US Patent Office Grants Massively More Patents Than Ever Before

from the this-is-not-good dept

Commerce Secretary Gary Locke has made it clear that he wanted to US Patent and Trademark Office (USPTO) to clear out some of the backlog on patents, and it quickly became clear early last year that the way the USPTO was doing this was by simply approving more patents while giving less scrutiny to the patents in question -- meaning that we're now getting a ton of bad patents approved. It seemed like an obviously bad sign when we passed the total number of patents approved in 2009 by October of 2010.

Now the final numbers are in: nearly 220,000 patents granted (219,614 if you want to be exact), a massive 31% increase over 2009 and still significantly more than ever before in history. The highest previous year was 2006, when a mere 173,772 patents were granted. So this is still 27% more patents granted in a single year than ever before. I find it hard to believe -- as USPTO supporters claim -- that the Patent Office suddenly figured out how to approve 30% more patents without decreasing the quality. Patently-O put together this lovely chart to demonstrate the pattern:
Prepare for the barrage of innovation delaying lawsuits... followed by a lot more patent applications. What's funny is that USPTO director David Kappos actually thinks that ramping up patent approvals will help with the backlog. It won't. Because all that will happen is that more people will try to get more ridiculous patents through, recognizing that the USPTO has massively lowered its standards, and they have a better chance of getting that magic monopoly with which they can sue other companies.

If you look at the numbers over the past thirty years, you see that massive jump in 1998 and heading on up until 2003, when the Supreme Court finally realized that the patent system was massively out of control and was doing a lot more to harm innovation than to help it. However, over the last few years, in response to pressure from those who abuse the patent system to squeeze money out of others, it appears we haven't just reverted to the way things were before, we've leapfrogged the trend line. I can think of no better way to massively slow down American innovation than this. What a shame. Kappos' legacy is going to be a pretty sad one when all is said and done. It's really too bad. When he was first put into the job, it appeared he actually understood how bad patents could be harmful.

And, it should be clear that it's not just the small "trolls" that are the issue here. A ton of big companies stocked up on tons of new patents. Not surprisingly, the top patent getters are companies that are heavily involved in massive value-destroying patent thicket lawsuits: IBM, Samsung and Microsoft. Apple -- which has gone all in on some patent battles -- received nearly twice as many patents in 2010 as in 2009. Ditto with Qualcomm.

Oh yeah, and guess which area of technology had the highest number of new patents? "Multiplex Communications." In other words, the mobile phone patent thicket of lawsuits we've discussed in the past is only going to get messier:


Reader Comments (rss)

(Flattened / Threaded)

  1.  
    icon
    Sneeje (profile), Jan 18th, 2011 @ 9:33am

    Interesting to correlate with other data

    Mike, might want to tone down the 'massive' use of 'massive'.

    It would be interesting to correlate this with patent lawsuits (which I think has been done and is hard work, I know) as well as with some analysis of the population of potential patent-generators. Patent applications shows how many patents are requested, but gives no reference for understanding how many could have or have been historically in relation to the population pool.

    The number of examiners only increased by 3% from 2008 to 2009. I wonder if this is partially motivated by a need to grow fee revenue and therefore the examiner pool. A vicious cycle.

     

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

  2.  
    icon
    :Lobo Santo (profile), Jan 18th, 2011 @ 10:23am

    Re: Interesting to correlate with other data

    Of money and metrics...
    The USPTO is funded by fees from patent applications... and the number of granted patents is a nice hollow number which can be pointed at to say "See! We're innovating now more than ever!"

     

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

  3.  
    identicon
    -, Jan 18th, 2011 @ 10:25am

    Interesting. I wonder when how does it affect international patent system. I see that these are valid in the US. But how about international ones?

     

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

  4.  
    identicon
    Eric, Jan 18th, 2011 @ 10:34am

    but think of the lawyers!

    But think of how this will benefit the lawyers!$!$!$

     

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

  5.  
    icon
    Hephaestus (profile), Jan 18th, 2011 @ 10:44am

    When charted out ...

    It Kinda looks like the curve just before the internet bubble collapsed, or the housing market, or etc, etc, etc.

     

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

  6.  
    identicon
    Willton, Jan 18th, 2011 @ 10:46am

    Re: Re: Interesting to correlate with other data

    and the number of granted patents is a nice hollow number which can be pointed at to say "See! We're innovating now more than ever!"

    It's also a nice hollow number that some use to say "Look, the USPTO has lowered their standards!!"

     

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

  7.  
    icon
    Mike Masnick (profile), Jan 18th, 2011 @ 10:51am

    Re: Re: Re: Interesting to correlate with other data

    It's also a nice hollow number that some use to say "Look, the USPTO has lowered their standards!!"


