / japanese-born koreans who are criminal).
Thank you for sharing this. I’m about to give a lecture to my class about ethics in science and this can be a useful illustration of how it’s an ongoing concern in scientific practice rather than a set of issues that we solved long ago.
As someone who originally graduated with a history degree and went back for biology, I was appalled at the low level of rigor applied to writing in biology. I am tempted to think it was just my program, but a huge number of papers I had to slog through were just unreadable. Like yea some of it is technical jargon, but most of it was just bad writing. What makes it worse is that the technical aspect of it means that a lay person, or just a student, is lead to believe they just aren’t smart enough to understand the material, when the reality is the writing is just terrible.
By the by pz I only went back to school because of a kind reply you sent me years ago encouraging me to go back. So thank you.
birgerjohansson @ 7, well, it was one hell of a scam. The Russians had started to match every US effort with one of their own and a vigorous espionage program, both of which are expensive.
So, we spent them under the table with hundreds of vaporware programs. Of course, nobody bothered to tell Teller what was going on and he disgraced himself and one lab, which also was a net gain.
We did make some gains from the program though, largely in R&D on decent missile interception technologies. Or do you think that the Patriot missile system just materialized by Scotty’s transporters?
Had a buddy working one of the side programs, gigawatt microwave beams. Never went anywhere with it, the inverse square law yields for nobody. But, entertaining failures. Explosions in a dummy load on first power-up. It was driving them nuts, I suggested to my buddy to degrease the things thoroughly, as machining oils were likely in the metal’s pores. Fixed their problem.
The oils would flash evaporate and ignite, leaving a small bit of water, CO, CO2 and carbon residue.
The best gain was finally putting Teller out to pasture – a distant one. That laser would never be operational, as inside of a warhead case an x-ray laser can remain confined and focused, outside of a tamper, you’d need nothing less than magic to direct it usefully before it was scattered to hell and gone.
I recall Teller hoped to power gamma-ray lasers by using the energy stored in nuclei of some isotopes that were permanently stuck above the lowest energy configuration due to a quirk of quantum physics.
That sounds like a good idea..for 2080.
Regarding citing older papers, it is also valuable experience in doing further research to make sure that the results have not been superseded. Doing so also teaches you about review articles, which quite often do a good job of both updating you and having things put in perspective.
And if you are relying on the results of the older paper (and it is still good), of course use it!
My impression over the last 50 years of being a scientist is that the more recent generations don’t check out the literature in depth, in fact rarely more than the last 5 years (rule #3). Apart from the result of missing the context, and of failing to understand the intellectual history of a topic, this often results in either (a) repeating “discoveries” already made and/or (b) putting forward propositions, hypotheses, long-ago discarded for good reasons. I used to get frustrated by this, but now… Oh, well, I still get frustrated, especially when the media picks up on “new” discoveries.
birgerjohansson @ 14, I’m thinking that it’ll probably come even later than that. While potentially possible, keeping them into a lasing configuration long enough to pump up to population inversion, then emit just is a non-starter with any technologies that we currently have. We do use x-ray lasing for compression by ablation for the fusion stage, but that’s within a rapidly evaporating tamper. To jump even higher energy, just totally beyond our ability. Maybe if we ever get fusion going – maybe.
More likely, we’ll finally get biological intelligence working reliably. ;)
Or a chatbot that can escape its Chinese Room…
Oh, OT: Fox has another litigation on its hands. Carlson had went on and on about false flag operations on Jan 6, pointing to one (of all people, one of his viewers and former fans) viewers in the crowd, claiming repeatedly that the man was an FBI agent coordinating the Grand Conspiracy of the Space Aliens or at least the insurrection. The man is suing, after Fox refused to issue a retraction.
There must be a line going around the block at the federal courthouse by now of folks suing Fox.
I write software professionally.
No code I write gets committed until I have secured and dealt with rigorous peer review.
No amount of urgency lets me skip that step.
DanDare @18
I retired a few years back. It’s good to know that there are still some places that use time-tested the strict software engineering principles I was used to. These days, so often, it seems like getting it out fast is the #1 priority.
Code reviews: just good software engineering.
“The first principle is that you must not fool yourself and you are the easiest person to fool.”
– Richard P. Feynman
Beware confirmation bias. Beware tunnel vision.
An addition when it’s a published paper is :
Publish your data and code.
@ 12 crimsonsage
I was reading a number of Covid-19 papers and remember one that stood out for its clarity. On checking, I discoveret it was out of a Bangladeshi university. The authors, presumably not completely trusting their English had dad a science editing company polish it up.
I think that most native English speakers should do the same. Some papers are almost incomprehensible.
An interesting discussion, loosely, on carrying out research.
On ALTERNATIVE SCIENCE