The Case for Patents
One open source guy's opinion on whether or not we should have these things at all
Patents get a bad rap here in open source land. And in my opinion, as many others, rightfully so. Nothing stifles innovation like some big company squatting on their IP to better control the market and extract as much money from it as possible, advancement of the species be damned. But maybe I’m getting ahead of myself. Maybe there’s some kind of justification for the patent system…? Well, get out your tinfoil hats, kids, cuz we’re going to dig into this!
photo courtesy of http://www.paulduane.net/tag/tinfoil-hat-tuesday/
First, let’s normalize our data set. The patent system goes back to 1790, the first incarnation allowing for 14 years of “I own this and you don’t” from the time of issuance. In 1836 it became 21 years, in 1861 it became 17 years, and in 1994 it became 20 years from first filing (although, design patents are 15 years). But you don’t just get to keep the patent for that long. You’ve also got [maintenance fees](https://www.uspto.gov/learning-and-resources/fees-and-payment/uspto-fee-schedule#Patent Maintenance Fee) to pay every few years (again, except for design patents). We’ll get back to the maintenance fees in a bit. But getting a patent in the first place can take upwards of $30,000 and three years.
At this point I’ll let you in on a little secret: I’m not totally against patents. Not in principle, anyway. I mean, if an entity is going to go the trouble of creating something that takes a substantial amount of time and money to pull off, then they ought to be able to recoup their investment and a little extra green to make it worth their while. This is, of course, assuming that the product in question doesn’t suck and they actually can make their money back. But I get it, and I support it. But 20 years?? It seems gratuitous to me after five. Maybe even after one. If I were king of the patent office, you’d get one year. If you can’t get your investment back after one year, your product probably sucks and you should have tried harder.
So that’s my first big beef with the MAN and his patent system: 20 years is 19 too long. You’re effectively holding the general population hostage and crippling our technological advancement intentionally to let big companies extract the most money possible out of us (I did tell you to get out your tinfoil hat for this). Definitely uncool.
I pick one year somewhat (meaning completely) arbitrarily, but my number is definitely not zero. Why? Because if a company doesn’t stand to turn an obvious profit, they’re not going to bother in the first place, and we consumers would be worse off for it. The number one goal of a company is to turn a profit, and the primary levers to do so are 1) maximize retail price and 2) minimize cost. (And there are lots of peripheral means - customer service and such - but those are the big two, IMO.) Maybe a privately-held company could demonstrate something like altruism, but certainly a publicly-held company will only operate in its own interests. No? Look at pharmaceuticals: It’s more lucrative for them to pay off opioid death settlements than it is to stop making the drug and save lives. Look at Volkswagen: they made more money by lying about their emissions to keep their vehicles performing well (as in fun to drive) while polluting the environment than they would by actually reducing emissions and making the cars slower. But they got caught. D’oh!
My point is that public companies don’t work for your benefit, they work for their own. As such, it’s all about the bottom line and how to make it grow. And a 20-year patent term only works to their benefit, not yours. But we still want them to be making things, so throw them a bone. OK, you can have three years. Then shut up and get back to work. Stop employing lawyers to make you more money. We want to incentivize the monsters to keep making good, original and innovative products, but it can’t be a hinderance to the rest of the world.
Now let’s talk about the “maintenance fees.” Beyond paying for the general services of the patent office, it is speculated that these fees will somehow spur innovation, presumably because the fees increase proportionally with the length of time the patent is held. But did you check that link to the fee schedule? Only in rare cases do they exceed $10,000. So, lemme get this straight: Google submits thousands of patents every year, and you’re going to force them to let a patent lapse and innovate over a $10,000 maintenance fee? Or a pharmaceutical company? No, you’re really not.
photo courtesy of http://www.quickmeme.com/meme/35qkzm
Now cinch down your tinfoil hats for a sec. Why do those fees seem so proportionally low? Because the same companies that have to pay them lobby (ok, bribe) to keep them that way. More than that, with the number of patents they have to service, the patent office is still bringing in a hefty sum of cash - around three billion dollars in 2016. Sounds like a little quid pro quo, don’t you think? The government is still getting their cut through the patent office, and anyone that might be in a position to do something about it is getting their cut through lobbying efforts (yeah, bribes). Foxes guarding the henhouse.