    Can you present any evidence that explains such an increase that didn't involve lowering standards?

     

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

  8.  
    identicon
    Willton, Jan 18th, 2011 @ 10:57am

    Confirmation Bias

    I find it hard to believe -- as USPTO supporters claim -- that the Patent Office suddenly figured out how to approve 30% more patents without decreasing the quality.

    And yet, you lack evidence that proves otherwise.

     

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

  9.  
    icon
    Vincent Clement (profile), Jan 18th, 2011 @ 10:59am

    When everyone is suing, well, everyone, doesn't that destroy the argument that patents are required in order to invent and innovate?

    Also, there is usually a lag between granting a patent and citing said patent in a lawsuit. Sometimes 10 or 15 years before the lawsuit is filed. I predict job security for the lawyers.

     

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

  10.  
    identicon
    Willton, Jan 18th, 2011 @ 11:00am

    Re: Re: Re: Re: Interesting to correlate with other data

    Can you present any evidence that explains such an increase that didn't involve lowering standards?

    Can you present any evidence that confirms the increase did involve lowering standards?

     

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

  11.  
    icon
    Mike Masnick (profile), Jan 18th, 2011 @ 11:01am

    Re: Confirmation Bias

    And yet, you lack evidence that proves otherwise.


    Wait, so you're suggesting the proper thing to do is to assume that the Patent Office magically figured out how to be 30% more efficient in one year?

    I would think that lower standards is a much more reasonable explanation. You don't?

     

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

  12.  
    identicon
    Willton, Jan 18th, 2011 @ 11:02am

    Re:

    When everyone is suing, well, everyone, doesn't that destroy the argument that patents are required in order to invent and innovate?

    That was never the argument. The argument is that patents encourage invention and innovation.

     

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

  13.  
    icon
    Marcus Carab (profile), Jan 18th, 2011 @ 11:06am

    Re: Re: Re: Re: Re: Interesting to correlate with other data

    To my knowledge, the number of patent officers did not increase by anywhere near the amount that the number of patents did, and there has been no news of procedural changes that would allow them to maintain levels of scrutiny while reviewing more patents - so simple logic would seem to dictate that the only possible way of increasing the output this much is by decreasing standards.

    If you have another explanation, though, I'd be glad to hear it.

     

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

  14.  
    icon
    freak (profile), Jan 18th, 2011 @ 11:07am

    Re: Re: Re: Re: Re: Interesting to correlate with other data

    Yes. The same number of people are approving an increasing number of patents.

    The cogent assumption is that either they were slacking off before, or the quality of patents is decreasing.

    In any case, even before this year, we are seeing a lot of patents which are harmful to innovation because the quality is so dang low already. Granting more, even if the quality stayed the same, could still be a very bad thing.

     

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

  15.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 11:07am

    Re: Re:

    How?

     

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

  16.  
    identicon
    Joe Smith, Jan 18th, 2011 @ 11:09am

    defensive patents

    If you are a big high tech firm then you need to try to patent every idea that any of your engineers come up with to protect yourself from someone else patenting it and using it against you.

     

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

  17.  
    icon
    Mike Masnick (profile), Jan 18th, 2011 @ 11:10am

    Re: Re: Re: Re: Re: Interesting to correlate with other data

    Can you present any evidence that confirms the increase did involve lowering standards?


    Yes. More patents approved by about the same number of examiners means that somewhere standards were lowered. I'm trying to figure out any other explanation. How do you approve more in less time without lowering standards?

    You have yet to present a single credible alternative.

     

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

  18.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 11:13am

    Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    Perhaps workforce productivity gains? Perhaps those examiners are getting more skilled and experienced, and tools for researching the patents they are studying are improving (e.g. internet) so that they are more efficient. Just being the devil's advocate here, literally.

     

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

  19.  
    icon
    Josh in CharlotteNC (profile), Jan 18th, 2011 @ 11:14am

    Re: Re:

    The argument is that patents encourage invention and innovation.

    Yes, because no one would have figured out how to do online press releases if someone didn't patent it.

    http://www.techdirt.com/articles/20110113/03334712652/reddit-digg-fark-slashdot-techcrunch-ot hers-sued-over-ridiculous-online-press-release-patent.shtml

     

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

  20.  
    identicon
    Willton, Jan 18th, 2011 @ 11:18am

    Re: Re: Confirmation Bias

    Wait, so you're suggesting the proper thing to do is to assume that the Patent Office magically figured out how to be 30% more efficient in one year?

