Slashdot Mirror
Let Researchers Try New Paths (nature.com)
Writing for Nature journal, scientists and professors Tolu Oni, Fabio Sciarrino, Gerardo Adesso, and Rob Knight, discuss an issue researchers have been facing a lot lately. The scientific enterprise is stuck in a catch-22, they say. Researchers are charged with advancing promising new questions, but receive support and credit only for revisiting their past work. They say that often times while examining one thing researchers are able to uncover several other important things, but deviating from the path is something frowned upon for various reasons among the industry. From the article (condensed): Most striking are the barriers to achieving impact. Our research often led us to questions that had greater potential than our original focus, typically because these new directions encompassed the complexities of society. We realized that changing tack could lead to more important work, but the policies of research funders and institutions consistently discourage such pivots. When reviewers assess grants or academic performance, they focus largely on track records in a particular field. Young scientists, who must focus on developing their careers, are thus discouraged from exploration. Our own experiences provide a glimpse of the well-intentioned forces that can keep researchers from trying other paths. This challenge is not new. Physicist-turned-structural biologist Venkatraman Ramakrishnan, who is president of the Royal Society, worked for several years in a job with funding that was contingent on a steady stream of publications. This forced him to ask safe but incremental questions. To pursue what became his Nobel-prizewinning work (on the structure of the ribosome), he moved to another institution where he could ask the questions that interested him, irrespective of the chances for publication. As he describes in his Nobel biography, the decision required an international move and a large pay cut.
Time to try a new path
It's not my duty to give you money for whatever you want. Sell your ideas to me and if they pique my interest, I'll give you money. This is easier than ever now with Kickstarter and other online funding tools. So quit complaining and treat your work like a business (or if you are unable or unwilling to learn some business skills, hire an agent to do that for you).
The every single deviation and sub-invention is separately filed for the corporate patent chest.
Look, I get it, everyone wants to come up with the next huge breakthrough. As you become an expert in a field, however, you find that the most value often does come from pushing the boundary forward a little bit. Everyone doing this together keeps a steady march forward. As you establish yourself as a competent researcher who knows your field, it becomes easier to push the envelope. If you do have a eureka moment, that's awesome, but you'll have to work extra hard to support it, and that involves greater risk. The problem with deviating from your hypothesis or grant focus to chase shiny is that those neat things are a little more likely to be outside your area of expertise (and not really shiny) or even worse, capitalizing on chance (opposite of shiny - it's a trap). In the age of big data, I would much rather researchers be a little frustrated than report on every single unhypothesized correlation in their matrix. Sorry to be a wet blanket!
Some time ago tenure system was devised to protect researchers who explore new paths. They could not be fired just because they seemingly accomplish nothing for years for a chance that they may suddenly revolutionize their field or something.
Nowadays universities in USA have turned into money making businesses which are all focused on whether a professor can bring grants or profitable patents disregarding long term benefits for exploring new paths.
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
If I had mod points, I'd mod this up.
SJW: a person who perceives an injustice, and while correcting it, commits a greater injustice.
Journals are partly a relic of the past, when sharing data and algorithms was far more difficult. If I develop a data set and a tool for creating and analyzing the data, sharing that with documentation is generally far more valuable to science than a publication. Peer review has a place, but it's fraught with problems. Let's make publications secondary and evaluate researchers based on the data and tools they release.
This is due to publish or perish mentality in academia, which is pervasive. Each university professor is required to publish a certain number of papers (as a first author or a corresponding author). Making a paper is in itself a time consuming task. If you are required to make at least three papers, plus teaching, plus mentoring, plus pro-bono service tasks, you're pretty much done with your time---let alone venturing toward "new and exciting" questions.
On top of that, most of scientists are heavily dependent on grant funding. Grant is typically highly conservative and one must show that the proposed grant is going to work by showing preliminary data and prior related experience. If you don't have prior related experience, make sure that you have a team member who do. If there is no preliminary data, chances are your grant will not be even scored (a.k.a., tossed outright as being non-credible).
If I had modpoints I'd SHOVE THEM UP MY ASS!
What we're seeing is the result of capitalism's reach getting to scientists. The focus of institutions has moved from discoveries of research to the monetary benefits of research. The reason for this is plain as day, a lack of funds. The question is, who is restricting funds and what is their motive. If you find this, you'll discover the problem.
