Of Smart Phone Wars and Software Patents

By Dennis Crouch

Stuart Graham (USPTO’s Chief Economist) and Saurabh Vishnubhakat recently published an interesting short paper entitled Of Smart Phone Wars and Software Patents. The paper largely defends the USPTO’s examination of software patenting by showing that its approach in the software arts is essentially the same as in other fields.

The two charts below come from the article. The first shows the percentage of first office actions that are first-action allowances. This is calculated for each fiscal year as the (# of first action allowances) / (# of first actions). The second chart looks at the first “final” action in a case. For their study, a final action is either (1) a final rejection or (2) an allowance. And, the first final is whichever one of those came first.

The point of the article was to show that software patents have been treated approximately the same as other patent applications. That is an important take-away. However, what is amazing to me is the dramatic change in practice that we have seen over the past few years.

41 thoughts on “Of Smart Phone Wars and Software Patents

  1. 38

    First, juries find patents valid over “good art” all the time.

    Patent numbers please.

  2. 36

    “Huh? Software is not patent eligible. Why are any software patents being allowed? Any?!”

    Did I miss something? Was there a Court case that expanded the Judicial Exceptions even further to explicitly include software?

    Seriously!

  3. 32

    Malcolm,

    How is totally ignoring my on-point response to you and you merely repeating a question for which the response provided was better than a yes or no response any way to engage in a conversation?

  4. 31

    probably shouldn’t have been granted in the first place

    Aside from your opinion, which of course you are free to have no matter how wrong it is, can you find any legal doctrine, rule, or law that restricts alienability as you would want?

    I can give you the law: 35 U.S.C. 261

    As for “useful shorthands,” there are plenty that do not have the pernicious effects on conversations like the pejorative “Troll.”

  5. 29

    “I was getting at the fact that this paper has nearly zero credibility and that I could continue to point out the failings.”

    Well, I wouldn’t go that far. After all looking at objective measures and saying you can’t tell a difference from these objective measures is an interesting result.

  6. 28

    “The point of the article was to show that software patents have been treated approximately the same as other patent applications. ”

    Approximately but not really 🙁 If we went ahead and went 100% there wouldn’t even be anything under discussion.

  7. 27

    I would guess that the quality of examination should not be correlated to the allowance rate.

    Even if 100% of filed claims were allowed, anon “guesses” that the examination quality should not be questioned.

    Good to know. That’s what I thought you’d say, anon.

    Now go ahead and tell me you’re saying something substantially different. Or better yet, just answer the question I asked “yes” or “no.” Somehow you avoided doing that. I wonder why.

  8. 26

    You should be careful with the use of such loaded words as they indicate (apparently) more meaning than you intended.

    It additional unintended meaning wasn’t apparent to me.

    I’ll let Casual Observer clarify, but I assumed he meant that patent owners who hide behind shell companies and shake down others for infringing patents that probably shouldn’t have been granted in the first place are deserving of derogatory names like “trolls.”

    The term “troll” is a useful shorthand, anon. Wealthy people manipulating the patent system for their personal gain aren’t a traditionally and systematically persecuted minority group. They can “take it”. Trust me. They seem to have weathered the Great Recession perfectly well, unlike a great many far less fortunate people, including some of the small businesses they have chosen to threaten with their lawsuits.

  9. 25

    Whether or not the gross level changes should not have any indication on the rate level

    That right there is some basic math”, anon, but it’s your problem, not mine. “Basic math” is applicable to simple systems with simple components.

    In contrast, I’m looking at the functioning of an overwhelmed government agency that processes very complicated documents and decides whether to grant powerful and potentially industry-destroying rights to applicants. This agency performs this task under relentless pressure from “watchdog” lobbyists and interest groups who want that task to be accomplished as quickly and inexpensively as possible and who raise holy h— whenever a perceived attempt to reduce that pressure is proposed. These same people will object to the pace of examination out of one side of their mouth and on the other side of their mouth they will complain if the PTO raises rates to support the hiring of more examiners.