    No, the proper thing to do is to not pretend that an increase or decrease in gross grant totals is an indication of efficiency or quality.

    I would think that lower standards is a much more reasonable explanation. You don't?

    No, I don't, especially in light of the fact that the requirements for patentability have become more onerous since KSR, and my opinion that examiners just love to issue rejections. What I think has happened is that (1) examiners are less ham-fisted in issuing their rejections, which is likely due to the change of USPTO leadership; and (2) applicants are drafting their claims more narrowly, which are more likely to be allowed.

     

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

  21.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 11:20am

    Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    They made a deal with the devil. That's how they've done it.

     

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

  22.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 11:20am

    Re: When charted out ...

    You heard it here folks, the patent bubble is about to collapse.

     

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

  23.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 11:24am

    Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    By that logic every time TD published articles at a higher rate than previously then their quality standards are lowered?

     

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

  24.  
    identicon
    Willton, Jan 18th, 2011 @ 11:29am

    Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    Yes. More patents approved by about the same number of examiners means that somewhere standards were lowered. I'm trying to figure out any other explanation. How do you approve more in less time without lowering standards?

    You have yet to present a single credible alternative.


    How about this: the examining corp is actually doing its job. I don't expect a non-patent lawyer to know this, but in the days of Directors Dudas and Doll, examiners were encouraged to reject-reject-reject, regardless of whether the was a legitimate reason for doing so. Now examiners are actually willing to work with applicants in finding what the actual patentable subject matter is. Take this quote from an examiner posting on Patently-O:

    Speaking from my experience in the 705 art class I can say that something happened last year to the Examining Corp that I had not seen in a long time: they would call and discuss the case in person and explore where there might be common ground, instead of rejecting everything out of hand - no matter how far fetched the argument. That means that there was finally some movement in cases that had pending for 7 - 8 years. Some of these cases were on their SECOND appeal b/c the prior Examiners would play games with the rules/rejections. I suspect the willingness to talk to Applicants and look for allowable subject matter (instead of "inventing" reasons to reject) is largely responsible for a lot of this increase.

     

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

  25.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 11:31am

    Re: Re: Re: Confirmation Bias

    "No, the proper thing to do is to not pretend that an increase or decrease in gross grant totals is an indication of efficiency or quality."

    Uh, what?

    Before: Process A outputs 30 units per hour.
    After: Process A outputs 100 units per hour.

    What could explain this? Only reasonable explanations are:

    1- Process was stripped of useless steps (made more efficient)
    2- Process was stripped of useful steps (quality drops)
    3- Both!
    4- Magic!

    Can't have a machine (or humans) outputting more unless you change the process or use some sort of magic.

     

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

  26.  
    icon
    Killer_Tofu (profile), Jan 18th, 2011 @ 11:36am

    Re: Re: Re: Confirmation Bias

    No, the proper thing to do is to not pretend that an increase or decrease in gross grant totals is an indication of efficiency or quality.

    You are right, as we see here all the time, a good just about most approaching all of patents are crap and meant to hinder innovation or to be used as defense from such. The number that is passed doesn't change that.

     

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

  27.  
    identicon
    Willton, Jan 18th, 2011 @ 11:45am

    Re: Re: Re: Re: Confirmation Bias

    Okay, fair point. My argument would be that there was an increase in efficiency due to a few factors:

    (1) Junior examiners becoming more experienced and therefore able to handle a higher case-load; and
    (3) Kappos beginning to scrutinize rejections by examiners to determine if they were legitimate. This gave examiners an increased incentive to work with applicants to find common ground, instead of conjuring reasons to reject an application.

     

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

  28.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 11:45am

    Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    Not true. The number of patents processed could be because some of the patents are easier to read, new systems in place, perhaps more efficient work habits. As an example, if a patent examiner works 40 hours a week, and user to spent 20 hours on "administration", but improvements in documentation and such drops that to 10, they could be much more effecient. If systems have been improved to allow for faster searches of existing patents (more information online and indexed, example), or if they are doing a better job in doing "patent triage" to blow off patents that will not pass, perhaps they are addressing more applications.

    See, the problem is, we don't know. You take a single number "patents are up" and assume they are just rubber stamping them and not doing their jobs. If you want to push that concept, bring something to support it.

    What do you call it? Due process? Perhaps you would like to understand their they are innocent until proven guilty of your charges, not the other way around.

     

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

  29.  
    identicon
    Willton, Jan 18th, 2011 @ 11:48am

    Re: Re: Re: Re: Re: Confirmation Bias

    Whoops, should be a (2), not a (3).

