Opinion: A Simple Solution for the Patent Mess

Discussions

News: Opinion: A Simple Solution for the Patent Mess

  1. Opinion: A Simple Solution for the Patent Mess (33 messages)

    As a recent patent applicant, I have given some thought about how to deal with the stupid patent process, as evident by the recent Kodak-Sun patent saga over Java. I have seen a lot of complaints but no real solutions. I believe I have come up with a simple solution and would like to receive your feedback.

    To keep my writing simple and short, I will just list my points.

    The problem and analysis of the situation:
    0. The current patent process is stupid. Period. (Google "patent stupidity" for more.)
    1. The process must be fixed and there must be a working solution. Simply complaining is not a good solution.
    2. I do believe the protection value of patents.
    3. Small inventors, like me, particularly need the protection and marketing helps of patents.
    4. I do believe computer software should deserve as much protection as inventions in any other engineering fields. (In particular, it is already hard to make people believing the value of software than the value of a house or a car.) Therefore, I go against not allowing software or business method patents.
    5. There is no simple algorithm to determine inventiveness and obviousness. Any invention will become obvious and trivial to someone once revealed. (The Theory of General Relativity could become trivial to one in the field of physics after the person reads about it.)
    6. So there is no generally accepted and perfect method to determine the inventiveness and non-obviousness of the invention.
    7. Most patents have no real market value. They are just there to trap other people into lawsuits.

    The Proposed Solution: Exponential Patent Fee

    How it works:
    1. You pay a small fee (currently $800, could be $1000) to file for a patent.
    2. If the patent is granted, you get 3 years of protection.
    3. You can extend the protection period at anytime. The patent fee grows exponentially with respect to the number of years of protection. (For example, if the fee doubles for each additionally year and initial fee is $1000, then you pay $2000 for a total of 4 years of protection, $4000 for 5 years, etc.) Obviously, the fee schedule must be determined carefully, but it should be generally exponential.
    4. There is no limit to protection period, as long as the patent holder pays. (Currently, it is 20 years fixed.)
    5. Now it takes several years of waiting to get a patent granted, it should be reduce to no more than one year. (Maybe some patent-pending search algorithm would help.)

    The Philosophy behind the solution:
    1. Let the market decide the value of an invention.
    2. So basically what’s wrong is not the patent process itself per se (even though there should some small fixes to it.) What’s wrong is the protection coverage provided by the patent system. It provides the same protection coverage to stupid trivial patents and the truly inventive useful patents and allows big companies to build patent assets for the goal of filing frivolous lawsuits.
    3. Any invention, whether filed by small investors or big corporations, should get some minimum protection.
    4. Truly useful inventions deserve more protection.
    5. The inventor and the market decide how much the invention is worth it. If a patent can generate a lot of revenue, then the inventor would buy more protection.
    6. Eventually, the patent fee goes up so high that nobody, not even Microsoft, can afford, then the inventor will release it to public domain.
    7. The government can collect more money than it does now with the minimum patent filing fees.
    8. Big corporations can still build patent wall but they must judge the benefits against the costs. Now it costs almost nothing to them. The fee schedule must be designed to minimized patent walls.
    9. Small inventors are protected as well, especially when the filing period is shortened.

    The next steps are to validate this solution and try to push for it, if the people in the congress would listen.) That would require your helps and supports.

    Threaded Messages (33)

  2. A Simple Solution for the Patent Mess[ Go to top ]

    I don't see how this addresses the core problem, at least in the States. USPTO officials grant applications for obvious patents, or patents with prior art, and then it takes years of litigation to sort it out. 'Some patent-pending search algorithm' isn't going to cut it.
  3. A Simple Solution for the Patent Mess[ Go to top ]

    Yes, but one of assumptions (which I agree with) is: "There is no simple algorithm to determine inventiveness and obviousness. Any invention will become obvious and trivial to someone once revealed".

    Therefor, this suggestion atleast minimizes the damage.
  4. Yes, but one of assumptions (which I agree with) is: "There is no simple algorithm to determine inventiveness and obviousness. Any invention will become obvious and trivial to someone once revealed".Therefor, this suggestion atleast minimizes the damage.

