c) teach non-programmers to view the world like that so that they can recognize these situations, submit an intelligently written ticket that actually explains the problem, and have the presence of mind to mention possible exceptions (transferred to a different hospital, etc) when talking about the possible solution.
I don't think medical facilities are really willing to pay nurses to write and debug code or scripts. Even if they didn't already cost more per hour than IT, they would take many more hours to solve the problem.
Step 2: If a process, follow the process analysis.
Step 3: Does the claim recite an abstract idea, law of nature or natural
phenomenon (a judicial exception)? If no, eligible.
If yes, proceed.
Step 4: Is the claim as a whole directed to a practical application of the abstract
idea, law of nature or natural phenomenon? If no, ineligible.
If yes, proceed.
A man-made tangible embodiment with a real world use is evidence of a
practical application.
Step 5: Does the claim cover substantially all practical applications of the
exception?
Is innovation based on the abstract idea, law of nature or natural
phenomenon foreclosed?
If yes, ineligible. If no, the claim qualifies as eligible subject matter.
An algorithm never deserves a patent. Practical application, tangible embodiment: patent eligible (for now).
Before you accept your next job you have an iris/retina/fingerprint scan (your choice) taken at a local government office. It gets sent to the IRS as a hash, using a function that will always identify you as you but will only identify other people as "not you" 98% of the time or so. It makes confirming your identity easy but creates too many false positives to be used forensically like a fingerprint. If you're afraid the government is still using the info to track you, choose the retina scan. You prove that you are you to the government once, then you're done for, say, 20 years. To use the system the IRS sells dirt cheap scan/hash devices to employers.
So when you start your new job:
New employer does the same scan and inputs your name and taxpayer ID into the device. Device sends the hash to the IRS. IRS confirms you're in the database and starts the W2/1099 paperwork or rejects you, and then sends the decision back to the employer within two or three minutes. The device could be a modified $50 smartphone, so it would be cheap enough for hiring nannies, fast enough for hiring day labor, and used on every job site or farm that can get a cell phone signal. Employers could buy ID checks (with encryption keys) the same way people buy minutes for pay as you go phones.
All of the NIH funded research is available after 1 year. So you don't have to pay once more if you are willing to wait.
I think university libraries (the principal customers for these publishers) will be the ones who successfully force a transition to either open access or cheap-access publications; budgets are too tight for them to be able to afford to keep supporting the current model.
Not many, but that's also a completely different assertion. Just because you discovered a target or created an animal model doesn't mean you've invented a drug. It also doesn't mean you've done most of the work involved in inventing a drug, the riskiest work involved in inventing a drug, or the hardest work in inventing a drug.
the cure isn't competing with the price of a dose of the treatment: it is competing with the entire cost of treating your disease until you die
Wrong, wronger, and wrongest. One imagines a "cure" is only given to people who have an actual medical problem (presumably to develop an actual cure, the mechanism of disease is fully exposed). Uncures are not so narrowly constrained.
Statins are consumed (or potentially consumed) by hundreds of millions of people with nothing more than a statistically elevated risk of possibly developing heart disease according to some rather arbitrary marker.
Those markers aren't to determine whether the patient has the disease, they're to determine whether the treatment is (maybe) appropriate for the patient.
A cure for atherosclerosis (what statins aim to treat) would be sold not just to everyone currently prescribed a statin for high chloresterol, but pretty much everyone, full stop. Not everyone has atherosclerosis due to high LDL chlolesterol (the indication for statins), but everyone has (or will have) atherosclerosis. Arterial plaque starts in teenagers and is almost universal in the elderly. It directly causes most cardiovascular disease. The market for such a cure would be pretty much everyone over 40 with access to a deep enough pocket to pay for it. The main risk would not be that it would capture only a small part of the high-cholesterol market; the risk would be governments all over the world overriding the patent after the normal "highest price the market will bear" marketing scheme is announced.
In order for your analysis to hold water, you need to cure pharma of diluting immense benefits to the few into an ocean of revenues from the many.
Not at all. There will be times when lots of people have the same symptom but only a few have the same root cause. A cure for that root cause would profitably treat those few - and leave the rest of that market for that symptom intact.
Now they are out of targets where search space scales linearly with findings. They have an exponential decay in findings to fight with, and you can't with with an exponential by brute force.
Do you mean blocking protein-protein interactions as opposed to blocking protein-small molecule interactions?
