Patent Policy on the Back of a Napkin


Are all functions in Economics continuous, differentiable and unimodal?

It gets worse. Sometimes they are straight lines with a textual disclaimer that they have a couple of assindotes.

It's a fair question, but based on my experience in several messy domains other than economics, it seems to me that c., d., and u. are reasonably common in real optimization problems, at least at the scale that matters. (Few things are truly continuous if you have the resolution to track individual atoms and photons, but it's seldom practical to work at that scale.) If you want to pick on the underlying assumptions of the post I'd nominate one-dimensionality of "patent strength", one dimensionality of "innovation", and all the details swept into an implicit ceteris paribus as more likely to be real flaws than continuity, differentiability, and unimodality.

That functions not continuous: if you look closely, you see that he has discontinuities drawn in at all the places where there are molecular bonds.

That's a laugher! ;) This is what they call a 'masculine napkin.' I hate to see what is on the front of the napkin: TGS vs WeRFd?

I'd argue that we're farther to the right, i.e., we'd have more innovation if there were no patents at all (although, the nature of the innovation would probably be a bit different)

I am quite skeptical on that one; would you care to elaborate? In particular, I'd like some examples of the sort of innovation you think is being suppressed by strong patent law but would flourish without patent protection at all. Note that, for example, Samsung wasn't interested in publishing an open-source design for a rectangular-button smartphone that any Malaysian sweatshop (or geek with a 3-d printer) could produce at will; they wanted a patent of their own.

Generally speaking, even current patent law is not a major bar to non-commercial innovation. Most countries don't even allow patent enforcement against non-commercial users; the United States technically does but it is fairly rare - if for no other reason than that patent enforcement is done by civil suit and non-commercial parties usually have shallow pockets.

So what we're looking for is a large body of commercial, for-profit innovation (as opposed to profitable exploitation of someone else's open-source work), where the innovator's profit doesn't depend on any level of patent protection. There are certainly niche cases where this dynamic exists, but I am skeptical that it dominates the landscape of commercial innovation.

Also, an important caveat: particularly in the commercial world, the alternative to patent protection is generally not open source and free flow of information. The alternative to patent protection is often trade secrecy. This can encourage innovation, as everyone who needs a better mousetrap goes about trying to invent their own, but not necessarily socially beneficial innovation. What we want is for everyone to be able to buy, or build, the best mousetraps that anyone has ever invented. License fees flowing to the inventors can arrange for this to happen. What else do you have?

I agree with Thomas. Take a look at how companies are buying junk companies just because they have a lot of patents and gives them additional ammo in a patent war.

If I had a great idea for a new device I wild skip it due to patent headaches. I can't believe things like "one-click" or clicking on a phone number to dial are patentable.

Great graphical representation, but I would move the point as suggested by Thomas.

Then there's biotech where companies are patenting gene sequences where they don't even know the function. In other areas, patents are being issued for many things that are so obvious that someone fresh out of college. The whole patent system is out of control and stifling innovation.

I don't see the part of that little sound-bite rant where you actually answer the question.

"If I had a great idea for a new device I wild skip it due to patent headaches"; fine. I'm not sure we need to be discussing innovation with someone who has never had even one good idea, but it hardly matters because ideas aren't the point. Ideas are all but worthless, in the economic sense of supply vastly exceeding demand. If you don't have any, rest assured that there are others who have far too many. What matters, is implementation. What the patent office calls "reduction to practice[1]". The idea, plus the actual willingness and ability to disappear into a properly-equipped lab or workshop or whatever for a couple of years to turn the idea into reality.

I've actually done this, repeatedly. And at least once your ideas get beyond the level of trivial bits of software, the "headaches" of obtaining and defending a patent are in-fucking-significant compared to the headaches of actually inventing something worth patenting. If you tell us that you are put off by the patent bit, you're really telling us you were never a contender.

But, still, you're willing to put yourself in the headspace of an imaginary inventor, so let's run with that. You've got an idea for, say, a better fuel injector that will improve the fuel efficiency of any car by half a percent. As with all such ideas, there's maybe one chance in ten it will actually work. Developing and testing the prototype, that means quitting your job and blowing your life savings on a garage laboratory/workshop. If you're under 35, you've also got to hit up your parents for their lives' savings. Either way, you've got to live on Ramen for the next few years while you work at it.

