Where did you get that premisse? Nobody is arguing about a patent reform that automatically grants patents for everything, so I'm inclined to see this as a false dilemma.
How about this? "Inventions are now patented which would not have been patented previously because researchers could not *automatically* get patents for federally-funded research."
That was the point I was responded to with "I have no idea what you're talking about." You then saw fit to ride in with a "RTFA" response, which, you should now see was a hasty mistake on your part. Naturally, I won't hold my breath waiting an apology or admission of error.
(1)by the produced articles, as you said (and which they themselves say is difficult to pinpoint, but this IS the way in academic circles to put a value onto it, so it's not like they're making the criterion up)....
It might be a way to measure U.S. contribution to science vs. that of the world. It's a much bigger leap to infer that the Bayh-Dole Act or licensing is the cause of that. That was my point.
(2)They also use the data of the total number of FDA-approved compounds and biologics in a current year, (3)but also at how many priority NMEs are making it through...
...which metrics they off-handedly reference but do not cite, and merely suggest are "worse" and cherry-pick a particular fact. No verification; no sources that we can use for cross-referencing.
I mentioned in another comment why Fortune chose not to report its source for these metrics - because the metrics vary a lot depending on which fields you look at, and how you define your terms. This isn't obfuscation; it's genuinely hard to correlate NME data to pharmaceutical company productivity. See my other comment, and better still the article I referenced there, for more info.
And lastly, they looked at the university-linked businesses which are using these patents, to see what they actually produce in innovative products, which in most cases have almost nothing to show for.
That's an excellent circular argument. "As proof of pharmaceutical company productivity, let's look at university business. Oh, hey, it's way up! They earn great licensing revenue and get a lot of patents! Well, that just proves that the Bayh-Dole Act is strictly a distraction; all of that business activity is happening in spite of the proven low productivity of pharmaceutical companies (see above.)"
He (the parent poster) clearly was talking about research flowing *back* into the *academic commons*." Notice the word research.
Research "flows?" What an imagination you have. You're using the fanciful part of your mind to envision some magical, intangible substance called "research" that graciously wafts back and forth between the gleaming towers of academia. If you close your eyes, and maybe ingest some hallucinogens, you can almost see the twinkling streams of "research," rippling as if brushed by a light breeze, flowing happily from one university to another. But here comes the big bad cave troll called Business, or maybe Patents, which gobbles up all of the flowing research and leaves behind a charred wasteland. No more flowing "research!" The landscape of research has been forever polluted, twisted into a blackened desert of greed!
Back here in the real world, "research" is an activity that takes place in a lab. It doesn't "flow" in any way, shape, or form, so it's impossible for "research" to "flow" back into a lab. Money flows in; results flow out - that's about it.
In other words, the literal meaning of that statement is nonsensical. I was trying to translate the original poster's nonsensical statement into what he probably meant, in order to respond to it. Again, you hastily rode in with a condescending "RTFA" response - without stopping to consider that the meaning you claim I missed is nonsense.
This was your second mistake. Again, I won't await an admission of error. I will just recommend that you quit while you're... not farther behind.
I'm sorry, but the two parent posters which already responded to you with a "RTFA" are right: you *should* RTFA.
I did - last night - and I didn't see any mention of some kind of change in the patent process, whereby researchers used to "automatically" receive patents for anything. Show me where the article mentions that.
RTFA. It clearly indicates that there IS a considerable impact on research. I mean, what, it has two pages which deal with several actual criteria to show the downfall in innovative research...
Criteria? Try "criterion," singular. The only metric they cited about "impact on research" is a really crappy one: "While the number of journal articles produced by American researchers has risen slightly since 1988, the rest of the world has raced ahead (see chart)."
The conclusion from this single metric that U.S. research is slowing down due to the Bayh-Dole Act lies somewhere is thoroughly specious. The rest of the world has spent the last 20 years finally getting into biotech. And as any actual scientist can tell you, publication quantity != knowledge quality. Etc.
He (the parent poster) clearly was talking about research flowing *back* into the *academic commons*. Federal funding isn't flowing back; it's simply payed by our tax-money.
How was it "flowing back" before? What magical mechanism existed to reinvest money in successful labs that was tragically shredded by the Bayh-Dole Act and the patent system? There wasn't one. Since that number was $0 before 1980, it can't have "sharply decreased," now can it?
I like how "RTFA" has become the defacto "STFU N00B" comment. It has about the same amount of persuasiveness, too.
It claims that new drugs (not old drugs repatented under a different formula and target) have come on the market less frequently than before the Bayh-Dole act.
Did you see any metrics supporting that claim? No. Go back and look at page 6 - they dance around the issue, discussing life expectancy and pharmaceutical company revenue. The closest data that they provide to show that pharmaceutical development is unchanged is... wait for it... a suggestion that it's "worse." They make a hand-wave reference to charts that they don't cite. Notice that?
Why would Fortune fail to include some numbers that support its core point? I spent some time this afternoon looking for metrics on the number of new drugs, or (as a different metric) approved NMEs, over the past decades. I didn't find any. It's suprisingly unavailable. Go ahead, look for yourself.
There's a reason for that. This article attempts to analyze the rate of new drug developments (in order to answer an unrelated question about the FDA approval process.) From his article:
"There is considerable variation across diseases--even diseases in the
same broad disease groups--in the extent and timing of increases in the stock of available
drugs. This is illustrated by Figure 6, which shows, for two conditions--diseases of the
thyroid gland and diseases of other endocrine glands--the number of drugs available to
treat the condition in year t, as a percent of the number of drugs available to treat the
condition in 1979.23 Between 1979 and 1984, the number of drugs available to treat
diseases of the thyroid gland increased 29%, while the number of drugs available to treat
diseases of other endocrine glands increased only 13%. However, between 1984 and
1998, the number of drugs available to treat diseases of the thyroid gland did not increase
at all, while the number of drugs available to treat diseases of other endocrine glands
increased 33%."
In a nutshell - it's difficult to measure the rate of new drug development, because the results vary depending on how you define key factors. Does this not raise some suspicion about Fortune's vague conclusion that drug development has gotten "worse?"
The costly experimentation in cell lines and animals that you think corporations do is actually done by the research scientists receiving those grants.
You're either being ignorant or clever. This preclinical research is done by academics - but under industry sponsorship, not under federal grants. Federal grants do not pay for drug validation. You couldn't get funding from the government to do that. The for-profit companies pick up the costs at that stage.
Before your next post, try googling "industry-sponsored research."
Once you show some hard data that the Bayh-Dole act lead to an increase in productivity, we'll talk again. Until then, the evidence points firmly to an unintended consequence of stifling innovation by reducing the incentive for scientists to collaborate.
Again, nice trick. You're faced with two unproven assertions, and you choose to believe one without proof unless I can prove the other. That's not very rational.
If your premise is correct, why were new drugs and treatments coming out before 1980?
Because rather than licensing a novel idea from a university (which wasn't an option), pharmaceutical companies conducted their own in-house research. This was expensive, slow, unreliable, and generally incompatible with the business realities of for-profit companies like those in the pharmaceutical industry. It was a bad fit, and as a result, drug development was slower.
In your bullet points you forgot marketing (including courting the doctors), on which is spent more money than all your other bullet points above combined.
I didn't forget it - I just didn't consider it a necessary precursor to putting a drug on the drug store shelf. Certainly there are a hundred similarly extraneous costs: follow-up research (including Phase IV clinical research), administration, facilities, scientific publications, legal support for the process.
I agree that pharmaceutical marketing is pretty egregious. It's a weird trend that I don't fully understand.