Back home at SFE, we don’t do patents. We don’t keep any IP. We hardly keep any secrets. Running a company in this environment is hard. Competition is ever-present and always pushing us to do our best because it’s the only way to differentiate ourselves. But the pursuit is less about the bottom line and more about moving us forward as a species. The more that’s commonly known, the less likely that we somehow slip back into the dark ages.
That said, we absolutely do need some amount of protection. A company that’s willing to go out on a limb to make something that requires relatively greater amounts of time and money should be able to make back their investment and more to insure their own growth. But 20 years is a big chunk of a lifetime, and technology moves fast. It’ll move much faster if we actually let it.
Sparkfun is in a unique situation because most of its original products require engineering efforts measured in weeks, maybe a few months at most. This is in sharp contrast to other tech companies with products (e.g., jet engines, iPhones, CAD software) requiring dozens or hundreds of man-years of effort. Put another way, if it took $2M and two years to come up with a new break-out board, you'd be pretty upset when clones showed up on Alibaba a month later. Maybe upset enough to look for legal remedies.
Another datapoint: The current pharmaceutical industry requires clinical trials of their (fully disclosed!) products before they can sell them. These trials cost up to $100M+ and drag on for years before getting approval to sell their products. The industry wouldn't exist without patent protection.
Pretty much all of the pharmaceuticals are developed using NIH grants, which means taxpayers are footing the bill, at least part way. I think those companies should be given a choice, if they use government money, they have to license their patents to competitors - any competitor, and grant the US government a license that can be used to produce meds for indigent populations -- or they can forego NIH grant money and foot the bill themselves.
I have been a patent attorney for more than 20 years. I also have been an inventor with my own patents in electronics, both in the US and in Japan and worked for small and large companies over the years and have a deep understanding from this from varying viewpoints. Most of the comments here are very good and insightful and I agree with.
I want to point out a few things:
all the low hanging fruit (invention of electric motors, internal combustion engine, transistor, basic plastics, antibiotics, airplane, laser, jet engines etc. etc. has been harvested 50-100 years ago. Very little really useful stuff the last 10 years (a few breakthrough cancer and biology treatments come to mind, that's it).
most patents nowadays serve large rich companies and help them screw the little guy (I have many examples but no time here)
the globalists have altered the US patent rules so much over the last 20 years that it really does discriminate against the small inventor in favor of the multinational (again, i have many examples, but no time here)
overall in my opinion, things have changed so much that for most technologies (not including chemical and biological treatments of disease) we are better off without a patent system
I have seen large multinationals use patents as a way to impede progress where they are making money on something old. Overall these patents hurt development more than they help.
finally, in the area of electronics, particularly hardware, the Chinese are generating MORE than HALF of all new inventions/patents. Because the US patent system is so expensive they dont file here but you need to read Chinese to find out more than half of the new inventions in our area. AND, most of the remaining half is from the rest of Asia (Japan/Korea/Taiwan mostly)
The American empire has rotted out mostly by globalists, who use law to create their neo feudalism order where most of us are held back from reaching our potential by a few on top. All of the legal apparatus, including patent law has been subverted and the relationship between patenting and inducement for new electronics inventions that we had in the golden age of the 1920s-1950s is long gone.
We need to focus our new technology efforts on small resilient self sufficient community development. That is the next step in evolution of our civilization and relevant inventions in the IoT space, in private communications and in renewable power for non-multinational and non-government use are particularly needed. For this space we can organize in small groups to create low cost patents to keep globalists out of our face.
One year sounds like a great idea, especially for things like a new kind of telescope mirror that takes three years to pour, cool, and grind. Or something with a lot of tooling a special processing in the supply chain, or that requires a special building or chemical processing with permits and studies and infrastructure support.
