Holding back innovation

Holding back innovation

Summary: This post on Techdirt does a great job of summarizing the problem with patents: The patent system isn't designed to "protect the little guy." It's designed to promote innovation -- and that's what it needs to be judged on.

SHARE:
TOPICS: Patents
36

This post on Techdirt does a great job of summarizing the problem with patents:

The patent system isn't designed to "protect the little guy." It's designed to promote innovation -- and that's what it needs to be judged on. Patents may make some sense in cases where a concept is truly unique and non-obvious -- but if others are coming up with the idea independently and are better able to bring it to market, then the patent holder is holding back innovation. The other companies didn't "steal" the idea, because they came up with it independently (suggesting that it wasn't unique enough to deserve patent protection anyway).

Note that the post is actually about patents in general, but I'm specifically interested in software patents. If the patent system is in place to enourage innovation, it's failing miserably.

If someone can come up with a system for software patents that truly encourages innovation, I'd like to hear about it. What we have now, a system that stifles innovation at every turn, and where the patent application volume overwhelms the patent office, needs to be dealt with.

Topic: Patents

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

36 comments
Log in or register to join the discussion
  • Re: Software patents

    This is a topic which in my opinion will for a long time spark very serious debate. I tend to agree that the immense proliferation of software patents has the potential to bring the world as we know it to its knees. Serious strides have been made in sofware innovation over the past few decades, this however is being hampered by the fact that a "budding" software architect is being hamstringed by developments within large corporations/'development teams', which may lead to said architect being challenged on the basis of patents which are unknown to him/her. This leads IMHO to larger "teams" (or a person with a string of patents) having the power to basically stifle the efforts of "budding" deveopers. In other words, the individual software architect may have an excellent idea for implementation, however he/she may not bother with the attempt and subsequently only produce run-of-the-mill works, or no works at all based on the fear that his/her innovation may be taken away on the basis of patent(s) held by third parties. This I think is aided and abbetted by the proliferation of the internet (less actual brainwork done by a lot of individuals ... google it), which exacerbates the tendencies of us humans to (in general) want to be spoonfed.
    Now I am not advocating a free-for-all on patents/technologies, however I think a new (maybe radical) approach has to be taken in current patent practices/policies.

    Any suggestions ?
    IanX
    • Re- Sotfware patents

      Actually on second thought a free-for-all system may not be such a bad idea. I know I may be shot down for this, however products sell based on marketing AND innovation.
      What if the marketing aspect could be separated from the innovation? In other words what if the advertising industry could be "radically" revamped to take into account the actual products as opposed to the products with the greatest financial backing?
      Nigh on impossible with the current scheme of things, however "what if?"
      IanX
  • The US acts as a Banana Republic..

    ...in what the patent system is concerned.
    Rubber stamping patents (few hours of examination) so that "details" are dealt with in court by the inventors/companies. The European Patent Office is a great example to follow for they invest heavilly on computer tools and examiners to do serious and deep search.
    The US is more advanced in technology despite its poor patent system because venture capital system and entrepeneurship mentality is decades ahead of the Europeans.
    Conclusion: stop spending money on lawyers and invest it instead on engineering!
    P.S.: and think again about the US ridicule software and business methods patents...
    Francisco Reis
    • What's the cost for European patents?

      [i]The European Patent Office is a great example to follow for they invest heavilly on computer tools and examiners to do serious and deep search.[/i]

      How much does it cost to apply (and get) a patent from the European Patent Office? Is it comparable to the US? Note that the cost is two fold: what the applicant pays and what the government subsidy is.

      If the cost is comparable (or within an acceptable range) then the US should get with the times. If the cost is significantly higher such that the small, independent researcher is barred from entry then there is a problem.
      __howard__
      • Cost differences between the US and Europe

        The filing costs at the European patent Office are in line with the USPTO but there are huge extra costs when translating the patent into the many Europen languages.
        The two big advantages of the EPO are:
        1) the very good global databases and search tools.
        2) the highly qualified staff (scientifically and culturally, speaking many languages) which stay for a lifetime at the EPO (which turn them into real experts in their fields).

