4 ms·

There are several good reasons to reject the argument that costs make massive government intervention necessary. First, a good telescope in the days of Galileo

by zeteo 13y ago

There are several good reasons to reject the argument that costs make massive government intervention necessary. First, a good telescope in the days of Galileo or an electric generator in Tesla's were still pretty expensive high-tech toys. They got financed through private means anyway. Second, the number of subfields that truly demand massive capital investment is small, and mostly confined to physics. Many, if not most projects these days can be conducted with a few graduate students / postdocs, who are paid close to minumum wage and use existing equipment. Still, this kind of "cheap" research is nonetheless conducted overwhelmingly with government funds. Third, money doesn't guarantee quality. Major scientific advances of the past were achieved not so much by people who were very well funded, but rather who had leisure to think about their field in novel ways (Maxwell, Darwin, Mendel, Einstein etc.). The constraints, stress and uncertainty of the grant-award process are not conducive to a similar environment. Finally, increased costs are also a symptom of lack of innovation [1]. Maybe we're just on the final legs of existing logistic curves and have no new ones to latch on to? [1] http://en.wikipedia.org/wiki/Diffusion_of_innovations http://en.wikipedia.org/wiki/Diffusion_of_innovations

astrodust 13y ago
A "large hadron collider" funded by private industry would never have happened, or if it did, it would have a significantly different mandate. As far as we can tell, here's no money to be made with Higgs bosons. That is, there's no money to be made today in that. Eighty years in the future we might realize some commercial applications, just as how we're still mining theoretical research done eighty years ago when building new integrated circuits or optical switching technologies. There is an enormous amount of government funded work out there that, while not always glamorous or even practical, provides a huge boost to future initiatives that can leverage it. Developing commercial applications is made easier, more frictionless, because researchers don't have to do this work themselves. Increased costs are not a sign of a lack of innovation. It's a sign that people are compelled to tackle larger problems. All the easy stuff has already been done. A hundred and fifty years ago you could tinker around with random radioactive compounds in your garage and win a Nobel Prize. Today you'll have to work pretty hard to find something that hasn't already been experimented with.
tensor 13y ago
Here is another data point for you. I went to grad school with public funding and obtained a PhD in CS. I'm one of these who worked for "close to minimum wage". Would I have done it if my research was to be used to further the profits of a corporation instead of being released free to the world? Not a chance. What if they were willing to pay 10x what I earned? Would I do it then? Maybe, though perhaps not even then. I'd sooner work for myself. I don't think I'm alone either. Among my peers, nobody wanted corporate grants. The sentiment was that they would rather just quite and get a job for full pay if a corporate grant where their research went to the corporation was the only option. Further, I'd argue that corporations today simply do not do any science at all. They do engineering, sure, but very little science. You might counter and point to drug companies, but the work they do is highly criticized as is with claims that they outright won't work on cures because cures don't make them ongoing profits. They also won't work on any disease that is not extremely common. The days of Bell Labs and Xerox Parc are largely over with few holdouts still doing interesting work that is released for the public good (Google, Microsoft). Public funding has yielded huge numbers of scientific and mathematical advances over the years. Many people here have given plenty of examples already, though there are a great many more too.
- dnautics 13y ago
  why do you assume that private funding has to go to a corporation? There are nonprofits too. And then there are guys like this: http://en.wikipedia.org/wiki/Peter_D._Mitchell#Independent_researcher http://en.wikipedia.org/wiki/Peter_D._Mitchell#Independent_r... And of course, Salk made the polio vaccine, which was no small feat, with zero federal funding, and he even refused to patent the thing.
  - tensor 13y ago
    This is why I was careful with my words about where the research would end up, and where the funding would come from. These examples along with government funding are distinct in that they are not motivated by profit, the overall theme of the discussion. Suffice to say, if people won't support science via government funding, I suspect they are equally if not more unlikely to do it via nonprofits. And I say that based on observations of opt-in vs opt-out that come from behavioural economics (which ironically may never have been discovered without public funding). In short, I don't see these ever sustaining the amount of scientific work that is currently going on.
jjoonathan 13y ago