Sparkfun doesn't need any IP because you sell things made with other people's IP (and you wisely avoided things that need your support, like Linux). Without the long terms patents, you would have nothing to sell but bar magnets for the ever popular dynamo and motor kits.
A 14-year patent length was not an issue back when tech changed every 20-25 years. It only became a problem when the tech cycle became slower than the patent cycle. Maybe the answer is to add more categories with different lengths. - software, maybe 3 years, electronics similar, wagon wheels 20 years and so on.
My big problem is what they allow to be patented. For example, components of the human genome. I know for a fact that I at least was around before any of those patents, so did my parents demonstrate prior art? It's one thing to patent a novel use of chromosomes, but the chromosomes themselves?
But if you want the real IP abuser, look at copyrights. Nothing has gone out of copyright since the 1930's (coincidentally when a certain rodent, now a resident of Florida, was first copyrighted).
I have 5 issued patents and 5 more pending. The main cost is not the filing and maintenance fees paid to USPTO but the cost of paying patent lawyers to understand the invention, draft the application, and defend it from clueless examiners armed with an undergraduate degree and a keyword search engine. The process is painful and three to five years pass before the patent is granted. But that's ok because it also takes three to five years for me to sell the idea to management, assemble a development team, complete the design, test it, and get it into production. Only then does the idea start making money, and by then there are many costs to recoup besides the lawyers fees. The things I invent are very specialized and don't sell in high volume. It can take up to three years just to pay off the tooling. Given all that, you can understand why it might be better not to publish the recipe. But I do, because patents are the way I exchange ideas with the other four inventors who drive this particular industry. It's a dialog that takes years, but patents are the only way it can happen: we each work for a different company and will probably never speak to one another in person. But reading their patents (I'm the youngest one in the group) is how I learned the prior art well enough to join the club and contribute my piece. Sometimes I get frustrated that there are good ideas I can't use without a negotiating a technology license. But that's still better than not being able to use those ideas because they're locked up in a competitor's vault. The process of studying those ideas, even when it's just to find a non-infringing work-around, is another way of spurring the next incremental advancement.
The real problem is not the patent duration. The real problem is that patent applications are ridiculously broad and convoluted as an attempt to patent technology and techniques that are in the public domain or already existing. The patent clerks should deny them but they don't because clearly they make a considerable amount of money for each patent granted. The companies then become patent trolls making a living from suing people who have every right to use a design by don't have the money to defend themselves in court..
The advantage of the Patent system is disclosure. While I may not be able to commercially manufacture the widget you've designed - your Patent tells me (mostly) how it works so I can potentially build one for my own use. I may adapt it, innovate and perhaps file a patent for an improved Widget - or sell it - or just enjoy it myself!
The disclosure element [can] spark ideas and push innovation as much as it stifles it.
I would recommend one of the first solutions to fixing the issue would be to REDUCE costs to file for a patent and maintain it. Next, I would try to create a good solution for being able to defend it. At this point, if a company has a lot of money (and lawyers) they are at a huge advantage over small start ups.
Didn't see your comment when I posted mine... It appears we share the same sentiment, DAS01. My startup is currently in the process of obtaining a patent, and it has been nothing but a giant hole in the ground to pour money into. The USPTO lives to screw you over on the most trivial points. They are the real trolls here...
This is because Congress passed a law requiring the patent office to be self-supporting. This also means they have a financial incentive to grant patents, which gives them the incentive to speed through patents for companies that apply for the most, like IBM, Apple, etc. I don't know if they do that, but I do believe that what gets rewarded gets done.
I'd get behind that, but only after a reduced term. Even in the realm of small business, I feel like $30k ought to be do-able. Not comfortable, but do-able. Good solution for defending? I got nuthin. But I'll support anything that doesn't involve lawyers. Le sigh. I really do long for a simpler way of life.
What's wrong with the patent system? Amazon being able to patent 1-click checkout. What's right with the patent system? Apple should be able to patent the in-total design of the iphone.
Reasonable time to hold a patent: I doubt one year of patent for innovation like the iPhone is justified. It is just too small a timeframe. Why? Because it takes time for people to realize the true value of new innovation. I think 7 years is ideal - it also ties with the roughly 7-year business cycle.
