Patents... what are they good for?

For the serious discussion of weighty matters and worldly issues. No off-topic posts allowed.

Moderators: Azrael, Moderators General, Prelates

moimoimoi
Posts: 9
Joined: Fri Jul 17, 2009 2:06 am UTC

Patents... what are they good for?

Postby moimoimoi » Sat Aug 29, 2009 5:48 pm UTC

Convince me that patents of any kind are a good thing. I think talk is cheap, and ideas are too.

To illustrate my idea, here are some scenarios of an invention with and without patents.

patents exist 1
Thomas Jettison invents a better mousetrap. He's granted a patent and sells 100,000 traps at $10 each. His son figures out how to sell 1,000,000 traps at a buck each, but legally cannot due to his father's patent. The father dies a millionaire.

patents exist 2
Thomas Jettison invents a better mousetrap. He's granted a patent and sells 100,000 traps at $10 each. Nikola Tellsya also invents a better mousetrap. He tries to sell them but is sued by Jettison. The courts deem Tellsya's mousetrap too similar to the patented one. Tellsya dies penniless, and Jettison dies a billionaire.

patents don't exist 1
Thomas Jettison invents a better mousetrap. He whines that no one values ideas and doesn't bother to build an implementation of it. He dies penniless 30 years later. His son finds the idea and builds it anyway, selling 100,000 traps at $10 each, and then 1,000,000 traps at a buck each. The son dies with modest income.

patents don't exist 2
Nikola Tellsya also invents a better mousetrap (having never heard of Thomas Jettison). He sells 10,000,000 traps at a dime apiece. He dies with modest income.

These scenarios may be contrived, but I think they illustrate all of the pros and cons of patents. Furthermore, I don't find the first two scenarios any more appealing than the latter.

User avatar
karkaputto
Posts: 29
Joined: Fri Aug 21, 2009 4:54 am UTC
Contact:

Re: Patents... what are they good for?

Postby karkaputto » Sat Aug 29, 2009 5:57 pm UTC

Patents do not exist
T spends years invents a mousetrap and begins selling them at a profit
U, V, and W enter the market selling T's mousetrap at cost and T's profit declines precipitously
A, B, and C decide that it's not worth it to spend their time and money on R&D, since there's nothing to prevent D, E, and F from copying their ideas almost immediately, and the pace of invention declines precipitously

Patents exist
T spends years inventing a mousetrap and selling them at a profit
U, V, and W cannot enter the market for 20 years
A, B, and C see that T makes a large profit when he invents something useful to society, and they also decide to put money in to R&D

moimoimoi
Posts: 9
Joined: Fri Jul 17, 2009 2:06 am UTC

Re: Patents... what are they good for?

Postby moimoimoi » Sat Aug 29, 2009 6:16 pm UTC

karkaputto wrote:A, B, and C decide that it's not worth it to spend their time and money on R&D

This is the part I have trouble believing. As long as the market for mousetraps exists, there will be a small profit available to trap-makers. Add to this the fact that people are optimistic and pursue 'worthless' ideas; I doubt that no one would see the value in R&D (even 'R&D' in someone's basement).

User avatar
karkaputto
Posts: 29
Joined: Fri Aug 21, 2009 4:54 am UTC
Contact:

Re: Patents... what are they good for?

Postby karkaputto » Sat Aug 29, 2009 6:46 pm UTC

the problem is, most r&d nowadays is done by huge corporations accountable to shareholders as most of the ideas that can be made in a basement anymore are pretty damn crappy (slanket anyone?). r&d in a basement just ain't what it used to be. furthermore, assuming a perfectly competitive market, economic profits are always 0. even in non-perfectly competitive markets, economic profit can be zero or very close to it(though they don't have to be).

User avatar
Vaniver
Posts: 9422
Joined: Fri Oct 13, 2006 2:12 am UTC

Re: Patents... what are they good for?

Postby Vaniver » Sat Aug 29, 2009 7:40 pm UTC

R&D spending across the world matches IP protection pretty well, I believe.

The problem isn't that there won't be a profit to investing in R&D- there will be. Unless somebody steals your idea. I mean, consider this: in America, now, it's customary for researchers who get a patent to have their name on the patent, and receive $1 from their employer, who now owns the patent. That seems rough on the researcher, until you realize they got a salary and a lab to experiment in. What happens now if the researcher, after years of working for someone else specifically to develop a product, decides to leave when it's finished and prototyped and start his own company to sell them? The person who employed him has no recourse against him, since the previous employer doesn't 'own' the idea, and received nothing for all the years he paid the researcher and furnished his lab.