     

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

  30.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 11:50am

    Re: Re: Re: Re: Confirmation Bias

    Let's see:

    1. processes streamlined.
    2. more data indexed and accessible through computer rather than paper files.
    3. more "pre-processing" to make sure that examiners are only looking at proper applications
    4. simplified administration methods, lower paperwork overhead, etc.
    5. better collaberation between examiners.

    If you look at the chart in the article, it is clear that the number of patents approved goes up each year. In straight line terms, the increase is well within the normal growth rates (20 year trend, example).

    There is also no indication of "accepted versus declines", nor any indications of the number of patents examined. It is very likely that they have become more effecient, as the trends seem to indicate.

    What TD is doing is drawing a conclusion without any facts, almost entirely based on his personal opinion. That is the very worst way to make a point.

     

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

  31.  
    icon
    freak (profile), Jan 18th, 2011 @ 11:51am

    Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    If the number of writers stayed constant and they did not spend more time writing stories?

    Than I would answer yes. You can clearly see cases where the quality is obviously sub-par and the journalism is iffy, particularly during busy days. But because Mike and the others do respond to the community and do correct their mistakes, and do link back to source material, it's much easier to forgive.

     

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

  32.  
    identicon
    Willton, Jan 18th, 2011 @ 12:02pm

    Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    If the number of writers stayed constant and they did not spend more time writing stories?

    Than I would answer yes.


    So you would foreclose the possibility that the writers at TD were actually getting better at writing? Interesting.

     

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

  33.  
    icon
    freak (profile), Jan 18th, 2011 @ 12:08pm

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    Well, I've noticed, (nothing empirical, mind you, so it could be a availability bias), that the journalism and writing is spottier on days when more articles are written.

    So, if TD had a 30% increase in articles very much all of a sudden, as with patents, no, I would not assume they had magically gotten 30% better.

    With patents, it's quite worse, because I see no reason to believe that the quality of examination was ever very good.

     

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

  34.  
    icon
    Stephan Kinsella (profile), Jan 18th, 2011 @ 12:10pm

    Patent Shills Halling and Quinn should be happy

    This is what they called for in recent blog posts: patent granting inflation.

    See my post Patent Shills want to make patents “incontestable”.

     

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

  35.  
    icon
    Vincent Clement (profile), Jan 18th, 2011 @ 12:36pm

    Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    When was the last time a government operation became MORE efficient over time?

     

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

  36.  
    icon
    Vincent Clement (profile), Jan 18th, 2011 @ 12:48pm

    You can tell which companies are successful by the number of arrows pointing to them (RIM and Apple) and which companies are in decline by the number of arrows pointing away from them (Kodak).

    But it is all about invention and innovation, right?

     

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

  37.  
    identicon
    AJ, Jan 18th, 2011 @ 12:56pm

    IMO

    IMO

    Mike has a point. Barring some kind of “magical” breakthrough in the science of patent approving, which I can’t seem to find anything on online, there doesn’t seem to be a logical explanation for the dramatic increase in approved patents.

    I can totally understand why any normal person would come to the conclusion that they are using the “when in doubt, approve” approach. They have suddenly become 30% better at doing something as complicated as approving a patent while, going to the lengths of actually engaging the persons/company applying for the patent, as another poster stated? You would think that would be big news for the patent office, and would be thrown in everyone’s faces at every opportunity.

    So if you combine the dramatic increase in approved patents, with a non-equivalent increase in patent examiners, pull out the possibility of a breakthrough in patent examining as we all know it would be tossed in everyone’s face at every opportunity, you could understand why someone would think they were just rubber stamping the patents.

     

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

  38.  
    identicon
    Willton, Jan 18th, 2011 @ 1:07pm

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    So, if TD had a 30% increase in articles very much all of a sudden, as with patents, no, I would not assume they had magically gotten 30% better.

    I did not ask whether you would assume that TD got better. I asked if you would FORECLOSE that possibility.

    With patents, it's quite worse, because I see no reason to believe that the quality of examination was ever very good.

    Two things:
    (1) I doubt you are the appropriate person to judge whether an examining corp is doing a good job or not, as I find it hard to believe that you have the sufficient knowledge to make such a judgment; and
    (2) if it were true that examination was bad, then it likely would not take much to improve to average or good, would it? Perhaps a change in the examination process would do the trick, or maybe just an additional year of experience. Perhaps that could account for an increase in efficiency.

     

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

  39.  
    identicon
    Wolfy, Jan 18th, 2011 @ 1:15pm

    The politicization of key posts in all federal agencies was a priority under Reagan. Under Bush, it got taken to a new level. Priority No.1 was the Justice Dept. No.2 was Interior. That's one Gov't operation that got more efficient over time.