They did buy a few orders of magnitude of brute force by using antibodies as drugs: immune systems and recombinant molecular biology are a lot faster than chemists at coming up with new drug candidates.
It would be naive to not understand that like most large businesses, the pharmas are driven financial motives which drives their research and product development cycles.
Absolutely, which is why i said they would prefer to sell a cure.
Say it will take 8 years and 3 billion dollars in R & D to get your next product to market. It will be either a cure or a treatment for a chronic disease: your pick.
1. The treatment will compete with all of the other treatments on the market for marketshare. The cure won't have marketshare: it will have the market. There will be no competitors - until another cure is approved, that is.
2. Price. As far as the accountants at your insurance company are concerned, the cure isn't competing with the price of a dose of the treatment: it is competing with the entire cost of treating your disease until you die (or become someone else's problem). As long as the cure comes out cheaper than a decade of doctors bills, hospitalizations, tests, and lots of different pills, it's a good deal for your insurance company. The treatment, on the other hand, could only hope to command a portion of that revenue stream
3. Risk/time value of money. Would you rather be paid your next 10 years salary today or once a month over the next 10 years? Someone who buys your cure pays you in full, today. You book all of that revenue while you are still CEO and take home your bonus. Someone who buys your treatment pays you a little at a time until they switch to a competitor's drug. Or until they die. They are an uncertain revenue stream, not a sure thing.
The researchers find a target the drug companies take it from there, but increasingly it is the first part that is most expensive.
Ok, my turn to demand a source: Which target took $4 billion to identify?
Right now the industry side spends $135 billion on R&D for which it gets ~30 new drugs approved per year plus new research on already approved drugs. Most of that is spent on phase II and III clinical trials, which are costing up to $100M each these days. For pretty much all drugs the vast majority of money and man hours are spent on developing and proving the drug (in industry), not on the target.
It may in fact be cheaper for society to do all this on the government dime, there is a lot of waste in the drug industry a lot of it from its very nature as private research. Fixing this would involve the government massively increasing research funding and deliberately killing an industry, not likely in the short run.
A little of that waste in private research is due to university research: most of the targets identified in the literature turn out to be irreproducible or unusable:
I think there's a lot of room for an enlightened government to more efficiently turn dollars into drugs than the present system, but I'm not convinced yet. Especially not now with the congressmen in charge of the NIH dumping peer review for their own religious and political views.
Sorry, a couple of years ago I looked at a year's worth of drug approvals and came up with 15%. The actual data (1998-2007) say 24% came from academia:
Of course more and more university research is funded by Pharma these days, especially the efforts that are most likely to lead to new drugs. Which column would you put that drug in?
No. The difference is that digital streaming like YouTube or Netflix must compete. Cable systems own their regions and milk their subscribers. Streaming systems will never be able to jack rates like cable because they're fungible.
Even worse than that: the monthly caps that cable systems place on subscribers will primarily hit streaming companies like Netflix, especially as HD streaming becomes more common. On the other hand, a streaming service that is partnered with your cable/ISP will not count toward your cap. Neat, huh?
What exactly would this "Premium Content" be? What do they have in that crappy little window that is so wonderful and "Premium" that I will gladly pay them for it?
I'd guess stuff owned by movie and TV studios that have competitive reasons for not wanting to be on Hulu, Netflix, or Amazon Prime, and aren't aligned with a cable/ISP company that is developing its own streaming system.
It would be nice to have some tests to determine if the problem is related to some chemical imbalance.
Give it fifteen years. Your doc will have you snort a dose of a labelling compound up your nose (privileged route past the blood brain barrier) like it was a line of cocaine or inject you with the labelling compound attached to little piece of a rabies protein ( another way to get stuff past the BBB). Then they'll pop you in an MRI machine and generate a 3D map of serotonin concentration in your brain.
A slightly paranoid person might buy into the theory that Big Pharma doesn't WANT to cure patients. Instead, they want to hook people on life-long "cures" that prove to be very lucrative.
Actual cures would be much more lucrative. If pharmas had a choice, they'd develop a cure rather than a treatment for any serious chronic disease. Unfortunately, for most non-infectious diseases a "cure" would mean making a fundamental and permanent change in how your body operates. Which pretty much means gene therapy, something which we really don't have a handle on yet.
And that most of the research money the Pharma companies spend is on doing clinical trials to see which ones actually work in humans after the university researchers have found potential candidates testing in cell cultures and animal models.