Or, even if you ignore betrayal, let's say I come up with a simple new product that improves life. Should I get to capture some of that improvement through restriction on who can make my product, or should anyone who wants to sell the product at cost be able to? The first seems fair to me; the second doesn't.
I mostly post over at LessWrong now.

Avatar from My Little Pony: Friendship is Magic, owned by Hasbro.

General_Norris
Posts: 1399
Joined: Fri Apr 17, 2009 12:10 pm UTC

Re: Patents... what are they good for?

Postby General_Norris » Sat Aug 29, 2009 8:49 pm UTC

What Vaniver said. Why am I going to bother to research something if I can't get anything from it.

icanus
Posts: 478
Joined: Mon Aug 13, 2007 1:19 pm UTC
Location: in England now abed

Re: Patents... what are they good for?

Postby icanus » Sat Aug 29, 2009 9:14 pm UTC

General_Norris wrote:What Vaniver said. Why am I going to bother to research something if I can't get anything from it.

In itself, that's fine, but you can then stall any further developments to your idea until your patent expires.

Say Mr Jettison's "better mousetrap" is patented, but Mr Tellsya looks at it and thinks "Hey, if I just change the shape of the flange and paint it green, it'll still catch mice just as well, and also cure cancer!". No-one except Jettison has an incentive to innovate any further until the patent expires, because they can't profit from their invention if it's based on an existing patent.

Which is a problem.

As is the situation where Better Moustraps Inc. starts patenting every mousetrap variation the guys in the lab can think of, with no intention of ever producing any of them, just so they can pre-emptively muscle out any smaller company that tries to get into the pest-control-hardware business. At best, a huge chunk of everyone's resarch budget gets diverted to patent lawyers, at worst, there's not enough left over to make it worthwhile trying to innovate.

General_Norris
Posts: 1399
Joined: Fri Apr 17, 2009 12:10 pm UTC

Re: Patents... what are they good for?

Postby General_Norris » Sat Aug 29, 2009 9:27 pm UTC

icanus wrote:In itself, that's fine, but you can then stall any further developments to your idea until your patent expires.


No because your competiton is reserching other, better mousetraps and thus, if you don't research too they will get their better mousetrap next year and you will be out of bussiness.

As is the situation where Better Moustraps Inc. starts patenting every mousetrap variation the guys in the lab can think of, with no intention of ever producing any of them, just so they can pre-emptively muscle out any smaller company that tries to get into the pest-control-hardware business. At best, a huge chunk of everyone's resarch budget gets diverted to patent lawyers, at worst, there's not enough left over to make it worthwhile trying to innovate.


Then a company has a monopoly and we have anti-trust laws for that (Or we should!)

I think there are laws concerning your second point but I don't know them. I think that if there are enough changes it's a new patent but changes like "better cheese" are irrelevant in the end. It however teaches us that patents must not be very broad as Newcoven died poor because his steam machine ran against Savery's patent that was a patent for every kind of steam machine.

User avatar
Vaniver
Posts: 9422
Joined: Fri Oct 13, 2006 2:12 am UTC

Re: Patents... what are they good for?

Postby Vaniver » Sat Aug 29, 2009 10:10 pm UTC

You can patent improvements on previous patents. If you design a better mousetrap, you can patent your better mousetrap.

The primary issues of contention arise in the subtler points, particularly when you consider things like genes. If I discover that a certain gene has a certain property, what rights should I have over that knowledge? [Note- this thread is not about these subtler points. It's about whether or not the patent system should exist.]
I mostly post over at LessWrong now.

Avatar from My Little Pony: Friendship is Magic, owned by Hasbro.

User avatar
Strilanc
Posts: 646
Joined: Fri Dec 08, 2006 7:18 am UTC

Re: Patents... what are they good for?

Postby Strilanc » Sun Aug 30, 2009 6:08 am UTC

How have software patents not been mentioned yet? They're the perfect example of how to turn patents bad. The terms are too long relative to the pace of the industry, too many obvious patents are granted, and research costs are comparatively low (you generally don't need million-dollar equipment).
Don't pay attention to this signature, it's contradictory.

User avatar
karkaputto
Posts: 29
Joined: Fri Aug 21, 2009 4:54 am UTC
Contact:

Re: Patents... what are they good for?