The typical drug company TV spot usually sounds like this: "Take Plavitor! It will improve your life! - Warning: Do not take Plavitor if you have any of the following 36 conditions..."
What are they accomplishing with this? They're not informing their customers of what the product actually does. And it's not like they're successfully brainwashing patients into ambling down to the local Walgreen's and buying a prescription of Plavitor - it's a controlled substance. I guess they might be creating an army of zombie patients who wander into doctors' offices moaning, "Plavitor... must have Plavitor..." But presumably (hopefully) the doctors know that Plavitor specifically cures gout, and declines the requests of the 99.999% of patients who don't have gout. So presumably, this tactic doesn't really increase sales.
I have a very strong objection to pharmaceutical companies lobbying doctors, congressmen, and insurance companies. That has a clear risk of biasing the system, and at the very least reeks of the "appearance of impropriety." This needs to change.
However, I've never seen metrics for the amount of money spent on pharmaceutical advertising/lobbying vs. drug development, and your suggeston that the former outweighs the latter is a little hard to believe. Can you cite some figures? If this is true, I'd like to have them on hand.
Heh - thanks for the support, but that's not me. David Stein isn't exactly a unique name!
In fact, I do have two advanced degrees - but neither one confers the title of "doctor," unfortunately. I'm a patent attorney, and I have a masters in computer science. I'll probably go back and earn a PhD in artificial intelligence, if only out of curiosity, but that's down the road.
But I do appreciate the support. The ad hominem stuff is out of place, of course - but this being the Internet, I'm used to looking past it. I usually write it off as the last refuge of the defeated pundit.;)
The main idea behind patents was that companies don't hold the ideas secret so that other can build further inventions upon them.
Yes, the idea is to compel a public disclosure of the invention so that others can built on it - eventually, i.e., after the patent expires. That's exactly how it works - even in the pharmaceutical industry. Now that the patent for Prozac has expired, world+dog is making variations on it.
But none of this happens unless you can compel the inventors to disclose the invention. That's what a patent does: a patent must be sufficiently clear and detailed to teach the public about the invention and how to use it. But you have to give the inventor something to compel that disclosure. If the reward doesn't exceed what he could make by keeping it as a trade secret, he won't disclose it.
So the question is: How do you reward inventors? The drafters of the Constitution stridently debated this point. The best competing idea was that the government would pay some kind of reward fee, but they realized that the federal government couldn't afford to give sufficiently large incentives. They settled on the patent monopoly - a protected right to have the only product on the market for a few years, and charge higher prices as a result.
Now, let's get back to your suggestion. If we institute a compulsory licensing scheme, where patentees must grant licenses to an invention, then we essentially short-circuit the reward system of patenting. Companies will stop applying for patents, and will instead keep the formula for Zoloft as a trade secret - giving them an monopoly on it until someone else happens to reverse-engineer it. And those companies will then waste resources by creating reverse-engineering methods, e.g., adding a thousand inert materials to their pills in order to frustrate analysis. That is inefficient and arbitrary.
Today, the Census Bureau reported that 45 million Americans lacked health insurance in 2003, up by 1.4 million from 2002 and 5.2 million from 2000. The report states that this increase is "statistically significant." *
45 million people = 15% of the population. The other 255 million Americans are the ones who, by and large, have no idea how much drugs cost.
They probably vaguely know that drugs are expensive, and that healthcare costs are crushing America, but they really don't see much connection with their drugs. It's the same as with taxes: everyone knows we have a huge federal deficit, but no one is personally willing to pay $1 more in federal taxes (or give up $1 in federal benefits) to fix it.
The government doesn't do this because our Congress-critters are completely beholden to big drug companies and insurance companies to fund their campaigns and political party infrastructure, and doing this would make a dent in the drug company and insurance company profits.
You are entirely correct. There is rampant corruption in government due to lobbying. It's interesting that even Adam Smith, the spiritual icon of the laissez-faire capitalists, warned of grave danger in letting companies get too large, since they would use their profits to corrupt the political system. He was correct.
The answer to that problem is pretty simple: we need more responsible voters. We need people to pay attention to elections and evict politicians with undue ties to private organizations. Unfortunately, the deep political apathy of most Americans defeats that effort.
I don't have a solution to that problem. I can only agree with you here, both to the existence of this problem and the terrible consequences.
So perhaps we should modify the system so the development and licensing rights go to the highest bidder?
In theory, that's exactly what happens. The university holds the rights to the technology, shops it around, and licenses it to the company willing to pay the most. And the licensing revenue gets reinvested in the academic research of that university.
In practice, it's not quite that effective. First, pharmaceutical companies often strong-arm their way into cheaper deals. Frequently, they donate a little bit of money or some reagents to a researcher, in exchange for rights to any invention that comes out of the research. Unless the researcher or the university's technology transfer office sees this as a problem and fights against it, the university gets very little revenue from the invention.
Also, when it comes to licensing, there's a clear gap of expertise between pharmaceutical companies and academic institutions. Universities can't pay their licensing officers much, so they usually get guys fresh out of school who know very little about IP, biotech, and licensing. Many pharmaceutical companies take advantage of this.
The latter - licensing - obviously makes them more money...
Yes, I'm sure you know more about the economics of licensing than the collective wisdom of the entire pharmaceutical industry combined. We'll just accept your statements at face value.
Since you have this brilliant idea for a new method of marketing pharmaceuticals, then you should go start a new pharmaceutical company - maybe Open-Source Drugs, Inc. I presume I'll hear your name in 10 years as the man who revolutionized pharmaceuticals, and when I do, I'll come back to this thread and admit defeat. Good luck with that!
Violate fiscal responsibility? Gimme a break. That is the LAST thing ANY corporate management gives a shit about.
Right. That's why they spend so much time monitoring the stock price and financials of their companies. Not every company is Enron, you know.
You're just another clueless management type that can't comprehend that risk equals reward.
Wrong. I've never managed anyone in my life. I am in the trenches. I do the actual nuts-and-bolts work of negotiating licenses. And I do it on the side of academia, not big pharma - I have plenty of vitriol for the pharmaceutical industry, and I could go into great detail about how they maliciously oppress innovation and academia. Where you and I differ is that I don't see them as just monolithic Bad Guys - my overall view of the licensing process is that it's logical and it works, even if they have distorted it in several ways for personal gain.
But you've seen fit to question my credentials, and now you know them. Might I ask you to return the favor, and share with us some of your credentials? Turnabout is fair play.
I shall start this great organization. I shall create it, for all of us. And I shall call it... government.
And the current system is completely amenable to that.
The federal government has a right to use, and even develop products for, every invention developed with federal funds - whether patented or not. If it so decided, it could develop a drug to compete with Viagra; it could even license the right to do so to another company.
It has that power today. It could start doing it immediately.
So why doesn't it? Answer: Because the federal government is not interested in developing drugs. That obvious fact, and obvious failure of the government to commercialize inventions, was the whole reason for the Bayh-Dole Act.
If you want to change the situation, then I encourage you to do so. Enter the ranks of the NIH, reach some level of prominence where you can compel change, and create a program for federal-government-brand pharmaceuticals. The legal system is amenable to this, and it might be an interesting result. Good luck! We'll look forward to your success.
Government research into basic technologies for the space program, NASA, military (especially aviation) programs and the department of energy...
...which, together, produced a total of zero new prescription drugs. The Bayh-Dole Act is primarily aimed at basic biological research with the goal of creating new diagnostic and therapeutic drugs. Last I checked, neither NASA, the defense department, nor the department of energy was much interested in manufacturing pharmaceuticals.
I am being forced to invest in your company by paying your R&D costs.
No one is ever going to force you to buy a pharmaceutical. You always have the choice of not buying it.
