What are the constraints on research progress?

[Greg Bernhardt]

While thinking of my previous thread here:
https://www.physicsforums.com/threads/how-is-new-math-researched.935167/

I've come up with another general question on research.

My airbnb guest is obviously brilliant and goes to a big and generally respected university of Georgia Tech. I got to thinking just how many researchers there are in the maths and sciences. It must be tens of thousands worldwide? Hundreds of thousands? If that is the case, you'd think major discoveries are being made every day. I don't know, maybe they are. My general question is what factors control the speed of research progress if there are hundreds of thousands of brilliant researchers working day in day out.

 

[Ygggdrasil]

Economists recently took a look at this issue and put out a paper on the topic:

In many growth models, economic growth arises from people creating ideas, and the long-run growth rate is the product of two terms: the effective number of researchers and their research productivity. We present a wide range of evidence from various industries, products, and firms showing that research effort is rising substantially while research productivity is declining sharply. A good example is Moore’s Law. The number of researchers required today to achieve the famous doubling every two years of the density of computer chips is more than 18 times larger than the number required in the early 1970s. Across a broad range of case studies at various levels of (dis)aggregation, we find that ideas — and in particular the exponential growth they imply — are getting harder and harder to find. Exponential growth results from the large increases in research effort that offset its declining productivity.

https://web.stanford.edu/~chadj/IdeaPF.pdf

This paper was recently discussed on the Freakonomics podcast: http://freakonomics.com/podcast/no-new-ideas/

 

[strangerep]

[...] My general question is what factors control the speed of research progress if there are hundreds of thousands of brilliant researchers working day in day out.

...thousands of researchers working on variations of the same dead end, because "that's what lots of other people are working on".

I guess you've probably seen Sabine's recent post on this subject ?

 

[mfb]

Most of the easy things have been already - now we are working on things that are harder to create or test. It is difficult to compare scientific progress over long timescales, but if we had the same researchers as today in 1900, I‘m sure we would have made faster progress simply because we would be more.

Keeping the rate of scientific discoveries high requires more researchers over time - but I think we can afford that. The amount of work needed to sustain a given living standard is going down due to research, we can have more people doing something that doesn’t immediately lead to a product. Research creates its own funding, basically. For NASA there are studies showing that every dollar invested in NASA research leads to 5-15 dollars for the economy. Article about it.

I create threads about scientific news once in a while if I find the topic noteworthy, interesting personally, and if I can write something about it. Without the second and third part there are news every day. Wikipedia has a list, phys.org has several news per day, ...
Sure, the vast majority of them are not as revolutionary as quantum mechanics or discovering the structure of DNA, but these breakthroughs were always rare. We still have a lot of recent big news. The whole field of exoplanet research is very new, CRISPR is so new we don’t know yet where the limits are, machine learning made huge progress in beating humans in computer games in the last three years, gravitational wave astronomy started two years ago. Closer to applications, self-driving cars are improving quickly, rockets are getting reusable and large-scale private spaceflight might become a reality. And so on.

 

[ZapperZ]

While thinking of my previous thread here:
https://www.physicsforums.com/threads/how-is-new-math-researched.935167/

I've come up with another general question on research.

My airbnb guest is obviously brilliant and goes to a big and generally respected university of Georgia Tech. I got to thinking just how many researchers there are in the maths and sciences. It must be tens of thousands worldwide? Hundreds of thousands? If that is the case, you'd think major discoveries are being made every day. I don't know, maybe they are. My general question is what factors control the speed of research progress if there are hundreds of thousands of brilliant researchers working day in day out.

The number of research topics or problems being worked on does NOT commensurate with the number of people working working in research. How many people worked on the discover of the Higgs, which to many, is considered to be just ONE topic. How many people are working in climate science, which to many people is considered to be only ONE topic? How many people are working, STILL, on finding the accepted theory for high-Tc superconductivity (since 1987) which to many people is considered to be only ONE topic? How many people worked on LIGO over so many years, and discovered gravitational waves, which to many people is considered to be only ONE topic? How many people worked on ITER and trying to ignite that darn fusion thing, which to many people is considered to be only ONE topic? How many people are continually working on finding the cure for HIV/AIDS, which to many people is considered to be only ONE topic?

I can go on and on and on and on...

On the other hand, there are many topics in which the funding, and consequently, the number of people working in them, are VERY small. Here, the topics or problems may be varied, but the number of efforts being put into it is tiny when compared to the big problems.