Those aren't good reasons, and at the risk of sounding more patronizing than I would prefer, I suspect that you aren't very well acquainted with the research world outside of Math/CS. > First, a good telescope in the days of Galileo or an electric generator in Tesla's were still pretty expensive high-tech toys. They got financed through private means anyway. 100% of the research done x number of years ago was funded using the funding methods of x years ago; the existence of successes is no indicator of the efficiency of the process. Plenty of alchemists threw their funding away on projects that would never meet the standards demanded by the NSF, NIH, and DOD. We both know that science funding would fall precipitously if it were left to the private sector. Do you really think that money would go to a better use? > Second, the number of subfields that truly demand massive capital investment is small, and mostly confined to physics. This is the reason why I accused you of not being well acquainted with modern research. The set of subjects in which science is cheap looks more like {CS, Math} than ~{Physics}. Chemists have to buy (pay the depreciation of, if you insist) their spectrometers (NMR, IR, mass), chromatographs, calorimeters, Schlenk lines, proprietary reagents, and gloveboxes. Biologists rely on sequencers, synthesizers, designer enzymes & antibodies, synchrotrons (monochromatic X-ray sources for crystallography), nanofab tools for microfluidics (all insanely expensive because of the semiconductor industry), patch clamps, and a million specialized commercial assays and imaging techniques. Astronomers depend on tons of custom made hardware: mirrors, specialized ASICs/FPGAs/CCDs, lenses, buildings in the most inhospitable places on earth, not to mention their space platforms. Ecologists and geologists need to travel to exotic locations almost by definition (requiring specialized transport & shelter) to set up sensor networks and the like. Although data-crunching is important, it doesn't even constitute the bulk of the scientific process. You can't replace academia with a bunch of weekend-warriors on laptops (outside of CS, that is). > Third, money doesn't guarantee quality. Major scientific advances of the past were achieved not so much by people who were very well funded, but rather who had leisure to think about their field in novel ways (Maxwell, Darwin, Mendel, Einstein etc.). Theorists have the most impressive long-term returns. By definition, they define their fields. This says very little about science in general or the utility of intermediate physical results, which inevitably guide the development of the underlying theories. Thought alone can't tell you which theories are right; at some point you have to test them. Einstein would have been ridiculed had it not been for the availability of ludicrously precise measurements of Mercury's precession and the Michelson-Morley experiment, both of which failed to match any metric of usefulness before their importance was understood. Darwin was inspired by his travels and Maxwell essentially had the E&M problem dropped in his lap by the experimentalists. Quantum Mechanics managed to derive macroscopically useful results (MO theory, semiconductor & condensed matter models, etc) by puzzling over observed atomic spectra. Who would have wondered why the rainbow had holes in it if nobody bothered to go look? Nobody. The "huh?" always precedes the theoretical explanation and the "huh?" usually comes from experiment, as well as countless guiding nudges along the way that eliminate possible explanations. Theory & experiment exist symbiotically. I don't think focusing on theory to the exclusion of experiment will benefit science as a whole. > The constraints, stress and uncertainty of the grant-award process are not conducive to a similar environment. Agreed, but I'm pretty sure the process of begging money off of rich individuals is even less conductive to scientific progress. Grant committees are much less susceptible to buzzwords & unrealistic promises than uninformed* private investors and even so they sometimes falter. NSF = due dilligence. It sucks, but someone has to do it, and I'm not convinced that the private sector could do it cheaper/better, given that it doesn't have any incentive to: science yields much more favorable ROI to society than individuals because the goods it produces aren't excludable and take 100 years to perform a proper valuation on. * = not an expert in everything, I don't mean to imply investors are stupid. > Finally, increased costs are also a symptom of lack of innovation [1]. Maybe we're just on the final legs of existing logistic curves and have no new ones to latch on to? That's exactly the case. It's also usually the case. When a breakthrough happens it only takes 5-10 years to climb the curve you linked to (I'm basing this claim off of the development of QM). The academic world has an extraordinary capacity for innovation (and not in the "look I built a CRUD app that makes somebody's life infinitesimally easier" sense of the word), although this capacity typically goes unappreciated due to the 99.9% failure rate of experiments conducted in the interim between periods of readily available low-hanging fruit (where failure := negative ROI). It's unfortunate that breakthroughs happen much less frequently than every 5-10 years but there really isn't any way to shortcut the process, at least not one I can think of (and not one that any of the world's experts in the field can think of, else they would simply go pursue that line of inquiry on their own). I don't think defunding America's scinetific apparatus would accelerate scientific progress and I think we would run a terrible risk of handing the next wave of industrial dominance to China & India if we tried. They're rapidly catching up in terms of publication counts/funding and it's no secret that new industries are born out of academic discoveries and that early entry into a new market conveys a huge advantage. The bulk of the spoils go to the victor, so I desperately hope the USA stays in the race.