    >>2. If the patent is granted, you get 3 years of protection.
    >>3. You can extend the protection period at anytime.

    Is this you call damage minimization?
  5. Yes - damage is minimized[ Go to top ]

    <quote>Is this you call damage minimization?</quote>

    Yes, because, as stated, the expiration date for the patent these days is 20 years or so. With the proposed new system, after a few years, it would no longer be applicable for a company to keep renewing the patent (excluding MS ofcourse ;-)

    I do agree with the other post that perhaps the initial fee should be increased, or perhaps the rate in which it rises.

    Also, the new system puts the pressure on the patenting entity (company/individual) to reep as much benefits from the patent as fast as possible, before it (inevitably) goes into public domain.
  6. What about marketing skills?[ Go to top ]

    You are assuming that a good patent will sell itself thus make money for the patent holder and afford him/her the extended payments. A small inventor might not have the marketing mussle to sell his ideas in 3 years while 2,4 or even 100000 dollars for a big firm woulr be a drop in the bucket. This will probably mean that small inventors who can't market the idea fast will be priced out of their own invention.

    Additionally, it would not even give big companies the incentive to purchase IP rights from the inventor since they can just sit around until the inventor can't affort the payments anymore and market the idea under the public domain clause.
  7. What about marketing skills?[ Go to top ]

    "A small inventor might not have the marketing mussle to sell his ideas in 3 years while 2,4 or even 100000 dollars for a big firm woulr be a drop in the bucket"

    That's true - and I said I definitly agree that perhaps the fees should be different based on the entity type (individual or corporation) and their financial capability.

    what you are arguing about are small parameters that can be twicked and changed - but you are missing the bigger picture. Look how MS tries to patent the FAT system just to make sure Linux goes out of the game. These are foul plays that ara considered legal today (although for now 10x to pubpat the patent was not yet approved). Sure, the proposed system is not perfect - but I think it has more to offer in terms of, well, justice ;-)

    That said, I do agree the parameters need to be modified to make sure the "little man" can still be affored enough breathing space to reap the values of his invention.

    "Additionally, it would not even give big companies the incentive to purchase IP rights from the inventor since they can just sit around until the inventor can't affort the payments anymore and market the idea under the public domain clause."

    I do not agree. Every invention should become public domain after a few years - the amount of years can be argued about, true, but I do not believe the current 20 years timeout is justified. It is just too long - atleast for the software industry. After 20 years most patents (with emphasis on "most") are useless (and don't give me examples about 20 years ago from today because the really big boom started only about 10-15 years ago).

    P.S.
    Off topic - how do you quote another post here so it appears framed? I've tried <quote> but...well...it shows ;-)


    Cheers.
  8. What about marketing skills?[ Go to top ]

    Off topic - how do you quote another post here so it appears framed? I've tried <quote> but...well...it shows ;-)

    Use 'blockquote' instead of 'quote'.
  9. What about marketing skills?[ Go to top ]

    You are right about the marketing skill concern. That's why I tried to emphasize the importance on designing the fee schedule. There could be other measures to give individual or small business more time leverage. But if the inventor cannot successfully market the idea in a few years (which is quite long in today's world,) he should either get help or give up.

    But I have read patents seemingly from individuals or small businesses that look as trivial and inelegant as those filed by Microsoft or IBM. We should discourage those patents no matter who file them.
  10. Yes - damage is minimized[ Go to top ]

    <quote>Is this you call damage minimization?</quote>Yes, because, as stated, the expiration date for the patent these days is 20 years or so. With the proposed new system, after a few years, it would no longer be applicable for a company to keep renewing the patent (excluding MS ofcourse ;-)
    And plenty of others.
    it (inevitably) goes into public domain.
    How come? (see list of exclusions)
    If it was really enevitably in 2 years I would support the idea.

    PS: knowing how is far from enough. I know how to build TV or PC, but I buy them.
  11. You want to patent your idea, implement a working prototype that can be verified.
  12. Lowering the Entry Barrier[ Go to top ]

    You want to patent your idea, implement a working prototype that can be verified.