     

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

  40.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 1:41pm

    Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    They do all the time, we just don't realize it. You have to compare longer periods of time to see improvements. Example would be the DMV without computers, and the DMV with computers. Perhaps the IRS as well. Those are great examples of singificantly increased productivity.

    The main point is this: There are other possibilities besides rubber stamping / lower quality, but TD has decided and that is that. It is incredibly simpleminded to do that.

     

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

  41.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 1:43pm

    Re: IMO

    You only have to look at the chart to see. Draw a line from 1990 to 2010. It touches the tops of each of the years (except very recently). There seems to be similar levels of gains average each year. 2000 had twice as many as 1990. Was 2000 magical too?

    It galling to see someone who gets all over anyone else for coming to a conclusion "without facts" to suddenly jump to a conclusions.

    Did anyone else notice that most of the posts today on TD are pretty angry? I have to think that something went horribly wrong somewhere.

     

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

  42.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 2:04pm

    Steam

    I recall reading here that the innovation on steam engines truly got going just after the relevant patents expired. Is the same going to happen with mobile phones, with the innovation truly gaining steam (heh) only after the bulk of the patent thicket has expired?

     

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

  43.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 2:32pm

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    http://www.jsonline.com/business/80244907.html

    A Journal Sentinel investigation published in August [2009] documented how congressional diversions of the agency's income from 1992 through 2004 left the Patent Office incapable of keeping pace with the volume and complexity of the applications it receives. The backlog has grown to more than 1.2 million applications, which the agency has said could take at least six years to get under control - assuming it receives the funding to hire and train new examiners.

    But a budgetary provision that could have allowed it to spend up to an additional $100 million during the current fiscal year was stripped on Dec. 9, the final day of budget negotiations.

    "We are currently operating on a barebones budget that makes it very difficult to attack our application backlog," said Sharon Barner, the agency's deputy director.

     

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

  44.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 3:40pm

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    Congrats, you have explains why they did so poorly in the last few years. Perhaps you would want to see what changes they have made in the last 24 months to see how things worked out.

     

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

  45.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 3:42pm

    Re: Steam

    yeah, because we haven't any any innovation since the Motorola Brick phone, have we?

     

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

  46.  
    icon
    TheRegulator (profile), Jan 18th, 2011 @ 4:44pm

    Maybe "Massive" Legal Action is required...

    Nothing inhibits small innovators (and even larger ones),both here and around the world, more than lurking patents of dubious validity.

    Perhaps it is time for a class action against USPTO, or if that isn't possible, a steady stream of lawsuits challenging the most egregious patents that showcase the PTO's abysmal disregard of prior art and obviousness.

    Another approach might be to routinize the use the threat of interference litigation as a bargaining chip for negotiating with the holders of annoying patents.

    An interesting source of plaintiffs and funding might be developing nations, who may be the worst victims of what is, in effect, IP colonialism.

     

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

  47.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 4:48pm

    Re: Re: Confirmation Bias

    would think that lower standards is a much more reasonable explanation. You don't?

    A reasonable explanation is a long way from any form of proof. I mean, based on your posts here, it would be a reasonable explanation that you are fronting for pirate websites, and perhaps are fronting for the Pirate Party to form in the US. It is a reasonable explanation for your words and your opinions.

    It would likely be wrong, but it is a reasonable explanation.

    Then again, even when you have facts, you try to obscure them. The "patent thicket" graphic is a great example. It is created to have the most confusion possible. By spraying the green boxes (they only sue) around in random places, you can create the longest gaggle of lines possible. The entire deal could be represented in a much neater and cleaner way, but that isn't you intention. A reasonable explanation would be that your are trying to add some FUD to the situation, which isn't as bad as you make it out to be.

    See? Reasonable explanations.

     

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

  48.  
    icon
    Modplan (profile), Jan 18th, 2011 @ 5:00pm

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    (1) I doubt you are the appropriate person to judge whether an examining corp is doing a good job or not, as I find it hard to believe that you have the sufficient knowledge to make such a judgment;

    Do you have sufficient knowledge to make such a judgement as to how well the USPTO does in reviewing patents?

    (2) if it were true that examination was bad, then it likely would not take much to improve to average or good, would it? Perhaps a change in the examination process would do the trick, or maybe just an additional year of experience. Perhaps that could account for an increase in efficiency.


    And yet the only thing we have on record as having actually significantly changed is a focus on speeding up patent reviews with no other strategy provided like better efficiency, with an increase of 30%, in itself an increase of over 27% over the previous patent record, and excluding various other points in the graph where similar massive jump occurs like 1998 - 2000, and a massive decline for 2006.