Postby karkaputto » Sun Aug 30, 2009 8:08 am UTC

Well, that's partly the fault of an incompetent patent board. Patent law states that patents should not be granted for something that's prior art (i.e. the obvious patents), and yet, because the bureaucrats at the patent office don't know anything about the current state of software innovation, they grant these ridiculous patents.

User avatar
Strilanc
Posts: 646
Joined: Fri Dec 08, 2006 7:18 am UTC

Re: Patents... what are they good for?

Postby Strilanc » Sun Aug 30, 2009 12:23 pm UTC

karkaputto wrote:Well, that's partly the fault of an incompetent patent board. Patent law states that patents should not be granted for something that's prior art (i.e. the obvious patents), and yet, because the bureaucrats at the patent office don't know anything about the current state of software innovation, they grant these ridiculous patents.


True. I guess my point is that, in practice, too much patenting can slow an industry down.
Don't pay attention to this signature, it's contradictory.

Sharlos
Posts: 720
Joined: Fri Dec 26, 2008 9:26 am UTC
Location: Straya

Re: Patents... what are they good for?

Postby Sharlos » Sun Aug 30, 2009 3:16 pm UTC

wouldn't copyright be more appropriate for software?

User avatar
Vaniver
Posts: 9422
Joined: Fri Oct 13, 2006 2:12 am UTC

Re: Patents... what are they good for?

Postby Vaniver » Sun Aug 30, 2009 3:35 pm UTC

Sharlos wrote:wouldn't copyright be more appropriate for software?
Sort of. The question is whether it's the block of code that's relevant (if I write code in Java to duplicate your code in C++, am I violating your copyright?) or the concept- and if it's the concept, that seems like it would fall under patents.
I mostly post over at LessWrong now.

Avatar from My Little Pony: Friendship is Magic, owned by Hasbro.

Dazmilar
Posts: 67
Joined: Fri Mar 28, 2008 11:37 pm UTC

Re: Patents... what are they good for?

Postby Dazmilar » Sun Aug 30, 2009 5:07 pm UTC

In the no-patent examples, companies could just start spending their R&D budgets on corporate espionage. Why research when you have an easier time stealing?

I think the mousetrap example is a bit simplistic, and assumes that there's only one basic kind of mousetrap that everyone in the market will want. The market doesn't necessarily want that specific mousetrap, they just want whatever is best at catching mice. So Jettison's company has their own patented model of mousetrap design, and Tellsya's company has their significantly different but also patented design. Both companies have incentives to improving their mousetraps, because they're competing for the same group of mice-fearing consumers.

A more realistic example might be the market for purifying/cleaning water. There may be multiple ways to do this: various chemical processes, ion-exchange resins, a biological-based solution, whatever. They all compete for the same market, power plants and industrial plants that need to clean their used water. They all compete using very different patented methods. Bob of Bob's red algae water purification may be the only one with the patent on the process, but he still needs to spend money on R&D because Joe's Nuclear Plant doesn't care how the water gets cleaned, just how cost-effective the process is. And if Steve's Ion-Exchange Resins can do it better for cheaper; Steve's getting the contract.

Admittedly, there are problems with the patent system and patent abuse. Sometimes patents are granted that are far too general, or people go patent fishing. There isn't a problem with DuPont having patents for Kevlar, but there would be if they were granted a patent for "protective clothing."

Outchanter
Posts: 669
Joined: Mon Dec 17, 2007 8:40 am UTC

Re: Patents... what are they good for?

Postby Outchanter » Sun Aug 30, 2009 8:04 pm UTC

Vaniver wrote:
Sharlos wrote:wouldn't copyright be more appropriate for software?
Sort of. The question is whether it's the block of code that's relevant (if I write code in Java to duplicate your code in C++, am I violating your copyright?) or the concept- and if it's the concept, that seems like it would fall under patents.

Wouldn't a patent on, say, the FFT algorithm have delayed the advancement of the entire field of digital signal processing by decades?

qbg
Posts: 586
Joined: Tue Dec 18, 2007 3:37 pm UTC

Re: Patents... what are they good for?

Postby qbg » Sun Aug 30, 2009 10:03 pm UTC

Strilanc wrote:How have software patents not been mentioned yet? They're the perfect example of how to turn patents bad. The terms are too long relative to the pace of the industry, too many obvious patents are granted, and research costs are comparatively low (you generally don't need million-dollar equipment).

