Empirical Research Reveals Three Big Problems With How Patents Are Vetted (arstechnica.com) 94
An anonymous reader quotes a report from Ars Technica: If you've read our coverage of the Electronic Frontier Foundation's "Stupid Patent of the Month" series, you know America has a patent quality problem. People apply for patents on ideas that are obvious, vague, or were invented years earlier. Too often, applications get approved and low-quality patents fall into the hands of patent trolls, creating headaches for real innovators. Why don't more low-quality patents get rejected? A recent paper published by the Brookings Institution offers fascinating insights into this question. Written by legal scholars Michael Frakes and Melissa Wasserman, the paper identifies three ways the patent process encourages approval of low-quality patents:
-The United States Patent and Trademark Office (USPTO) is funded by fees -- and the agency gets more fees if it approves an application.
-Unlimited opportunities to refile rejected applications means sometimes granting a patent is the only way to get rid of a persistent applicant.
-Patent examiners are given less time to review patent applications as they gain seniority, leading to less thorough reviews.
None of these observations is entirely new. But what sets Frakes and Wasserman's work apart is that they have convincing empirical evidence for all three theories. They have data showing that these features of the patent system systematically bias it in the direction of granting more patents. Which means that if we reformed the patent process in the ways they advocate, we'd likely wind up with fewer bogus patents floating around.
-The United States Patent and Trademark Office (USPTO) is funded by fees -- and the agency gets more fees if it approves an application.
-Unlimited opportunities to refile rejected applications means sometimes granting a patent is the only way to get rid of a persistent applicant.
-Patent examiners are given less time to review patent applications as they gain seniority, leading to less thorough reviews.
None of these observations is entirely new. But what sets Frakes and Wasserman's work apart is that they have convincing empirical evidence for all three theories. They have data showing that these features of the patent system systematically bias it in the direction of granting more patents. Which means that if we reformed the patent process in the ways they advocate, we'd likely wind up with fewer bogus patents floating around.
A solution (Score:2, Insightful)
Eliminate patents and copyrights. Get rid of them altogether. This also has the benefit that IBM can no longer bully open source opponents with patent litigation, plus onerous licenses like the GPL are invalidated. No more low quality patents, plus many other issues get fixed.
Re: (Score:1, Troll)
Eliminate patents and copyrights.
That's a great solution if you want to enjoy mediocre technology and entertainment. Because of course why would anyone bother spending R&D dollars when they can just wait for someone else to do it then copy their product.
Re: (Score:3)
the *best* outcome an inventor can hope for, is for a paltry salary from a giant corp that then goes on to make millions in profit off of what is solely your work.
Exactly. Elon Musk is making billions using the AC induction motor in his cars and trucks. However, the inventor of the AC motor, Nikola Tesla or is his estate, won't see a penny, thanks to patents being only 20 years long. If the patent system were fair, Tesla Corp. would pay a sales royalty of 5% that would further be divided up to pay all the inventions their cars use.
Patents are bait (set by the govt and big companies) to trick the working class person to do R&D for 10x to 100x his typical salary. O
Re: (Score:1)
Because when you release your fine product on Monday, by Tuesday evening the market is flooded with legal Chinese knockoffs at price which makes it unprofitable for you to have done the research and development in the first place. And next time you don't do it.
Re: (Score:2)
Depends on your reputation, the quality and price of your product, the quality and price of the clones etc..
If your product is high quality and you have a good reputation, you can charge a premium for it even if cheaper options are available.
If your product is sufficiently cheap it won't be profitable to make clones of it, as the clones would only be marginally cheaper at best, or might even be more expensive due to economies of scale.
Most people will pay a little extra for a quality item from a reputable b
Re: (Score:2)
You missed the part where your product is high quality but you don't have a good reputation (or massive sales force), yet.
This article is all well and good but extra scrutiny now would just be closing the door after all the horses are outside.
There's now so many junk patents that we really need to start over.
Re: (Score:2)
Depends on your reputation, the quality and price of your product, the quality and price of the clones etc..
Counterpoint: Wal-Mart exists.