You are patenting a scientific fact (the principal by which a medication or class of medications works).
Nope. You can't patent scientific facts. Every patent application must demonstrate that the invention is useful for a specific purpose. In fact, this is particularly true for biological compounds - recently, the USPTO and the courts have raised the threshold for patenting new drugs. Applicants can't just state "may be useful to treat cancer, or MS, or diabetes, or Lyme disease, or HIV, or ingrown toenails, or..." Rather, they have to present some data that would persuade "one of ordinary skill in the art" that it might actually work for a specific purpose.
As someone who has spent most of the last 10 years with either utterly shitty prescription coverage or no coverage at all...
I think that's terrible and needs to be fixed immediately. But according to the U.S. Census, 85% of Americans have some form of health insurance. Many who don't can receive prescription drug assitance from the federal government. So most Americans are not in your situation.
Inventions are now patented which would not have been patented previously because researchers could not *automatically* get patents for federally-funded research. There was a complex process that wasn't invoked for relatively trivial or cumulative research.
I have no idea what you mean here. The process of getting a patent has not fundamentally changed since the Patent Act of 1952. Researchers had to jump through exactly the same hurdles to get a patent in 1979 as they did in 1981 (and today.)
The change in law led to a cash grab, and a research culture in which universities encouraged staff to patent any new developments, even those growing out of collaborative research. This led to commercial barriers being formed (licensing) which inhibited research and industrial application.
It hasn't impacted research much because (a) it's extraordinarily difficult to enforce any patent against the private research of another company, (b) there's little incentive to do so anyway, and (c) those other companies enjoy an "experimental use" exemption to patent infringement (which is narrow, but is being expanded.)
And while it did create barriers to industrial application, it had the result of giving all of the rights to one company, instead of spreading them so thinly across a field that no company has an interest in developing it. That was the whole point - even though everyone had a right to use those inventions, no one was actually using them. The Bayh-Dole Act changed that.
The amount of research flowing back into the academic commons is being sharply reduced.
Completely untrue. The amount of federal funding flowing into the academic commons has grown over time - particularly in the last ten years. It's not keeping pace with the growth of research in academia, but by no metric has it been "sharply curtailed" - quite the opposite.
You *do* pay twice, because you are paying licenses to use technology which was never previously patented because it grew out of the public domain and so there were greater barriers to patenting it because a bureaucratic approval process was required.
Okay - tell you what. Pick a disease, any disease, for which there is a patented drug. I guarantee that you can spend some time in PubMed and find a few dozen other ideas for biological mechanisms for curing the disease. They don't infringe the patent, and you can freely use them! All you have to do is develop your own drug applying that concept. Or, you can buy some kind of non-FDA-approved herbal supplement that the Acme Corporation swears is effective for that purpose. Some ginsana for your cancer, perhaps?
Very little, I believe - as I have executed approximately 2,000 IP-related licenses for an academic institution. (My experience is irrelevant to this discussion, but since you brought it into play, I figured I'd answer it.)
To get to the heart of your response: Let's say Merck develops a great new drug. Now they have two options: they can develop it themselves under a patent monopoly, or they can develop their version and license it to other companies to develop alternatives. Which one makes them more money? If it's the latter, why aren't they doing it already? And if it's the former, wouldn't the latter violate the company's fiscal obligation to give stockholders a return on their investments?
If you rely on a patent to make you money, you will shortly go out of business or spend most of your monopoly profits fighting legal battles in court.
That's awfully funny. Yeah, I'm sure that's why companies have tremendously cranked up their patent investments in recent decades - because patents completely don't work. A simple glance at the state of any large industry will demonstrate how wrong you are. I could explain why, but you wouldn't read the explanation.
The argument that you need so much money to bring out a drug is bogus! The biotech industry can do with less, but because it is so easy to pump the clientel it is done!
Seriously. Do you do any kind of research before you make such ridiculous statements?
Do you know anything about the path between licensing a drug concept from a university, and actually putting a drug on the market? In part, it involves:
Research into chemicals that can effectuate the idea licensed from the university.
Research into the potential side-effects of this drug in the human body in a wide variety of circumstances.
Research into potentially adverse interactions with every other drug on the market.
Research into drug formulation: assessing various chemical structures for maximum effectiveness, dosage and frequency tweaking, additives to prevent spoilage, and additives to reduce side effects (upset stomach, etc.)
All of the infrastructure for actually manufacturing the drug on a very large scale.
Costly experimentation of the drug in cell lines and animals to make sure it works as expected.
Phase I clinical trials: tests of low doses in a small group of humans to test for serious adverse events - stuff like, you know, death.
Phase II clinical trials: months or years of long-term testing on hundreds of patients, as a long-term efficacy test. And since these are double-blinded, they require a large control group of patients to receive placebos.
Phase III clinical trials: several years of testing on thousands of patients for more thorough testing and tweaking before the product goes on the market.
And, of course, the infrastructure to submit all of this data to the FDA and to push for approval of the drug for clinical use.
By necessity, this is a lengthy, complex, costly process. If you shortcut the process, you get, you know, thalidomide.
I detest patents because they don't promote innovation. Cars became popular because a Mr Henry Ford broke the car patent and mass produced a car!
Your example is not an example of a "patent-free world." It's an example of a deceptive patent system that lures a company to develop a product by a promise of monopoly that turns out to be illusory. And you're right - that works very well... until those developer companies realize that it's consistently illusory, and stop developing new products. Then you're screwed.
I like capitalism, but bio-tech is a vulture preying on the lives of humans. They say, "Oh you want to live? Well here is a drug, but guess what it is going to cost you!"
For most Americans, the answer is: very little. Usually it's like a $2 co-pay, thanks to the miracles of health insurance and Medicare prescription drug benefits.
Of course, it's nowhere near this cheap. It's just that the costs are defrayed through other mechanisms. Those mechanisms absolutely need an overhaul - they're buckling under the weight of pharmaceutical company markups and other sleaziness - and we need to extend them to all Americans. But the drama you see in the sitution just doesn't exist for most Americans.
In other words you have proved the argument that by copying we get a vibrant industry! You get real innovation!
That's a more plausible argument in software, where the development costs are essentially nil. It's less persuasive in biotech because of the large initial costs.
In fact, the field of pharmaceuticals proves your suggestion catastrophically wrong. Do you have any idea how many great ideas for drug targets are unpatented, unexploited, and ready to be tried? Many thousands - probably closer to a million. We have hundreds of ideas for drugs that would probably cure every disease on the planet. They're not patentable, so they can be freely used!
In light of this fact, do smaller drug companies choose to develop new drugs based on them? No. They create knockoffs of Viagra and
I think you're missing something-if that idea was developed using public money, mine and yours, isn't that idea publically owned, mine and yours?
That's what we had before the Bayh-Dole Act: such inventions were owned by the federal government on behalf of the people. It didn't work out too well. No company could afford to step up to the plate, take the idea off of the shelf, and develop it into a successful product - not without the ability to prevent competitors from immediately copying it. It was a completely unfeasible business model. It was so bad that it prompted massive federal legislation, in the form of the Bayh-Dole Act, to repair it.
In essence, if you want to dispense with the monopoly-drug business model, you'll have to replace the entire second half of the development track for drugs. You'll have hundreds or thousands of raw, freely usable ideas for diagnosing and curing disease. It will be up to you to figure out how to use them. And good luck with that - it's usually a pretty hard technical problem, taking several man-decades of experimentation. Alternatively, you can just buy a formulation on the store shelf with no FDA testing... perhaps a Wikipedia page containing dozens of (anonymous) recommendations of which drugs cure which diseases....