How dire is the need for software patents? I'd be surprised if companies stopped creating new software innovations just because software patents did not exist. For your boxed software, companies would still have incentive to innovate by creating new features to entice customers to upgrade. For contracted software/in house software, innovation will still happen as long as it is useful for solving a problem.

User avatar
Vaniver
Posts: 9422
Joined: Fri Oct 13, 2006 2:12 am UTC

Re: Patents... what are they good for?

Postby Vaniver » Sun Aug 30, 2009 11:52 pm UTC

Outchanter wrote:Wouldn't a patent on, say, the FFT algorithm have delayed the advancement of the entire field of digital signal processing by decades?
Probably. There appear to be a number of FFT algorithms, so maybe not?

The primary question is whether the algorithm is an invention or a discovery, I believe- and generally processes fall under inventions, but with math the line is, at best, hard to pin down. You can't patent natural laws or mathematics, to the best of my knowledge.

qbg wrote:How dire is the need for software patents?
It depends on what they'd be replaced with. My guess is the need is pretty marginal, and that what espionage does happen could be dealt with through copyright law. But I am not an expert on the issue, so I can't say anything definitive.
I mostly post over at LessWrong now.

Avatar from My Little Pony: Friendship is Magic, owned by Hasbro.

User avatar
karkaputto
Posts: 29
Joined: Fri Aug 21, 2009 4:54 am UTC
Contact:

Re: Patents... what are they good for?

Postby karkaputto » Mon Aug 31, 2009 12:08 am UTC

The primary question is whether the algorithm is an invention or a discovery, I believe- and generally processes fall under inventions, but with math the line is, at best, hard to pin down. You can't patent natural laws or mathematics, to the best of my knowledge.

algorithms are patentable
and we should definitely not deal with software secrets through copyright law, since currently copyright on anything extends to nearly a century, whereas patents expire after 20 years. i'm rather opposed to even 20 years for something like modern software, much less a century.

BattleMoose
Posts: 1993
Joined: Tue Nov 13, 2007 8:42 am UTC

Re: Patents... what are they good for?

Postby BattleMoose » Mon Aug 31, 2009 12:10 am UTC

Just because a patent is awarded doesn't mean it cannot be challenged or that it will hold up in a court of law.

I would like to talk about the pharmaceutical industry.

Such companies spend fortunes on developing drugs, often life saving or life enhancing drugs, think HIV and ARVs. To fund their R&D they have to charge a great deal more than the cost of production. There is just simply no way that these companies can possibly fund R&D if there are companies out their willing to charge 5% on top of cost of production. The result would quite simply be zero drug research, which would be a completely sucky universe to live in.

Patents ensure that technology evolves, develops and that there is R&D. Often with bureacreacy and silly mousttrap IP property rights and fights but it ensures technology is developed.

Also I would much rather live in a society that over reweards innovation and development than in which one that IP is not protected. Even if it does mean that Person A makes billions from doing a half days work by designing and patenting a better mousetrap. But this means that society has a better mousetrap (yay) and he makes a fortune by selling a product that the consumer base values. (double yay)

Iv
Posts: 1207
Joined: Thu Sep 13, 2007 1:08 pm UTC
Location: Lyon, France

Re: Patents... what are they good for?

Postby Iv » Mon Aug 31, 2009 8:56 am UTC

Strilanc wrote:How have software patents not been mentioned yet? They're the perfect example of how to turn patents bad. The terms are too long relative to the pace of the industry, too many obvious patents are granted, and research costs are comparatively low (you generally don't need million-dollar equipment).
Software patents are what made me believe that every patent was necessarily evil. Now I am more moderate in my approach, but I still think that a no-software-patent approach would be better than what we have now. The current implementation is broken on so many counts that it can kill innovative businesses.
karkaputto wrote:algorithms are patentable
Not in Europe. There are provisions againts mathematical formulas being patented and algorithms are considered as mathematical formulas.

karkaputto wrote:and we should definitely not deal with software secrets through copyright law, since currently copyright on anything extends to nearly a century, whereas patents expire after 20 years. i'm rather opposed to even 20 years for something like modern software, much less a century.
Copyright seems like the good tool, and this is also a tool that needs a serious lifting. And serious definitions as well. Re-writting an algorithm will result in different code (different variable name, order of initialization, order of conditions, etc...) how much of difference is a difference ? IP protection tools are one of the most insidious monopoly tool that exists today, we should be more careful about them.