But we also need to look at it in another way. MANY problems ARE being solved, and being solved very frequently, or at least on the way to being solved. Every single publication in research journals aims to address a problem, and offer a progress in solving or tacking such a problem. So how many papers are published each week, or each month, in every single research/scientific discipline in the world? What you don't see often are these wholesale, abrupt jumps in discovery, which really, in the history of science, do not occur very often. Scientific knowledge and progress are overwhelmingly done in small, snail-pace steps. It is why you see so many scientific papers on various topics over so many years.

Zz.

 

[BillTre]

At least until recently, there was a problem of orphaned diseases/medical problems which had very few people working on them and little funding.

Some projects, if successful, would affect many disease treatments like working out how to use mRNA to treat disease states.

 

[Choppy]

In no particular order...
(Sorry - started typing this out last night, but didn't get a chance to finish if until now, sorry if there are any overlaps of points)

1. Standards, Ethics and Codes
   In terms of outright productivity, when it comes to medical research, for example, there are a lot more rules in place now than there were years ago. You can't just start experimenting on patients or even animal models because you have a good idea. Patients are entitled to a standard of care and must give informed consent for any trial. Identifying patients who are willing to consent to a clinical trail can be very difficult. Animal research must be vetted by committees and certain standards need to be in place to run a lab that uses animals. These rules are important. They've evolved out of a general desire in the scientific community to do better, to act more ethically and humanely, and avoid both incorrect results, accidents and cases of mistreatment. This slows the pace of research down, and for good reason.
   
   
2. Ownership of Information
   From a scientific/medical research point of view, it would be great if we just had one world wide repository of information that tracked everything about everybody. Sequence your genome, track your diet, habits, radiation exposure, identify when you develop cancer, track every intervention and follow up with quantifiable outcomes and put this all in a giant database that any scientist can access... but you can't. People are entitled to a certain level of control over their information. And, as a healthcare provider, if you collect information on a patient, there are laws that dictate with whom you can and cannot share that information.
   Another dimension to this is industrial research and development and the reluctance to share information because of the potential to capitalize on intellectual property. Again, I'm not arguing that this is a bad thing necessarily. But it means you have different groups working on exactly the same thing, none of whom are aware of what anyone else is doing, particularly in the commercial sector.
   
   
3. The Low Hanging Fruit Is Scarce
   I think its fairly safe to say that a lot of the "easy" scientific problems that face our world have been solved already. A lot of very smart people have spent a lot of time and money working on them. The problems that remain are the more difficult ones, problems that will require cooperation and collaboration from people with many different specializations in order to make progress. That's not so say that all the "easy" ones are gone, but they're likely a lot more rare than they used to be.
   
   
4. More Education/Training Needed
   Building on point 3, in order to push the envelope, the individual scientist needs to learn a lot about what's been done before. The problem is that more and more has already been done. So, from an educational point of view, we've gone from scenarios where someone with a bachelor's degree could become an effective researcher, to scenarios where guys who've completed multiple post-docs are still learning the ropes of their chosen discipline. Ultimately if the time needed to train a researcher goes up, the amount of time that researcher spends pushing the envelope goes down.
   
   
5. Technology Curve
   Technology is changing quickly, but in most cases any given technology has a learning curve associated with it. If you happen to be in a group on the cutting edge of a given technology, you can gobble up all the low hanging fruit that comes with its development. But once that happens, or if you're a little behind the curve, you don't get nearly the output as you might have. You spend just as much time on the learning curve, not doing research, and when you can, everything that can be done has been done. So you start the cycle over again.
   On this note there is also the amount of time it takes to learn/develop technology to get your lab group onto the cutting edge at all.
   
   
6. Greener Pastures for the Brightest Bulbs
   There just aren't that many academic opportunities out there compared to professional or commercial ones. For a student with a 4.0 GPA and all the other "high" credentials, there are other options that lead to more stable, higher paying careers - medicine, for example. If the "scientist" option remains "living just above the poverty line until you're thirty, then fight for a series of temporary jobs, then you get a chance to compete against 50 other highly qualified people for a job that you still might lose after five years"... we're going to continually loose a lot of gifted people to other options.|
   
   
7. Funding Per Scientist
   This one is speculation on my part, but I suspect that the amount of funding available for basic science research simply hasn't tracked with the growing number of PhD graduates. This would make the major grants more competitive to get and those who do get them have to spend the bulk of their time preparing grant applications instead of doing research.
   
   
8. Pressure To Publish
   You would think that this would lead to an increase in major discoveries on the surface. Instead, it seems to have lead to an explosion of low quality journals, and a tendency for groups to publish much smaller incremental steps in their work. If has a group, you put out say, four papers per year on a topic instead of one "big" paper, you end up doing four times the writing, four times the back-and-forth review with referees, four times the figure preparation... all of which leads to less time actually doing the research.