    Patents were created specifically so that this wouldn't be necessary. Here's a scenario: Someone has an idea for a new medical treatment device. If it weren't for the patent system, some company with lots of money could steal that idea and implement it right away, while the poor sap is stuck and gets nothing. The intent of patents was to give the inventor some breathing room to either get the resources needed to implement it or license the idea to someone else.

    The problem is now the system has evolved so that works for the companies with the most money, and thus the most lawyers. So we need to find a way to fix the system and make it work for the little guy again.
  13. I think that's an excellent idea. You've got my signature! ;-)
  14. I suggest you try to patent this idea. I'm sure it's worth the $800-$1000 it would cost you right off the bat :-)
    Jeez - talk about ridiculous!
  15. I agree with your idea.
    Adding to that, increasing the patent protection fee can or should be used to clear the patent-pending process challenges, waiting long years to get the patents approved.
    I have a question. Is this fee $800 a constant amount for individuals and Corporates or does it vary?
     Why USPTO shouldn't fix the fee based on the financial status and/or based on the value of the patent?
  16. 4. There is no limit to protection period, as long as the patent holder pays. (Currently, it is 20 years fixed.)
    And I like that, I mean 20 year expiration period. Now, for example, anyone can make Trinitron CRTs. If they still can find reasoning for CRTs at all ;-)
  17. Your little guy is now a dipshit creep who specializes in "submarine" paptents : patent the obvious and fail until a company that implement an idea ( not stolen but evolved from its business and technology) and then sue them for millions.

    Remember the creep who sued to protect his "idea" of plugins. Or the one suing everyone for streaming media.

    An idea should legally be construed as nothing until is implemented.
  18. Damn straight! Patents should protect one's ability to generate profit from their invention, and not merely obstruct others in doing so. Some working prototype and maybe proof of attempt to bring a product/service to market should probably be mandatory. Maybe "sleeper" patents should have a much earlier expiration date, i.e. if the patented invention is not made part of a product/service that was brought to market in this shorter timeframe, the patent should automatically expire. The burden should be on the inventor to prove that the patent is "legitimate" i.e. is part of a product/service that is selling (this is not to encourage another set of hefty fees to be imposed onto the inventor).
  19. 5. Now it takes several years of waiting to get a patent granted, it should be reduce to no more than one year. (Maybe some patent-pending search algorithm would help.)

    Huh? Evaluating some of the submissions from IBM et. al. is a lengthy, complex process involving a lot of people. This is utopia.
    4. Truly useful inventions deserve more protection.

    Why? And how is "usefulness" measured? Impossible, of course.
    The government can collect more money than it does now with the minimum patent filing fees.

    Which is a good thing?
  20. 4. Truly useful inventions deserve more protection.
    Why? And how is "usefulness" measured? Impossible, of course.

    "by the market"
    The government can collect more money than it does now with the minimum patent filing fees.
    Which is a good thing?

    Yes! The government does need money! Have you not checked your tax return? And it may give them more incentive to reform. The proceeds can go back to the patent office so they can self survive without a lot fundings.
  21. I too believe that patents should exist to protect small inventors, but the real problem silly patents being awarded, bogged down patent offices without proper resources or competence to evaluate filings etc etc.
    I believe the right solution is rather different and much simpler:
    1. make all patent applications public from the moment they are received, easily searchable by the public.
    2. Let people from the general public (such as researchers in the same area) easily submit objections to patents during an "objection period" of say 6 months, stating prior art and other obvious causes for refusal.
    3. Let the patent office employees evaluate the validity and weight of all objections.

    This would free research resources from the patent offices, instead letting the the general public do the prior art research for them, á la Groklaw.
    Not a perfect solution, but the best I could think of in less than 3 minutes. :)


    4. There is no limit to protection period, as long as the patent holder pays. (Currently, it is 20 years fixed.)

    This kind of defeats the whole purpose of the patent system: patents are given a limited time of protection in exchange for them to be public knowledge(patent records) for others to be able to build on once the protection runs out. Unlimited protection time would probably result in a much slower rate of progrss and intellectual property strategies of companies focusing on hindering competition and progress rather than working on better solutions.
    Utter stupidity if you ask me..
  22. Back to first principles[ Go to top ]

    The patent system largely works for engineering inventions. So this discussion is limited to software, not the patent system in general. First of all, does the patent system differentiate between technical fields when granting patents? If not, that's quite a big hurdle to get over before considering anything else. Then, we might consider first principles: why do we (the people) allow these temporary monopolies (all monopolies are bad, some are just less bad than some of the alternatives, e.g.: no technological advancement). We believe that it offers real inventors incentive to invent more. So, in order to tap into the pool of resources, we need to make it cheap to apply, but filter out the chancers (all companies will take a chance, we make them that way, remember).