     

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

  49.  
    icon
    Mike Masnick (profile), Jan 18th, 2011 @ 5:23pm

    Re: Re: Re: Re: Re: Confirmation Bias

    (1) Junior examiners becoming more experienced and therefore able to handle a higher case-load; and (3) Kappos beginning to scrutinize rejections by examiners to determine if they were legitimate.


    You think that explains a 30% leap? I don't see how that's likely. First of all, there's also turnover among experienced patent examiners, meaning that they're replaced by junior examiners, so (1) is sort of canceled out.

    As for point (2), again, that does not explain a 30% leap in a single year at all. If the leap were gradual, you might have a point, but it was not. In fact, if they were scrutinizing rejections more carefully then it would DELAY, rather than speed up patenting.

    So, no, I'm afraid we're still left with quality being the clear issue here.

     

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

  50.  
    icon
    Modplan (profile), Jan 18th, 2011 @ 5:28pm

    Re: Re: Re: Re: Re: Confirmation Bias

    (1) Junior examiners becoming more experienced and therefore able to handle a higher case-load; and


    This is likely to be a consistent, ongoing process that would not have significant impact unless you can point to an increase in junior examiners that lowered the number of patents being approved or lowered their quality.

    (3) Kappos beginning to scrutinize rejections by examiners to determine if they were legitimate. This gave examiners an increased incentive to work with applicants to find common ground, instead of conjuring reasons to reject an application.


    This would make the system slower, not faster/more efficient. It's true that in this case more patents may be granted, but it also seems true to me at least that this process is slower than outright rejection, leading to an increased workload on the same number of examiners, along with an assumption that the new patents being granted are of any higher or the same quality than the ones already being granted, which doesn't appear to be likely at all, meaning the system is still worse through having granted more bad patents.

     

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

  51.  
    icon
    Modplan (profile), Jan 18th, 2011 @ 5:38pm

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    What are those changes? Where can we see/find out about these changes?

     

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

  52.  
    icon
    Modplan (profile), Jan 18th, 2011 @ 5:46pm

    Re: Re: IMO

    I'm not sure about 1990 to 2000, but what about 98 to 2000? or 2006 to 2007? Were those efficiency gains too? It's remarkable just how many sharp efficiency gains the USPTO makes without any lasting, consistent effect.

     

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

  53.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 6:52pm

    Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    "Perhaps those examiners are getting more skilled and experienced"

    On average, people change jobs every two to three years or so, and I see no reason why the USPTO wouldn't follow that average. and I doubt those people who work at the USPTO have worked at previous PTO's before working at the USPTO, so a change in job is likely a change in career in this case. Not to mention, the USPTO is known to have a particularly high turnover rate.

    So it's highly unlikely that patent examiners are more experienced on average than what they used to be. Just like with any business that's been in business for a long time, the overall experience of the employees reaches an equilibrium as experienced people either leave their job or retire and new less experienced people enter. So it is reasonable to conclude that the experience level at the USPTO is about the same as it has been for quite some time now. It's not like they suddenly hired a bunch of new experienced examiners out of thin air.

     

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

  54.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 7:00pm

    Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    "Now examiners are actually willing to work with applicants in finding what the actual patentable subject matter is."

    That's always allegedly been the case, but it never seems to be true.

    "the examining corp is actually doing its job."

    Given the history of the USPTO why should I assume that they've suddenly decided to do their job. because of some statement by one of them that they will?

     

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

  55.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 7:02pm

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    "There are other possibilitie"

    There are always other possibilities but we are merely concluding with the most likely possibility.

     

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

  56.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 7:03pm

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    I'm sure they would have gladly announced those changes if they were made. The reason none of them have been announced is because they never happened.

     

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

  57.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 7:06pm

    Re: Re: Re: Re: Confirmation Bias

    1) More patents in less time
    2) About the same number of patent examiners
    3) ?????
    4) HIGHER QUALITY!!!!

     

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

  58.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 7:16pm

    Re: Re: Re: Re: Re: Confirmation Bias

    "What TD is doing is drawing a conclusion without any facts"

    We do have facts. Patent quality has always been low. More patents are being approved by about the same number of examiners. Those facts suggest that patent quality will only get worse. Sure, they don't prove it with absolute certainty but we could never arrive at any conclusions if absolute certainty is a requirement. Science would cease to exist.

    Yes, there are other possibilities, but the problem is that those other possibilities lack evidence. The evidence is consistent with what we would expect if our conclusions are true. The problem is that your theory lacks falsifiability. With your logic one can conclude that it is illogical to conclude that gravity causes us to fall because it's always possible that magic fairies do so. The problem? The evidence points to gravity and there is no evidence that magic fairies are the cause of us falling to the ground. Your theory needs falsifiability.

