Shouldn't The Patent Office Be Able To Reject A Bad Patent Application For Real?

from the just-a-thought dept

A few people have been submitting various versions of the story about how online gaming company OnLive is claiming a patent on its system for cloud-based video games. You can see the actual patent (7,849,491), if you'd like to dig into the details.

However, rather than dig into the specifics on this particular patent, I did want to dig into one somewhat offhand comment that Dean Takahashi made in his writeup of this patent over at VentureBeat, where he gets in a little dig at the Patent Office:
it was originally filed in December, 2002... The patent was granted last week, showing just how far behind the patent office is in acknowledging inventions.
Similarly, in the WSJ article where OnLive's Steve Perlman claims this is a "pivotal patent," he also complains about how slow the patent office is, and how annoying it was that he had to wait so long to get this patent.

Now, I'm not known for standing up for the Patent Office on much, and everyone seems to agree that the USPTO has a huge backlog that it takes them a while to get through, but the implication here is that if you just submit an application, you'll have to wait 8 years to get that patent approved. That's not really fair or accurate. I went and looked at the history of this particular patent and it's a lot more complicated than that. First of all, while it was initially filed in 2002, the application wasn't actually complete until April of 2003. Then, it appears that a whole bunch of additional information was filed over the years -- which is perfectly fine (as new things come out, you're supposed to file such information disclosure statements), but each of those need to be reviewed as well.

But what's a lot more telling is that the patent was "rejected" by the patent examiner not once, not twice, not three times, not four times, but five times before it was finally approved. However, the way the patent system works is that there's no such thing as a real rejection of a patent application. Even if there is something called a "final rejection" (the second rejection here was officially a "final rejection"), that's clearly misnamed, as the applicant is able to keep requesting new examinations, perhaps with adjustments to the patent, or after the applicant (or, more likely, a patent attorney they hired) argues that the patent should be approved.

So, perhaps it's not just that the patent office is "behind" on "acknowledging inventions," it's that patent applicants can just keep trying and trying and trying over and over and over again until they finally convince the examiner to approve the patent. Of course, all of this does contribute to the actual backlog -- because each time the examiner has to deal with a "rejected" application request for another examination, and hear them out on why it should be approved, or review the changes, that's time that could be spent reviewing a different patent application.

What I still don't understand is why we still let the system work this way. I can understand being able to make your case as to why a patent should be approved after an examiner initially rejects it -- and perhaps potentially an opportunity for an appeal -- but why do we let the system be so open ended? And, to be clear, I'm not saying if this particular patent is a "bad patent." I really don't know (though, I assume some of you may have some opinions on that). I'm just pointing out that, contrary to the claim in that article, the delay here might not have just been about a slow patent office, but about the fact that the patent was "rejected" so many times, and each time OnLive was able to come back and get another crack at the apple...


Reader Comments (rss)

(Flattened / Threaded)

  •  
    identicon
    Anonymous Coward, Dec 16th, 2010 @ 11:44am

    FTFY

    "crack at Apple...and Microsoft...and Google...and Blizzard...and..."

     

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

  •  
    icon
    Designerfx (profile), Dec 16th, 2010 @ 11:46am

    simple solution

    don't allow patent resubmittals after the second rejection, period.

     

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

  •  
    identicon
    Cowardly Anon, Dec 16th, 2010 @ 11:50am

    Yes. It's a bad patent.

    It follows your favorite equation: old idea + new technology = get a new patent.

    It's depressing actually.

     

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

  •  
    identicon
    J.D., Dec 16th, 2010 @ 11:52am

    This got me thinking. At which point then is something considered prior art? Is it when the application was initially filed or when it was approved?

     

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

    •  
      icon
      Marcus Carab (profile), Dec 16th, 2010 @ 11:57am

      Re:

      At which point then is something considered prior art?

      I believe it usually dates from the filing, not the approval. However a much bigger problem is that the patent office seems to have very little interest in actually looking for prior art - many patents have been passed even though a quick Google search of the concept reveals busloads of prior art.

       

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

      •  
        icon
        Marcus Carab (profile), Dec 16th, 2010 @ 11:59am

        Re: Re:

        very little interest

        Actually I should be more fair - very little time is probably the real issue. I don't know enough about the innerworkings to say what's actually causing the problems - but it certainly seems that nobody is bothering to search for prior art.

         

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

        •  
          icon
          crade (profile), Dec 16th, 2010 @ 12:05pm

          Re: Re: Re:

          "Actually I should be more fair - very little time is probably the real issue"
          So if we don't have enough time to review a patent is valid or not it just defaults to passing? Seems like the real issue would be policy then.