by zeteo 13y ago

astrodust 13y ago

A "large hadron collider" funded by private industry would never have happened, or if it did, it would have a significantly different mandate. As far as we can tell, here's no money to be made with Higgs bosons. That is, there's no money to be made today in that. Eighty years in the future we might realize some commercial applications, just as how we're still mining theoretical research done eighty years ago when building new integrated circuits or optical switching technologies. There is an enormous amount of government funded work out there that, while not always glamorous or even practical, provides a huge boost to future initiatives that can leverage it. Developing commercial applications is made easier, more frictionless, because researchers don't have to do this work themselves. Increased costs are not a sign of a lack of innovation. It's a sign that people are compelled to tackle larger problems. All the easy stuff has already been done. A hundred and fifty years ago you could tinker around with random radioactive compounds in your garage and win a Nobel Prize. Today you'll have to work pretty hard to find something that hasn't already been experimented with.

tensor 13y ago

Here is another data point for you. I went to grad school with public funding and obtained a PhD in CS. I'm one of these who worked for "close to minimum wage". Would I have done it if my research was to be used to further the profits of a corporation instead of being released free to the world? Not a chance. What if they were willing to pay 10x what I earned? Would I do it then? Maybe, though perhaps not even then. I'd sooner work for myself. I don't think I'm alone either. Among my peers, nobody wanted corporate grants. The sentiment was that they would rather just quite and get a job for full pay if a corporate grant where their research went to the corporation was the only option. Further, I'd argue that corporations today simply do not do any science at all. They do engineering, sure, but very little science. You might counter and point to drug companies, but the work they do is highly criticized as is with claims that they outright won't work on cures because cures don't make them ongoing profits. They also won't work on any disease that is not extremely common. The days of Bell Labs and Xerox Parc are largely over with few holdouts still doing interesting work that is released for the public good (Google, Microsoft). Public funding has yielded huge numbers of scientific and mathematical advances over the years. Many people here have given plenty of examples already, though there are a great many more too.

dnautics 13y ago
why do you assume that private funding has to go to a corporation? There are nonprofits too. And then there are guys like this: http://en.wikipedia.org/wiki/Peter_D._Mitchell#Independent_researcher http://en.wikipedia.org/wiki/Peter_D._Mitchell#Independent_r... And of course, Salk made the polio vaccine, which was no small feat, with zero federal funding, and he even refused to patent the thing.
- tensor 13y ago
  This is why I was careful with my words about where the research would end up, and where the funding would come from. These examples along with government funding are distinct in that they are not motivated by profit, the overall theme of the discussion. Suffice to say, if people won't support science via government funding, I suspect they are equally if not more unlikely to do it via nonprofits. And I say that based on observations of opt-in vs opt-out that come from behavioural economics (which ironically may never have been discovered without public funding). In short, I don't see these ever sustaining the amount of scientific work that is currently going on.