     

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

  59.  
    identicon
    Anonymous Coward, Jan 18th, 2011 @ 7:21pm

    Re: Re: Re: Confirmation Bias

    "A reasonable explanation is a long way from any form of proof."

    Do you have any other reasonable explanations? Ones with supporting evidence? Our explanation is the only explanation with supporting evidence and it's the only reasonable explanation. and, regarding proof, science doesn't deal with absolute proof. If you want absolute proof then you can never come to any explanation about anything.

     

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

  60.  
    identicon
    Willton, Jan 18th, 2011 @ 11:02pm

    Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    "Now examiners are actually willing to work with applicants in finding what the actual patentable subject matter is."

    That's always allegedly been the case, but it never seems to be true.


    Believe me, from someone who practices in front of the USPTO, it's apparent.

    Given the history of the USPTO why should I assume that they've suddenly decided to do their job. because of some statement by one of them that they will?

    That's better than assuming without evidence that they've lowered their standards. Plus, there are a number of practitioners and applicants that would agree with the examiner's statement. In addition, you don't have a basis to say that the examiner is wrong.

     

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

  61.  
    identicon
    Willton, Jan 18th, 2011 @ 11:17pm

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    Here's a quote from Director Kappos back in August 2009:
    http://www.patentlyo.com/patent/2009/08/director-kappos-patent-quality-equals-granting-those-claim s-the-applicant-is-entitled-to-under-our-laws.html

    "One key is to expeditiously identify and resolve issues of patentability—that is getting efficiently to the issues that matter to patentability in each case, and working with applicants to find the patentable subject matter and get it clearly expressed in claims that can be allowed. The examiner and the applicant share the responsibility for the success of this process.

    On the subject of quality, there has been speculation in the IP community that examiners are being encouraged to reject applications because a lower allowance rate equals higher quality. Let's be clear: patent quality does not equal rejection. In some cases this requires us to reject all the claims when no patentable subject matter has been presented. It is our duty to be candid with the applicant and protect the interests of the public. In other cases this means granting broad claims when they present allowable subject matter. In all cases it means engaging with the applicant to get to the real issues efficiently—what we all know as compact prosecution.

    When a claimed invention meets all patentability requirements, the application should be allowed expeditiously. ... [B]y engaging with applicants early on, we certainly can get to the point more quickly, and efficiently allow those claims that are entitled to patent protection. "

     

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

  62.  
    identicon
    Willton, Jan 18th, 2011 @ 11:30pm

    Re: Re: Re: Re: Re: Re: Confirmation Bias

    You think that explains a 30% leap? I don't see how that's likely. First of all, there's also turnover among experienced patent examiners, meaning that they're replaced by junior examiners, so (1) is sort of canceled out.

    First, a 30% leap is not that big. That's about 8-9 more grants per examiner.

    Second, if you can find the turnover statistics for the USPTO's examining corp in 2010, then feel free to bring it into the conversation. I will also inform you that the USPTO made a concerted effort to hire former examiners in 2010, as has been documented on Patently-O and perhaps even here.

    As for point (2), again, that does not explain a 30% leap in a single year at all. If the leap were gradual, you might have a point, but it was not. In fact, if they were scrutinizing rejections more carefully then it would DELAY, rather than speed up patenting.

    Disagree. The pace of examining is not the problem with examination efficiency; it's the volume of bad office actions that slow things down. The effect of scrutinizing more rejections means that primary examiners are less likely to make bad ones. Thus, before making a decision, an examiner would likely take a more focused look at the claims of the applicant and consider granting interviews with the applicant. Doing this may slow down the examining aspect of things, but if it leads to fewer bad office actions, which applicants have 3 months in which to respond, it can quicken the pace to issue for applications with merit.

     

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

  63.  
    identicon
    Anonymous Coward, Jan 19th, 2011 @ 12:40am

    Re: Maybe "Massive" Legal Action is required...

    As a private placement investor, I can tell you that you're spot on. Increasingly the lawsuits and those that bring them are multi infringement with a half dozen or so ridiculous patents. That can spell the end of life for budding companies.

     

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

  64.  
    icon
    Mike Masnick (profile), Jan 19th, 2011 @ 1:31am

    Re: Re: Re: Re: Re: Re: Re: Confirmation Bias

    First, a 30% leap is not that big. That's about 8-9 more grants per examiner.


    Um. And you don't think that's a huge increase?

    Wow.