And there's no such thing as patent protection. If and when it does work, a year after the first car so equipped comes on the market someone with a Malaysian sweatshop will be offering aftermarket versions for half the price, because innovation isn't their thing but optimized sweatshop assembly is.

So explain to us how, in the Brave New World of No Patents, the improved fuel injector winds up under the hood of every new car in America. Because it should, but I'm genuinely not seeing how it would.

[1] If anyone is looking for ways to improve, rather than eliminate, the patent system, more rigorous enforcemnt of the "reduction to practice" rules would be a good start.

Lay aside that most inventions aren't made in garages anymore, most inventions are made by companies within there research departments of people getting paid for it.

But let's still take your hypothesis.
Would he even be able to invent a improved fuel injection without infringing already existent patents?
I guess not and that's where the whole problem lays, most inventions nowadays aren't something completely new (for sure not all, but the amount of real inventions compared to iterative inventions is almost laughable) they are iterative and some sort of improvement of something already existing and even if the improvements make it a completely new product (or sort of) it can be sued.

No, let's not lay aside that inventions are often made by corporate R&D departments rather than individuals, because that makes it worse. The mathematics of profit vs. loss work the same way for individuals and corporations; the difference is that an individual might put his own money into something that will predictably be a net loss for personal reasons. A corporation won't.

As for infringing existing patents, the way to avoid that is to license the prior patent. That's the way it is supposed to work. If you run a corporation that invested fifty million dollars almost-perfecting the injector (10% more efficient than old-school carburetors) and now sells them at $25 each to among other things recover your R&D costs, how is it a good thing that I can disappear into my garage with a few samples of your fuel injectors, come out a year later having squeezed out an extra 0.5% efficiency that you missed, and sell the resulting product at $10 because I only have a few $100K of investment to recover?

Yes, great that the world gets a 10.5% improvement in fuel efficiency at $10/cylinder, but you are out $50 million and the model is not sustainable.

So yes, the inventor who makes a v1.1 version of someone else's invention, genuine inovation that it may be, still gets sued if he sells it without licensing the v1.0 patent from the original (possibly corporate) inventor. That's not the problem. That's the solution. The problem is, how do we get lots of seperate teams all working to advance the same technology when, in an unrestricted market, only the one guy selling the latest and greatest version would generate any revenue?

The solution we have involves lots of lawyers negotiating multi-party cross-licensing agreements, and as every Shakespeare fan knows we're going to be killing off the all the lawyers any day now. And there will be much rejoicing, but then we are going to need a new solution to the problem.

Your predecessors in this subthread are conspicously dodging the call for new solutions; what have you got?

>>>>"In particular, I’d like some examples of the sort of innovation you think is being suppressed by strong patent law but would flourish without patent protection at all."

It's a very good question, and one that anarcho-capitalists and sundry other anti-patent types rarely answer. When they do, it's often with examples from the food industry or cosmetic industry. Yet when we examine the labels on such products, we often see the identical ingredients except in a slightly different order: Sabra Hummus has chickpeas first, oil second, spices third; a competitor lists spices first, chickpeas second, oil third.

Pure copycatism. Apparently, that's the sort of "innovation" the anti-patent crowd admires.

At root, of course, is their fallacious notion that ideas — specifically, those that are developed enough in their details to be a practical recipe for creating a product — are not a form of property.

Hank Rearden would not approve.

Well said. BTW you can indeed patent a recipe but most rely on trade secret hence celebrity chefs rely on trademarks rather than patents--easier to enforce. So a world without patents is like the Middle Ages--when guilds zealously guarded trade secrets--and btw concrete--invented by the Romans and then LOST, was "reinvented". This is the world without patents--a world of secrecy even more so than today. All in the name of 'fighting monopoly' aka price differentiation which economists think is such a bad thing. They don't realize that a world of perfect competition requires zero profits (Arrow-Debreu conditions) that most people would not settle for.

"zero EXCESS profits"

Yes. They never answer this question.