Hell no. About 15% of drugs come from academic research, the rest are invented by biotech or pharma companies. For the most part academic labs identify new drug targets. Most of the compounds they develop to test their hypotheses are for the most part useless as actual active pharmaceutical ingredients due to toxicity, bioavailability, and metabolism.
Considering the "I'm willing to go thermonuclear war on this" patent wars Apple has instigated in the smart phone and tablet markets, I wouldn't exactly say they gave them away.
Microsoft wants to copy the success of Apple and Google in mobile,
Microsoft wants to copy the success of RIM/BlackBerry. The success of RIM in the Pre-Iphone/Android era, that is: selling business to business in a locked down environment.
personally I am waiting for a 3D printer that can do things like rubber o-rings, seals, gaskets. that will be huge if you can get all the companies to supply the right sized drawings.
The last issue will be supplying the drawings. someone will have to draw them all out and that will take decades.
Gaskets and o-rings would have to be redesigned to be made out of thermoplastics to be 3D printed... I guess you could use the printer to make an injection mold or part of the die cutter though.
How does a manufacturing company generate a profit from supplying drawings so that you can reproduce (parts of) the products they sell?
DNA chemistry and sequencing is an example here - how many biologists understand the chemistry of the analyzers? How many chemists understand the software?
The answer is both most and not enough.
I don't think you could get through an undergraduate biology degree without being introduced to the basic chemistry underlying traditional (Sanger) sequencing, PCR, etc. Most chemists end up having to do at least some basic scripting if they're going to use automated analytical or synthesis equipment.
On the other hand, lots of biology papers have proven to be fatally flawed because of poor understanding/poor usage of statistics. Brain functional MRI studies, gene array studies, and a lot of other fields have published loads of irreproducible crap because of this.
The carriers want you to agree to a new phone/contract, not keep using your current phone. Preventing your current phone from running apps that require Android 4.x by preventing you from upgrading to 4.x is a great (great as in "heads I win, tails you lose", "you" as in "you've never heard of Slashdot") way of motivating you to get a new phone and a new contract.
The subscription model I want: I'll pay them $1 per hour of original programming, and somewhat less for repeats. They pay me $1 per hour of commercials that I'm forced to watch during those shows. That works out to about 40 - 75 cents per hour for them.
Slashdot Mirror
I think the real answer is:
c) teach non-programmers to view the world like that so that they can recognize these situations, submit an intelligently written ticket that actually explains the problem, and have the presence of mind to mention possible exceptions (transferred to a different hospital, etc) when talking about the possible solution.
I don't think medical facilities are really willing to pay nurses to write and debug code or scripts. Even if they didn't already cost more per hour than IT, they would take many more hours to solve the problem.
Why blame the examiners?
http://www.uspto.gov/patents/law/exam/101_training_aug2012.pdf
An algorithm never deserves a patent. Practical application, tangible embodiment: patent eligible (for now).
Before you accept your next job you have an iris/retina/fingerprint scan (your choice) taken at a local government office. It gets sent to the IRS as a hash, using a function that will always identify you as you but will only identify other people as "not you" 98% of the time or so. It makes confirming your identity easy but creates too many false positives to be used forensically like a fingerprint. If you're afraid the government is still using the info to track you, choose the retina scan. You prove that you are you to the government once, then you're done for, say, 20 years. To use the system the IRS sells dirt cheap scan/hash devices to employers. So when you start your new job:
New employer does the same scan and inputs your name and taxpayer ID into the device. Device sends the hash to the IRS. IRS confirms you're in the database and starts the W2/1099 paperwork or rejects you, and then sends the decision back to the employer within two or three minutes. The device could be a modified $50 smartphone, so it would be cheap enough for hiring nannies, fast enough for hiring day labor, and used on every job site or farm that can get a cell phone signal. Employers could buy ID checks (with encryption keys) the same way people buy minutes for pay as you go phones.
But Orwell and hackers, so it will never happen.
1. Done:
http://publicaccess.nih.gov/
2. Done:
http://www.ncbi.nlm.nih.gov/pmc/
(if you discount the value of immediate access to research, that is)
Taxpayer funding does stipulate that - when it comes from the NIH.
I think university libraries (the principal customers for these publishers) will be the ones who successfully force a transition to either open access or cheap-access publications; budgets are too tight for them to be able to afford to keep supporting the current model.
Not many, but that's also a completely different assertion. Just because you discovered a target or created an animal model doesn't mean you've invented a drug. It also doesn't mean you've done most of the work involved in inventing a drug, the riskiest work involved in inventing a drug, or the hardest work in inventing a drug.
