Investigating The Obviousness Of Software Patents

from the can-you-look-at-obviousness? dept

In many of our discussions about the patent system, one of the big debates between those who think the system is just dandy and those who have problems with it concerns the question of "obviousness." One of the stated tests for whether or not a patent should be granted is whether it is non-obvious to a person who is skilled in the art. However, the way the patent office currently looks at things, they only look for prior art to prove obviousness. In most cases, prior art either means earlier patents or published journals. There can be plenty of other prior art, which is a problem, but it's worth examining whether or not prior art is actually needed for obviousness. Just because something hasn't been done before, it doesn't mean that it's not the next "obvious" progression in the art. However, those who fight against this view (often patent lawyers) claim that it's impossible to have any test for obviousness because those who are "skilled in the art" will think anything is obvious in retrospect (perhaps out of professional pride for not having thought of it first) and because it's impossible to define obviousness for someone skilled in the art. The first claim seems to take a dim view on human nature. The second raises questions about all sorts of other similar tests in the law (how do you define "reasonable" in "reasonable person" tests, for example).

So, is it possible to look at the obviousness of a patent? Tim Lee, over at Tech Liberation Front, is starting a new project each week to examine a different software patent. This week, he's looking at the flight simulator patent that was used to try to stop the distribution of Google Earth. Lee points out how the patent meets the qualifications of obviousness, while also showing just how broad made of the claims are -- going well beyond protecting a single invention. It seems like a good demonstration of how someone can explore the obviousness of a patent, without necessarily delving into the specific prior art -- and it certainly does not appear to be a case of Tim's "professional jealousy" trying to invalidate the patent. Rather, he points out that the various claims are fairly obvious, and just about any decent programmer would use the same approach in creating a similar flight simulator system. So, even if this specific solution had not been done before, anyone attempting to tackle a flight simulator program would likely do the same thing. In that case, what benefit is there to granting a monopoly to just one player?


Reader Comments (rss)

(Flattened / Threaded)

  •  
    identicon
    Thomason, Jun 16th, 2006 @ 11:03am

    Obvious, please

    Is there any software patent that does not use programming steps already known to programmers? If the patent law didn't require inventions to be non-obvious, then would not every combination of these known programming steps be arguably an obvious combination?
    Back in the old days, when memory and computing power were more limited, the real innovators could write software that was amazingly more efficient than what hacks could write. It's comparable to prose writers, who can say in a sentence, what others run on about for paragraphs.
    That's where I suppose Mike and I differ about what defines a patentable invention. Mike wants it to be earth-shattering, or pioneering, or previously-unthinkable. I prefer to award exclusive rights to those who make things better, more efficient, wiser, cheaper, easier, more-troublefree - the difference between a work of inventive genius, and something ordinary and workmanlike can be seen and has to be appreciated. Give that man a patent!!

    P.S. I agree that patents on "concepts" that might be implemented in software are a bad thing, as are all patents or copyrights on "concepts" or unimplemented "ideas."

     

    reply to this | link to this | view in chronology ]

    •  
      identicon
      mthorn, Jun 16th, 2006 @ 11:49am

      Re: Obvious, please

      Patenting a specific implementation of something can be innovative. Patenting a broad concept reguardless of how obvios it is or not should not be allow. You should have to be specific in what you are patenting and you should be required to produce something for the patent within a year or you will lose it.

       

      reply to this | link to this | view in chronology ]

    •  
      identicon
      Anonymous Coward, Jun 16th, 2006 @ 2:03pm

      Re: Obvious, please


      Is there any software patent that does not use programming steps already known to programmers?


      Is there any book written that doesn't already use words and grammatical rules already known to authors? Does that mean that there is no "innovation" in writing anymore?
      I realize we're talking about the difference between 'copyright' vs 'patent' but in terms of code, I think the post regarding 'follow the money' hits the nail on the head.

       

      reply to this | link to this | view in chronology ]

  •  
    identicon
    Tim, Jun 16th, 2006 @ 11:25am

    And you think a particularly elegant hack can be described using page after page of legalese?

    That guy already have protection: it's called copyright law. If you copy his code, he can sue you for infringement. How does patent law offer any additional protections?

     

    reply to this | link to this | view in chronology ]

    •  
      identicon
      Anonymous Coward, Jun 16th, 2006 @ 11:43am

      Re:

      No additional protections unless you count the amount of money that you can sue for.

      The amount received from copyright infringment is small compared to what can be received from patent violation.

       

      reply to this | link to this | view in chronology ]

  •  
    identicon
    anonymous coward, Jun 16th, 2006 @ 12:01pm

    The obviousness of the solution and the PTO's inaction shows the PTO's inability to understand the obviousness of anything...

    Standing next to a bucket of water with your show on fire? I could get a patent for the process of inserting your foot into the bucket to extinguish the flames.

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    Mitch the Bitch, Jun 16th, 2006 @ 12:12pm

    Immersion Technology Force Feedback

    A few years ago we had competition for different Force Feedback devices from different companies, MS comes to mind. Joysticks Steering wheels medical devices etc.

    Since that time Immersion Technology out of San Jose has won all of it's patent lawsuits and has pretty much ruined any innovation from devices that provide tactile feedback to ANY peripjeral connected to ANY computing device.

    Ive begged and pleaded anyone to fight this nonsensical patent but nobody ever seems to even care or try.

    As an ordinary citizen what can we do other than pay some scumbag lawyer? Why do WE (the public)have to LOSE out on innovation because of one companies greed?

    Does ANYONE actually believe Immersion deserves such a "BROAD PATENT" (other than lawyers)?