    In short, “basic math” is the wrong math to be applying here. Let’s take a similar hypothetical situation, as one’s investment in the Patent Casino is known to cloud one’s views of the matter.

    Imagine a government agency whose job it is to evaluate the levels of lead in fish sticks served in public school lunches. Imagine that the agency is overburdened. It rejects 20% of fish stick shipments as having too much lead in year one. The government establishes a new standard that sets maximum allowed lead levels lower than before. The fish stick lobby is furious. In year two, the fish stick lobby manages to install a former fish stick industry mogul to take charge of the agency. The number of fish sticks delivered to school lunches triples in year two. The number of fish stick shipments rejected by the agency drops to 15% in year two.

    If your “basic math” tells you that nothing *ahem* “fishy” is going on here, then probably you should try some “advanced math” for a change.

  10. 24

    considering the dramatic rise in patent application filing over that same time period

    Malcolm, you seem to be having a recurring problem with basic math. Whether or not the gross level changes should not have any indication on the rate level that Prof. Crouch shared (or the comments on point to that rate level).

    I would like to think that you are not simply trying make a vacuous comment or, even worse, that you are trying hide the true meaning of something, ro (mis)use the data in some silly attempt to prop up one of your outlandish accusations. Oops – too late, you were.

  11. 23

    As I said – other than the fact that a relation was not disclosed (yes, I got the lack of transparency), your use of the troll card just does not fit. In the one case, the transparency is problematic – and I provided a more fitting analogy, in the other, well, too many assumptions come into play.

    You should be careful with the use of such loaded words as they indicate (apparently) more meaning than you intended.

    Thanks.

  12. 22

    I would guess that the quality of examination should not be correlated to the allowance rate.

    That was kind of the point made by bja.

    It was also the point (in heavy duty emphasis) made by me.

    This thing that I did – taking in bja’s response and augmenting it – it’s called discussion.

    I heard that that is all the rage.

    That thing you did: “(although you’ll likely still avoid answering)” is called gratutious nonsense. It doesn’t help discussions.

    Please desist. Thank you.

  13. 21

    I guess my point was not clear. When Mr. Graham presented at the Patent Assertion Entity workshop at the Federal Trade Commission he spoke to the need for more clear disclosure of who the real person of interest was in litigation involving non-practicing and patent assertion entities calling for more transparency.

    Here is an article touting USPTO’s fairness with respect to the treatment of software patents and he and Mr. Visnubhakat obfuscated their real role and the source of their paychecks form direct rolls in implementing the AIA.

    So much for transparency.

  14. 20

    .IF examination is done to aim at ANY rejection rate, then yes, that examination rational is terrible.

    That’s not the question. Let’s put it this way (although you’ll likely still avoid answering):

    All other things being equal, if starting next week the USPTO began allowing 100% of filed claims in first actions, would you guess that the quality of examination of those claims had increased or decreased? What if they allowed 95%?

    It’s a yes or no question.

  15. 19

    40% of final actions were rejections in 2003.
    60% of final actions were rejections in 2009.

    That doesn’t seem “amazing” to me at all considering the dramatic rise in patent application filing over that same time period. Or did human beings suddenly become more innovative?

    I also note that the “high” rate of rejections in 2008 did little to deter people from filing applications.

    What’s more “amazing” is that in the wake of Supreme Court cases that theoretically should have made it more difficult to obtain a patent, it’s apparently becoming as easy as its ever been to obtain a patent.

    The simplest explanation is that in the wake of the housing bubble bursting in 2008, the PTO was advised to open the spigots so the investor class could have more fun at the Patent Casino. Instead of repairing roads, sewers, bridges and giving nice things to the middle class and poor people, our great country decided to let those people hang (the rich, wise white men on TV call this “sharing the pain”) while the well-to-do and their lawyers keep themselves busy arguing over “new” methods of fondling a touch-screen.