    How to filter the chancers? Perhaps the application fee is high, but removed from the first annual payment if the patent is granted. Perhaps the fee would be non-refundable if the application was deemed not to fit the legal requirements, but would be refunded if only rejected due to prior art (it's the patent system's responsibility to identify prior art, not the inventor's). Of course, clear guidelines as to what is patentable, and adequate resources to find searches for prior art are assumed.

    Then, once the patent is granted, how to make sure that it's used for technological advancement, not for monopolistic business practice? I like the idea of an escalating scale of fees, I think that will achieve its goal because it plays directly to the economics of the problem, i.e.: valid but valueless patents are not worth the annual payment after a few years, whereas valid, valuable patents are indicated as such by the repeated renewal of the patent. And if you pay lots for a patent, you make sure you commercialise it. Of course, a hard limit of somewhere between five and ten years is also a necessary tool: a robust implementation of the term "temporary", as it applies to legal monopolies.
    Also, in order to renew a patent, the owner must show a) that there has been an effort to commercialise it, and b) answer to any complaints regarding abuse (this is the implementation of the term "limited" in the defining description of a patent). Any abuse claims can be referred to a court. The law must make competition trump "technical innovation" when there is even a shadow of doubt, so if the claims stand up in court, the patent falls into the public domain.

    PS: We, the people, grant and are ultimately responsible for patents. Therefore the patent application process must be funded from public resources. If the value is not returned in increased taxes from the commercialisation of technical innovation, then the system isn't working.
  23. Back to first principles[ Go to top ]

    The patent system largely works for engineering inventions. So this discussion is limited to software, not the patent system in general. First of all, does the patent system differentiate between technical fields when granting patents?

    Agree, for example, a biotech patent should deserve more time to test its values since it usually take more time to get testing, approval etc. for the implementation. But the general exponential fee schedule should apply to patents of all fields.
  24. To save readers from stupidity like this ...
  25. If you can invent an algorithm to filter them, you will make huge amount of money. ;-)
  26. No Software Patents![ Go to top ]

    An european point of view I agree with:

    http://www.nosoftwarepatents.com/en/m/intro/index.html
  27. 5. There is no simple algorithm to determine inventiveness and obviousness. Any invention will become obvious and trivial to someone once revealed. (The Theory of General Relativity could become trivial to one in the field of physics after the person reads about it.)
    6. So there is no generally accepted and perfect method to determine the inventiveness and non-obviousness of the invention.

    You don't realize that without restrictions regarding inventiveness and obiousness, there will be abstract patents covering vast areas of products and there will be patents that only changes minor things to existing patents/products just to undermine others' businesses. Patents used solely for generating money and not as an advantage to competitors are bad for the economy as a whole.

    In fact, your post makes me wonder what kind of patents you have filed. It would not surprise me if they are quite obvious.

    My suggestion:
    1. Require more inventiveness and make sure there are no prior art.
    2. Limit the patents to two, extendable to three years from the day granted. The software industry is evolving rapidly and the abstraction level is increasing, which means patents that seem specific today will be general tomorrow.
    3. Limit the damages that can be issued in court. If the suing party don't have a product, he hasn't lost any income.
    4. Don't use a jury in court when multiple nationalities are involved, they usually are biased.
    5. Hire more competent people in USPTO. Europe's IP practices are much better in terms of determining inventiveness and obviousness (although they too tend to have to low requirements).
  28. Your suggestion for limiting the timeframe for a software patent is definitly good: 20 years is ridiculous. Limiting the damages in court is also good: today it costs on avg $2M to defend yourself in a US court for a patent violation. As for european IP practices I think you should take a look at this link.
    You will probably change your mind after that. Software patent are still illegal in the EU, FYI.