Yeah... but in the case of Apple, I hardly think they're going to suffer too very much after a short-term patent expires. They've got so much brand recognition that I think it would carry them pretty far. But not so far that they wouldn't have to innovate to stay out in front.
I'm extremely wary of arguments that any complicated system that has evolved over time can be quickly fixed by the application of common sense or a person who isn't tainted by association with the existing system. This is how we've ended up with a government run by the current set of inexperienced, incompetent clods. Your analysis seems to be informed only by your experience in electronics design, which does change rapidly enough to make a shorter patent term seem reasonable. On the other hand, the whole point of the patent system is to encourage publication of solutions to problems so that others can build on them. In a field that is rapidly advancing, patents should be made obsolete relatively rapidly by new inventions, which has the natural effect of devaluing the older patents. In a field like pharmaceuticals, which is my field, the long time between invention and the completion of all the testing required to market a new drug in the US makes the 20 years after filing go very fast indeed. Did you know that it usually takes about 10 years to get a drug through all the necessary trials? I think the biggest issue around patents is in the issuance of overbroad claims and subsequent rent-seeking , which as been a huge issue in software recently. I take some encouragement from the fact that in the much more mature field of pharmaceuticals, this has been pretty close to eliminated by careful examination of claims by the examiners and a whole lot of lawsuits over the years, which have eliminated wildly broad claims. It may just take time for new technologies to be processed by our system.
The "disclosure" contained in software patents has yielded nothing. No one reads software patents except patent attorneys. Infringement of software patents is almost always independent invention. Software patent litigation typically goes something like this -- we patented a concept and then outlined a few ways to achieve this concept. Now you've independently implemented the same concepts so lets dig into all of your code and see if we can convince a jury you "copied" our implementation. Of course the company being accused of "copying" has never heard of this patent until the opposing lawyers arrive. The core problem with software patents is that they are patenting mathematical concepts something that was not supposed to be patentable.
Something is clearly wrong in that there is a single area, software, where you can patent and copyright the same piece of code. It should be one or the other, not both. The industry is not going to give up copyright so patents have to go.
Another clue, consider searchability of patents. Things like chemistry patents are easily searchable which makes it easy to find applicable patents. Now try doing a patent search on software patents. Your head is going to explode. There is no way to search software patents other that wade through them one by one with a team of lawyers and consider if they apply. That's a big clue that software should not be patentable.
As far as my argument goes, you're certainly right to be wary. Because, as you say, my analysis is informed only by my experience, which includes reading a lot of news. But sure, I'll cop to being a heathen and not knowing what I'm talking about in most regards. But also, a "complicated system that has evolved over time" often (invariably?) gives way to a bureaucracy that tends to favor a few over many. Will my hick-like opinions fix the system quickly? Of course not. But neither do I believe that all we need to do is give the existing system time. The system will just become more entrenched, I think. A fundamental shift in thinking needs to occur.
I believe that reverse engineering has often been employed to aid in creating patentable ideas that may not impinge upon existing patents. Munro & Associates, Inc. dismantled a BMW i3 and offer for sale a package of reports of their findings ($500,000 buys all!), including parts and manufacturing costs.
The 'little guy' may be willing to locally manufacture their own little widgets tediously, but it requires big bucks to create the machinery to mass produce widgets at lower cost, creating the profit necessary to pay back manufacturing start up costs. I never watched to its ending, the motion picture, "Flash of Genius" about inventor Robert Kearns and his patent infringement cases against Ford Motor Company and Chrysler Corporation, but that's gotta be a rough road to follow.