    Disagree. The pace of examining is not the problem with examination efficiency; it's the volume of bad office actions that slow things down. The effect of scrutinizing more rejections means that primary examiners are less likely to make bad ones.

    This is wishful thinking. I've been talking to folks who work with the USPTO all day today and almost to a person, they've all said that what they see is that it's easier than ever to get questionable claims through. The examiners know they're under the gun to approve more patents, so if you push hard enough, stuff gets through.

     

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

  65.  
    identicon
    Anonymous Coward, Jan 19th, 2011 @ 6:47am

    Re: Re: Re: Re: Confirmation Bias

    Considering that the previous period measured was down, perhaps a significant number of patents were approved early in January, example. We don't have the data, we don't know. As someone else pointed out, this is only 8 or 9 patent approvals per inspector difference. There are plenty of ways that could happen.

    Perhaps the approvals are a significant number of "extentions" to existing patents. Perhaps the quality of patent submissions went up (we have this thing called the internet to help people get it right). Perhaps they have inserted a triage system that narrows the inspectors time down to applications that are properly completed. Perhaps they have changed their systems or methods of work.

    If you want absolute proof then you can never come to any explanation about anything.

    I don't want absolute proof. I want something more than a single number. There isn't anything here except a single data point and TD's massive bias.

    Critical thinking is a thing that TD teaches you not to have. If you have it, you can't enjoy the koolaid anymore!

     

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

  66.  
    identicon
    Anonymous Coward, Jan 19th, 2011 @ 8:40am

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    "Believe me, from someone who practices in front of the USPTO, it's apparent."

    Evidence please.

    "without evidence"

    Evidence has been provided. Your assumption is the one with no evidence. My assumption is better than yours.

    "In addition, you don't have a basis to say that the examiner is wrong."

    Yes I do. and besides, what is he supposed to say, that patent quality will only get worse? Of course he will say it will get better.

    Do you honestly expect a corporate spokesperson to admit that quality control will get worse? Likewise, I doubt a USPTO spokesperson would admit to such a thing. It's their job to claim that they will do a good job.

     

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

  67.  
    identicon
    Anonymous Coward, Jan 19th, 2011 @ 8:42am

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    "It is our duty to be candid with the applicant and protect the interests of the public."

    But this has always been the case, yet patent quality has always been low. So what's changed exactly?

     

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

  68.  
    identicon
    Anonymous Coward, Jan 19th, 2011 @ 8:45am

    Re: Re: Re: Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    "without evidence"

    It's called historical evidence. You know that history is indicative of the future right? If someone has a history of wasting money that I give them on drugs, it is reasonable for me to assume that they will continue to do so. Their history is evidence of that. Likewise, if the USPTO has a history of poor quality, that is evidence that they will continue on this trend.

     

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

  69.  
    identicon
    Anonymous Coward, Feb 4th, 2011 @ 3:43am

    Re: Re:

    Hmmm...is that why everything is produced in Asia now because patents saved manufacturing in the U.S.?

    Which is the system used in mobile phones in the U.S. again?

    CDMA(a Japanese invention)?

    Didn't patents actually diminished the number of innovations in the drug industry? and by that I mean they launch less new products every year and spend less in Research and Development.

     

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

  70.  
    identicon
    Anonymous Coward, Feb 4th, 2011 @ 3:46am

    Re: Re: Re: Re: Re: Re: Interesting to correlate with other data

    Mike c'mon clearly the USPTO created a better software that analyzes patents and it is being automated, but it is a secret because it wasn't able to self grant a patent for itself yet.

     

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

  71.  
    identicon
    Bioresearcher now with tied hands, Sep 22nd, 2013 @ 6:32am

    Re: Re: Re: Confirmation Bias

    "Applicants are drafting their claims more narrowly?" How can you claim that with examples like this:

    http://patft.uspto.gov/netacgi/nph-Parser?Sect2=PTO1&Sect2=HITOFF&p=1&u=/netaht ml/PTO/search-bool.html&r=1&f=G&l=50&d=PALL&RefSrch=yes&Query=PN/8354258

    Indeed a trial to patent nature's own ability to produce structures, and yet again previous published work on the same applications.

     

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

  72.  
    identicon
    Limited scope, Sep 22nd, 2013 @ 6:37am

    Re: Re: Re:

    Well, no... everything is made in Asia because US companies outsourced their production, not because they are copying everything. You have to understand that when corporations find better conditions for profit they will move. They don't care about how it affects employment or the economy at all.

    Apple is "proudly" an American company, and yet again, almost everything is "assembled" in China, hmmm....

    Patents won't save manufacturing jobs... don't be naive.

     

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


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