    That being said, I do not agree with any suggestion for fixing the software patent problem. It is too broken to be fixed. First of all, SW patent are relatively recent: software existed for decades before it became patentable in the US. There was no shortage of innovation in the software industry during this time as far as I know, so the argument that some people take that patents foster innovation is simply ridiculous. Second, copyright is sufficient for IP protection. What takes time and money for the development of a new product is not the "invention" itself, it is the development, debugging, testing and so on. This is protected by copyrights and trade secrets (if you don't want others to see how we do this, don't publish your code). If that wasn't true then the world would use Open Office instead of MS Office, because Open Office would have a perfect file compatibility. Third, software patents are incompatible with open source, which is now a main driver for innovation and standardization.

    There is only one solution to fix the software patent problem: drop it.
  29. .....

    There is only one solution to fix the software patent problem: drop it.

    I think that we may have found a solution.
  30. I think the solution to the patent process lies in turning it against itself in the following way:

    1) If I apply for a patent then I get it almost immediately.

    2) If someone already had the patent, I get fined some amount, and the money goes to the original holder.

    3) If prior art is found then the patent is offered to the holder of the prior art. The problem with this is that licenses obtain from the previous patent owner become invalid. So perhaps a restriction can be placed on the transferred patent that the new owner can only demand licensing from the previous owner.

    4) If earlier prior art is found then the patent is transferred again, and the restriction is that the newest owner can only demand licensing from the previous owners.

    I'd also amend the patent process so that a patent can be owned by multiple people. So if I apply to patent an idea and before it's granted someone applies to patent the same idea then both of us will acquire ownership. This requires that applications that are not yet granted remain a secret. (I don't know if this is a feature of the present system)

    All of this transfers the burden of discovering prior art from the patent office to the applicant, the applicant can of course forgo this but by doing so they open themselves up to a risk.
  31. In addition to some of the good ideas mentioned in the parent posts, how about adding the idea of a limit on the number of patents that a single entity can hold at one time. The limit to be determined through appropriate analysis.

    This would help to level the playing field (i.e. works against the large wealthy "patent producing machine" type companies that file for 1000's of patents per year). Companies and individuals would only patent what they truly believe is worth their investment. A good side effect is that it would lessen the work of the Patent Office.

    Of course, the downside is that it may slow down the introduction of innovation into the marketplace by truly inventive, above board companies/individuals that should legitimately hold more than the maximum allowed number of patents.

    But, as Sun says, "nnovation is something that happens elsewhere." That is, no one entity has a monopoly on innovation. Thus, this would distribute the innovation among more entities. Which would therby, help the "little guy" get back into the game.
  32. Good book to read about patents[ Go to top ]

    You should read "Free Culture" by LAWRENCE LESSIG.
    Can be found here.
    It makes an interesting read, and the mess clears up a bit (you understand that you can't compete with the mega-big bucks of corporate america).

    Peace,
    Edward
  33. Software is special[ Go to top ]

    Software is special because it is infinitely composable. Perhaps more so than in any other field, today's inventions in software become the building blocks of tomorrow's inventions.

    The US Constitution makes clear that the purpose of patent and copyright law is ultimately to benefit the *public*--yet holding a patent on a particular algorithm may
    actually deprive the public of a new breakthrough, or push them overseas where there are no such restrictions. For example, if the patents on public-key crypto had expired sooner, SSH might have been built in America instead of Finland, and our e-mail clients might have more tightly-integrated encryption.

    Therefore the need to reward and encourage invention must be balanced against the public benefit provided, and perhaps a shorter time period (5-10 years?) in software would be warranted.

    Also, several problems with the patent system are, to some extent, a product of the US tort system, which is in bad need of reform in general. "Loser pays" laws like most of the rest of the world has would discourage frivolous lawsuits, such as when there is obvious prior art on a patent.
  34. Software is special[ Go to top ]

    Software is special because it is infinitely composable. Perhaps more so than in any other field, today's inventions in software become the building blocks of tomorrow's inventions.
    Software is not special, the same is true for music, art, mathematics, physics, everything!. All living creatures and humans among them advance by copying peer's behavior and improve upon past achievements.
    Patens work against human nature and biological laws, so they have to be dropped.