I read this IEEE paper abstract and the thesis is wrong: that somehow software is different than hardware because it's not grounded in physics. Truth is, you can come up with a much faster algorithm in s/w just like you can build a faster car in h/w. Google Quicksort by Tony Hoare. Now having said that, for historical reasons nobody patented s/w so perhaps we should have 'transition rules' to help ease the transition, as well as special rules for grandfathering, and of course better patent examiners.

These are very straight lines and beautiful handwriting for writing on a napkin.
Must have been an expensive restaurant. Or was it at the Romney fundraiser?

Dear Sir:
We have found that the above-referenced napkin is in substantial infringement of Patent # 897,346,923,457,692,987,234,659: "Method and Diagram for Persuading Elevator Audiences that a Government Policy has Gone Too Far," held by Arthur Laffer ...

Thread winner.

You just won the internet for the day.

No he did not win the internet for a day. "Al Gore invented the Internet" (TM, SM). (c) 2012 by Ray Lopez. All rights reserved.

Just to be clear this mostly revolves around software patents, yes?

Design patents can be pretty terrible as well.

Don't forget business method patents.

Let's just go for all patents...

Stacey @Interaction

But I think we can all agree that Apple deserves a patent on a rectangular cell phone that has rounded corners.

Steve Jobs has just sued God claiming God has used corners, rounded, flat, and white without Apple's permission.

Computer hardware architecture patents are also bad.

There is no fundamental difference between software patents and chip design patents. Both are idea patents that don't involve taming any laws of nature. Same for business-model patents. It's these idea patents that we need to get rid of.

Well patent laws clearly aren't holding you back. Five-minute lectures, stylish econ-geek napkins (which actually imitate a long tradition of economics on napkins)...In addition, I would like to request a TGS-inspired cookbook. I think that would nicely round out the social experience of taking an MRU course.

And you left out the continuous gathering of any and all data which can be used for metrics to further enhance the MR University experience, such as receiving the appropriate fund raising material for various causes.

A business process (a rose by any other name...) which, to my knowledge, has not been patented yet - though if MR University was ever to offer one click books, they would be violating a patent.

The American patent system has been in epically bad shape ever since the RSA decision.

'The patent for the RSA algorithm (U.S. Patent 4,405,829) was issued on September 20, 1983, exclusively licensed to RSA Security Inc. by the Massachusetts Institute of Technology, with an expiration date of September 20, 2000. RSA Security maintained a standard, royalty-based licensing policy that could be modified for special circumstances. In the U.S., a license has been needed to ‘‘make, use or sell’’ products that included the RSA algorithm. However, RSA Security has long allowed free non-commercial use of the RSA algorithm, with written permission, for academic or university research purposes.'

Ironically, there was prior art in 1973 - except it was classified by the UK government until 1997.

Which leads to the entire subject of 'secret' patents - this becomes particularly notable when dealing with TEMPEST (this link is indicative - -

Patent number: 3897591
Filing date: Aug 27, 1942
Issue date: Jul 2, 1975 )

The perennially aggrieved - their knickers are always in a knot.

I believe the modern, hip, internet term is "butthurt".


" prior_approval is butthurt about MRUniversity."

Not in this case - since truly, Amazon does own a patent on one click ordering (in the U..S.), and if the author of the Great Stagnation was, for example, to offer a one click download of his own book, that would be a violation of that patent.

Or at least give Amazon grounds to go to court.

Strange how one can't even mention the environment in which MR University is swimming in when one of its founders talks about patents, and then attempting to point how patents now can reach into online education, which is absurd on the face of it. (Though the patent for education over long distance involving interactive screen technology is just waiting for someone to get it through the American patent process.)

And, the Lord saeth to His obedient followers on this site:

"Behold this Alex Napkin, For It Contains All the Economic Economic Truths You Need to Know About Patents,


Ignore the Ketchup Stain.

That is all."

So saeth the Lord.

OOh- groupthink burn!!

Now excuse me while I pop over to Krugman's blog comments for some lively, open-minded back-and-forth.

According to that diagram, we are still experiencing *slightly* more innovation than if there were no patents at all.

I disagree. :p

What printer do you have? Mine doesn't accept napkins.

