Are we going about Research the Wrong Way?

[Luminal]

Well, not wrong per se, yet not the most effective or swiftest form of progressing science either. (Skip to the bottom for the short version)

 

I have a "focus fire" concept on how to approach expanding fields of science and technology.

 

Essentially, universities, government agencies, and corporations (involved in the appropriate field, of course) could come together, and focus all of their funding and efforts on a small number of projects. These projects would be selected in advance based on how much they will advance that area of science, likely through a vote of those pulling the strings.

 

For example, computer research groups (IBM, Bell Labs; universities such as Carnegie Mellon, MIT; the Air Force, and so on) would select a project to focus all of their attention and resources on. This would be a project that they believe would push forward computer science more so than anything else in the works.

 

Let's say these groups chose quantum computing. Thousands (millions possibly, if this was an international effort) of engineers, programmers, and researchers would be recruited. Collectively, many billions of dollars would be invested in a time frame of a year or two. Brainstorming of ideas and discussions would be encouraged at all levels. Every breakthrough would be passed onto all of the other agencies immediately. Then, the collective breakthroughs made in this time period would be applied to all other aspects of computing.

 

Let's say that progress was increased at a five-fold rate in quantum computing breakthroughs, while other areas of computer engineering lagged behind quite a bit. The advances of quantum computing would be applied to the other areas that lagged (normal miniaturization of microchips).

 

Ideally, this would compensate (perhaps even exceed) the normal progression in those area had their been no focus on quantum computing advances Of course, quantum computing itself would be a decade ahead.

 

Afterwards, these groups would select another area of computer science and focus on it vigorously for several years, such as brain-computer interfacing (or anything).

 

In other words, when you are facing an enemy in an strategy game, you do not attack all foes simultaneously by spreading your damage; rather, you focus everything on one at a time, which weakens the entire fighting unit of the opposing force. Likewise, in science, breakthroughs in one area almost always benefit every other area tremendously.

 

Of course, I realize that the current system works fine, and probably will not change anytime soon. Doesn't stop me from pondering a more effective system.

 

The short version: There are five subfields in a particular field of science: A,B, C, D, and E. Instead of 10 years of progress in each, what if we devoted all resources to 2 intense years of A research, then 2 years of B research, and so on. After 10 years, you'd still have approximately the (at least) same level of technological achievement in each field, except that you could leverage new technologies off highly advanced research in A only after 2 years.

[someguy]

why are you in such a rush to discover new things? maybe the way we research is wrong because it is based on money and therefore those things that make the most money are discovered first. perhaps a better way would be to have what serves mankind the best discovered first. and that does not coincide with what people want the most or can afford. like seinfeld's standup comedy joke where he wonders why all these scientists have "wasted" their time discovering making seedless watermelons rather than something more productive because really, spitting out seeds is just too much of a pain. certainly i agree with you though that combining things together and intelligently orchestrating them would be a more efficient way of doing things but what makes our world go around is the marketplace and i think for your idea some of that would be compromised. we need separate companies fighting each other for our current society to work well. combining them eliminates that and removes incentive to make more money and steal a larger share of the market from their opponents. they are not looking to invent and design the best things, just the things that will make their stock prices go up the most. to change that would require fundamentally changing the whole of society on the entire planet. still not a bad idea in my opinion.

[Paralith]

I think a very large stumbling block to this approach would be deciding on the subject which is to be focused on. Especially if you're talking about an entire industry - getting all the people involved, even if you only focus on the leaders of the various groups, to agree on the one thing that is most important to work on would be very difficult. Many people will have different preferences and views on what exactly should get top priority. Not to mention the complexity of the sciences themselves and narrowing them down to finite subcategories. I can't speak for too many other disciplines, but I work in biomedical research - and I can't imagine looking at the vast depth and breadth of subjects that are currently being researched, and either trying to lump them together in various awkward groups or picking just one of them. Even if you focused on, say, heritable diseases. There are TONS of them, and they all work in different ways. And of the many inherited diseases that afflict humans, which deserves to be cured first? What group of people with these diseases will be given respite while the other group waits, and suffers and dies, for their disease's turn? I understand that with your method we might at least be able to knock down a disease or two right away, as opposed to waiting several years for all the diseases - but the difficulty of making the choice of where to start remains.

 

Then there's the issue of researchers who specialize in different fields. What will they do when their field of expertise is not the one chosen? Will we have any specialists when all the researchers have to change what they focus on every couple of years? And is it practical for every researcher out there to stop and learn all the necessary concepts every time a new subject is assigned? I think there's a lot we might loose with your proposed method, that may or may not outweigh what we would gain.

[someguy]

not only that but, it's kind of an awful thought but.. if you think about it being vulnerable to disease right now, before colonizing other planets is feasible, maybe disease is not really such a bad thing. i would prefer disease controlling our numbers than something like war, and really we need some sort of control for our numbers. or else at some point something may just give and a huge catastrophe might occur. i think it would better to control numbers as we go rather than in one big lump. (i'm not supporting like a big epidemic like the black death or something that to me would be one big nasty lump) so maybe the best thing would be to research colonization technology first.

 

i agree with you also that choosing what way to best use the r&d of mankind is somewhat of a problem. but i think if everything was controlled rather than free market it could work if it was controlled by the right people with the right information. it's sort of scary to think that nobody's at the helm of what mankind does as a whole. just free market and bureaucracy. personally i agree with plato that the philosopher kings would be the best way. in my opinion it would be a great day when i could feel safe and secure that the people deciding the fate of the world are the smartest ones alive rather than people like George bush. i would be pleased to trust them to make the right decision with where to focus our energies. the problem lies with having a system that can assure that philosopher kings is what we get and not selfish exploitative and corrupt kings.