I think there is still a case to be made that drug companies are getting something of a free ride at taxpayers' expense.
In general, I agree with you. Like oil companies, pharmaceutical companies achieve annual net profits that can only be described as indefensible.
I like your idea about channeling some big pharma profits back into basic research. But I think that's an indirect route to solving the core problem with big pharma: the complete lack of a normal capitalist buyer/seller market for prescription drugs. No patient in America knows or cares how much drug companies actually get paid for their prescription, since they never pay the price directly; the costs are completely hidden by the impenetrable bureaucrasies of health insurance and federal healthcare payments. It's a ridiculous scandal, but one that most Americans will only see in retrospect after it has collapsed.
Pressuring big pharma to sell drugs at more realistic prices would tremendously reduce the crushing healthcare burden on the federal government. Those spare tax dollars could be used in many other ways - for instance, more federal funding of basic research.
I did RTFA. Like all Fortune articles, it's ten pages long, comprising four pages of anecdotes, two pages of buzz-clip quotes, four pages of unsupported supposition, and zero pages of logical reasoning. Fortune caters to the C?O set, so it eschews attention-span-draining critical analysis for talking points.
And yes, you DO pay twice (or X or multiple times X) when the company is granted a monopoly on a drug invented by someone else. The issue - and the point of the problem the article has with Bayh-Dole - is that those with patent monopolies tend NOT to license or cross-license their products - which raises the cost of those products to monopoly rates and limits the spread of the IP - which is supposed to be the point of patents in the first place.
Of course they don't license their products. If you spent $5 billion developing and testing a new drug, would you let your competitors buy the rights to sell their own versions? And if you did, how quickly do you think your company's investors - the people who paid money to help build your company - would sell their stocks and run?
Again, I'm not here to defend the pharmaceutical industry, which I view as undeniably corrupt. But their product strategy is simple reality, owing to the very simple fact that copying is easier than innovating. Isn't that why everyone (rightly) criticizes Microsoft - because it lets everyone else take the R&D risks, and then just copies their ideas? (Stacker/DoubleSpace, Netscape/IE, Eudora/Outlook, WordPerfect/Word, etc.)
...and I'll follow up that post with some observations about the problems with academic research.
Basic academic research is a phenomenally important beast - perhaps the most important component of our long-term economy and the long-term health of our nation. It's impossible to overstate this or to use enough hyperbole here.
And contrary to the tiresome pop-culture tirades, America doesn't have a terrible shortage of scientists. We could use more, sure, but our labs are jam-packed with Ph.D.s and incoming students. In fact, we have a slight overabundance of them, which is reflected in their unappreciatively low salaries.
When I talk to scientists about the factors holding back their research - which I do frequently, in fact, as part of my job - I get a pretty consistent answer.
Grants are getting more competitive. While federal funding overall is growing, it hasn't kept pace with the explosion in life sciences research in labs across the country. Life sciences research complexity and data have experienced Moore's-Law-like rates of growth; just look at the sizes of GenBank and the Protein Data Bank over time. But because all universities are increasingly focusing on federal research dollars (in part because their academic rankings are based on it), it's becoming harder every year to get a grant funded. Scientists must show more data, and argue more convincingly that the research is useful.
Contributing factors:
The federal government has taken a stronger role in guiding research. A growing proportion of federal funds is earmarked for research on specific topics - e.g., AIDS or bioterrorism defenses. This necessarily diverts funds away from other areas of research.
Also, the administrative oversight of federal grants has grown. Universities are now held to much stricter rules about how those federal funds are spent. If a grant proposal indicates that a scientist will commit 40% of his employment to a project, he must actually commit that 40%. Funds must be carefully tracked by university admins - no more "I want to buy a Light Cycler with my unrelated grant money" discretionary BS.
These are real rules, and are strictly enforced. The penalties for violating them include paying the grant money back to the federal government, large penalty fines, and potentially getting barred from future federal funding. As a result, scientists and university admins have to spend tremendous time on administrative tasks - which detracts from research.
Those are the factors that scientists most frequently cite. Surely there are others.
Ah yes, another good "let's bash patents" thread on Slashdot.
The irony in this case is that the Bayh-Dole Act has almost nothing to do with patents. The overlap is essentially nil. Before the Bayh-Dole Act of 1980, universities could patent inventions; after the Act, universities can still patent inventions, in exactly the same manner (foregoing the completely unrelated changes in patent law.)
What the Bayh-Dole Act changed is who gets to commercialize federally-funded inventions.
Before the Act, all inventions created by federal grants, which of course are almost all academically-created inventions, were owned by the federal government. But the government had a terrible track record of commercialization. As a result, the technologies and patents just got locked away inside the government and went unused. Eventually, the patents expired; but no company was willing to spend R&D developing a brand-new but unpatented product - if it succeeded, competitors would just copy it.
The Bayh-Dole Act recognized and remedied this functional problem. Universities were allowed to keep and commercialize the rights to inventions created by their employees. That commercialization usually entails patenting the invention, and then exclusively licensing the patent(s) to a single company, so that they can develop a product. In fact, the Bayh-Dole Act places affirmative obligations on the university to commercialize it in a timely manner: if they don't, the federal government can seize the rights to the invention and patents.
Now let's talk results. If you want to see the benefits of the Bayh-Dole Act, look at pharmaceuticals. How many new, extremely useful drugs have hit the market in the last 25 years? I am not a defender of big pharma - their enormous lobbying efforts and enormous profit margins reflect an obese and gluttonous industry - but it's undeniable that they have created a host of useful drugs since the Bayh-Dole Act passed.
So that's the Bayh-Dole Act in a nutshell. Hopefully it's apparent that this has essentially nothing to do with patents. It pertains to who is responsible for commercializing an invention: the university that created it, or the federal government. We can certainly discuss the positive and, yes, the negative aspects to pressuring universities to commercialize their technologies; but that's a completely different discussion than the effect of patents on innovation.
(One common but flawed Slashdot comment that I'd like to head off at the pass: the ridiculous notion that taxpayers are somehow "paying for the invention twice" by first funding the research, and then paying monopoly prices for the drugs. That reasoning is specious, because the public is paying once for two discrete steps. First, an institution creates an idea for curing a disease, like "let's inhibit this protein as a cure multiple sclerosis." Taxpayer dollars cover that step. Second, a company takes that idea and spends $$$$$$$ to develop a drug that executes this principle. That's paid for by customer purchases and health insurance. They're two distinct steps - not "the same thing" being bought twice.)
But I propose strictly GPL and copying here, not sublicencing.
Huh? Sublicensing is the whole point of the GPL.
Look -
Person A originally creates some code and released it under the GPL.
Person B takes the code, improves it, and passes it along.
Person C downloads the modified code from person B.
In this case, the GPL requires B to grant certain rights to C. C receives rights from B, flowing down through the license between A and B. That's what we call a sublicense.
Now, let me explain why your suggestion matches the Seattle Computing scenario. Obviously, A = Seattle Computing; B = Microsoft; C = IBM.
The GPL permits A to give a copy to B, and A may charge B for the cost of delivering the original program. However, A may not forbid B from redistributing the code, in original or unmodified form, or from charging for that distribution.
You think this scenario is fine. You recommended changing it only by having A charge a very large sum for the one-time transfer. That is exactly what Seattle Computing and Microsoft did. Microsoft, having some actual business sense, redistributed the modified version on very different terms. The profound results naturally followed.
The rule, in lay terms, is that if you implement a patent in GPL'd code, then you implicitly license that patent to any users of the GPL'd code...
If it's implicit, then why are you asking me to read the license?