        This last point is due to the very high salaries
        at the EPO and that in the USPTO as soon as the patent examiner gains some expertise they resign to work for patent attorneys offices where the big money is!
        Francisco Reis
        • Sounds like congress should look at EPO for USPTO reforms

          [i]filing costs at the European patent Office are in line with the USPTO[/i]

          The filing costs are certainly one part of the total cost -- and certainly the important one to the patent filer. It's probably impossible (okay, too difficult to be worth the effort) to determine what the government subsidy is for the patent office per patent.

          [i]huge extra costs when translating the patent into the many Europen languages[/i]

          For now the U.S. doesn't have to worry about that ... only translating ballots and things :)

          [i]the highly qualified staff (scientifically and culturally, speaking many languages) which stay for a lifetime at the EPO[/i]

          I completely agree that this is a major problem with the USPTO. You have under paid civil servants where the good ones gain experience and leave for private industry and the mediocre to bad ones are almost guarenteed a job for life.

          I've been told -- but do not know -- that the internal reward/recognition system within the USPTO is geared towards encouraging patent approval. I can see an award system based on number of applications processed, but only if it does not take into consideration the disposition of the patent request.
          __howard__
          • What about a World Patent Office...

            ...when countries agree on the real purpose of patents!
            Many US companies, like IBM, do some of their first filings of an invention at the EPO. This is usually when they really want to know if it is a TRUE invention. Unfortunately many of the huge companies file hundreds of worthless patents at the USPTO only to keep competition away.
            The printers field is one of the examples of HP, CANON, EPSON and LEXMARK coming to agreements among them on their hundreds of patents in the field.
            My conclusion: the patent system everywhere really deserves a re-thinking.
            Francisco Reis
          • World Patent Office would be political

            Given how everything touched by the U.N. eventually turns into a politcal football, I think we're better off with the system we have now.

            The USPTO should clean their house. But I'm very much against throwing the baby out with the bath water.
            __howard__
  • Real research like RSA should be patentable

    The problems with the USPTO are several fold, but the general ignorance of software by the patent clerks is clearly exacerbating the problem. Patents are not supposed to be issued for obvious inventions or obvious extensions to existing inventions. Since nothing in software is obvious to those that are ignorant of software, too many obvious software ideas are issued patents.

    There are times where software patents should be issued. There was a significant amount of time and research spent to come up with the non-obvious RSA encryption algorithms. Granting them a patent was a fair and just thing to do.

    Sending advertisements over an RSS feed is an obvious extension to RSS and Google should not be granted a patent for it.
    __howard__
  • Software Patents

    Software per se should not be patenable. If one writes an application (program) that contains unique code and process, they should keep the process a trade secret. The end product of the process can be replicated through other means or code and should never be the object of the patent. All software patents should be revisited and examined as to whether the real patent is on the resulting end product or the code itself. If the patent is on the end product such as "One Click" it should be disallowed. This falls into the look and feel domain and language compilers replication.
    SamBirnbaum
    • Software given to customers can not be protected by trade secret

      [i]If one writes an application (program) that contains unique code and process, they should keep the process a trade secret.[/i]

      That isn't feasible. Any software you deliver to a customer can be reverse engineered and then your trade secret is out in the open.

      Furthermore, if you take something like the RSA encryption, you have to expose part of the trade secret in order to get it to function. You certainly would have to in order to get people to trust it.

      [i]If the patent is on the end product such as "One Click" it should be disallowed[/i]

      I disagree. If the patent is on an [b]obvious[/b] end product like "one click" it should be disllowed. If it is on something radically different or non-obvious it should be allowed.

      Using my RSA encryption analogy. RSA encryption was not obvious at the time and took brains and money to come up with. Once explained, it could be readily copied. Therefore it should be patentable.

      On the other hand, once DES is out in the public domain, applying DES 3 times with different keys to produce the encrypted results is an obvious extention of DES -- and should not be patentable.
      __howard__
      • I suppose DRM could protect trade secrets

        Okay, in thinking about it, if there was a really strong Digital Rights Managment (DRM) in place, then it would be possible for software to be protected by trade secrets, but only then -- and only until the DRM is cracked.

        Frankly, I wouldn't want my computer to turn into just one giant DRM system.
        __howard__
  • Completely, BOGUS claim

    Look around you, the tech industry is NOT lacking innovation at all. In fact you can't pick up a trade mag without there being dozens of new products, both hardware and software. The fact is innovation is happening so fast it's anyone can do to try and keep on top of it.