You could have posted that 5 minutes ago.

On an unrelated note: Is anyone interested in purchasing a slightly used printer?

We will invent a printer for you to print on napkins.

However, with the cost of development and the fear of knock-off because we view any patent as less secure, we will be unable to do this unless you pay up front and commit to a three year purchase at minimum quantities.

Intuitive graph Alex. I like it. I will probably never forget it. Goes to show a picture is worth a thousand words. What software did you use to make the entire thing though? I'm assuming you didn't draw it on a real napkin or did you? The lines seems too straight and dark, and the ink isn't bleeding which I associate with napkins drawing.

I'd like the evidence that patents are ever a net positive, innovation-wise.

(Tyler and Alex should really establish a "vote up" system here.)

I agree, for some reason I am always reaching for the upvote button on this site, whereas on most I don't miss it!

Pharma can provide some pretty bullet proof examples of technologies that would not be available without patents.

I agree with the general sentiment of this napkin. I disagree with the placement of the dot.

I disagree -- you're wildly overgeneralizing.

For software and business methods, you're probably right. But for pharmaceuticals, we're to the left of the hump. But the economically neat solution -- different terms and strengths for different fields and different extents of innovation -- are barred by TRIPS.

Personally, I think a sui generis IP protection for software and business methods (like those available for boat hulls and semiconductor masks), with a much shorter term and weaker scope, would be far better than a full-fledged patent. This would also avoid the TRIPS issue.

Finally, as an economist, I'm disappointed that you're speaking about the patent system in the abstract, rather than the marginal substitutes. Is Trade Secret -- with theoretically perpetual duration, messier laws, and NO disclosure to the public -- really better to put things into than the patent system, where just about everything becomes public 18 months after it's filed, whether or not a patent is granted? And to have companies place even more more resources into reverse engineering others' innovations, rather than creating their own? (And let's just note that in terms of compositions used to treat conditions, mass spec basically makes Trade Secret worthless for pharmaceutical patents, as chemical structure is rather easy to reverse engineer.)

Absolutely, get rid of idea patents (software, business methods) and things will be much better. No need to replace them with anything new. Copyright will do nicely.

You really want to put in criminal penalties for IP infringement, massive statutory damages per act of infringement, constructive notice provisions from registration, and replacing a substantive examination system with a perfunctory registration system that doesn't even purport to check for originality or innovation?

Not to mention that you can't protect the ideas in copyright, just a particular expression.

Like I said, I don't want to put in anything new. Just continue using copyright the way it is used today and has been used for software longer than patent protection.

"Not to mention that you can’t protect the ideas in copyright, just a particular expression."

That's the whole point.

Software should be protected for the life of the author plus 70 years? Good to know where you stand.

You're apparently trying to be smart but it's not clear what your point is. Copyright protection for software is not a "should", it's the status quo. The excessive protection period you mention is not a problem for software as it is usually obsolete after about a decade anyway.

Actually our patent laws were much stronger in the 1980s and 1990s, when our economy was doing much better. Countries with stronger patent laws do better economically than those with weak or non-existent patent laws. You cannot find a single example of a country with patent laws that are too strong.

How could you tell if Patent laws were "too strong"? What the hell does that even mean? What could possibly serve as evidence which is more than just anecdotal?

See NTP v RIM,_Inc.#RIM_patent_infringement_litigation

Jurisprudence related to patent damages have changed significantly since then, making situations like this impossible now.

I took it to mean they are too strong when compared against a theoretical benchmark.

Cant tell if you are trolling or serious.

Here's a little anecdata regarding software patents:

Back in the 80s and early 90s when software patents were first becoming part of scene for high-tech companies, I was the R&D department for a smallish software company (sales were a few 10s of millions per year).
Before software patents were on radar, *everything* I did was protected as a trade secret. Management's answer when I wanted to publish was always "No! That's a trade secret."

After software patents arrived, that changed completely. Either we applied for a patent and then published (not too frequently, as the cost of pursuing a patent looked high to our management), or we published quickly to establish prior art before someone else could patent our innovation out from under us. Either way, the entire output of our (small) R&D department went from completely hidden to completely open.