           

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

          •  
            icon
            Marcus Carab (profile), Dec 16th, 2010 @ 12:37pm

            Re: Re: Re: Re:

            Yeah, for sure. I'm not trying to defend the system - it's clearly broken. I just didn't like my initial implication that the problem is the ineptitude or apathy of the patent officers themselves (though maybe it is - I really don't know)

             

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

          •  
            identicon
            Anonymous Coward, Dec 16th, 2010 @ 3:16pm

            Re: Re: Re: Re:

            Well, the applicant has a duty to disclose prior art as well. If they don't, the patent may end up being unenforceable (although that would require a showing of willful deception or witholding).

             

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

            •  
              identicon
              Anonymous Coward, Dec 16th, 2010 @ 10:25pm

              Re: Re: Re: Re: Re:

              It would also require a lawsuit that many aren't willing to bother with due to expense.

               

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

      •  
        identicon
        Anonymous Coward, Dec 16th, 2010 @ 1:18pm

        Re: Re:

        That which serves as prior art is determined as of the date that an application is filed. Importantly, it is not limited solely to publications.

        It is easy to understand people wondering about the scope of searches and sources relied upon for conducting searches. One reason for the confusion is that searches are conducted in association with the claims being presented in the application. There may be "tons" of information that pertains in general to the subject matter area of an invention, but those "tons" invariably reduce down to mere "ounces" when it is realized that the claims are what drives the search for prior art.

        Are searches perfect? Of course not. It is impossible for anyone to locate throughout the world each and every piece of prior art that is actually relevant and important for judging whether or not a claim meets the legal criteria for its allowance. For example, just how likely is it that either an examiner or an applicant is aware that a PhD thesis that is a "claim killer" is catalogued and available to the relevant public in the library of a university in Latvia? How likely is it that either of them are aware of a similar "claim killer" that resides in a classified vault at a DOD R&D facility and relating to a TOP SECRET DOD program?

         

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

        •  
          icon
          Marcus Carab (profile), Dec 16th, 2010 @ 1:46pm

          Re: Re: Re:

          What you say certainly makes sense in some situations, but it seems like patents get approved in way more ridiculous circumstances. I'm not talking about prior art in the form of some obscure thesis or secure document, nor am I talking about entirely general information about the field.

          I'm talking about cases where even the specific claims in the patent are *clearly* preempted by widely available examples of prior art... Techdirt has a few recent examples, like the patent on online music distribution or the one on putting a press release online.

          And that's just the tip of the iceberg... in the field of software patents especially, it certainly seems like patents get granted for inventions where there is not only prior art, but in fact widespread and common usage of that prior art by countless parties.

          A great example is the double linked list patent, where the claimed invention was a basic concept that had already been fundamental to computer programming for more than a decade. The patent was granted despite the fact that it is not novel in any way whatsoever.

           

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

  •  
    icon
    Michael (profile), Dec 16th, 2010 @ 11:56am

    Public Patent

    Whenever a patent is rejected, it should be converted into a new patent owned by no-one, similarly to how the Public Domain works.

    It would effectively un-patent that concept. For good?

     

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

    •  
      icon
      Michael (profile), Dec 16th, 2010 @ 12:42pm

      Re: Public Patent

      I mean, in the same way that you shouldn't be able to patent something that is already patented, couldn't this make a faster or simpler way of checking if someone is submitting repetitively?

       

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

  •  
    icon
    crade (profile), Dec 16th, 2010 @ 11:59am

    Cloud computing is just a new name for dumb terminals.

     

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

  •  
    identicon
    Patrick, Dec 16th, 2010 @ 12:17pm

    I am not sure this would fix anything

    I am not sure adding a true "Final Rejection" would actually solve anything. As long as companies believe there is an economic incentive to hold patents they will do their best to game the system. Now they push the system by filing appeals and requests for continued examination (which can be expensive to do frequently).

    If a true "Final rejection" was issued I think patent strategy would just change a bit. I think you would see many more patents filed with narrower and possibly overlapping claims. If one gets rejected so what? We have 3 others. Also what is to stop someone, one they receive a "Final Rejection" to just remove or change one claim and then refile as a "new" patent application?