User avatar
Peanutchair
Posts: 15
Joined: Fri Aug 28, 2009 9:29 pm UTC
Location: London.

Re: Patents... what are they good for?

Postby Peanutchair » Mon Aug 31, 2009 11:01 am UTC

Before patents what happened? You invented something and sold it, however, as soon as someone found out how it worked they could sell it themselves, often putting the inventor himself/herself out of business. Before patents, people were almost scared to invent because it was an incredibly risky profession.

The creation of patents gave rise to a huge surge in inventions. People were no longer afraid to innovate and invent because what they invented could now be theirs. And theirs alone. This meant that people could now be inventors by profession with no (or at least considerably less) fear of their inventions being stolen. This surge of inventions that patents created did everything from making people happier to increasing productivity. In fact, economic growth in western Europe and the US took off 200 years ago, right after patents became widely enforced. That isn't a coincidence. For the first time in history, a secure financial incentive existed for someone to use their brain to innovate.

I firmly believe that they are a good thing.
Richard Dawkins was put on this Earth to test our faith. Like fossils... and facts.
-
Stewart Lee.

Iv
Posts: 1207
Joined: Thu Sep 13, 2007 1:08 pm UTC
Location: Lyon, France

Re: Patents... what are they good for?

Postby Iv » Mon Aug 31, 2009 11:46 am UTC

We must be careful, however, that patents don't go completely the other way : preventing people from inventing because a field is encumbered with obvious patents.

In my opinion, the problem is that a lot of commercial R&D (especially in IT) is made by engineers, not inventors : they solve a real life problem that a company is confronted with. Ask two different competent engineer to solve the same problem, and they will solve it the same way. While I agree that R&D investment should have some kind of protection, I am a bit confused about the notion that the first person to solve a problem should have the monopoly of it.

The proverbial "good idea" is cheap, requires zero investment and should not give any monopoly to the first person who has it (or, more precisely, who protects it). What should be protected is designs that take some work to create but a lot less to copy : a CPU design, a robust embedded system, a water-proof device...

User avatar
Clumpy
Posts: 1883
Joined: Thu Jul 19, 2007 4:48 am UTC
Contact:

Re: Patents... what are they good for?

Postby Clumpy » Mon Aug 31, 2009 5:47 pm UTC

Y'know, a certain about of "reverse engineering" goes into any innovation based on existing technology. The companies aren't paying the people who actually worked on the thing the big bucks or making them celebrities anyway. I'd be in favor of any law that discourages corporations from stunting human innovation by locking off entire areas of research, while giving people enough protection over their own innovations to allow people to profit from their ideas. This type of social change is probably not what patent law has in mind as all commercial law is eventually co-opted by corporations with the time and resources to figure out how to take full advantage of it.

I think we can at least agree that patent trolls need to go.

nitePhyyre
Posts: 1280
Joined: Mon Jul 27, 2009 10:31 am UTC

Re: Patents... what are they good for?

Postby nitePhyyre » Wed Sep 02, 2009 7:36 am UTC

It seems to me the simplest option is to change patents from protecting ideas, to protecting specific individual implementations of products.

For example I invent a product where piece 'A' fits in with piece 'B', piece 'AB' goes around piece 'C', and everything is held together with piece 'D'. This product accomplishes function 'X'.

My patent on this device should not protect:
1) Any of the pieces 'A', 'B', 'AB', 'C' or 'D'
2) Any combination of these pieces that is not for function 'X'
3) Products that combine 'A', 'B', 'C', or 'D' that accomplishes function 'X' in new and novel ways. (To be decided upon after infringing by the courts)

In this scheme only the last "sub"-patent would count. i.e.: If I have a patent of product 'A', if I use 'A' in the device listed above the patent for 'A' alone would be null and void. The greatest part of this scheme, is that patents would essentially have to be blueprints. Anyone could use them for non-commercial purposes.

AFAIK, this is the way patents were designed to be used?
sourmìlk wrote:Monopolies are not when a single company controls the market for a single product.

You don't become great by trying to be great. You become great by wanting to do something, and then doing it so hard you become great in the process.

Iv
Posts: 1207
Joined: Thu Sep 13, 2007 1:08 pm UTC
Location: Lyon, France

Re: Patents... what are they good for?