Re: (Score:2)
Counterpoint: Wal-Mart exists.
Walmart has the reputation for having the lowest prices. I can go to Walmart and compare the price to the supermarket literally next door and Walmart often has the lower price.
For something like ass wipes that you use once and flush down the toilet why would you pay the higher price?
Re: (Score:2)
Well maybe you should read the thread you're replying on and realize we're not wiping anything here:
Why should that result in mediocrity?
Because when you release your fine product on Monday, by Tuesday evening the market is flooded with legal Chinese knockoffs at price which makes it unprofitable for you to have done the research and development in the first plac
Re:A solution (Score:5, Interesting)
I'm curious, do you (or anyone here) know of any substantial evidence that actually supports this claim? Are there any good, relevant examples of historical evidence that show that a patent/copyright system is required for and/or effective at fostering innovation?
Re:A solution (Score:5, Interesting)
Actually, there are a few examples of the contrary - I think it was the Dutch that eliminated patents at one point, and ushered in years of technological growth and prosperity, lasting until they reinstated patents.
How about the Computer Software Industry? (Score:5, Interesting)
Software was not patentable until about 2000. But there was huge innovation before that.
Bill Gates said something like " If software was patented when the fundamental priniciples of computer science were developed, the industry would be at a standstill today".
Software patents reduce innovation. They add unknowable technical risk to any innvovative project.
Re: (Score:1)
Not True! Software Patents under US law were granted in the early 1970's. Do you really think IBM wouldn't try to get patents in the Corporate mainframe era?
https://en.wikipedia.org/wiki/Software_patents_under_United_States_patent_law#History
Re: (Score:2)
Re: (Score:2)
war on drugs is for people on drugs (Score:2)
The "war on drugs" was a "common sense" policy
No it wasn't, and still isn't. It has been demonstrated that it is immensely more cost-effective to address the problem on the buy side; the cost to get a single user clean is a lot lower than the cost of disrupting the production and distribution pipeline for the amount of drugs that the drug user would have consumed.
Economically it doesn't make sense to focus on the sell side because every seizure or disruption leads to a price adjustment. Either that or the production costs are lowered, which means the s
Re: (Score:2)
Re:A solution (Score:5, Insightful)
Many corporations use patents defensively, as weapons against other corporations suing them. It is very common for companies to form "patent pools" to share technology among themselves while excluding outsiders.
This indicates that most companies see little or no inherent value in patents, and consider them more of a Prisoner's Dilemma. They are forced to collect patents because others are collecting, but they could be collectively better off if patents didn't exist.
There is little reason to believe that the patent system is a net contributor to innovation. For every inventor that is protected, there are many more that are stifled. You can't improve and extend what you aren't allowed to use.
Also, copyrights and patents are very different, and should not be lumped together.
Re: (Score:2)
Simply because companies develop defensive patent portfolios does not mean that they have _no_ legitimate, valuable patents. I think you'll find that it's not that the companies see little or no value in patents per se. Rather, it's rather that _most_ of the patents, especially software patents, have little individual value. It's a small percentage of software patents that are significant, enforceable, and worth spending resources to bring to court for enforcement. They also have value to the owner of the p
Re:A solution (Score:5, Interesting)
My solution; Make intellectual property an inalienable personal right of the creator, like free speech. This means that we stop treating patents as real estate by making them non-fungible. The inventor of work gets a patent on it, and that patent stays with the creator until it expires. Any company wishing to exploit the patent would have to maintain a contractual relationship with the creator during the period of validity. That would mean no more exploiting inventors and then kicking them aside. No more defensive patent portfolios to stifle innovation, which is the opposite of what was intended by Article 1, Section 8.
Imagine what kind of societal dystopia we would have if we could sell off our right to free speech? The wealthiest corporations would buy out the speech rights of their critics and steadily increase their power.
Re: (Score:2)
For every inventor that is protected, there are many more that are stifled. You can't improve and extend what you aren't allowed to use.