Capitalism has it's place but using it everywhere will lead to disaster.
Anons need not reply. Questions end with a question mark.
The root problem (I speculate) is the same one that afflicts companies looking ahead only to the next quarter, schools teaching only prescribed and minimally challenging material, the slow strangulation of endless safety regulation, etc. We're short-sighted, and can't fathom even a slight risk of negativity. As long as the next increment turns out OK, we figure we'll be just fine. That works, so long as the path you're taking doesn't lead you off a cliff or to stagnation, but if you have to get over some barrier (financial loss, global competition with rising powers, new discoveries) to get onto the optimal track, the purely local gradient-based search won't work. Even if eventual failure of a system is demonstrated, we'll keep doing the same thing because we're too fearful of the unknown to do anything else.
On the other hand, the real breakthroughs have never been supported by conventional thinkers or their backers. Kuhn, etc. Would taxpayers accept 99% of research funding to add up to nothing for the remaining 1% to pay off 1000-fold or more? I doubt it. The angels and VCs might risk those odds, but not the standard research funding apparatus.
Sure, as soon as we stop government from giving corporations subsidies in the form of patent protection.
You are welcome on my lawn.
This is a terrible idea. There are types of research that aren't immediately profitable to those doing the research, but are definitely necessary. For example, the government has a legitimate interest in providing accurate weather warnings for both military and civilian interests. There are good reasons why a business wouldn't be interested in issuing warnings, one of which is liability. Government has sovereign immunity, protecting it from lawsuits when the system doesn't work quite right. It would be a risky move for a business to take on that liability and start issuing warnings. But those warnings are necessary to protect life and property, including other government interests, so there's a valid reason for them to be involved here.
It's also in the interests of the government for those warnings to be accurate. That requires that we understand the phenomena that we're issuing warnings for, what causes them to occur, and are able to make reliable observations and predictions of them. Much of that research is conducted in academia, through funding awards provided by the government. However, those awards aren't limited to academia; they can be given to businesses. The fact is that businesses generally aren't interested in applying for these awards, though nothing prohibits them from doing so.
There is research that should and is conducted by the private sector. But there's also necessary research that, for reasons like what I described above, is not and often should not be funded by the private sector. The chance of a particular businesses being struck by a tornado is low. Why would they find that research when it's not all that beneficial for them and they would probably get sued if they tried to issue warnings? On the other hand, it makes complete sense for the government to do this.
And thereby you kill fundamental research, which is the basis of all other research, completely. Stupid. And incidentally at the very root of the current problems as you currently have to promise research results for funding. If you already know the results, it is not research. It is development, and it has no long-term value, unlike actual research.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
A greater abuse is the Bayh-Dole Act, which allows universities that receive taxpayer funding to patent inventions generated by that funding. I have no problem with patents if they cover something that's novel and not trivial, while also being used. It's not particularly rare for universities to sell their patents to cover their budgets. Patent trolls do buy up patents in some of those auctions. That means there are almost certainly patent trolls who are abusively suing other businesses based on patents that were granted for federal/taxpayer funded research.
That is the single, most stupid take on research I have ever heard. If people like you were running the show, we would still be living in caves.
Not really, when that first researcher came along and proposed making an artificial man-made "cave" wherever wood could be found, it probably would have piqued the interest of the grant committee. :-)
By the way, the National Science Foundation is training researchers in marketing and other traditional business skills. They want to improve the success rate of moving research out of the lab and into the marketplace so they are teaching researchers to do customer discovery, an iterative product development cycle, realistic planning to move from early adopters to a more mainstream market, etc.
https://www.nsf.gov/news/speci...
Private entrepreneurs, hoping to profit from research by promising researchers should be the ones financing it.
Not quite. The National Science Foundation provides an interesting "bridge" between basic research and an investible opportunity. The goal is to help NSF funded research "escape" from the laboratory. Sometimes a commercialization effort is too early or too high risk for the private investment community. This is where the NSF steps in with SBIR, to help scientists get from pure research to a point where private investors see opportunity. NSF SBIR is a bridge from the lab to Angels and VCs.
https://www.nsf.gov/eng/iip/sb...