Snarky comment aside:
You are essentially suggesting that if
(a) Microsoft releases a product derived from GPLed code, and
(b) the derivative product embodies a patentable software invention, and
(c) Microsoft chooses to patent that invention, ...then the entire world has the right to use that patent.
I don't believe that the GPL reads that way. Not implicitly, and absolutely not explicitly.
Here's how it really works. The novel portion of the GPL pertains to flow-down rights. If you take a GPLed product, improve it, and pass it along, the users of that derivative product have certain rights. I think what you're interpreting as "implicit" is that those rights may supersede any patents held by the creator of the derivative product - which is different.
Even if we accept this supersessory flow-down right as "implicit," it's not what's happening in the Microsoft/Eidos scenario I suggested. Eidos doesn't derive superseding rights from Microsoft's derivative product - because Eidos isn't using Microsoft's derivative product at all. Instead, it has developed a product that embodies an invention also embodied by a product patented by Microsoft that, incidentally, was developed under the GPL. That's a completely different situation.
(Aside from that, the discovery issue still remains extremely problematic. I shouldn't need to elaborate on that.)
And that is exactly why, if your work is really innovative and valuable, you may and should put very high price on "transfer" of it, orders of magnitude higher then end-user proprietary software "licencing" price - your customer should think about it as an investment.
Let me tell you a story about a company called Seattle Computer Products.
In 1980, Seattle Computer Products spent some time creating a really neat disk management package called QDOS. They then operated on exactly the model you just suggested. They offered the software for sale at $50,000 apiece, with the free right to sublicense it.
Guess who bought it? A little upstart called Microsoft. Microsoft then sublicensed it exactly once, to IBM, as a new product called MS-DOS - but the sublicense reserved royalties for Microsoft.
Where are Microsoft and IBM today? And where is Seattle Computer Products?
Slashdot Mirror
How about this? "Inventions are now patented which would not have been patented previously because researchers could not *automatically* get patents for federally-funded research."
That was the point I was responded to with "I have no idea what you're talking about." You then saw fit to ride in with a "RTFA" response, which, you should now see was a hasty mistake on your part. Naturally, I won't hold my breath waiting an apology or admission of error.
(1)by the produced articles, as you said (and which they themselves say is difficult to pinpoint, but this IS the way in academic circles to put a value onto it, so it's not like they're making the criterion up)....
It might be a way to measure U.S. contribution to science vs. that of the world. It's a much bigger leap to infer that the Bayh-Dole Act or licensing is the cause of that. That was my point.
(2)They also use the data of the total number of FDA-approved compounds and biologics in a current year, (3)but also at how many priority NMEs are making it through...
I mentioned in another comment why Fortune chose not to report its source for these metrics - because the metrics vary a lot depending on which fields you look at, and how you define your terms. This isn't obfuscation; it's genuinely hard to correlate NME data to pharmaceutical company productivity. See my other comment, and better still the article I referenced there, for more info.
And lastly, they looked at the university-linked businesses which are using these patents, to see what they actually produce in innovative products, which in most cases have almost nothing to show for.
That's an excellent circular argument. "As proof of pharmaceutical company productivity, let's look at university business. Oh, hey, it's way up! They earn great licensing revenue and get a lot of patents! Well, that just proves that the Bayh-Dole Act is strictly a distraction; all of that business activity is happening in spite of the proven low productivity of pharmaceutical companies (see above.)"
He (the parent poster) clearly was talking about research flowing *back* into the *academic commons*." Notice the word research.
Research "flows?" What an imagination you have. You're using the fanciful part of your mind to envision some magical, intangible substance called "research" that graciously wafts back and forth between the gleaming towers of academia. If you close your eyes, and maybe ingest some hallucinogens, you can almost see the twinkling streams of "research," rippling as if brushed by a light breeze, flowing happily from one university to another. But here comes the big bad cave troll called Business, or maybe Patents, which gobbles up all of the flowing research and leaves behind a charred wasteland. No more flowing "research!" The landscape of research has been forever polluted, twisted into a blackened desert of greed!
Back here in the real world, "research" is an activity that takes place in a lab. It doesn't "flow" in any way, shape, or form, so it's impossible for "research" to "flow" back into a lab. Money flows in; results flow out - that's about it.
In other words, the literal meaning of that statement is nonsensical. I was trying to translate the original poster's nonsensical statement into what he probably meant, in order to respond to it. Again, you hastily rode in with a condescending "RTFA" response - without stopping to consider that the meaning you claim I missed is nonsense.
This was your second mistake. Again, I won't await an admission of error. I will just recommend that you quit while you're... not farther behind.
- David Stein
I did - last night - and I didn't see any mention of some kind of change in the patent process, whereby researchers used to "automatically" receive patents for anything. Show me where the article mentions that.
RTFA. It clearly indicates that there IS a considerable impact on research. I mean, what, it has two pages which deal with several actual criteria to show the downfall in innovative research...
Criteria? Try "criterion," singular. The only metric they cited about "impact on research" is a really crappy one: "While the number of journal articles produced by American researchers has risen slightly since 1988, the rest of the world has raced ahead (see chart)."
The conclusion from this single metric that U.S. research is slowing down due to the Bayh-Dole Act lies somewhere is thoroughly specious. The rest of the world has spent the last 20 years finally getting into biotech. And as any actual scientist can tell you, publication quantity != knowledge quality. Etc.
He (the parent poster) clearly was talking about research flowing *back* into the *academic commons*. Federal funding isn't flowing back; it's simply payed by our tax-money.
How was it "flowing back" before? What magical mechanism existed to reinvest money in successful labs that was tragically shredded by the Bayh-Dole Act and the patent system? There wasn't one. Since that number was $0 before 1980, it can't have "sharply decreased," now can it?
- David Stein
I like how "RTFA" has become the defacto "STFU N00B" comment. It has about the same amount of persuasiveness, too.
It claims that new drugs (not old drugs repatented under a different formula and target) have come on the market less frequently than before the Bayh-Dole act.
Did you see any metrics supporting that claim? No. Go back and look at page 6 - they dance around the issue, discussing life expectancy and pharmaceutical company revenue. The closest data that they provide to show that pharmaceutical development is unchanged is... wait for it... a suggestion that it's "worse." They make a hand-wave reference to charts that they don't cite. Notice that?
Why would Fortune fail to include some numbers that support its core point? I spent some time this afternoon looking for metrics on the number of new drugs, or (as a different metric) approved NMEs, over the past decades. I didn't find any. It's suprisingly unavailable. Go ahead, look for yourself.
There's a reason for that. This article attempts to analyze the rate of new drug developments (in order to answer an unrelated question about the FDA approval process.) From his article:
"There is considerable variation across diseases--even diseases in the same broad disease groups--in the extent and timing of increases in the stock of available drugs. This is illustrated by Figure 6, which shows, for two conditions--diseases of the thyroid gland and diseases of other endocrine glands--the number of drugs available to treat the condition in year t, as a percent of the number of drugs available to treat the condition in 1979.23 Between 1979 and 1984, the number of drugs available to treat diseases of the thyroid gland increased 29%, while the number of drugs available to treat diseases of other endocrine glands increased only 13%. However, between 1984 and 1998, the number of drugs available to treat diseases of the thyroid gland did not increase at all, while the number of drugs available to treat diseases of other endocrine glands increased 33%."
In a nutshell - it's difficult to measure the rate of new drug development, because the results vary depending on how you define key factors. Does this not raise some suspicion about Fortune's vague conclusion that drug development has gotten "worse?"
The costly experimentation in cell lines and animals that you think corporations do is actually done by the research scientists receiving those grants.
You're either being ignorant or clever. This preclinical research is done by academics - but under industry sponsorship, not under federal grants. Federal grants do not pay for drug validation. You couldn't get funding from the government to do that. The for-profit companies pick up the costs at that stage.