[swansont]

- As has been touched upon, basic university research is about more than the subject matter, it's about training future researchers. When it came time to research "X" you'd find nobody trained in that field because there was no money to support it in the interim. You'd have had a drain of the accumulated expertise in these fields, and no way to regain it once its lost. You'd find yourself reinventing the wheel, over and over again; it takes many years to develop high-level expertise.

 

- Industry research is proprietary, and you'd have a difficult time getting them to share knowledge.

 

- Research is often incremental and sequential. If you have five groups investigating five pathways (one each) toward some goal, with the hope that one pans out, you gain little by having twenty-five groups, with five researching each pathway. You just tend to have more groups reaching the same dead end. There's a saying that it takes a woman nine months to make a baby, but you can't get nine women to make a baby in one month. Some things just take time, and throwing more resources at the problem will not necessarily speed it up. You reach a point of rapidly diminishing returns, so spreading the resources out often gives you a much better overall return.

[Paralith]

- Research is often incremental and sequential. If you have five groups investigating five pathways (one each) toward some goal, with the hope that one pans out, you gain little by having twenty-five groups, with five researching each pathway. You just tend to have more groups reaching the same dead end. There's a saying that it takes a woman nine months to make a baby, but you can't get nine women to make a baby in one month. Some things just take time, and throwing more resources at the problem will not necessarily speed it up. [emphasis mine - paralith] You reach a point of rapidly diminishing returns, so spreading the resources out often gives you a much better overall return.

 

That's another good point I forgot to mention. My boyfriend is a software programmer, and he has this book called The Mythical Man Month. Basically the book describes why having only one person work on a project will actually get the work done faster than it will by throwing five people on the project. Simply adding more people does not necessarily reduce the time needed, and can even prolong it as each person who makes progress has to stop and make sure the other people involved understand and are up to speed, decisions take more time, etc. This book focuses on the software industry, but obviously this phenomenon is not just limited to computers.

[swansont]

That's another good point I forgot to mention. My boyfriend is a software programmer, and he has this book called The Mythical Man Month. Basically the book describes why having only one person work on a project will actually get the work done faster than it will by throwing five people on the project. Simply adding more people does not necessarily reduce the time needed, and can even prolong it as each person who makes progress has to stop and make sure the other people involved understand and are up to speed, decisions take more time, etc. This book focuses on the software industry, but obviously this phenomenon is not just limited to computers.

 

I'm sure it's not. Adding people can be a benefit to some types of projects, especially if they bring different skills to the table; you'd rather not have e.g. your electronics whiz spending his/her time bolting a vacuum system together. But sometimes more cooks just make thing mediocre. And when you reach a certain size you have to take extra steps to coordinate the information, as you point out, and that often means meetings (ugh). There isn't a one-size-fits-all solution.

[Bluenoise]

I think the biggest problem with this argument is that most discoveries rely on discoveries in seemingly unrelated projects and fields. It's not just a matter of doing a lot of research on something to discover it. That's a very naive perception. To discover some thing the supporting environment and infrastructure has to be there. Notice how many major discoveries are often happen to be reported by the different groups at the same time? It's because the world was finally in a state where science had progressed to the point where the discovery had become virtually inevitable. In essence science had become primed for the discovery.

Now because research is a search for something unknown and there are often many hypothesis or none at all how can you accurately say "we have the tools and background knowledge necessary to archive this goal". Well you can't! Because you can't know exactly what the necessary precursors are till you're there.

Really often you must have some minimal understanding of the system as whole to progress an individual aspect of it.

 

Take cancer research for example. There has been progress definatley, life spans are slowly increasing. However to the volume of money that has been pumped into it is disproportionately large. I'd argue that the lack of solid progress on finding a "cure" isn't due to such an thing being completely impossible (at least in every sense of the word). But more so due to the fact that underling technology and infrastructure doesn't exist to allow for such a discovery. However unfortunately most government and populations are very poorly educated about what it takes to allow a scientific discovery to happen. So we have a disproportionate large sum of funding going directly to cancer research as apposed to research aimed at biologically engineering techniques that are desperately needed to advance the ultimate goals. Even physics and chemistry are very important to this field. Take biophotonics without this cancer research would go nowhere. Or what about all those underfunded organic chem labs designing synthesis techniques that are ultimately used to build the small molecules that we currently used to treat cancer?

 

Even take your example of quantum computing. It's likely a matter of not having proper material and techniques that is due to the slow rate of implementation, not a matter of a lack of brain power

It's really sad how little credit the guys working on the techniques and design and engineering get compared to the ones that apply it.

 

I say you put a disproportionate amount of funding into one thing. And you go nowhere fast.

[swansont]

Even take your example of quantum computing. It's likely a matter of not having proper material and techniques that is due to the slow rate of implementation, not a matter of a lack of brain power

It's really sad how little credit the guys working on the techniques and design and engineering get compared to the ones that apply it.

 

I say you put a disproportionate amount of funding into one thing. And you go nowhere fast.

 

Laser cooling and trapping, which is an enabling technology for things like quantum computer research, really took off when laser diodes became available, since they could be easily tuned and were inexpensive. Those came about as CD players were being developed. Enabling technologies are often co-opted from other research, and sometimes discovery is just accidental and has no roadmap. You'd lose these stepping stones if you concentrated on just a few areas.