  16. 18

    where offensive scientific facts about the world are deleted from textbooks so as not to offend the sensitive people who live there

    Change a few words and that statement is really funny.

    Would you say that if the rejection rate was 0%? What about if it was 5%?

    IF examination is done to aim at ANY rejection rate, then yes, that examination rational is terrible.

    Of course, you could use the archives to see that this has alwasy been my position on the matter.

  17. 17

    if the art is good, you’re not going through trial anyway.

    First, juries find patents valid over “good art” all the time. Some juries, for example, are educated in Texas where offensive scientific facts about the world are deleted from textbooks so as not to offend the sensitive people who live there. Then they are asked to opine as to whether the patent owned by the guy with the cowboy hat was infringed by the California company that serves its employees arugula for lunch.

    Second “bad patents” often make it very far indeed.

    Look at the claims in Prometheus. They went up to the Supreme Court. They were incredibly bad claims. Not only was the subject matter issue off the charts, but the claims were anticipated/obvious. Look at how far Myriad has gotten with its claims, which also suffer from severe problems in view of the prior art. What do parties like Prometheus and Myriad have in common? A big bank acccount.

    the rejection rate is a terrible measure of examination quality.

    Would you say that if the rejection rate was 0%? What about if it was 5%?

  18. 16

    1. Litigation is the only true measure of examination quality.

    a. Shouldn’t we account for broadest reasonable interpretation versus actual interpretation. It’s much easier to find invalidity with broadest reasonable interpretation than with the actual meaning of the claims.

    b. The cost of giving up your invalidity claim is much higher than presenting good art. And if the art is good, you’re not going through trial anyway.

    c. Then what’s the right way to look at it? If 100 are taken to court, 90 settle before trial because the risk of infringement is just too high (i.e., no good prior art is found). 10 make it to trial, and of those 10, 80% are valid (8/10), that’s a rough validity rate of about 98/100 or 98%. Not bad. The bad patents never make it that far. You show the owner the good art and they go away. If you don’t have good art, then it’s valid, so there’s no point in whining.

    2. Yes, the rejection rate is a terrible measure of examination quality.

  19. 15

    “You didn’t point to a single distinguishing feature between software and any other type of patent.”

    What difference does that make? They are the ones using a completely unsupported assumption about quality of examination by comparison to OTHER art units. That comparison is silly on its face. It assumes facts that are not really in evidence:
    1. That other art units are in fact getting high quality examination.
    2. That other art units have claim scope and subject matter scope similar to those being examined in the software area
    3. That the searching, art, sources, language, etc. are somehow equivalent

    The report is just gibberish on that front.

    “Again, your ilk has fallen back to I know it when I see it, but I can’t produce anything objective so I’d like a test where I can just pronounce the outcome.”

    I have no idea what that means. I wasn’t attacking that patentability or anything like that. I was attacking the foundations of the “study” published in this very brief paper.

    “and on and on”
    I was getting at the fact that this paper has nearly zero credibility and that I could continue to point out the failings.

  20. 14

    What is silly is the rampant assumptions and absence of anything actually meaningful in your post (except your feelings, of course – those are special).

    I am curious as to if you have a right way of patent enforcement, or do you think that patents should not be allowed at all?

  21. 13

    Correlation doesn’t prove causation.

    It would be more telling to see these statistics for each unit separately (biotech, chemistry, mech eng, etc.). If the PTO (or examiners) have goals or limits on the % of first OA allowances (or final rejections) then its no surprise that the #s look the same for most areas.

    If true this is a stat about the PTO and says very little about the value of software patents.

  22. 12

    And on and on what? You didn’t point to a single distinguishing feature between software and any other type of patent.

    Again, your ilk has fallen back to I know it when I see it, but I can’t produce anything objective so I’d like a test where I can just pronounce the outcome.