     

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

    •  
      icon
      jjmsan (profile), Dec 16th, 2010 @ 12:25pm

      Re: I am not sure this would fix anything

      They could do that, but the date of filing would remove the " it took eight years for this to be approved" argument. Plus any prior art that showed up in the time that the initial patent was approved would be revelant

       

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

      •  
        icon
        Chronno S. Trigger (profile), Dec 16th, 2010 @ 12:32pm

        Re: Re: I am not sure this would fix anything

        I like that idea. Let's do it. A final rejection is a finale rejection, if you want to try again you have to start the entire process all over again, fees and all.

         

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

        •  
          icon
          Free Capitalist (profile), Dec 16th, 2010 @ 12:37pm

          Re: Re: Re: I am not sure this would fix anything

          fees and all.

          Ding ding!

          And increase the fees relative to the sum of the all statutory awards granted to applicant.

           

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

          •  
            identicon
            Anonymous Coward, Dec 16th, 2010 @ 2:01pm

            Re: Re: Re: Re: I am not sure this would fix anything

            That wouldn't help the individual inventor of something truly revolutionary who spent all of his money creating it.

             

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

            •  
              icon
              Chuck Norris' Enemy (deceased) (profile), Dec 16th, 2010 @ 2:17pm

              Re: Re: Re: Re: Re: I am not sure this would fix anything

              truly revolutionary

              It wasn't 'truly revolutionary' if the patent office rejects it...twice! Especially with how low the bar is set for patent approval these days.

               

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

  •  
    identicon
    Anonymous Coward, Dec 16th, 2010 @ 1:04pm

    It would have been appropriate to note that the original application was filed in 2002, but the first Office Action for the application was not issued until 2008.

    BTW, Office Actions cover a wide range of issues. Claims can be rejected for any number of reasons having nothing to do with the prior art (non-statutory subject matter, restriction requirement, lack of enablement, etc.). For example, in this case one Office Action was principally directed to the sufficiency of the disclosure contained in the "specification" of the application (what is known as a "112" rejection).

    After having looked over the file history of this application, I can well understand why the patentee talked about the delay from filing to issue. An application should not have to wait six (6) years before it is taken up for the first time. Clearly, the delay in this case was at the USPTO's end, and not that of the applicant (now patentee).

     

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

    •  
      identicon
      Anonymous Coward, Dec 16th, 2010 @ 1:33pm

      Re:

      It is so refreshing when we get someone in the comments that actually knows what they are talking about. Thank you.

       

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

  •  
    identicon
    Anonymous Coward, Dec 16th, 2010 @ 1:08pm

    How do you find how many times it was rejected?

     

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

  •  
    identicon
    Anonymous Coward, Dec 16th, 2010 @ 1:58pm

    Exponentially increasing the fees with each rejection should have an interesting effect on the patent system. But I have no idea if it would be a net positive or a net loss for the public. On one hand it would either induce people to abondon bad claims. On the other hand it would induce them to charge huge licencing fees once a petent was granted. On the gripping hand it would give the patent office a financial incentive to reject patent claims.

     

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

  •  
    identicon
    Anonymous Coward, Dec 17th, 2010 @ 12:54am

    I have a dozen dated audiotapes of discussing this exact concept in excruciating detail with 2 of my IT type friends in 1995. Does this mean that if I apply enough times I can get a patent to beat these nimrods over the head with?

     

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

  •  
    identicon
    Darryl, Dec 17th, 2010 @ 2:54am

    There is no limitations on the number of times you can lodge a patent. That would be crazy.

    Of course you should be able to lodge a patent application any amount of times, I can only see a valid reason for rejecting a patents, if there is allready a patent on that, or its obvious, or its prior knowledge.

    Where does it say anywhere, that you cannot lodge any number of patents.

    And of course, again, it does not matter how many times it was rejected, and im sure you understand that each application would be different, or written, worded in a better way to convey the invention.

    Once they got the way of explaining their invention in terms the patent office requires, the PTO then issues them a patent.

    It would be quite stupid to have a limitation on how many times you could lodge, DIFFERENT patents, to patent the same invention.. once you write a patent application that meets the requirements of a patent, then they give you one.

    if it's allready been patented, then they reject it on those grounds, or they can find it invalid for many technical reasons, but they cannot find a patent invalid, just because its not worded right, or not in a correct form, they can (and do) ask them to try again, and again and again.. I does not matter how many times, all that matter is the one that is accepted is right.