And the NSF has a training program to make scientists more likely to succeed when they are nearing the "bridge". NSF I-Corp trains researchers in basic product development and business tasks so that they are more likely to succeed with SBIR or private investment.
https://www.nsf.gov/news/speci...
Back when you could look up at the night sky with a home-built telescope and make ground-breaking discoveries, you could argue that science was more pure in that the scientist was accountable only to himself. Now it costs billions of dollars to make scientific instruments that are capable of detecting phenomena past the boundaries of the possible. Mathematics has never been a turn-the-crank discipline. Biology is also harder than it was in the day of the gentleman-scientist because it's very capital intensive. Money doesn't grow on trees and you eggheads don't get to hike our taxes to fund your (statistically speaking) fool ideas just because you've come up with a more socially acceptable way of saying that you deserve everyone else's money because you're doing God's work.
Breakthroughs don't happen overnight, never have, not really. Most published science is not reproducible, especially in the life sciences. And you want to move from an incremental model to a fund-everything model and expect better results? Nope. Breakthroughs always come at the margins. Nose to the grindstone and stop complaining. Statistically speaking, not a single one of you special little snowflakes has an ego-to-competence ratio below unity. Given the tenor of this post, I'd surmise that it's somewhere several dB north of unity.
I totally agree that researchers should go down new and interesting paths, most of which will lead nowhere. The problem is the opportunity cost of letting them do so. Every $1 you spend down a dead end is $1 you haven't spent on something that will actually make a difference, like filling potholes in roads.
However I would say that a lot of funding is currently miss-allocated towards politically expedient research rather than something actually useful. About 90% of climate research funding, for example - a huge amount of money.
Please mod me down. Thanks.
Please, explain, why one must depend on the other.
If anything, the opposite must happen — those entrepreneurs spending their money on research need reassurance, the fruits of their investment will be theirs to rip.
In Soviet Washington the swamp drains you.
Non-sequitur. Does not follow.
Why would not an entrepreneur — like Bezos, Gates, or Musk — invest in fundamental research?
But, if nobody would do it voluntarily, why do you think it fair to compel you and me to pay for it at gun-point?
In Soviet Washington the swamp drains you.
Because I figured that I could actually get time to get some work done, because I wouldn't need millions in equipment, just a few computers.
Then I realized that you're expected to be a "small business owner" funding a posse of grad students, and I imagine if you didn't keep grinding the grant mill for that, you'd be forced out through all kinds of nasty subtle tricks that academic departments have to force people out these days, whether it's denying tenure track or doing a negative performance review process even after tenure.
It would just be no fun at all, in this day and age, so I went a different path.
Where are we going and why are we in a handbasket?
Ok, why would anyone fund Galileo's telescope? He didn't find anything anyone could profit from for 100years.
Who would build the large hadron collider?
Why would anyone fund any of our current large telescopes? There is no immediate profit to uncovering some esoteric anomaly in the diskoseismology of the accretion disc around a black hole. But maybe 100years from now someone will figure out a new theory of gravity based on that knowledge that will allow us to have flying cars or unlimited energy.
Nobody has that sort of risk capital. So we gathered around and said well just like how everyone has to pay for the army that protects us, if you wanna live in civilization then you gotta pay for things the majority thinks is necessary.
I am mostly libertarian in the sense government shouldn't compete with private industry .. but there are some societal needs government should fill -- if it doesn't prevent private industry from competing.
[Researcher] Give me your money, it's for sciency stuff, you wouldn't understand. [Politician] Oh, okay, I like you, here you go.
Help! I am a self-aware entity trapped in an abstract function!
https://www.youtube.com/watch?...
The shepherds did so well protecting the flock that the sheep no longer believed that wolves existed.
So I propose a basic repository for these types of research avenues. Researchers submit their project and stay within their desired approved scope. Then as they find other interesting paths to branch they 'suggest' them to other research teams to peruse. That way work would not need to be repeated except to verify results. Of course this will probably never happen due to the ambiguity of the credit.
Galileo funded his own endeavors hoping, it would pay off. A good example.
The same guys, who are paying for travel to Mars — they aren't going to live long enough to travel there themselves either. And yet, they do it — with their own monies.