Before your next post, try googling "industry-sponsored research."
Once you show some hard data that the Bayh-Dole act lead to an increase in productivity, we'll talk again. Until then, the evidence points firmly to an unintended consequence of stifling innovation by reducing the incentive for scientists to collaborate.
Again, nice trick. You're faced with two unproven assertions, and you choose to believe one without proof unless I can prove the other. That's not very rational.
- David Stein
Because rather than licensing a novel idea from a university (which wasn't an option), pharmaceutical companies conducted their own in-house research. This was expensive, slow, unreliable, and generally incompatible with the business realities of for-profit companies like those in the pharmaceutical industry. It was a bad fit, and as a result, drug development was slower.
- David Stein
I didn't forget it - I just didn't consider it a necessary precursor to putting a drug on the drug store shelf. Certainly there are a hundred similarly extraneous costs: follow-up research (including Phase IV clinical research), administration, facilities, scientific publications, legal support for the process.
I agree that pharmaceutical marketing is pretty egregious. It's a weird trend that I don't fully understand.
The typical drug company TV spot usually sounds like this: "Take Plavitor! It will improve your life! - Warning: Do not take Plavitor if you have any of the following 36 conditions..."
What are they accomplishing with this? They're not informing their customers of what the product actually does. And it's not like they're successfully brainwashing patients into ambling down to the local Walgreen's and buying a prescription of Plavitor - it's a controlled substance. I guess they might be creating an army of zombie patients who wander into doctors' offices moaning, "Plavitor... must have Plavitor..." But presumably (hopefully) the doctors know that Plavitor specifically cures gout, and declines the requests of the 99.999% of patients who don't have gout. So presumably, this tactic doesn't really increase sales.
I have a very strong objection to pharmaceutical companies lobbying doctors, congressmen, and insurance companies. That has a clear risk of biasing the system, and at the very least reeks of the "appearance of impropriety." This needs to change.
However, I've never seen metrics for the amount of money spent on pharmaceutical advertising/lobbying vs. drug development, and your suggeston that the former outweighs the latter is a little hard to believe. Can you cite some figures? If this is true, I'd like to have them on hand.
- David Stein
In fact, I do have two advanced degrees - but neither one confers the title of "doctor," unfortunately. I'm a patent attorney, and I have a masters in computer science. I'll probably go back and earn a PhD in artificial intelligence, if only out of curiosity, but that's down the road.
But I do appreciate the support. The ad hominem stuff is out of place, of course - but this being the Internet, I'm used to looking past it. I usually write it off as the last refuge of the defeated pundit. ;)
- David Stein
Yes, the idea is to compel a public disclosure of the invention so that others can built on it - eventually, i.e., after the patent expires. That's exactly how it works - even in the pharmaceutical industry. Now that the patent for Prozac has expired, world+dog is making variations on it.
But none of this happens unless you can compel the inventors to disclose the invention. That's what a patent does: a patent must be sufficiently clear and detailed to teach the public about the invention and how to use it. But you have to give the inventor something to compel that disclosure. If the reward doesn't exceed what he could make by keeping it as a trade secret, he won't disclose it.
So the question is: How do you reward inventors? The drafters of the Constitution stridently debated this point. The best competing idea was that the government would pay some kind of reward fee, but they realized that the federal government couldn't afford to give sufficiently large incentives. They settled on the patent monopoly - a protected right to have the only product on the market for a few years, and charge higher prices as a result.
Now, let's get back to your suggestion. If we institute a compulsory licensing scheme, where patentees must grant licenses to an invention, then we essentially short-circuit the reward system of patenting. Companies will stop applying for patents, and will instead keep the formula for Zoloft as a trade secret - giving them an monopoly on it until someone else happens to reverse-engineer it. And those companies will then waste resources by creating reverse-engineering methods, e.g., adding a thousand inert materials to their pills in order to frustrate analysis. That is inefficient and arbitrary.
- David Stein
45 million people = 15% of the population. The other 255 million Americans are the ones who, by and large, have no idea how much drugs cost.
They probably vaguely know that drugs are expensive, and that healthcare costs are crushing America, but they really don't see much connection with their drugs. It's the same as with taxes: everyone knows we have a huge federal deficit, but no one is personally willing to pay $1 more in federal taxes (or give up $1 in federal benefits) to fix it.
- David Stein
You are entirely correct. There is rampant corruption in government due to lobbying. It's interesting that even Adam Smith, the spiritual icon of the laissez-faire capitalists, warned of grave danger in letting companies get too large, since they would use their profits to corrupt the political system. He was correct.
The answer to that problem is pretty simple: we need more responsible voters. We need people to pay attention to elections and evict politicians with undue ties to private organizations. Unfortunately, the deep political apathy of most Americans defeats that effort.
I don't have a solution to that problem. I can only agree with you here, both to the existence of this problem and the terrible consequences.
- David Stein
In theory, that's exactly what happens. The university holds the rights to the technology, shops it around, and licenses it to the company willing to pay the most. And the licensing revenue gets reinvested in the academic research of that university.
In practice, it's not quite that effective. First, pharmaceutical companies often strong-arm their way into cheaper deals. Frequently, they donate a little bit of money or some reagents to a researcher, in exchange for rights to any invention that comes out of the research. Unless the researcher or the university's technology transfer office sees this as a problem and fights against it, the university gets very little revenue from the invention.
Also, when it comes to licensing, there's a clear gap of expertise between pharmaceutical companies and academic institutions. Universities can't pay their licensing officers much, so they usually get guys fresh out of school who know very little about IP, biotech, and licensing. Many pharmaceutical companies take advantage of this.
- David Stein
Yes, I'm sure you know more about the economics of licensing than the collective wisdom of the entire pharmaceutical industry combined. We'll just accept your statements at face value.
Since you have this brilliant idea for a new method of marketing pharmaceuticals, then you should go start a new pharmaceutical company - maybe Open-Source Drugs, Inc. I presume I'll hear your name in 10 years as the man who revolutionized pharmaceuticals, and when I do, I'll come back to this thread and admit defeat. Good luck with that!
Violate fiscal responsibility? Gimme a break. That is the LAST thing ANY corporate management gives a shit about.
Right. That's why they spend so much time monitoring the stock price and financials of their companies. Not every company is Enron, you know.
You're just another clueless management type that can't comprehend that risk equals reward.
Wrong. I've never managed anyone in my life. I am in the trenches. I do the actual nuts-and-bolts work of negotiating licenses. And I do it on the side of academia, not big pharma - I have plenty of vitriol for the pharmaceutical industry, and I could go into great detail about how they maliciously oppress innovation and academia. Where you and I differ is that I don't see them as just monolithic Bad Guys - my overall view of the licensing process is that it's logical and it works, even if they have distorted it in several ways for personal gain.
But you've seen fit to question my credentials, and now you know them. Might I ask you to return the favor, and share with us some of your credentials? Turnabout is fair play.
- David Stein
And the current system is completely amenable to that.
The federal government has a right to use, and even develop products for, every invention developed with federal funds - whether patented or not. If it so decided, it could develop a drug to compete with Viagra; it could even license the right to do so to another company.
It has that power today. It could start doing it immediately.
So why doesn't it? Answer: Because the federal government is not interested in developing drugs. That obvious fact, and obvious failure of the government to commercialize inventions, was the whole reason for the Bayh-Dole Act.
If you want to change the situation, then I encourage you to do so. Enter the ranks of the NIH, reach some level of prominence where you can compel change, and create a program for federal-government-brand pharmaceuticals. The legal system is amenable to this, and it might be an interesting result. Good luck! We'll look forward to your success.