Postby Iv » Wed Sep 02, 2009 7:50 am UTC

Any process requiring a judge ruling can take several years to conclude. We need a faster process. Two years in the software industry, for example, is more than an unfair advantage. I would have no problem with the current process if it took one week and only a moderate amount of money instead of one or two years to question the validity of a patent.

qetzal
Posts: 862
Joined: Thu May 01, 2008 12:54 pm UTC

Re: Patents... what are they good for?

Postby qetzal » Wed Sep 02, 2009 2:12 pm UTC

@nitePhyyre,

As I understand the current US patent system (IANAL or a patent expert), it does protect implementations rather than mere ideas. To get a patent, one must show novelty (it hasn't been done before and it's not obvious from what has been done before), utility (it's useful), and reduction to practice (you've specifically implemented, as you put it). If you are the first to combine A, B, C, & D in a particular way to accomplish the useful function X, that's patentable. If all you have is an idea for a product that does X, that's not patentable.

Whether any of A, B, C, or D are patentable (alone or in combination) is decided the same way - are they new & useful and have they been reduced to practice? Suppose in your process, you had to come up with a clever new way to shape D so that it would hold A, B, & C together in just the right way. Without your clever new idea, the whole thing would never work. Under current law, I believe that is patentable. You have a new kind of piece D and it's useful for holding A, B, & C together in a new way that allows you to make a product that does X.

Do you think D should be patentable or not? If not, how would your proposal distinguish between patentable X and non-patentable D?

User avatar
Velict
Posts: 609
Joined: Wed Dec 24, 2008 9:07 pm UTC
Location: Icecrown Citadel

Re: Patents... what are they good for?

Postby Velict » Thu Sep 03, 2009 2:42 am UTC

Iv wrote:Any process requiring a judge ruling can take several years to conclude. We need a faster process. Two years in the software industry, for example, is more than an unfair advantage. I would have no problem with the current process if it took one week and only a moderate amount of money instead of one or two years to question the validity of a patent.


That's not really a good business strategy, though. Patent infringement penalties hold the infringing corporation viable for between one and three times the profit the corporation alleging the claim would have made with the patent (source). Judges also issue injunctions towards the corporation accused of patent infringement in most cases that possess merit, at the preliminary stages of the trial. That means that companies can't just prolong the legal process and continue to infringe the patent.

Iv
Posts: 1207
Joined: Thu Sep 13, 2007 1:08 pm UTC
Location: Lyon, France

Re: Patents... what are they good for?

Postby Iv » Thu Sep 03, 2009 7:59 am UTC

I may not have been very clear : I am suggesting that by getting a patent accepted for an obvious invention (it is easy to do) you can gain an unfair advantage and even attack (and win!) trials against competitor, preferably smaller ones, while your patent is in the process of being revoked, which could take two or three years.

Reasonable patents in the software industry should be around 2 or 3 years and not exceed 5. Therefore, the capability to revoke an obvious patent before a court is useless if it is not faster.

nitePhyyre
Posts: 1280
Joined: Mon Jul 27, 2009 10:31 am UTC

Re: Patents... what are they good for?

Postby nitePhyyre » Thu Sep 03, 2009 11:54 am UTC

qetzal wrote:As I understand the current US patent system (IANAL or a patent expert), it does protect implementations rather than mere ideas. To get a patent, one must show novelty (it hasn't been done before and it's not obvious from what has been done before), utility (it's useful), and reduction to practice (you've specifically implemented, as you put it). If you are the first to combine A, B, C, & D in a particular way to accomplish the useful function X, that's patentable. If all you have is an idea for a product that does X, that's not patentable.


I agree with that. It just seems that it doesn't work that way. For instance:Microsoft patents 'Page Up' and 'Page Down' or Method of swinging on a swing or Method of exercising a cat.

Yes, that last one is the patent for 'waving a laser pointer near your cat'.

qetzal wrote:Whether any of A, B, C, or D are patentable (alone or in combination) is decided the same way - are they new & useful and have they been reduced to practice? Suppose in your process, you had to come up with a clever new way to shape D so that it would hold A, B, & C together in just the right way. Without your clever new idea, the whole thing would never work. Under current law, I believe that is patentable. You have a new kind of piece D and it's useful for holding A, B, & C together in a new way that allows you to make a product that does X.

Do you think D should be patentable or not? If not, how would your proposal distinguish between patentable X and non-patentable D?


I would think the rule of thumb should be: Does it actually do something by itself?

If D is a fastener, i.e. it holds "things" together, it should be patentable. (Think tie wraps)
If D is a case, i.e. it holds "A, B, C" together, it should not be patentable by itself.
Simply put, if it doesn't do anything by itself, it shouldn't be patentable by itself.

