What is in the Final Rejections: Eligibility

by Dennis Crouch

Though there have been some improvements, initial office actions regularly require a fair amount clean-up and fine-tuning. This process often includes rectifying typographical errors and clarifying loose claim language. It’s also common for the examiner to misconstrue aspects of the invention.  However, by the time the final rejection stage is reached, these issues are usually addressed, and the lines of difference are more clearly drawn.  So, for my study here, I decided to look solely at final office actions and ask the question of what percentage of these involve eligibility rejections.

The chart above reports the percentage of final office actions that include an eligibility rejection.  The uppermost (orange dashed) line on the chart represents applications examined by the eCommerce art units (specifically the 3620s, 3680s, and 3690s). Notably, in 2023 more than 90% of the final rejections for the 3690s (finance, banking, insurance) included an eligibility rejection.  The middle (grey) line in the chart pertains to non-financial data processing inventions (2120s). The main focus of 2120s examination has now shifted predominantly to issues related to AI  simulation and modeling. The third and last (double blue) line in the chart represents all applications. This totals line illustrates that around 9% of all final rejections include an eligibility focus. This is an increase from the 2% rate in 2012.  Notably, about half of the USPTO art units have an eligibility rejection rate of around 0%.

One issue with my study here approach is the time factor. The majority of cases that receive a final rejection are at least three years removed from their effective filing date, and so more recent filings might be doing a better job.  This study was conducted based on a sample size of 100,000 final rejections and is limited only to publicly available files.

105 thoughts on “What is in the Final Rejections: Eligibility

  1. 14

    I do not if there is any particular significance to this observation, but the experience that Kevin Ahlstrom describes in soliciting invention disclosures from women at Meta sounds a lot like the experience that Ray Millien describes in soliciting invention disclosures from Swedes at Volvo. I wonder if the same techniques were also tried at Volvo?

  2. 13

    Nice caveats:

    This study was conducted based on a sample size of 100,000 final rejections and is limited only to publicly available files.

    How was this 100K sample pulled across technology art units? Proportional to count of applications per unit?

    Any (side) effort for estimating “flavor” of the non-pub effect (size? Art unit distribution?)

    Will this apply (be attempted to be applied) to patent law…?

    link to scotusblog.com

  3. 12

    Ugh,

    Another in the virtue signaling serious of “equity.” (the one newer post than this one — and of course, the comments are turned off)

    That article is SO full of B$.

    If a change in the system works to invigorate more participation, why in the world would it be limited to under represented groups?

    You cannot do that and then try to claim “improvements in relative degrees of participation” because you have changed the playing field and are now comparing apples to oranges.

    1. 12.1

      “If we leik totally hold women’s hands through inventing, then they become inventors and start to see themselves as inventors! Because it “worked” this one time at band camp (sample size 100)! Is going to be gendar parity nao!” -soiboi simp no. 1102502

      ^ Leftoids getting their hopes up for the thousandth gazillionth time. When it mysteriously costs a lot, relies entirely on men gibbing (probably double what the actual inventing required), doesn’t scale, and their one time at band camp was a fluke they’ll blame the ebil white cis hetero christian capitalist patriarchs and “systems of systemical oppression” as well as “republican kneecapping” (with no republicans anywhere in sight other than volunteering for the latest leftoid social engineering project) like always.

      Truly, all this sht is powered by people writing feel good articles and concocting social engineering plans like this that mysteriously fail 95%+ of the time and half fail the another 4% of the time. When will leftoids learn? There is no power on earth that is going to make women en masse want to engage in relatively low ROI, high up front cost activities (uncool man) when they could be on tiktok, onlyfans, twitch, youtube, tinder etc. making a half decent dollar (supar cool man) at relatively high ROI (super low investment, especially up front, very cool man) or meaningful work (family stuff etc). Save for “the whip” (metaphorical or real), which they supposedly don’t like. Teeny Tiny carrots like this will not scale. All the best of luck, we’ve only been trying the same thing for 30 years now, but I’m sure success is just around the corner!

    2. 12.2

      The gap is not in innovation.

      The gap might be with people willing to aggressively step forward to claim credit for work which might or might not even be theirs.

      Besides, no one knows who deserves credit for a corporate invention until the claims are written.

      The gap might be with managers who don’t know anything about the technology in the company’s products. They should know if something is innovative and maybe worth a patent application.

      The real gap is with the corporate patent attorney/law firm not asking the team “who did what on this invention?”

      Creative people who are developing actual products for the company’s business should not also have to be experts on patent law.

      The USPTO and the Courts certainly aren’t.

      1. 12.2.1

        NOiP,

        While correct, the actual treatment of, “Besides, no one knows who deserves credit for a corporate invention until the claims are written.” tends to vary considerably (by firm, national custom, and both).

        Generally speaking, all potential contributors are captured at the get-go. This way, such things as claim amendments and continuation claim sets tend to be “less burdensome” to handle (and also generally, we rely on explicit directions from inside counsel as to any changes to listed inventors).

        Separately, and why not since it’s Friday, I would point to the article (and comment) at:

        to pose an open question in this forum:

        link to ipwatchdog.com

        Is the ‘head-in-the-sand’ treatment of AI as (partial) inventor violating 17 USC 506(e)?