     

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

  •  
    identicon
    staff, Dec 17th, 2010 @ 6:34am

    due process

    'But what's a lot more telling is that the patent was "rejected" by the patent examiner not once, not twice, not three times, not four times, but five times before it was finally approved. '

    Those who actually prosecute patent applications know and expect that. It takes time to arrive at claims that are allowable. There is a bit of back and forth. However, patent attorneys will tell you that the PTO routinely rejects claims out of hand and not on merit. For inventions that have broad implications you'd better be ready for a fight. Keep in mind that extended prosecution is expensive for applicants. In some cases it can cost tens of thousands a year for an application. If they feel they are entitled to a patent, they will keep trying. If not, they will save their money. Under such circumstances to not allow extended prosecution would be a denial of due process. Next time, talk to a seasoned patent attorney or agent before you write.

     

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

  •  
    identicon
    Anonymous Coward, Dec 17th, 2010 @ 8:52am

    Six years to first office action is a very long delay for the USPTO which promised us 18 months pendancy.

    Such long pendancy prejudices the examination by allowing the invention to become immersed in what would be competing inventions which can and do distract the examiner (who may only look at the state of the world of invention at the time of patent filing).

    As to the number of times an examiner rejects before approval as something WEONG with the PTO and a sign of weakness? Such a conclusion shows an utter lack of understanding of the process itself.

    Each time an examiner rejects the inventor responds with a NEW application (typically) in which the claims are changed to overcome the examiners objections or new arguments about why the examiners objections do not have merit. This process minimizes the breadth of patents to protect the public.

    The people who seem to write and read this column don't seem to understand that a patent with very weak and limited claims is of little and limited value and cannot typically block or license many competitors. Getting a patent after many rejections is not a measure of the quality of examination or the strength of the rejections.

    The only way to determine how broad a patent is, is by careful review of the art and finally a test by fire in court.

     

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

  •  
    identicon
    Vic Kley, Dec 17th, 2010 @ 8:55am

    Six years to first office action is a very long delay for the USPTO which promised us 18 months pendancy.

    Such long pendancy prejudices the examination by allowing the invention to become immersed in what would be competing inventions which can and do distract the examiner (who may only look at the state of the world of invention at the time of patent filing).

    As to the number of times an examiner rejects before approval as something WEONG with the PTO and a sign of weakness? Such a conclusion shows an utter lack of understanding of the process itself.

    Each time an examiner rejects the inventor responds with a NEW application (typically) in which the claims are changed to overcome the examiners objections or new arguments about why the examiners objections do not have merit. This process minimizes the breadth of patents to protect the public.

    The people who seem to write and read this column don't seem to understand that a patent with very weak and limited claims is of little and limited value and cannot typically block or license many competitors. Getting a patent after many rejections is not a measure of the quality of examination or the strength of the rejections.

    The only way to determine how broad a patent is, is by careful review of the art and finally a test by fire in court.

     

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

    •  
      identicon
      Anonymous Coward, Dec 17th, 2010 @ 10:38am

      Re:

      You do realize, of course, that one who actually understands how the entire process works is unable to enjoy the satisfaction that comes from rants and tirades against the system?

       

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

  •  
    identicon
    Jerry Leichter, Dec 29th, 2010 @ 10:00am

    There are all kinds of reasons for patent rejection, but one thing to keep in mind is that a rejection is typically highly focused: The examiner finds *one* complaint, rejects the patent, and goes on to the pile of other work in front of him. For example, the examiner finds *one* piece of what he considers prior art, and that's it.

    Now, it may well be that the apparent prior art really isn't - the examiner misunderstood the claims (equivalently - it's a pointless debate - the claims weren't drafted clearly enough) and in fact the patent being submitted really is different. Why shouldn't the applicant be allowed to clear things up and try again? And ... given that the examiner will almost always only report one problem at a time - what's the sense in some arbitrary limit on the number of resubmissions?

    Are there abuses of this process? Absolutely. There are patents that are resubmitted over and over after rejections that really leave nothing standing. The patent office tries hard to have the same examiner deal with all the resubmissions, but some applicants seems to drag things out until they somehow get reassigned to a different examiner who may not be as diligent.

    *If* you're going to have a patent process at all, the role of the applicant is to convince the examiner that the patent meets all the requirements the system has. That's go to require so to-and-fro. You could take a trial-like approach - everyone in the room in front of a neutral arbiter who decides when enough is enough and the decision is final (subject perhaps to some fixed-length appeals process that in principle only looks at correctness of procedure, not at facts) - but that would be immensely more expensive than what we have today, and - if you look at court delays - would probably take even longer. The current process seems like a better approach - not that some better controls wouldn't help (though that comment seems to apply equally well to pretty much everything having to do with patents).

    -- Jerry

     

    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