- David Stein
I am being forced to invest in your company by paying your R&D costs.
No one is ever going to force you to buy a pharmaceutical. You always have the choice of not buying it.
You are patenting a scientific fact (the principal by which a medication or class of medications works).
Nope. You can't patent scientific facts. Every patent application must demonstrate that the invention is useful for a specific purpose. In fact, this is particularly true for biological compounds - recently, the USPTO and the courts have raised the threshold for patenting new drugs. Applicants can't just state "may be useful to treat cancer, or MS, or diabetes, or Lyme disease, or HIV, or ingrown toenails, or..." Rather, they have to present some data that would persuade "one of ordinary skill in the art" that it might actually work for a specific purpose.
- David Stein
I think that's terrible and needs to be fixed immediately. But according to the U.S. Census, 85% of Americans have some form of health insurance. Many who don't can receive prescription drug assitance from the federal government. So most Americans are not in your situation.
- David Stein
I have no idea what you mean here. The process of getting a patent has not fundamentally changed since the Patent Act of 1952. Researchers had to jump through exactly the same hurdles to get a patent in 1979 as they did in 1981 (and today.)
The change in law led to a cash grab, and a research culture in which universities encouraged staff to patent any new developments, even those growing out of collaborative research. This led to commercial barriers being formed (licensing) which inhibited research and industrial application.
It hasn't impacted research much because (a) it's extraordinarily difficult to enforce any patent against the private research of another company, (b) there's little incentive to do so anyway, and (c) those other companies enjoy an "experimental use" exemption to patent infringement (which is narrow, but is being expanded.)
And while it did create barriers to industrial application, it had the result of giving all of the rights to one company, instead of spreading them so thinly across a field that no company has an interest in developing it. That was the whole point - even though everyone had a right to use those inventions, no one was actually using them. The Bayh-Dole Act changed that.
The amount of research flowing back into the academic commons is being sharply reduced.
Completely untrue. The amount of federal funding flowing into the academic commons has grown over time - particularly in the last ten years. It's not keeping pace with the growth of research in academia, but by no metric has it been "sharply curtailed" - quite the opposite.
You *do* pay twice, because you are paying licenses to use technology which was never previously patented because it grew out of the public domain and so there were greater barriers to patenting it because a bureaucratic approval process was required.
Okay - tell you what. Pick a disease, any disease, for which there is a patented drug. I guarantee that you can spend some time in PubMed and find a few dozen other ideas for biological mechanisms for curing the disease. They don't infringe the patent, and you can freely use them! All you have to do is develop your own drug applying that concept. Or, you can buy some kind of non-FDA-approved herbal supplement that the Acme Corporation swears is effective for that purpose. Some ginsana for your cancer, perhaps?
- David Stein
Very little, I believe - as I have executed approximately 2,000 IP-related licenses for an academic institution. (My experience is irrelevant to this discussion, but since you brought it into play, I figured I'd answer it.)
To get to the heart of your response: Let's say Merck develops a great new drug. Now they have two options: they can develop it themselves under a patent monopoly, or they can develop their version and license it to other companies to develop alternatives. Which one makes them more money? If it's the latter, why aren't they doing it already? And if it's the former, wouldn't the latter violate the company's fiscal obligation to give stockholders a return on their investments?
If you rely on a patent to make you money, you will shortly go out of business or spend most of your monopoly profits fighting legal battles in court.
That's awfully funny. Yeah, I'm sure that's why companies have tremendously cranked up their patent investments in recent decades - because patents completely don't work. A simple glance at the state of any large industry will demonstrate how wrong you are. I could explain why, but you wouldn't read the explanation.
- David Stein
Seriously. Do you do any kind of research before you make such ridiculous statements?
Do you know anything about the path between licensing a drug concept from a university, and actually putting a drug on the market? In part, it involves:
By necessity, this is a lengthy, complex, costly process. If you shortcut the process, you get, you know, thalidomide.
I detest patents because they don't promote innovation. Cars became popular because a Mr Henry Ford broke the car patent and mass produced a car!
Your example is not an example of a "patent-free world." It's an example of a deceptive patent system that lures a company to develop a product by a promise of monopoly that turns out to be illusory. And you're right - that works very well... until those developer companies realize that it's consistently illusory, and stop developing new products. Then you're screwed.
I like capitalism, but bio-tech is a vulture preying on the lives of humans. They say, "Oh you want to live? Well here is a drug, but guess what it is going to cost you!"
For most Americans, the answer is: very little. Usually it's like a $2 co-pay, thanks to the miracles of health insurance and Medicare prescription drug benefits.
Of course, it's nowhere near this cheap. It's just that the costs are defrayed through other mechanisms. Those mechanisms absolutely need an overhaul - they're buckling under the weight of pharmaceutical company markups and other sleaziness - and we need to extend them to all Americans. But the drama you see in the sitution just doesn't exist for most Americans.
In other words you have proved the argument that by copying we get a vibrant industry! You get real innovation!
That's a more plausible argument in software, where the development costs are essentially nil. It's less persuasive in biotech because of the large initial costs.
In fact, the field of pharmaceuticals proves your suggestion catastrophically wrong. Do you have any idea how many great ideas for drug targets are unpatented, unexploited, and ready to be tried? Many thousands - probably closer to a million. We have hundreds of ideas for drugs that would probably cure every disease on the planet. They're not patentable, so they can be freely used!
In light of this fact, do smaller drug companies choose to develop new drugs based on them? No. They create knockoffs of Viagra and
That's what we had before the Bayh-Dole Act: such inventions were owned by the federal government on behalf of the people. It didn't work out too well. No company could afford to step up to the plate, take the idea off of the shelf, and develop it into a successful product - not without the ability to prevent competitors from immediately copying it. It was a completely unfeasible business model. It was so bad that it prompted massive federal legislation, in the form of the Bayh-Dole Act, to repair it.
In essence, if you want to dispense with the monopoly-drug business model, you'll have to replace the entire second half of the development track for drugs. You'll have hundreds or thousands of raw, freely usable ideas for diagnosing and curing disease. It will be up to you to figure out how to use them. And good luck with that - it's usually a pretty hard technical problem, taking several man-decades of experimentation. Alternatively, you can just buy a formulation on the store shelf with no FDA testing... perhaps a Wikipedia page containing dozens of (anonymous) recommendations of which drugs cure which diseases....
- David Stein
In general, I agree with you. Like oil companies, pharmaceutical companies achieve annual net profits that can only be described as indefensible.
I like your idea about channeling some big pharma profits back into basic research. But I think that's an indirect route to solving the core problem with big pharma: the complete lack of a normal capitalist buyer/seller market for prescription drugs. No patient in America knows or cares how much drug companies actually get paid for their prescription, since they never pay the price directly; the costs are completely hidden by the impenetrable bureaucrasies of health insurance and federal healthcare payments. It's a ridiculous scandal, but one that most Americans will only see in retrospect after it has collapsed.
Pressuring big pharma to sell drugs at more realistic prices would tremendously reduce the crushing healthcare burden on the federal government. Those spare tax dollars could be used in many other ways - for instance, more federal funding of basic research.
- David Stein
I did RTFA. Like all Fortune articles, it's ten pages long, comprising four pages of anecdotes, two pages of buzz-clip quotes, four pages of unsupported supposition, and zero pages of logical reasoning. Fortune caters to the C?O set, so it eschews attention-span-draining critical analysis for talking points.
And yes, you DO pay twice (or X or multiple times X) when the company is granted a monopoly on a drug invented by someone else. The issue - and the point of the problem the article has with Bayh-Dole - is that those with patent monopolies tend NOT to license or cross-license their products - which raises the cost of those products to monopoly rates and limits the spread of the IP - which is supposed to be the point of patents in the first place.