As a hypothetical: Someone designs a spark-plug. The patent essentially reads: "A small device designed to take a short electrical impulse and turn it into a visible spark". The creator of this patent is using the spark-plugs as a light show at an amusement park. Someone designs an internal combustion engine. The last problem the engineer has to solve is how to create the combustion. Eventually, whether or not he has heard about the spark plug, decides on the spark plug.

In this scenario, should the second inventor be barred from using the spark plug?


All that being said, I feel the answer to this thread should just be: "It is impossible to tell what patents are good for. Their true effectiveness is buried under a pile of ineptitude at the patent office"
sourmìlk wrote:Monopolies are not when a single company controls the market for a single product.

You don't become great by trying to be great. You become great by wanting to do something, and then doing it so hard you become great in the process.

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7588
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Patents... what are they good for?

Postby Zamfir » Thu Sep 03, 2009 12:25 pm UTC

nitePhyyre wrote:I agree with that. It just seems that it doesn't work that way. For instance:Microsoft patents 'Page Up' and 'Page Down' or Method of swinging on a swing or Method of exercising a cat.


I was under the impression that most patents are granted without much consideration, if they just fulfill some technical format requirements. Only when you actually sue someone for patent infringement will be the merit of your patent be considered. So the effort to determine the validity of your patent is only spend if there is an actual conflict.

So all those "silly patents" do not say much about the system if the case is never brought to court.

User avatar
Velict
Posts: 609
Joined: Wed Dec 24, 2008 9:07 pm UTC
Location: Icecrown Citadel

Re: Patents... what are they good for?

Postby Velict » Fri Sep 04, 2009 1:48 am UTC

Zamfir wrote:
nitePhyyre wrote:I agree with that. It just seems that it doesn't work that way. For instance:Microsoft patents 'Page Up' and 'Page Down' or Method of swinging on a swing or Method of exercising a cat.


I was under the impression that most patents are granted without much consideration, if they just fulfill some technical format requirements. Only when you actually sue someone for patent infringement will be the merit of your patent be considered. So the effort to determine the validity of your patent is only spend if there is an actual conflict.


According to this, there is a 0.44% acceptance rate of patent applications at the moment. This strongly implies that a good deal of consideration takes place before the USPTO grants the patent.

However, I'll agree that we need patent reform when it comes to challenging facially invalid patents. You can actually do so without an infringement case, but it's a very inconvenient process.

dosboot
Posts: 143
Joined: Sun Jul 01, 2007 5:26 am UTC

Re: Patents... what are they good for?

Postby dosboot » Sat Sep 05, 2009 8:12 pm UTC

Maybe there should be a separate patent system for things that need millions of dollars and decades of research. Give them special treatment so that the scope and length of the patent is decided on a case by case basis (like a judicial system). The current patent system would remain to handle the larger volume of less research-intensive inventions, and would have patent duration reduced to reflect this (5 years?). Reforms of the current patent system to eliminate obvious patents and/or software patents could be made on top of this.

The new system should probably require the party to register with the patent system when research begins and not when it produces a useful result. The purpose is so that research costs can be tracked and so that they can be forced to divulge all results of the research. A reasonable concession could be that all non-patentable results will be made public (instead of being withheld like it is now).

qetzal
Posts: 862
Joined: Thu May 01, 2008 12:54 pm UTC

Re: Patents... what are they good for?

Postby qetzal » Sun Sep 06, 2009 12:04 am UTC

@dosboot

They do a small token of that for pharmaceuticals. Specifically, a company that gets a drug approved can some times get some patent life extension based on the time required for regulatory approval. IIRC, they don't get credit for all the time spent in clinical trials (5-8 years, at least). Only for certain aspects of FDA review time.

That could be extended, but I'm not sure it should be. I doubt it would bring drug costs down much. Pharma would have more time to recoup development costs, so it could arguably spread that cost out more, but would it? They'd still have an incentive to charge as much as they could, except for even longer.

I'm not sure the idea of registering before developing something patentable would work. How would you know what to register for? Even if you know you're trying to develop something that does X, you don't know how you're going to do it yet. (If you did, you'd already be submitting your patent application.) And what if you're trying to develop, say, a blood-pressure drug, but instead discover a treatment for ED?

Also, it probably wouldn't work for areas like biotech that are supported by investment capital. Investors won't risk their cash unless the company already has patents on what they're doing.