Now, please, answer the question I posed — and you ignored:
So, it is your "mostly libertarian" thinking, that government knows better, than private citizens, what scientific endeavors should be financed — and has your permission to confiscate money from me to that end? That's not "Libertarian" — that's as Authoritarian as it gets...
In Soviet Washington the swamp drains you.
Because you're all about the private sector, but want government subsidies for corporations.
You don't see a little problem with that?
See, without government, without regulation, there are no markets, and the private sector would be little more than subsistence farmers. All of those private sector companies that are going to use Capitalism to do research wouldn't do a goddamn thing. Even the very idea of a "corporation" requires government to exist.
Do you want to see a list of technologies that are the result of government-funded research?
- the Human Genome Project
- the Internet
- GPS
- smartphones
- supercomputers
- Magnetic Resonance Imaging (MRI)
- Shale gas exploration
- Artificial intelligence
- The Google search engine
You're welcome.
You are welcome on my lawn.
Building a telescope for Galileo is cheap. It can be self funded.
"The same guys, who are paying for travel to Mars — they aren't going to live long enough to travel there themselves either. And yet, they do it — with their own monies."
READ THIS LINE CAREFULLY: NOBODY is building a rocket ship to Mars with their own money. Boeing, SpaceX/Elon's entire space program is funded by government buying rides on his rockets. Who do you think will fund 20+ billion to go to Mars except government(s)? How many private foundations even gave 25 cents to build the Large Hadron Collider? What about the Hubble Space Telescope? Did Boeing pay for it?
"if nobody would do it voluntarily, why do you think it fair to compel you and me to pay for it at gun-point?"
Because YOU are competing me to pay for things that YOU want but *I* don't need (examples will follow). And btw obviously a majority of voters are willing to pay for it with their taxes. Hey I don't want to build a wall to keep people out .. you're trying to compel me to pay for that aren't you? Hey I don't want to pay for imprisoning someone who stole from you. I got no beef with the guy. Why should I pay for that? Why should I pay for the things that you may want but I have no use for?
Because we need it for the things a majority of people have decided they want. You aren't compelled to pay. You can exit the contract at any point. Most of our fundamental science has come from government funding.
You aren't being compelled at gunpoint .. you can leave the social contract (country/economy) and not pay taxes.
You missed:
- Computers.
I expect that one would have happened eventually, but it got an awful lot of government money early on.
SJW n. One who posts facts.
Academic researchers are primarily funded and promoted by (1) government and (2) other researchers. "Industry" has little to do with it.
Industrial researchers tend to work on whatever is actually important to their company and tend to be flexible in terms of research directions.
This article describes precisely my reason for not pursuing a tenure-track position. Choosing tenure-track – aside from the known committee obligations, teaching, and so on – almost always results in most of your ideas being still-born.
That is, you get a startup package, and eventually manage to build up a several million $$$ capability for a single, specialized purpose. Soon enough, you have tenure. Soon enough, you solve the Grand Challenge in your subject area. You can write funding proposals on just about anything and win, as long as those are very closely related to what you have already done, and also use the expensive system that you've built up.
But hey, you are smart, and often converse with professors in other Departments or Disciplines. You see new 'cross-cutting' opportunities to go after other Grand Challenges, only they are not in your main field of renown. Pity. If you are young, you try a couple of times to enter those new scientific arenas because you have valuable ideas in those areas. Your proposals are rejected, and you are eaten by a grue. Three years later, someone in that field you wanted to enter publishes results of a clone of your proposed concept, insight, experiment, or work. You were shut out, and learn that you have wasted your time.
With experience, you learn that your tenured position is regarded by the Regents of the University as, "We boast a world-renowned expert in topic *********." And then you are stuck. You've probably got a mortgage and family by then, too. You not only lack the time to do what a research-concentrated professor should really do, but are actively discouraged from it by the system.
So you spend the rest of your career proposing incremental advances in an area that by-then bores you. There is no advantage nor profit in expanding your skill-set to other areas where you would have otherwise done great science.
Funny, it was your own example — now you are walking it back?
That government is one of the customers — even a major one — does not contradict the fact, that these are private companies invested in (and even sponsored by) private interests. Voluntarily. Earlier, USPS may have been a big customer of the airlines, but Wright brothers still funded their research privately.