    No, innovation is alive and doing just fine, except of course unless you are talking about open source being able to continue copying everyone...
    No_Ax_to_Grind
    • You're right, but there is a problem.

      I agree with your basic argument that there is plenty of innovation happening -- and that patents do (properly) hamstring people just trying to copy other people's product. The Chicken Little "no more innovation" claims are an exteme.

      However, there is a problem with software patents in that [i]obvious[/i] things are being patented; and this needs to be fixed. I don't believe new laws need to be enacted; but the patent office needs to get educated on what is and isn't obvious (i.e., hire software people) and stop issuing patents to obvious software inventions.
      __howard__
      • I'll go you one better.

        Congress needs to allow the patent office to use the fees they collect to hire knowledgable people and straighten out the mess.
        No_Ax_to_Grind
        • Re: I'll go you one better.

          [i]Congress needs to allow the patent office to use the fees they collect to hire knowledgable people and straighten out the mess.[/i]

          If Congress allowed Social Security to keep the deduction it takes from every worker the agency would be $billions in the black and there wouldn't be an issue with SS running out of money. But it can't - Congress is simply unable to restrain itself, generally.

          What Congress needs to do is speak up on an issue its remained silent about and direct the USPTO not to issue software patents.


          .
          none none
          • Tossing the baby with the bath water

            [i]If Congress allowed Social Security to keep the deduction it takes from every worker the agency would be $billions in the black and there wouldn't be an issue with SS running out of money. But it can't - Congress is simply unable to restrain itself, generally.[/i]

            Yes and no. Social Security was a Ponzi scheme from the start. Of course it helped that when started the retirement age was 65 and life expectancy was 63. And don't forget, there never has been (and probably never will be) a "lock box" that holds the funds collected by the Social Security tax.

            Though you're point about keeping Congress' hands off it would help. But it won't happen.

            [i]What Congress needs to do is speak up on an issue its remained silent about and direct the USPTO not to issue software patents.[/i]

            That would be throwing the baby out with the bath water. Probably 90% or more of software patents are for intuitively obvious things, but there are a few things that can be implemented in software that should be patentable (e.g., RSA encryption).
            __howard__
          • Defining the obvious.

            Sometimes professionals look at an idea and say, "Well, of course." That doesn't make the idea obvious. An innovation can be just an extension of already known ideas; most innovations are just that.

            Sometimes people advocating professionalism in patent evaluations expect that such a system would issue almost no patents. Substantial numbers should be issued because anvances can, to use No_Ax's term, appear to be evolutionary.
            Anton Philidor
          • Re: Defining the obvious.

            [i]Substantial numbers should be issued because anvances can, to use No_Ax's term, appear to be evolutionary.[/i]

            Usually this is called product differentiation. When all the razors have two blades, differentiate yours by adding a third. Should that be patentable? I guess the answer depends on whether the third blade would have been added anyway as a result of market forces, or whether only the promise of a monopoly would incent the investment in marketing, er... R&D.

            Because there's one inescapable fact: patents exist to benefit the public, not the inventor. If market forces and product differentiation would be sufficient to impel Microsoft to develop "Creating a note related to a phone call," or any other of it's 3,000 "inventions" this year then patenting these ideas is not in the public interest.

            We'll never know, though, because anything that's not nailed down is patentable. Thats the problem. We [i]do[/i] need fewer patents.


            :) <-- [url=http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PG01&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.html&r=1&f=G&l=50&s1=%2220050156873%22.PGNR.&OS=DN/20050156873&RS=DN/20050156873]Patent Pending[/url]


            .
            .
            none none
          • Escaping the inescapable.

            The principle behind patents is that benefitting the inventor benefits the public.

            The two main benefits are publication of the idea for others to use and the monetary incentive to invent. These are closely linked.

            You wrote:
            Because there's one inescapable fact: patents exist to benefit the public, not the inventor.


            The language establishing patents speaks of the "useful arts and sciences." Those are subjects for the inventor. The public benefits when something improving life is eventually provided.
            Again, benefitting the inventor benefits the public.
            Anton Philidor