Is this true?

[Erebus the Black]

[Randomizer]

YES

 

It's easier to get grant money to improve existing research than it is to get it for new and original research ideas. Mostly because the granting agencies are like Hollywood studios in that they don't want to risk money on something that is unproven even though it might have the potential to be more rewarding.

 

I know the method to create carbon 50 and carbon 60 molecules, buckminister particles or buckyballs, was from left over research money from another project. The researcher couldn't get money for the idea that lead to a multi billion dollar field producing these carbon nanoparticles and now carbon nanotube technology. He was discouraged from pursuing this and the university was embarrased with the success.

[Student of Trinity]

The funding agencies all pay lip service to the idea that you don't know whether research will work out. The usually say that they want to support risky projects. But everybody likes a radical breakthrough that is also a sure thing best of all.

 

Sometimes it's pretty much just like in that cartoon. Often, though, the point is that in order to write a successful proposal, you need to have done a lot of thorough and careful thinking out of just exactly what you're planning to do, and why it's a good idea to do that, and what alternative approaches there might be, and what other people have done and are doing. Without all that preparation, your proposal is going to sound half-baked and amateurish, and hence a poor investment of tax dollars or industry R&D money. So you won't get the money without a lot of prep work.

 

Of course that thorough preparation takes a lot of time and effort. Principal investigators are normally expected to have steady jobs, such as tenured professorships, so that the planning of research projects gets done on some nickel other than the project grants themselves. But that much preparation can easily amount, in the end, to a large fraction of the total effort required to get the final results. I mean, once you've nailed it down as firmly as you have to, to get the grant, then actually doing it can turn out to be comparatively straightforward. If all goes well, you can find yourself spending much of the grant money on exploring further implications of the basic results you correctly anticipated getting.

 

So it can be difficult to distinguish where one project ends and the next one begins, and research can easily end up looking kind of like the cartoon, even if you're not really doing it that way on purpose.

 

One of the things that I think is really smart about the German system is that professors get a certain amount of funding that just comes with their chair. It's not enough to run your whole group, but it's enough to incubate bolder projects that aren't yet ready to impress grant proposal referees. Not too many countries do this, though. As far as I know, it's mainly a German thing.

[Lilith]

often you can save money on the preparation phase of research by offloading most of it to a prospective grad student as their project, thus killing two birds in one stone by giving them something to do

 

and then a couple of months in it turns out that your search of the literature was inadequate and somebody else already did the research you were going to have your student do, and you and the student have to spend valuable time redesigning the project to build on the other group's research instead

 

not that i'm bitter or anything

[Student of Trinity]

That's a real issue, and it's potentially a huge pain. What I can say is, I know the sort of research that gets done by the groups whose papers always have the long reference lists, and who often write review articles, and always have an authoritative overview of what all has been previously done by people in their field.

 

It's boring stuff. Paper # N+1.

 

The stuff that shakes things up, and moves things forward, is more often done by people who didn't bother to look at what everyone else had done, because they took it for granted that what they would do would be something new. And much more often than not, it was.

 

Research is like everything else. There's a big majority of people following the beaten path, and a few pathfinders, flailing away in the undergrowth, occasionally breaking through. If you're part of the undergrowth team, and good enough to survive there, then you don't really have to worry about the beaten path.

 

If you take that approach, then, yeah, in principle you can sometimes get burnt. But in my experience, honestly: it's not a big risk. I read far too few papers by other people. Several times I've read paper titles and abstracts that made me think, Aaah! They've done what we're doing! But on closer inspection, it has always been: man alive, they've blatantly overhyped something far less than what we're really achieving. The worst penalty I've actually suffered for my negligence in keeping up with the literature is to occasionally be forced by a referee to add a bit to my paper explaining just how different what we're doing really is.

[Alorael at Large]

Do notice that the two flowcharts are the same, just with the cycle rotated slightly and some hidden steps made explicit in the second. You write a grant, you get money, you do research. You need some money and research to write a grant, and you'll use some grant money and some leftover research to jumpstart the next project, but the process as a whole works.

 

Despite what SoT says, virtually no research comes out of nowhere. Breakthroughs aren't the n+1 papers, but they do build on a solid foundation of methods and understanding. Often the trick is to apply something to a new target, or to modify a simple assumption that everyone has taken for granted. If you actually find yourself working on the same thing as someone else, well, the chances of it being literally identical are miniscule. More often you've just discovered a collaborator, which isn't always what you want but also isn't always a bad thing.

 

—Alorael, who says this while working on a manuscript identical to someone else's. Both are under review, probably for the same journals. One is going to make an impact; the other is going to be a "me too" in some far less prestigious journal. The free market of science has its downsides.

[Dantius]

One of the better Onion articles relevant to the subject.

[Student of Trinity]

Writing a cutting-edge paper on nothing more than a sudden rush of blood to the head is probably more possible in theoretical physics than in other sciences. But it happens, and the results are over-represented among classic works.

[Randomizer]

Originally Posted By: Lilith

often you can save money on the preparation phase of research by offloading most of it to a prospective grad student as their project, thus killing two birds in one stone by giving them something to do

and then a couple of months in it turns out that your search of the literature was inadequate and somebody else already did the research you were going to have your student do, and you and the student have to spend valuable time redesigning the project to build on the other group's research instead

not that i'm bitter or anything

At least you did it in the era where computer searching of data bases is common. My advisor only paid for one search and the rest of us did it the old fashion way of key word searches through journal indexes and using the Science Citation Index to find papers that referenced know paper on a topic.

Another professor offered a bounty to students that could find papers missing from his search list.

Making sure you aren't duplicating someone else's work is a pain when it isn't cutting edge.

[Lilith]

Originally Posted By: Student of Trinity

If you take that approach, then, yeah, in principle you can sometimes get burnt. But in my experience, honestly: it's not a big risk. I read far too few papers by other people. Several times I've read paper titles and abstracts that made me think, Aaah! They've done what we're doing! But on closer inspection, it has always been: man alive, they've blatantly overhyped something far less than what we're really achieving. The worst penalty I've actually suffered for my negligence in keeping up with the literature is to occasionally be forced by a referee to add a bit to my paper explaining just how different what we're doing really is.

well in the end the results we got directly contradicted their findings, and we had better-quality data than they did, so it actually turned out okay for us