I believe significant improvements on an existing patent are perfectly legal, though IANAL and could quite possibly be wrong. In my experience most technology patents seem to be pretty obvious to anyone who is in the industry, however. Patents involving playing frame by frame reverse MPEG video, for instance. There is literally only one way to do that, which is quite obvious from the technology itself. And yet, someone has a patent for it and will sue you if you try to offer such a feature. Patents inv
Re: (Score:2)
In my experience most technology patents seem to be pretty obvious to anyone who is in the industry, however.
This.
The central issue. Far too many patents are of the type "With A Computer", or "On The Internet", "on a touchscreen", or as you say obvious to someone in the industry. I am careful never to read any patent documents, as it's quite likely that some bright bulb wrote a patent about moving data, or packets, or whatever it is I build on a given day. If some dipshit company tries to sue the company I work for, at least we won't get hit with treble damages.
Generally speaking, if an architect or developer in c
Re: (Score:2)
You can patent an improvement to an existing, patented invention. However, if you are still using the tech described in the initial patent, you can't produce it without getting a license. So, for example, we have Edison,Swan, and the modern light bulb. Swan had a decent vacuum. Edison had a decent filament. Without cooperating, neither was able to manufacture a decent light bulb
Re:A solution (Score:5, Interesting)
I don't want to eliminate patents, but I'd be fine with eliminating certain categories of patents. Software patents, design patents, business method patents... all of these are overreaching beyond the original intent of protecting the rights of inventors.
Software can be copyrighted, designs are protected by trademarks. Business methods... well, MBAs who need a patent to stay competitive can go fuck themselves.
Re: (Score:2)
Software can be copyrighted...
Yes, but that only protects you against pirates: copyright only applies when someone makes an exact copy of your code. If a competitor simply reverse engineers it and rewrites it, copying all of your ideas but not your specific variable names, then they are not infringing your copyright.
Copyright is great for music and movie companies*, because people want to see the Avengers movie or listen to Kanye's new album rather than the "Revengers" knock-off version or my new album - they want that specific work. Bu
Re: (Score:2)
The fact that Zynga can rip off indie titles AND the fact that they have tons of patents should speak FOR elimination of software patents, not against it. The world is not perfect or fair, but software patents only make it less so. Indie developers can't afford patent projection, nor should they have to do so.
And as an indie game developer myself, I believe there's a special place in hell for Zynga. What they do is beneath contempt, but is technically not illegal. And as much as I hate to say it, tryin
Re: (Score:2)
The fact that Zynga can rip off indie titles AND the fact that they have tons of patents should speak FOR elimination of software patents, not against it. The world is not perfect or fair, but software patents only make it less so. Indie developers can't afford patent projection, nor should they have to do so.
So because some indie developers can't afford patent protection, Zynga should get to walk roughshod over all developers? That makes no sense, unless you're part of Zynga's management team.
And patents really aren't that expensive. About $25k over 3-4 years to get a patent, and what's the value of having a Zynga-proof monopoly, rather than merely being Dream Heights to their Tiny Tower?
And as an indie game developer myself, I believe there's a special place in hell for Zynga. What they do is beneath contempt, but is technically not illegal. And as much as I hate to say it, trying to prevent them from doing so goes down a slippery legal slope that I think would be devastating for the game industry. Could you imagine if the litigation nightmare if a company managed to "patent" the basic game mechanics of a simple first-person shooter? If no one could copy game ideas, Stardew Valley could not have taken inspiration from Harvest Moon.
Or they could come up with their own IP, or they could take a license for any elements they were reusing...
Re: (Score:2)
And patents really aren't that expensive. About $25k over 3-4 years to get a patent
You have a very big-corporation view of "aren't that expensive." It's true that to substantial companies, $25K isn't much. But as a small developer, I could pay for a patent, or I could pay for union actors to voice my small game, or maybe even have a marketing budget. Which do you think I'm going to choose?
Given that Zynga is mostly a problem for small, little-known developers, it seems a little odd for you to simply throw them under the bus, given that you seem to believe we somehow need them to protec
Re: (Score:2)
Yeah, in economics, that's what is known as "competition," and it's generally regarded as a good thing. Now, those in business generally prefer to not have competition, but it actually does often result in higher profits, because they have genuine motivation to streamline and improve.