jjoonathan 13y ago

Those aren't good reasons, and at the risk of sounding more patronizing than I would prefer, I suspect that you aren't very well acquainted with the research world outside of Math/CS. > First, a good telescope in the days of Galileo or an electric generator in Tesla's were still pretty expensive high-tech toys. They got financed through private means anyway. 100% of the research done x number of years ago was funded using the funding methods of x years ago; the existence of successes is no indicator of the efficiency of the process. Plenty of alchemists threw their funding away on projects that would never meet the standards demanded by the NSF, NIH, and DOD. We both know that science funding would fall precipitously if it were left to the private sector. Do you really think that money would go to a better use? > Second, the number of subfields that truly demand massive capital investment is small, and mostly confined to physics. This is the reason why I accused you of not being well acquainted with modern research. The set of subjects in which science is cheap looks more like {CS, Math} than ~{Physics}. Chemists have to buy (pay the depreciation of, if you insist) their spectrometers (NMR, IR, mass), chromatographs, calorimeters, Schlenk lines, proprietary reagents, and gloveboxes. Biologists rely on sequencers, synthesizers, designer enzymes & antibodies, synchrotrons (monochromatic X-ray sources for crystallography), nanofab tools for microfluidics (all insanely expensive because of the semiconductor industry), patch clamps, and a million specialized commercial assays and imaging techniques. Astronomers depend on tons of custom made hardware: mirrors, specialized ASICs/FPGAs/CCDs, lenses, buildings in the most inhospitable places on earth, not to mention their space platforms. Ecologists and geologists need to travel to exotic locations almost by definition (requiring specialized transport & shelter) to set up sensor networks and the like. Although data-crunching is important, it doesn't even constitute the bulk of the scientific process. You can't replace academia with a bunch of weekend-warriors on laptops (outside of CS, that is). > Third, money doesn't guarantee quality. Major scientific advances of the past were achieved not so much by people who were very well funded, but rather who had leisure to think about their field in novel ways (Maxwell, Darwin, Mendel, Einstein etc.). Theorists have the most impressive long-term returns. By definition, they define their fields. This says very little about science in general or the utility of intermediate physical results, which inevitably guide the development of the underlying theories. Thought alone can't tell you which theories are right; at some point you have to test them. Einstein would have been ridiculed had it not been for the availability of ludicrously precise measurements of Mercury's precession and the Michelson-Morley experiment, both of which failed to match any metric of usefulness before their importance was understood. Darwin was inspired by his travels and Maxwell essentially had the E&M problem dropped in his lap by the experimentalists. Quantum Mechanics managed to derive macroscopically useful results (MO theory, semiconductor & condensed matter models, etc) by puzzling over observed atomic spectra. Who would have wondered why the rainbow had holes in it if nobody bothered to go look? Nobody. The "huh?" always precedes the theoretical explanation and the "huh?" usually comes from experiment, as well as countless guiding nudges along the way that eliminate possible explanations. Theory & experiment exist symbiotically. I don't think focusing on theory to the exclusion of experiment will benefit science as a whole. > The constraints, stress and uncertainty of the grant-award process are not conducive to a similar environment. Agreed, but I'm pretty sure the process of begging money off of rich individuals is even less conductive to scientific progress. Grant committees are much less susceptible to buzzwords & unrealistic promises than uninformed* private investors and even so they sometimes falter. NSF = due dilligence. It sucks, but someone has to do it, and I'm not convinced that the private sector could do it cheaper/better, given that it doesn't have any incentive to: science yields much more favorable ROI to society than individuals because the goods it produces aren't excludable and take 100 years to perform a proper valuation on. * = not an expert in everything, I don't mean to imply investors are stupid. > Finally, increased costs are also a symptom of lack of innovation [1]. Maybe we're just on the final legs of existing logistic curves and have no new ones to latch on to? That's exactly the case. It's also usually the case. When a breakthrough happens it only takes 5-10 years to climb the curve you linked to (I'm basing this claim off of the development of QM). The academic world has an extraordinary capacity for innovation (and not in the "look I built a CRUD app that makes somebody's life infinitesimally easier" sense of the word), although this capacity typically goes unappreciated due to the 99.9% failure rate of experiments conducted in the interim between periods of readily available low-hanging fruit (where failure := negative ROI). It's unfortunate that breakthroughs happen much less frequently than every 5-10 years but there really isn't any way to shortcut the process, at least not one I can think of (and not one that any of the world's experts in the field can think of, else they would simply go pursue that line of inquiry on their own). I don't think defunding America's scinetific apparatus would accelerate scientific progress and I think we would run a terrible risk of handing the next wave of industrial dominance to China & India if we tried. They're rapidly catching up in terms of publication counts/funding and it's no secret that new industries are born out of academic discoveries and that early entry into a new market conveys a huge advantage. The bulk of the spoils go to the victor, so I desperately hope the USA stays in the race.