    This is just one example but it affects every gamer.

    Mitch

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    PATENT PUNK, Jun 16th, 2006 @ 12:18pm

    TO GRANT OR NOT TO GRANT...

    "A patent may be granted for a process, act, or method that is new, useful, and not obvious, for a new use of a known process, machine, or composition of matter or material, as well as for an asexually reproduced distinct and new variety of plant (excluding one propagated from a tuber), and for any new, original, and ornamental design for an article of manufacture."

    This is straight from the USPTO.

    It seems to me that something is clrealy missing and that is scope. What is the scope of any invention?

    If I patent a wire that is bent a certain way and is used to remove hardboiled eggs from boiling water does that include all wires bent in that manner even if they are used to say light pilot lights in gas boilers? Which is more novel an idea the bent wire or a new use for it?

    Do I, as the inventor, have the right to pursue the maker of the wire for pilot lights?

    I am convinced that some of this art is applied to too broad a scope. Patents should be application specific, if it falls in a grey area that is what the USPTO is for. But it seems that the USPTO grants a bent wire patent regardless of application and herein lies the problem.

     

    reply to this | link to this | view in chronology ]

    •  
      identicon
      Joe Smith, Jun 16th, 2006 @ 3:19pm

      Re: TO GRANT OR NOT TO GRANT...

      "A patent may be granted for a process, act, or method that is new, useful, and not obvious,..."

      It seems to me that since something must be both "new" and "not obvious" those two expressions must be intended to refer to different concepts. "New" is easy - if it is contained in the existing literature it is not new. But being "new" is not enough - that is by the ordinary construction of the language it is not enough that the idea cannot be found in the existing literature - the idea must meet a further test of not being obvious. The fact that deciding obviousness may be difficult is no reason for ignoring the wording of the law. It seems to me that any idea which will inevitably be arrived at by at least one skilled person within a reasonable time if a number of skilled people turn their minds to the problem is "obvious".

       

      reply to this | link to this | view in chronology ]

    •  
      identicon
      Anonymous Coward, Jun 16th, 2006 @ 9:04pm

      Re: TO GRANT OR NOT TO GRANT...

      To be honest, it all depends on how the wire is claimed. Taking a flip through patents you'll eventually run into inventions that are essentially that same but are claimed in totally different ways so they aren't the "same invention."

      In the wire example, it still depends. If it's claimed as a product, then probably rejected. If claimed as a method, then might be allowable. Never know. Again, it mostly depends on the claims. They can make or break an invention.

       

      reply to this | link to this | view in chronology ]

  •  
    identicon
    Giving away the bank., Jun 16th, 2006 @ 12:31pm

    All it would take...

    I know how to end all of this silliness. If one person could get the "business process" patent for "using the patent system to create numerous patents in the eventuality that someone violates them and reep the windfall of extroting them" (using proper leaglese, of course!) then anyone who does that in the future would be infringing on THAT patent.

    I'm looking for investors to get this business off the ground...

     

    reply to this | link to this | view in chronology ]

    •  
      identicon
      ben, Jun 16th, 2006 @ 2:49pm

      Re: All it would take...

      using the patent system to create numerous patents in the eventuality that someone violates them and reep the windfall of extroting them


      That's it! $$$ is on the way

       

      reply to this | link to this | view in chronology ]

  •  
    identicon
    az1324, Jun 16th, 2006 @ 2:23pm

    Now let's get all those Amazon patents re-examined

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    Just a programmer, Jun 16th, 2006 @ 3:12pm

    Patent a language, copyright the program

    Seems like a quick fix would be that you can patent a software language, but anything produced by the language should simply be a copyright.

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    Mike Mixer, Jun 16th, 2006 @ 8:10pm

    What I wouldn't give

    What a wonderful thing it would be to own a time machine and go back to say 1968 and apply for patents on GUI and mouse pointers and folder pictures. You get the idea. Hoard all of them in your attic somewhere then burst out in 2003 and shut down microshit and crapple or maybe license out to a young swede of our acquaintance back in the 90's

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    Seth Noble, Jun 19th, 2006 @ 8:14am

    Retroactive Obviousness

    Saying that experts will find a good invention obvious in retrospect has nothing to do with jealosy: it is practically the definition of a good invention. The biggest barriers to any scientific or technological breakthrough is knowing what is possible and what is practical. Once you know that something HAS been done, the fact that it CAN be done is obvious.

    So the test of obviousness is not "is it obvious NOW", its "was it obvious THEN". The only way to know that is to look at the literature from around the time of invention. Unfortunately, such background research rarely happens, and that's where we start to get into trouble.

     

    reply to this | link to this | view in chronology ]

  •  
    identicon
    Wizard Prang, Jun 19th, 2006 @ 8:28am

    Software should not be patentable!

    Patents are for protecting inventions. oftware is, by definition, something that is designed and written, not invented.

    The only exception that I can think of is where software is an expression of an invention - such as encryption, which can be expressed as an engineering process. In that case it is the process that is patentable, but the software implementation.

    Software=Copyright
    Inventions=Patents

    An questions?

     

    reply to this | link to this | view in chronology ]


Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here
Get Techdirt’s Daily Email
Save me a cookie
  • Note: A CRLF will be replaced by a break tag (<br>), all other allowable HTML will remain intact
  • Allowed HTML Tags: <b> <i> <a> <em> <br> <strong> <blockquote> <hr> <tt>
Follow Techdirt
A word from our sponsors...
Essential Reading
Techdirt Reading List
Techdirt Insider Chat
A word from our sponsors...
Recent Stories
A word from our sponsors...

Close

Email This