Wrong, wronger, and wrongest. One imagines a "cure" is only given to people who have an actual medical problem (presumably to develop an actual cure, the mechanism of disease is fully exposed). Uncures are not so narrowly constrained.
Statins are consumed (or potentially consumed) by hundreds of millions of people with nothing more than a statistically elevated risk of possibly developing heart disease according to some rather arbitrary marker.
Those markers aren't to determine whether the patient has the disease, they're to determine whether the treatment is (maybe) appropriate for the patient.
A cure for atherosclerosis (what statins aim to treat) would be sold not just to everyone currently prescribed a statin for high chloresterol, but pretty much everyone, full stop. Not everyone has atherosclerosis due to high LDL chlolesterol (the indication for statins), but everyone has (or will have) atherosclerosis. Arterial plaque starts in teenagers and is almost universal in the elderly. It directly causes most cardiovascular disease. The market for such a cure would be pretty much everyone over 40 with access to a deep enough pocket to pay for it. The main risk would not be that it would capture only a small part of the high-cholesterol market; the risk would be governments all over the world overriding the patent after the normal "highest price the market will bear" marketing scheme is announced.
In order for your analysis to hold water, you need to cure pharma of diluting immense benefits to the few into an ocean of revenues from the many.
Not at all. There will be times when lots of people have the same symptom but only a few have the same root cause. A cure for that root cause would profitably treat those few - and leave the rest of that market for that symptom intact.
Now they are out of targets where search space scales linearly with findings. They have an exponential decay in findings to fight with, and you can't with with an exponential by brute force.
Do you mean blocking protein-protein interactions as opposed to blocking protein-small molecule interactions? They did buy a few orders of magnitude of brute force by using antibodies as drugs: immune systems and recombinant molecular biology are a lot faster than chemists at coming up with new drug candidates.
It would be naive to not understand that like most large businesses, the pharmas are driven financial motives which drives their research and product development cycles.
Absolutely, which is why i said they would prefer to sell a cure.
Say it will take 8 years and 3 billion dollars in R & D to get your next product to market. It will be either a cure or a treatment for a chronic disease: your pick.
1. The treatment will compete with all of the other treatments on the market for marketshare. The cure won't have marketshare: it will have the market. There will be no competitors - until another cure is approved, that is.
2. Price. As far as the accountants at your insurance company are concerned, the cure isn't competing with the price of a dose of the treatment: it is competing with the entire cost of treating your disease until you die (or become someone else's problem). As long as the cure comes out cheaper than a decade of doctors bills, hospitalizations, tests, and lots of different pills, it's a good deal for your insurance company. The treatment, on the other hand, could only hope to command a portion of that revenue stream
3. Risk/time value of money. Would you rather be paid your next 10 years salary today or once a month over the next 10 years? Someone who buys your cure pays you in full, today. You book all of that revenue while you are still CEO and take home your bonus. Someone who buys your treatment pays you a little at a time until they switch to a competitor's drug. Or until they die. They are an uncertain revenue stream, not a sure thing.
The researchers find a target the drug companies take it from there, but increasingly it is the first part that is most expensive.
Ok, my turn to demand a source: Which target took $4 billion to identify?
Right now the industry side spends $135 billion on R&D for which it gets ~30 new drugs approved per year plus new research on already approved drugs. Most of that is spent on phase II and III clinical trials, which are costing up to $100M each these days. For pretty much all drugs the vast majority of money and man hours are spent on developing and proving the drug (in industry), not on the target.
It may in fact be cheaper for society to do all this on the government dime, there is a lot of waste in the drug industry a lot of it from its very nature as private research. Fixing this would involve the government massively increasing research funding and deliberately killing an industry, not likely in the short run.
A little of that waste in private research is due to university research: most of the targets identified in the literature turn out to be irreproducible or unusable:
http://blogs.nature.com/news/2011/09/reliability_of_new_drug_target.html
I think there's a lot of room for an enlightened government to more efficiently turn dollars into drugs than the present system, but I'm not convinced yet. Especially not now with the congressmen in charge of the NIH dumping peer review for their own religious and political views.
[source needed]
Sorry, a couple of years ago I looked at a year's worth of drug approvals and came up with 15%. The actual data (1998-2007) say 24% came from academia:
http://www.nature.com/nrd/journal/v9/n11/full/nrd3251.html
Firewalled, but there is a great discussion at In The Pipeline that breaks out the numbers:
http://pipeline.corante.com/archives/2010/11/04/where_drugs_come_from_the_numbers.php
Of course more and more university research is funded by Pharma these days, especially the efforts that are most likely to lead to new drugs. Which column would you put that drug in?