  23. 11

    A few things about this paper:

    1. Looking at litigation outcomes as to the “quality” of examination is silly.

    a. In litigation, there is a presumption of validity with a high standard of evidence to overcome.

    b. In litigation, there is a high, individual cost to maintain an invalidity challenge through trial. What’s more, this invalidity challenge is often coupled with potentially high cost of infringement.

    c. The fact that nearly 20% of those taken to trial (4/21) are found invalid is actually significant cost on the market. Looking at the 80% as evidence of success is actually the wrong way to look at it.

    2. The fact that the rejection rate is similar to that of other fields really doesn’t mean much. That’s certainly can’t be a good measure of the quality of examination.

    and on and on…

  24. 10

    I find this obfuscation kind of like when the trolls try to hide the real person of interest.

    Really? How? I mena other than the fact of a connection not being made public, how is the one like the other? I don’t see it.

    Grant you, you do raise a valid point here of what I would more liken to posting under third party influence. The fact that the authors may have a stake in what they say can initiate some serious obstacles to a meaningful, adult, intelelctually honest discussion. But the “troll card” does not fit the scenario.

  25. 9

    Don’t you find it a bit disingenuous that the authors are not exactly being forthcoming about the real nature of their relationship with USPTO.

    Graham, the Chief Economist at USPTO, identifies himself as an “Expert Advisor” at USPTO and is playing on his academic credentials and Saurabh Vishnubhakat advises the USPTO’s Chief Economist and senior management on a range of issues including genetic patenting, IP assignment, international trademark examination, and technology standards. As part of the USPTO’s implementation team for the Leahy-Smith America Invents Act, Saurabh helps administer a number of regulatory reforms, studies, and reports to Congress.

    I find this obfuscation kind of like when the trolls try to hide the real person of interest.

  26. 8

    I think there is an interesting story in the second graph with the steady rise of up to 50% between 2003 and the peek in 2009, with the result of greater than half the applications coming to a “Final” of rejections.

    It is more than symbolic of the breakdown of working together between the Office and the applicants in finding allowable material in patent applications. This has been ascribed as an Office policy of rejecting itself out of its backlog problems and MUST be understood in the bigger picture of the Taffas defeat.

    The recent spin of attempting to portray a rubber stamp of Approval through some maligning of Kappos and his link to IBM is shown to lack a factual correlation in the data.

    Yes, there really was a reason for Kappos’ quote of “Quality does not equal reject” no matter how hard some try to hide the truth of that statement either through the incessant posting of dubious patents Ina particular art area or through the flat out denial of historical factual information.

  27. 6

    Inventive capacity jealousy (sometimes known as Well,-it-should-take-real-genius) has been rumored to be much lower on the priority lists of certain individuals, and thus far easier to obtain. Often related to self-loathing and bitterness, the symptom has been highly correlated to a high QQ level.

  28. 4

    That’s why so many patent attorneys and their wives filed their own biotech applications when they saw an opportunity to get rich off the hot new trend

    Different barriers to entry. It is generally accepted that unless you have a PhD, you should not be doing biotech patent work. There are very, very, very few of those around. Moreover, to do your own inventing takes the type of research activites that cannot be performed by a patent attorney.

    You can invent plenty of mechanical/electrical/computer related stuff without the need for even an engineering degree. As such, it is much easier for the average patent attorney to come up with inventions involving those technologies.

    Semiconductor work is one that does not lend itself to patent attorneys coming up with their own inventions. Having access to a semiconductor fab isn’t easy. On the other hand, the average joe can program a computer or develop a golf-training device.

  29. 3

    However, what is amazing to me is the dramatic change in practice that we have seen over the past few years.

    Please explain.

  30. 2

    The point of the article was to show that software patents have been treated approximately the same as other patent applications.

    No kidding. Because they are the same, right? That’s why so many patent attorneys and their wives filed their own biotech applications when they saw an opportunity to get rich off the hot new trend.

Comments are closed.