Re: (Score:2)
Yeah, in economics, that's what is known as "competition," and it's generally regarded as a good thing.
Not when it involves copying your ideas. See, economists would say that that creates a disincentive to do R&D... Or conversely, that it creates an incentive to lock everything up in trade secrets, with NDAs and restrictive contracts that prevent customers from using the products they buy in any way they want. And society generally regards that as a bad thing.
Re: (Score:2)
Oh, please, the trade secret bit is a garbage post-hoc justification. If you wanted to rid society of the social ills of trade secrets, you would just get rid of the existing protections for them, not provide a system that still controls the usage of the knowledge, with zero incentive to patent anything that could reasonably be kept a trade secret for 20+ years.
As for the rest, progress happens faster when there are a) no patents in a field, or b) after key patents have expired. Patents are legal monopo
Re: A solution (Score:1)
I would agree with you IF todays patents where made by years of Hard research. But No One seems to be doing that anymore. Most patents today seems more like waking Up in the middle of the nigth and hey rounded corners
Re: (Score:2)
meh (Score:2)
Because, believe it or not, legal monopolies like patents actually HURT competition, which is where innovation actually comes from. One clue might be that the word MONOPOLY is a descriptor, and is the literally the opposite of competition.
no this is a fallacy. The competition is free to come up with a better product, not to steal the design of the existing one. Apple didn't steal Blackberry's design when they came up with the iPhone and yet they won the market (at the time). George Lucas didn't steal the plot from Logan's Run and yet Star Wars has been a success.
Patents and copyright can be abused, but they do serve a purpose.
Re: (Score:2)
It's not a fallacy (and you didn't even mention a specific one). Copyright and patents are crude economic tools that have no place in modern economies. They do not work, have never worked, and never will work. Where competition exists, it exists DESPITE copyright and patents, not because of it.
Legal monopolies were a common tool prior to democracy and capitalism, but were correctly recognized as generally harmful in the 17th and 18th centuries. That's why, outside of copyright and patents, we only hav
Re: (Score:3)
Not eliminate them, but put constraints on them and also demand higher quality on them. Unused patents should be put into public domain if they are unused to prevent anti-competitive behavior, used patents should be taxed.
Re: (Score:2)
They missed an important problem. (Score:1)
The reviewers are incompetent. The whole requirement of a patent being novel and non-obvious has gone by the wayside. Reviewers either fail to associate prior art because of an unfamiliarity with the field and grant the patent. Or they do the opposite and incorrectly associate prior art and reject it.
Re: (Score:2)
An AI can help find material for a human to review. That's the end of an AI's usefulness at this point.
Re:They missed an important problem. (Score:5, Interesting)
Obviousness is a fundamentally subjective measure which can only be judged by domain experts.
That is why patent lawyers always try to redefine the English language and create really convoluted ways to redefine obviousness in terms of prior art.
Re: (Score:2)
They're only researchers, not domain experts. Domain experts work for a living. If inferior processes were in commercial use you would have some grounds for claiming non obviousness, if your new method didn't depend on very recent technology to make it possible at all.
That's almost always the case though. Most patents are about being the first to realize the potential of advancing technology, rather than any invention which wouldn't have happened regardless. Patents reward "speculative schemers who make it
Wasn't that obvious? (Score:3, Informative)
Wasn't that obvious?
I remember in the 1990's (I think) the patent office changed its policies so that examiners were rewarded based on the number of patents they granted. This led to them granting marginal patents.
The legal theory at the time was "if there's a problem, let the courts work it out". This, of course, is highly inefficient because it puts an enormous burden and expense on everyone who disputes the patent, loads down the court system, and generally halts innovation.
It was obvious then and it's obvious now. There's a great deal of social expense surrounding bogus patents, infringement, and patent trolls.
It's too bad the lawmakers, on both sides of the aisle, don't have the best interests of the people at heart.
They might otherwise be moved to fix the problems.