It will go the same way as cable.
No. The difference is that digital streaming like YouTube or Netflix must compete. Cable systems own their regions and milk their subscribers. Streaming systems will never be able to jack rates like cable because they're fungible.
Even worse than that: the monthly caps that cable systems place on subscribers will primarily hit streaming companies like Netflix, especially as HD streaming becomes more common. On the other hand, a streaming service that is partnered with your cable/ISP will not count toward your cap. Neat, huh?
What exactly would this "Premium Content" be? What do they have in that crappy little window that is so wonderful and "Premium" that I will gladly pay them for it?
I'd guess stuff owned by movie and TV studios that have competitive reasons for not wanting to be on Hulu, Netflix, or Amazon Prime, and aren't aligned with a cable/ISP company that is developing its own streaming system.
In the rest of medicine when a group of symptoms tend to travel together without any observable root cause/latent variable they call it a syndrome.
It would be nice to have some tests to determine if the problem is related to some chemical imbalance.
Give it fifteen years. Your doc will have you snort a dose of a labelling compound up your nose (privileged route past the blood brain barrier) like it was a line of cocaine or inject you with the labelling compound attached to little piece of a rabies protein ( another way to get stuff past the BBB). Then they'll pop you in an MRI machine and generate a 3D map of serotonin concentration in your brain.
A slightly paranoid person might buy into the theory that Big Pharma doesn't WANT to cure patients. Instead, they want to hook people on life-long "cures" that prove to be very lucrative.
Actual cures would be much more lucrative. If pharmas had a choice, they'd develop a cure rather than a treatment for any serious chronic disease. Unfortunately, for most non-infectious diseases a "cure" would mean making a fundamental and permanent change in how your body operates. Which pretty much means gene therapy, something which we really don't have a handle on yet.
And that most of the research money the Pharma companies spend is on doing clinical trials to see which ones actually work in humans after the university researchers have found potential candidates testing in cell cultures and animal models.
Hell no. About 15% of drugs come from academic research, the rest are invented by biotech or pharma companies. For the most part academic labs identify new drug targets. Most of the compounds they develop to test their hypotheses are for the most part useless as actual active pharmaceutical ingredients due to toxicity, bioavailability, and metabolism.
Considering the "I'm willing to go thermonuclear war on this" patent wars Apple has instigated in the smart phone and tablet markets, I wouldn't exactly say they gave them away.
Microsoft wants to copy the success of Apple and Google in mobile,
Microsoft wants to copy the success of RIM/BlackBerry. The success of RIM in the Pre-Iphone/Android era, that is: selling business to business in a locked down environment.
personally I am waiting for a 3D printer that can do things like rubber o-rings, seals, gaskets. that will be huge if you can get all the companies to supply the right sized drawings.
The last issue will be supplying the drawings. someone will have to draw them all out and that will take decades.
Gaskets and o-rings would have to be redesigned to be made out of thermoplastics to be 3D printed... I guess you could use the printer to make an injection mold or part of the die cutter though.
How does a manufacturing company generate a profit from supplying drawings so that you can reproduce (parts of) the products they sell?
DNA chemistry and sequencing is an example here - how many biologists understand the chemistry of the analyzers? How many chemists understand the software?
The answer is both most and not enough.
I don't think you could get through an undergraduate biology degree without being introduced to the basic chemistry underlying traditional (Sanger) sequencing, PCR, etc. Most chemists end up having to do at least some basic scripting if they're going to use automated analytical or synthesis equipment.
On the other hand, lots of biology papers have proven to be fatally flawed because of poor understanding/poor usage of statistics. Brain functional MRI studies, gene array studies, and a lot of other fields have published loads of irreproducible crap because of this.
Works great for Pfizer, though they put their IP holding company in Ireland instead.
The carriers want you to agree to a new phone/contract, not keep using your current phone. Preventing your current phone from running apps that require Android 4.x by preventing you from upgrading to 4.x is a great (great as in "heads I win, tails you lose", "you" as in "you've never heard of Slashdot") way of motivating you to get a new phone and a new contract.
The subscription model I want: I'll pay them $1 per hour of original programming, and somewhat less for repeats. They pay me $1 per hour of commercials that I'm forced to watch during those shows. That works out to about 40 - 75 cents per hour for them.