User avatar
Vaniver
Posts: 9422
Joined: Fri Oct 13, 2006 2:12 am UTC

Re: Patents... what are they good for?

Postby Vaniver » Mon Sep 07, 2009 2:13 am UTC

Velict wrote:According to this, there is a 0.44% acceptance rate of patent applications at the moment. This strongly implies that a good deal of consideration takes place before the USPTO grants the patent.

However, I'll agree that we need patent reform when it comes to challenging facially invalid patents. You can actually do so without an infringement case, but it's a very inconvenient process.
Your decimal is misplaced; it's 44%, that is, one out of every 2.3.
I mostly post over at LessWrong now.

Avatar from My Little Pony: Friendship is Magic, owned by Hasbro.

User avatar
ByrneTofferings
Posts: 1
Joined: Mon Sep 07, 2009 11:58 pm UTC
Contact:

Re: Patents... what are they good for?

Postby ByrneTofferings » Tue Sep 08, 2009 1:20 am UTC

The argument that patents create innovation is one I don't think is automatically justified really.

I mean, imagine if I locked the brightest minds in the world in a room, and held a gun to their heads demanding they come up with some plans to improve our lives. Sure, I have no doubt it would drive innovation, but only through illegitimate force.

qetzal
Posts: 862
Joined: Thu May 01, 2008 12:54 pm UTC

Re: Patents... what are they good for?

Postby qetzal » Tue Sep 08, 2009 2:44 am UTC

I don't particularly see the relevance of that analogy.

The argument is that patents give inventors an incentive to invest time & money into things, because the inventors know that IF they invent something novel & useful, they have the rights to make a profit from it for a defined period. Patents also ensure that inventors fully disclose their invention. If someone invents a 'miraculous' cure for cancer or a fabulous new source of energy, they can't patent it without disclosing exactly what it is. Sure, they'll get to profit from it for 20 years, but after that, it's public domain.

qbg
Posts: 586
Joined: Tue Dec 18, 2007 3:37 pm UTC

Re: Patents... what are they good for?

Postby qbg » Sat Sep 12, 2009 12:59 am UTC

Whiles reading this (take it as you will), I came across this:
And patents are not necessary as an incentive to innovate. According to Rothbard, invention is rewarded by the competitive advantage accruing to the first developer of an idea. This is borne out by F. M. Scherer's testimony before the FTC in 1995 [Hearings on Global and Innovation-Based Competition]. Scherer spoke of a survey of 91 companies in which only seven “accorded high significance to patent protection as a factor in their R & D investments." Most of them described patents as "the least important of considerations." Most companies considered their chief motivation in R & D decisions to be "the necessity of remaining competitive, the desire for efficient production, and the desire to expand and diversify their sales." In another study, Scherer found no negative effect on R & D spending as a result of compulsory licensing of patents. A survey of U.S. firms found that 86% of inventions would have been developed without patents. In the case of automobiles, office equipment, rubber products, and textiles, the figure was 100%.


Seems interesting if it is true.

qetzal
Posts: 862
Joined: Thu May 01, 2008 12:54 pm UTC

Re: Patents... what are they good for?

Postby qetzal » Sat Sep 12, 2009 6:26 am UTC

@qbg,

That's a mildly interesting link, but I'm not at all persuaded. For example, I googled F. M Scherer and found this book by him. Page 21 contains the following passage:

Finally, if one despairs of reforming the patent system through halfway measures, there remains the possibility of abolishing it. This is not an alternative favored by the author, but that should scarcely bar considering it seriously. We have no reason to fear that such a step would bring all inventive and innovative activity to a halt, since there are many alternative incentives. Yet some contributions would undoubtedly be lost, unless substitute incentives were created.


Scherer clearly thinks the patent system needs reforming, and I agree. But even he doesn't favor abolishing patents.

Iv
Posts: 1207
Joined: Thu Sep 13, 2007 1:08 pm UTC
Location: Lyon, France

Re: Patents... what are they good for?

Postby Iv » Mon Sep 14, 2009 8:01 am UTC

I read in a French news site that clothing, haute-couture and mode was a domain without patents and copyrights. It would have been ruled in 1941 by the Supreme Court of the US that haute-couture creations could not be protected by copyrights and copying by other tailors was legal, causing the phenomenon of high competition and innovation, striving to create a new fashion every year...


Return to “Serious Business”

Who is online

Users browsing this forum: No registered users and 7 guests