Many folks I know who worked *in* *R&D* elsewhere in the software industry tell me they saw a similar pattern at the time. Significant increase in publication, plus a modest number of patent applications.

It's only anecdata, but it was uniform across the industrial research groups I had contact with. Other portions of the software industry may have seen other effects. Your mileage may vary. Insert disclaimer here.

The question is if this phenomenon increases knowledge transfer. According to my anecdotal evidence, hardly anyone in the software world looks at patents in search of new ideas. The content of most software patents is either too trivial or too superficial to be useful.


I'm not sure whether the increase in patent applications increases knowledge transfer, but I'm *certain* that the increase in defensive publication does. We completely stopped hiding advances from the rest of the world, in favor of publishing anything that might perhaps be patentable by some stretch of the imagination. The need to defensively establish prior art became a big deal to us.

This is a very strong point. Bad patents are highly visible, but the lack of information dissemination is mostly hidden. It may bias our perception of the relative benefits/dangers. I would be interested to see Alex address this point.

I think you mislabeled the x-axis as "wealth inequality."


I'd like to see the same analysis for copyright law. We'd need a much bigger napkin though since we'd be way, way out on the right axis, about 3 SNs (that's Standard Napkins - can I patent or copyright or trademark that or something?)

I published a book recently. Or course if I hadn't been sure my great-grandchildren would still be able to own the copyright it wouldn't have been worth my while.

Alex, your napkin contains a beautiful model, but from an empirical perspective, we have more innovative products and technologies today than in any time in human history? Doesn't this empirical reality falsify your model, or is it your argument that we would have even more innovation if IP law were weaker? if the latter, how could prove that assertion? After all, countries with weaker IP laws (e.g. China) produce less innovations than countries with stronger IP laws.

As someone who holds several patents and has 5 applications in process I have some observations.
First - that "back of the napkin" graph reminds me of a famous ecology graph that is shaped like an inverted U. Except that if you actually find real data points and graph them, you get a graph that circles back on itself in loops and wiggles all over. Such "concept graphs" are dangerous because they lead to wildly inaccurate conclusions.

Second - I have also done technical diligence for investors on deals. That includes examining patents. There is a problem, but it is mostly in lousy examinations. Some examiners are great. The are straightforward, know their stuff, know the area and help make your patent solid. Others are crap. They don't know their area so well, or don't care, or just want to make you suffer. Some of those appear to engage in deliberately trying to be impenetrable, sometimes I am sure to cover for their ignorance. Blood sugar monitoring is an example of an area with way too much overlap. There are patents submitted that appear to have deliberately withheld knowledge of prior patents because the filer was on the prior patent, but submitted one nearly identical later. Investors hate that because it's a cost.

Third - The flip side is that without patents companies can't get investment. There are areas i think shouldn't have patents. Software is one of them. But if you are working with any device, a drug, a biologic, gene therapy, what have you - the capital required to get those to market is just too high. Those areas of the economy just wouldn't work.

Last - The biggest problem I see with patents is that the system has moved away from its original purpose. It's intent was to ensure that those who think of things get paid for them, and that those who go first get paid for their effort rather then arrows in their back. That is beneficial for society. But we have gone in the opposite direction. This so-called 'patent reform' is garbage. It is legal license to steal without recourse. And that kind of litigation was rarely a problem. It occurred on an infinitesimal number of patents.

I think the solution is to legislate that patents can only be filed by true inventors as before. And, that a minimum of 50% ownership in the patent cannot be signed away by any legal contract to any party. That forces those who do the inventing to be paid for what they do. That encourages engineers and scientists who are the best to go into inventing.

I have also been toying with trying to generate a test case. Because the US Constitution states:
"To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries."
That means that nobody who files in the USA who is not the true inventor can receive patent rights. The revised law is unconstitutional.

IOW - There are some solutions.

1. Make patent examining a better paying profession. The patent office is a serious profit center for the Federal Government. So money isn't really the issue. Raise standards for examiners. Give them long-term benefits based on points accrued for value of patents granted vs bad patents the granted. Make it easier to get rid of a crummy examiner.