Punishing criminals — to deter future crimes — is one of the few legitimate roles of the government. The legitimacy-criteria is very simple: if, without doing this, the society/country ceases to exist, it is Ok to force tax-payers to pay it.
Crime-fighting qualifies. Military does too (though, I'm ready to admit, not to the extent the US currently spends on it). But "fundamental research" does not. Feeding the homeless — neither. And so on.
A scientist with an awesome — but expensive — idea can start a funding campaign to convince others to give him money. He does not get to compel us — such compelling is both corruption-prone and, as TFA underlines, still unsatisfactory.
That self-contradictory and thus automatically wrong. Government has no money of "its own" — it all comes from taxes, which are collected at gun-point. The gun-point is rarely explicit — until the armed deputies come in to evict you from the IRS-confiscated home — but it is always there implicitly.
Not until I've paid my taxes...
But, now that you've admitted it being acceptable for you to force others to pay for the things you want, I'm done with this argument. Thanks for playing.
In Soviet Washington the swamp drains you.
That's not true at all. Law-enforcement is a government's duty regardless of who is (or would be) the victim — a KKKorporation or a homeless. Can't you win an argument without false accusations?
Nice conflating "government" with "regulation" there. But you are caught red-handed.
Now you need to prove, that this research — and these nice things — would've been impossible without government funding. Not government cooperation (as in, yes, you can lay your pipes/cables/tunnels/tracks here), but government funding.
Meanwhile, I can present a few examples of what appeared before your kind took over the ruling elites: Ford's conveyor, Wright brothers' aircraft, flush toilet, natural gas stove and electric refrigerator.
In other words, things — including very nice things — were being routinely researched and invented in the era before big government. That things continue being invented still during this era, is not at all a sign, that it is somehow better this way.
One last thing about the future... The government compels us — at the above-mentioned gun-point — to spend billions every year on road-maintenance. Do you suppose, the flying car would've been available a few decades earlier, had the highways remained at the pre-Eisenhower levels and we used the (privately-owned) railroads and airplanes for the bulk of intercity travels instead?
In Soviet Washington the swamp drains you.
No, all I need to prove is that "these nice things" didn't happen until government got involved.
So yes, maybe they might have happened someday, just as SpaceX is almost to the point of putting a human into space, something that government did half a century ago. But the fact is that private industry didn't make them happen. And you can see from that list just how valuable that government-funded research has been to mankind.
And let's not lose sight of the fact that we're having this conversation on a system that was originally created due to government-funded research. Private industry took a shot at creating an interactive communications network, and you know what they gave us? Cable television.
You are welcome on my lawn.
No. For all we know, it could've happened just as well without the government's involvement. Maybe, it would've happened later. Or, maybe, earlier.
Correlation, famously, is not causation — you've listed some nice things, that got invented while the government was funding most of the research. You are yet to prove, the inventions would not have happened with government minding just the police, the courts, and the military — as it is supposed to.
That's a perfect example, actually. SpaceX is doing it now, when it could be useful. The government did it out of utter vanity (beat the Soviets!!!!) and for no benefit whatsoever. We went to the Moon for what exactly? Inspiration?.. The billions it took could've been spent much better.
Bullshit. Tesla predicted the Internet — and smart devices — in 1926:
Had it not been the government, it would've been created — in due time — anyway. The idea was there for the taking.
Sure it was. My argument was — and remains — that the research does not have to be government-funded to be valuable. And I've listed some awesome examples of the inventions of the pre-Big Government era.
Actually, it created telephone networks first. Then the government gave AT&T a telephone monopoly killing off innovation there (and delaying Tesla's predictions) for decades. Cable television was heavily regulated too. There is no telling, what it would've evolved into, had the government not given nice cozy regional monopolies to the cable companies.
In Soviet Washington the swamp drains you.
If you think the space program was for "no benefit whatsoever", you should go back and take a look at the list I provided earlier.
You are welcome on my lawn.
Not until you prove — or, at least, come close to proving — the listed advances would not have happened without government funding, with the funds remaining instead in the pockets of taxpayers free to spend them however they pleased.
In Soviet Washington the swamp drains you.