Re: (Score:2)
The patent office makes money from granting patents. If, when a patent was overturned in court, the office had to pay back four times the patent fee, there'd be an incentive to grant patents less than 25% likely to be overturned.
99% (Score:4, Insightful)
Re:99% (Score:4, Interesting)
For example, the bounce animation [youtube.com] Apple patented is familiar to almost every engineer as the system response of a slightly underdamped second order system [wikipedia.org]. It overshoots slightly before settling at a new steady state value. It's been implemented for thousands of years as a mechanical spring-mass-damper (most familiar as the springs and shock absorbers supporting your car wheels), and for over a century in electronics. There was literally zero innovation implementing this well-known system response in software. But because of our broken patent system, Apple was able to get a patent for it because they were the first to apply for an obvious idea implemented in software.
This wouldn't be a problem if patents were actually being used correctly. You cannot patent an idea [legalzoom.com]. You can only patent an implementation of an idea So even if it were new, nobody could patent the idea of a bounce animation when scrolling hits the end. Apple could patent a certain implementation of a bounce characteristic to iPhones, but other manufacturers could come up with different implementations which were slower, faster, bounced more, bounced less, bounced differently. Unfortunately, the courts have been way too generous in allowing patent infringement lawsuits against the same idea implemented in different ways.
Re: (Score:1)
"But because of our broken patent system, Apple was able to get a patent for it because they were the first to apply for an obvious idea implemented in software. "
If a patent was granted then the test for obviousness did not yield results consistent with your uninformed opinion. Furthermore, what you've described would not be a patent of a "slightly underdamped second order system" nor is there any reason to believe that's now it's implemented or that it's important. I bet you think you sound smart though
Re: (Score:2)
Patents cover specific implementations. That's the whole point. If you accomplish something entirely mechanically, and I do it electronically, we can each patent our implementations, even though they accomplish the same task. Apple's bounce animation is applying math to animate elements on a screen. The implementation bears no resemblance to any physical system, especially shock absorbers in a car.
Re: (Score:2)
Shame what's modded insightful around here.
99% sure you know nothing about patents and have no experience with them.
Re: 99% (Score:2)
That's correct, most of us are not paid shills for patent monopolists. The ownership of ideas is immoral.
Re: 99% (Score:2)
Property is theft, AC.
Re: (Score:2)
I know of industries that use a portfolio of patents as deterrence but not in the sense that they think the patents are any good. Competitors can be happy with a standoff situation where they have (mutual or not) potential patent infringement situations but they prefer to do nothing about them. And this is in part because many patents are unlikely to hold up in court against a good lawyer but as long as the patents are not defeated in court they have deterrence/exchange value. So it's a bit of a bubble.
Re: (Score:3)
99% sure you know nothing about patents and have no experience with them.
I have a few patents. My experiece is the first time you apply for one it always gets rejected. Then you give your patent attorney a boatload of money and a few years later the exact same this gets awarded a patent.
I worked at a place that developed a product and their patent attorney said it wasn't patentable for reasons X, Y, and Z. I don't want to go into it here. Our largest competitor copied the product and patented it, then sued us for patent infringement. That was a real kick in the nuts. Also that p
Problems (Score:2, Funny)
I...wait.
Only THREE??
Way #1 and #2 are inconsistent (Score:2)
-The United States Patent and Trademark Office (USPTO) is funded by fees -- and the agency gets more fees if it approves an application.
-Unlimited opportunities to refile rejected applications means sometimes granting a patent is the only way to get rid of a persistent applicant.
Contrary to point #1, and as shown in point #2, the agency gets more fees if they reject an application and allow applicants to refile the rejected applications - really, not so much "refiling", but filing a "request for continued examination" and a fee. Specifically, the issue fee [uspto.gov] paid when a patent application is granted is currently $960. But if an application is rejected and "refiled", then the request for continued examination fee is $1200 for the first RCE and $1700 for each one after that. Plus the a [uspto.gov]
Omnibus reply (Score:2)
I work in this field. Seeing a lot of bad information here. Consolidating several replies to save space:
Yes. There are empirical studios by economists that correlate a country's patent laws with increased investment in technology. Can't rem