2. Throw out 'patent reform'. It has nothing to do with fixing anything broken.

3. Give 50% of all license revenue to the actual inventors. That cuts into bottom line a bit, but it also would attract the best and brightest into inventing. That's what we want.

4. Establish special high-speed courts to deal with patent issues and clean up actual problems. For instance the 'dog in the manger' holdout problem. There are people who buy options on patents from poor scientists with the object of blocking use of a technology. A company that has rights to use most of the patents in an area should be able to force a deal down the throat of these jackals in 60-90 days, start to finish, for a few hundred dollars in fees. It should be open and shut - you show you have rights to most of the technology - if they won't deal the court sets their license fee for them and it can be zero. Another instance of something real - patent squatting. This is similar, but it's a patent that isn't getting to market and the owner won't deal. Same thing. In 60-90 days, start to finish, it should be possible to force a license that is backloaded based on actual sales at some future date. The license cost should be something reasonably supportable. The patent owner shouldn't be able to stop it if they aren't actively using the patent - period.

Do those four things and a whole lot of stuff breaks loose.

Well said Brian. I agree. BTW they do give inventors 'shop rights' and a non-transferable royalty right to inventors in Germany and Japan (of all places, which most people think is pro-employer). I would add to your wish list, a "Not Invented Here" defense for companies that actually invent the same thing as the patented invention without prior knowledge of the same, and/or, bring back the 19th century rule that an actual working model of the invention be filed with the Patent Office unless the invention is pioneering. This would prevent a Jerome Lemuelson type troll from harassing legitimate companies with tiny patented incremental improvements that the companies would inevitably have designed themselves. Sadly, I doubt the vast majority of readers in this well educated forum know what I'm talking about, much less the hoi polloi. So again society will depend on innovation "exogeneously" and from Good Samaritans and born-to-invent people, rather than introducing a stronger profit motive. Ironic that long-term growth in capitalism depends essentially on socialistic charity.

The Lemuelson troll problem was fixed a long time ago. That is why the examination trail is available and it is also why you cannot add claims unsupported by the original application to your patent. Trust me, if you file a provisional and something in your regular application isn't supported by the provisional the examiner will declare it invalid and to maintain rights to it you will have to split it off and file a new patent for it. Same goes for your regular application. I've had this happen. Once, I challenged it and the examiner accepted the claim.

Additionally, you have a limited number of office actions now before the patent application is closed. You can petition to re-open it. I did that twice on one application where I had a lousy examiner. Neither I nor my attorney could figure out what on earth he was objecting to in the language. When I contacted the PTO legal department to complain the PTO attorney listened and said he thought i was right. I don't know this for sure, but I think the might have said something internally to him. But it's a royal pain and the examiners look very carefully at whether or not the petition has merit. You can't do what Lemuelson did anymore and keep filing drivel to keep the application open in order to submarine.

Working models aren't feasible. The PTO can't possibly deal with them all. Many are dangerous. Many are very hard to demonstrate without a lot of other equipment. What is demonstrating a working model for a gene therapy? It can take a year to demonstrate clearly that something 'took'. This would also mean huge laboratory facilities, animal testing labs, etc. Also, in today's world you can't get money without a patent at least filed. Quite often you can't build the invention until it's patented. If we were going to use that rule i would say it should apply to extraordinary inventions only. For instance cold fusion. And the way that should work is that a patent could claim pioneering status and get an extra 20 years of life. But to get that it would have to show an extraordinary leap of technology.

The not-invented-here clause would not be constitutional. Patents are to protect the invention. If you say that inventing it later independently isn't covered then there is no patent system at all.

The "tiny incremental patent" problem is back at the examination level. Should this method or device be patentable at all? Is it non-obvious? Is it a design patent? Good examiners do a good job.

You're fighting a strawman about inventorship. Both under the new system and the old system, the inventor's assignee could file a patent in its own name.

All the current system does is say that rather than the invention usually belonging to the first to invent* (subject to diligent filing after reduction to practice, as shown in the arcane and labyrinthine Interference process) who files, it now belongs to the first filer who invents.

Despite what you may have heard, we do NOT have a first to file system. We have a first INVENTOR to file system -- just as, at least in part, we previously had a first FILER to invent system rather than a first inventor system.