Patents are a form of rent seeking in the market place and are totally unnecessary and only hurt innovation. A good example goes all the way back to the original steam engine patent Patents are a way for the non-creative class to make money off the creative class (engineers, scientists and etc). Trade secrets are enough protection, With today's speed and spread of information, nobody really invents something in a vacuum all by themselves, everybody is standing on the shoulders of giants. Unfortunately the non-creative class has way more power than the creative class in today's society so the patent system will continue, because the legal rent seeking industry makes too much money from it. There are some exceptions, but usually it is the investor that will make the big money not the inventor, who will get a mug and a t-shirt as a reward for inventing. All most all patents today are incremental improvements on existing technology, nothing revolutionary. Human beings are an inventive species, we do not require a patent system, which only slows down innovation. If society wants to speed up inventions it must get rid of the patent system.
The patent system is not only for our kind of "tech" it covers many systems from many bodies of knowledge. It is important that we remember that. We cannot just focus on the speed at which "tech" evolves.
A car manufacturer sometimes change the shape and tech in their vehicles ever "such" years, like BMW who mostly does like seven years or so, because it takes a lot to create a new model. they need to recoup those investments over a period, which may indeed take longer than 7 years, so a specific tech may find its may in successive models because its not financially feasible to redesign the tech after 7 years.
It is also the same for other organizations that develop items we use on a daily basis. A Lot of the investment that is poured into R&D is base don loans, sometimes with high interest rates because the loaner does not necessarily have quantitative evidence of your product's future performance so the investment in you "the developer" is a risk. The revenues to cover those repayments have to be recouped.
for organizations that are not seasoned and do not have a wealth of experience behind them have to invest heavily in talent to have their systems developed and even more to have them brought to market.
With the competitive forces within the market place at play and the slow dissolving geographical boundaries competition is being presented on many different fronts. this makes it even more difficult for an organization to make steady profits as you wont have all the customers or the right products. this means it takes much longer for company "x" to make a profit of "y" and they have to work much harder and longer. With the suggested 1 or 3 years, it may be insufficient for them to see a profit.
The next issues is that innovation does not only come from large corporations, single individuals and small "momma" and "poppa" companies innovate too and their situation begs for the 20 years.
The issue of the renewal fees follows the same pattern, patents were not just created for large "top-heavy" organizations who "poop" and "pee" money like google. yes they do lobby a lot, but what happens if you, or I apply for a patent?
I may want to bring a product to market, which may cost me 500K or a million to do so, plus the ancillary fees and legal stuff which are all additional cost, and the loan interest payments. Now i apply for a patent, that's another 30k (a loan you say?). How long will that take me to recoup? what should my maintenance fee be? 10K every few years, which eats into any profit i make after those huge bill i racked up trying to push that product to market are paid.
My point, the patent system and the associated maintenance fees were not devised simply for large organizations, they are the way they are so they may also apply to the small person.
I believe there could be tiers to the maintenance fees which looks at your gross income and does a calculation based on that, providing a reasonably low cost for the small individual and a more appropriate ("whatever that is") cost for larger organizations.
as for the term length, the 20 years is just fine. If what a man develops is so important that it can slow the progress of the species should it not be disclosed, he deserves to make a buck or two from it.
Imagine if you are a software developer, who uses licenses to lock your products (that's your bread and butter) and you are told that after 5 years, or even 10, your system (even your patented license or encryption scheme) has to be release to the public, would you want to innovate?
I think there is a place for patents, but I don't think patents should be allowed to block development. How can we do this? I think all patents must be licensable for a reasonable fee based on profits. That would mean no more monopolies, but it would guarantee that patent owners would get a fair return for their investment. It would mean competition between companies bringing products to market based on the patent, which would help keep prices down. Ultimately, it might increase the value and utility of patents, because products including the new technology would move into the marketplace more quickly and perhaps sell more broadly, generating revenues for the patent holder. Since non-profit production would be exempt, it would make it possible to produce patented medications for indigent populations at a reasonable cost.
The biggest problem with my idea is that Congress would have to pass a law and determine the "fair" percentage. It would be best if that task was done by an independent body not so easily influenced by the big corporations that would lose their ability to club small businesses with patent infringement suits.
I'm sure there are many other problems, but I think this would be a good place to start, rather than just disposing of patents altogether.