Of course they don't license their products. If you spent $5 billion developing and testing a new drug, would you let your competitors buy the rights to sell their own versions? And if you did, how quickly do you think your company's investors - the people who paid money to help build your company - would sell their stocks and run?
Again, I'm not here to defend the pharmaceutical industry, which I view as undeniably corrupt. But their product strategy is simple reality, owing to the very simple fact that copying is easier than innovating. Isn't that why everyone (rightly) criticizes Microsoft - because it lets everyone else take the R&D risks, and then just copies their ideas? (Stacker/DoubleSpace, Netscape/IE, Eudora/Outlook, WordPerfect/Word, etc.)
- David Stein
Basic academic research is a phenomenally important beast - perhaps the most important component of our long-term economy and the long-term health of our nation. It's impossible to overstate this or to use enough hyperbole here.
And contrary to the tiresome pop-culture tirades, America doesn't have a terrible shortage of scientists. We could use more, sure, but our labs are jam-packed with Ph.D.s and incoming students. In fact, we have a slight overabundance of them, which is reflected in their unappreciatively low salaries.
When I talk to scientists about the factors holding back their research - which I do frequently, in fact, as part of my job - I get a pretty consistent answer.
Grants are getting more competitive. While federal funding overall is growing, it hasn't kept pace with the explosion in life sciences research in labs across the country. Life sciences research complexity and data have experienced Moore's-Law-like rates of growth; just look at the sizes of GenBank and the Protein Data Bank over time. But because all universities are increasingly focusing on federal research dollars (in part because their academic rankings are based on it), it's becoming harder every year to get a grant funded. Scientists must show more data, and argue more convincingly that the research is useful.
Contributing factors:
The federal government has taken a stronger role in guiding research. A growing proportion of federal funds is earmarked for research on specific topics - e.g., AIDS or bioterrorism defenses. This necessarily diverts funds away from other areas of research.
Also, the administrative oversight of federal grants has grown. Universities are now held to much stricter rules about how those federal funds are spent. If a grant proposal indicates that a scientist will commit 40% of his employment to a project, he must actually commit that 40%. Funds must be carefully tracked by university admins - no more "I want to buy a Light Cycler with my unrelated grant money" discretionary BS.
These are real rules, and are strictly enforced. The penalties for violating them include paying the grant money back to the federal government, large penalty fines, and potentially getting barred from future federal funding. As a result, scientists and university admins have to spend tremendous time on administrative tasks - which detracts from research.
Those are the factors that scientists most frequently cite. Surely there are others.
- David Stein
The irony in this case is that the Bayh-Dole Act has almost nothing to do with patents. The overlap is essentially nil. Before the Bayh-Dole Act of 1980, universities could patent inventions; after the Act, universities can still patent inventions, in exactly the same manner (foregoing the completely unrelated changes in patent law.)
What the Bayh-Dole Act changed is who gets to commercialize federally-funded inventions.
Before the Act, all inventions created by federal grants, which of course are almost all academically-created inventions, were owned by the federal government. But the government had a terrible track record of commercialization. As a result, the technologies and patents just got locked away inside the government and went unused. Eventually, the patents expired; but no company was willing to spend R&D developing a brand-new but unpatented product - if it succeeded, competitors would just copy it.
The Bayh-Dole Act recognized and remedied this functional problem. Universities were allowed to keep and commercialize the rights to inventions created by their employees. That commercialization usually entails patenting the invention, and then exclusively licensing the patent(s) to a single company, so that they can develop a product. In fact, the Bayh-Dole Act places affirmative obligations on the university to commercialize it in a timely manner: if they don't, the federal government can seize the rights to the invention and patents.
Now let's talk results. If you want to see the benefits of the Bayh-Dole Act, look at pharmaceuticals. How many new, extremely useful drugs have hit the market in the last 25 years? I am not a defender of big pharma - their enormous lobbying efforts and enormous profit margins reflect an obese and gluttonous industry - but it's undeniable that they have created a host of useful drugs since the Bayh-Dole Act passed.
So that's the Bayh-Dole Act in a nutshell. Hopefully it's apparent that this has essentially nothing to do with patents. It pertains to who is responsible for commercializing an invention: the university that created it, or the federal government. We can certainly discuss the positive and, yes, the negative aspects to pressuring universities to commercialize their technologies; but that's a completely different discussion than the effect of patents on innovation.
(One common but flawed Slashdot comment that I'd like to head off at the pass: the ridiculous notion that taxpayers are somehow "paying for the invention twice" by first funding the research, and then paying monopoly prices for the drugs. That reasoning is specious, because the public is paying once for two discrete steps. First, an institution creates an idea for curing a disease, like "let's inhibit this protein as a cure multiple sclerosis." Taxpayer dollars cover that step. Second, a company takes that idea and spends $$$$$$$ to develop a drug that executes this principle. That's paid for by customer purchases and health insurance. They're two distinct steps - not "the same thing" being bought twice.)
- David Stein
Huh? Sublicensing is the whole point of the GPL.
Look -
Person A originally creates some code and released it under the GPL.
Person B takes the code, improves it, and passes it along.
Person C downloads the modified code from person B.
In this case, the GPL requires B to grant certain rights to C. C receives rights from B, flowing down through the license between A and B. That's what we call a sublicense.
Now, let me explain why your suggestion matches the Seattle Computing scenario. Obviously, A = Seattle Computing; B = Microsoft; C = IBM.
The GPL permits A to give a copy to B, and A may charge B for the cost of delivering the original program. However, A may not forbid B from redistributing the code, in original or unmodified form, or from charging for that distribution.
You think this scenario is fine. You recommended changing it only by having A charge a very large sum for the one-time transfer. That is exactly what Seattle Computing and Microsoft did. Microsoft, having some actual business sense, redistributed the modified version on very different terms. The profound results naturally followed.
- David Stein
If it's implicit, then why are you asking me to read the license?
Snarky comment aside:
You are essentially suggesting that if
...then the entire world has the right to use that patent.
(a) Microsoft releases a product derived from GPLed code, and
(b) the derivative product embodies a patentable software invention, and
(c) Microsoft chooses to patent that invention,
I don't believe that the GPL reads that way. Not implicitly, and absolutely not explicitly.
Here's how it really works. The novel portion of the GPL pertains to flow-down rights. If you take a GPLed product, improve it, and pass it along, the users of that derivative product have certain rights. I think what you're interpreting as "implicit" is that those rights may supersede any patents held by the creator of the derivative product - which is different.
Even if we accept this supersessory flow-down right as "implicit," it's not what's happening in the Microsoft/Eidos scenario I suggested. Eidos doesn't derive superseding rights from Microsoft's derivative product - because Eidos isn't using Microsoft's derivative product at all. Instead, it has developed a product that embodies an invention also embodied by a product patented by Microsoft that, incidentally, was developed under the GPL. That's a completely different situation.
(Aside from that, the discovery issue still remains extremely problematic. I shouldn't need to elaborate on that.)
- David Stein
Let me tell you a story about a company called Seattle Computer Products.
In 1980, Seattle Computer Products spent some time creating a really neat disk management package called QDOS. They then operated on exactly the model you just suggested. They offered the software for sale at $50,000 apiece, with the free right to sublicense it.
Guess who bought it? A little upstart called Microsoft. Microsoft then sublicensed it exactly once, to IBM, as a new product called MS-DOS - but the sublicense reserved royalties for Microsoft.
Where are Microsoft and IBM today? And where is Seattle Computer Products?
- David Stein