In both systems, you still need to have invented t(or received the assignment from the inventor) o qualify for the patent.

"Both under the new system and the old system, the inventor’s assignee could file a patent in its own name."
What alternate universe do you inhabit? Assignee's are never inventors. Patents are never in their name. Assignees can file for patents in other countries as owners, but not as inventors - not in any I am aware of. Assignees can pursue the patent completion for the inventor. But the inventor remains the inventor.

"it now belongs to the first filer who invents."
First, if someone "invents" something that was previously invented by someone else then it isn't an invention. It's a re-invention. Second, if you think that isn't a freeway for theft of IP I have a bridge to sell you. All that is required now to file as inventor is to claim you are. It is now required that everyone else prove you are not. Try it. You could invent cold fusion, I could copy your design notes and everything else, tweak it (if i care to bother) and submit it as mine.

After 8 trips through the patent system, three in the new system, I know a bit about it. Yes, I have a good attorney. So far every patent attorney I have talked to thinks the new system is garbage, and legalizes theft.

This is brilliant. Do one for copyright duration!

I don't think patent *strength* is the problem. The problem is that the growth in patent applications over the last two decades has vastly exceed the growth in resources available to the patent office, resulting in 1) long delays between patent application and granting (which can render patents completely pointless in fast moving industries), and 2) inadequate ability to examine the patent application for novelty, usefulness and non-obviousness. In short, the patent application process has gotten a bit loose and messy, making it more viable for people to exploit it in ways that were not intended (patents for hostage taking, trolls, etc.)

This is half true. It is true that the growth in applications has exceeded manpower allocated. It is not true that growth in demand has exceeded monetary resources. If the PTO was given all the money it collects and allowed to spend it as needed patent backlogs would virtually disappear, salaries could rise and examinations could greatly improve.

That said, it depends on the area of art. Backlogs are on the order of 10 - 36 months they get to your file. Office actions take a few months, maybe 8 or 9 in a highly backed up art unit. But - and this is a big qualification - I have never known a situation where the backlog was a problem. Patents are a lot of work on both ends. I rarely turn around an office action in less than a month and usually I go right up against the deadline. Time is not really not a problem.

The hostage-taking problem is separate and that is why I propose special patent courts. It should take 60-90 days to force a grant of license in such cases. The court should be able to decide, for a low cost, whether the dog-in-the-manger should get anything at all. You're right. An entity with most of the IP should not be stymied by a minor patent or one that shouldn't have ever been granted.

Everything should be oriented toward rewarding of true inventors and making the inventions available.

I’ve been thinking about R+D, and R+D pools, and it should be possible to get higher ROI on R+D by giving incentives for some patent pooling, and there are probably other policies that would help. non-WMD Basic R+D is definitely underfunded as there is uncertainty in the projects, and probably funders are adverse to uncertainty just like finance industry is. If it takes longer than 20 years to get some R+D to market, chaining together patents or extending them would help with basic R+D. There ought to be a formula for the optimal size of government procurement (buying tiny electric cars 5 years ago or building greenhouses in cold climates (using nanoplastics) based on expected market size and odds of success. Just have to stop running such deep deficits. Governments could pool to have GERD R+D done by the lowest interest rate gov.

Wow, one of the 1st computers was for the USA Census. My country shouldn't be cutting our Stats Canada. A well designed Crown (our Space Crown was only getting $250M/yr a few years back) can make up for private or public underinvestment. Canada is at 1.9% well USA is better at around 3.5% annual R+D spending. Without USA gov procurement, we might be at 1980s computers. M.Romney's M&A industry would still be at the 1987 computer program trading crash, level of computers. Fewer profits.
I finally beat Oregon Trials 21 years ago. Those settlers were a bad shot. If one of them was a prophet of god, shouldn't it have been easier to shoot Mcmeals?

In 20087 Michele Boldrin and David K. Levine published a book-length examination of the manner in which patents and copyrights stifly innovation; numerous economic studies are cited:

Interesting read on this topic.

How is innovation related to patent strength? What if there are no patents left in this world but innovations keep on happening?

Comments for this post are closed