From my 30 years experience as an engineer in the oceanographic instrument field, when I design something and we discover someone else has patented it, there is not even the suggestion that we might pay a dime to that competitor to use their patent. Instead we will spend much more money to find another way to do the job even if it makes for a clearly inferior product. As an engineer it irks me to make such an inferior product but management would never stand for paying patent royalties for something our people (re)invented.
The patent costs aren't too low; they're FAR too high. Far too high for your typical garage inventor to obtain a patent without mortgaging his house. Yet these fees are still low enough that the "evil corporations" have no trouble obtaining patents. It's almost as though these government regulations are designed to favor big business by stifling competition from small business...
I agree with Pete here, 20 years is way too long. One year, on the other hand is probably not long enough. There is probably a sweet spot somewhere in the 5-7 year range, no renewals. The biggest change I would push for is to make patents non-transferable. By that I mean, the original patent holder could license the patent to anyone for manufacture, implementation, improvement, etc, but could not be licensed to or sold to a non-practicing entity. Hopefully that would drive corporations and individuals to use patents to reward innovation rather than as a defense mechanism against patent trolls.
So, you get your plant built and running, sell things for 3 or 4 years then lay everybody off and shut down? How to you build another plant? How do you do the standard thing of selling the foreign rights when you are going to be copied at home in a couple years? In fact, why would anyone buy from you when cheaper versions will be here in a just a couple years?
The way to fix copyright is escalating fee paid renewals. First 20 years free and automatic. Anything past 20 years requires registration and fee payment. Something like $100 first renewal and doubling on each subsequent renewal. That lets Disney keep the mouse as long as it wants to and cures the orphan works problem. The current system holds culture hostage. Every photo and recording from WWII will be under copyright until my grandkids (not even born yet) are adults. That is just wrong.
Patents. Just get rid of software and business method patents.That will fix 90% of the problem. Next raise the bar on what can be patented. It is insane that over 250,000 patents apply to cell phones. https://www.patentprogress.org/systemic-problems/too-many-patents/
I can fix the patent system in five minutes. Limit the PTO to issuing 1,000 patents per month. Make the PTO decide what are the best 1,000 each month from their pool of applications and issue them. After six months in the pool without being issued, your patent is dead. Doing that will make those 1,000 patents truly valuable instead of the pile of patent spam we have today.
This is just nuts and not "promoting the progress of science and technology".... "RPX estimates that there are 250,000 active U.S. patents applicable to the smartphone. Assuming that smartphone patents have an average of 20 claims each, that is 5,000,000 restrictions on smartphone innovation. Researchers calculated that it would take roughly 2,000,000 patent attorneys working full-time to compare every software-producing firm’s products with every software patent issued in a given year. Moreover, any such search would be incomplete, because it cannot reveal applications that are not yet published (18 months from filing in most cases), let alone those that have not yet been filed."
You are an idiot (or a patent lawyer) if you truly believe there are 5,000,000 "inventions" in a smart phone worthy of patent protection.
I don't like the current patent system, but I like the idea of a patent clerk deciding if my innovation is one of the top 1000 even less. First off, if they were that good at tech, they'd be filing patents instead of granting them. Second, that's just asking for Russian-scale bribes and corruption.
That's the whole point of the 1,000 patent limit. Move the fighting to before the patent is issued instead of after like it is now.
That limit would do wonders for the quality of patents issued. Nobody will bother submitting junk applications like we have now. You need to be realistic about what should be patented too. Are we really making 1,000 inventions a month worthy of patent protection? I don't think so. See the above reference about 250,000 patents against smart phones. A phone can carry maybe a $40 patent load. That's $0.00016 per patent. Yet every one of those 250,000 patent holders wants a $1 or more royalty. And so we have endless cell phone patent litigation.
"Quality" is too subjective. One man's gourmet cheese is another man's stinking bucket of spoiled milk. So the hunt is now on for an arbiter who either likes cheese or whose opinion can be bought at a reasonable price.
Instead of a numerical cap, the standards for what constitutes a valid patent disclosure and where patents are applicable need to be better defined and/or enforced. Otherwise we'll be at the mercy of the corrupt and stupid.