  4. 11

    One more observation: the E-commerce rejection rate shoots up to 90% in the year following Alice, and then has been more-or-less steadily declining ever since. Maybe this reflects some improvement in the knowledge and skill of the patent bar over time, but I suspect that it mostly reflects a compositional effect. The knowledge is gradually diffusing through the industry that it is not worth the bother of trying to get a U.S. patent on an e-commerce invention, such that ideas that might once have been filed are not being filed. The result is that the few that are still being filed are the strongest and best of the lot.

    1. 11.1

      >>The result is that the few that are still being filed are the strongest and best of the lot.

      If you define “strongest and best” as the ones most likely to overcome 101, which has little to do with the value or merit of the invention.

      Also, those of us that have been doing this for 20 years in the ecommerce space may have different opinions. The 36xx AUs have lightened up a bit.

      1. 11.1.1

        Greg is too busy channeling the likes of Yglesias to bother with making cogent patent posts.

  5. 10

    Addy >>Although many prep/pros practitioners insist they can write and prosecute software-based innovation apps, precious few actually can.

    The typical practitioner will write the typical “use a computer to compare two things and make a decision based on the comparison” type claims, or some other approach that claims the “what” (i.e., what it does) of the innovation and not the “how” (i.e., how it works – algorithmic or other computational structural elements).

    They get the eligibility rejection and continue to argue the “what” back at the Examiner and then complain the Examiner is digging in their heels; no…the Examiner is doing their job. In a number of cases, the Examiner gives in and allows these claims to issue, and they predictably fall in litigation. Did “the patent system” fail? Yes and No. No – the practitioner failed and the Examiner failed. Yes – the courts did their job.<<

    Addy's post goes straight to the heart of the controversy. Addy is 100 percent wrong. The courts leave out the interpretation of the claims by a person skilled in the art and ignore the ladders of abstraction.

    1. 10.1

      Of course, if a skilled artisan looks at the claim and laughs at how ridiculously obvious it is because it describes basic logic at a 10,000 foot level (but uses some made-up trendy lingo), then THAT skilled artisan isn’t a True Skilled Artisan but some plainly some kind of anti-patent woke communist.

      1. 10.1.1

        Lots of nonsense again from “The Prophet”. And note that what he says goes to 103 and not 101.

        1. 10.1.1.1

          There is no doubt that Malcolm IS a Woke “communist.”

          There is also no doubt that Malcolm is also anti-patent.

          But do note how he attempts to project his own one-bucketing tendencies on you.

          Clearly, one can have anti-patent tendencies and NOT be a Woke “communist.”

          His “reply” — as usual — says more about him than about the person he is aiming his reply at.

          1. 10.1.1.1.1

            “Clearly, one can have anti-patent tendencies and NOT be a Woke “communist.””

            And vice versa.

            1. 10.1.1.1.1.1

              “And vice versa”

              While I certainly ascribe to NOT placing everyone in one bucket based on a single attribute, the “vice versa” path is a little more difficult to obtain, as one who is a Woke “communist” most likely will detest the very notion of personal property (and heaven forbid, using that personal property for one’s own self-enrichment).

              Malcolm of course routinely displays the cognitive dissonance of one who detests patents, while his (professed) professional livelihood is in helping to create something he personally detests.

          2. 10.1.1.1.2

            “ Clearly, one can have anti-patent tendencies and NOT be a Woke “communist.”

            Tell it to Wiper, the paid commenter who molests kittens at PetSmart.

  6. 9

    Notably, about half of the USPTO art units have an eligibility rejection rate of around 0%.

    That IS notable.

    Are there any surprises (to you) of Art units with 0% that you might think should have an appreciable amount?

    1. 9.1

      Just for the record I haven’t sent a 101 eligibility rejection out in probably a decade. Even then I only ever did like 5 cases that ever needed one. Maybe 10 total.

  7. 8

    That chart doesn’t appear to be meaningful. We need to see how many first actions have 101 rejections and then chart the first final office actions that have 101 rejections to see if the hypothesis has merit.

    101 rejections are just weird. The PTO shifts around 101 all the time.

  8. 7

    “However, by the time the final rejection stage is reached, these issues are usually addressed, and the lines of difference are more clearly drawn.”

    That’s a sentence that could only have been written by someone who hasn’t prosecuted a patent application at the USPTO in over a decade. Me, more and more often I’m seeing sloppy work from examiners. Failure to analyze the references cited. Calling something in a reference “x” when the reference calls it “y” and clearly states it doesn’t do what the examiner says it does. Referring to a publication by name when that publication hasn’t previously been introduced and so no one knows to what the examiner refers. Misstatement of obviousness doctrines. Sentences that aren’t written in English. I could go on and on.

    Those examiner mistakes – and intransigence in admitting them – needlessly cost applicants lots of time and lots of money.

    But it satisfies the examiners’ union, so all is well.

    1. 7.2

      AM,

      While I definitely recognize your point, individual mileage may vary, as I have actually seen of late much more diligent and reasonable examiners, with fairly solid office actions.

      Mind you, I still see some doozies (especially on 101), and I do not doubt that the discrepancies that you point out still exist on a large scale (and thus, ANY attempt at legislating that so-called “patent quality” NEEDS to be more explicit as to what should be being aimed at to be “patent examination quality,’ [and ‘Just Say No’ IS NOT QUALITY examination]), I cannot say that the overall tend has been sharply negative.

    2. 7.3

      “However, by the time the final rejection stage is reached, these issues are usually addressed, and the lines of difference are more clearly drawn.”

      That’s a sentence that could only have been written by someone who hasn’t prosecuted a patent application at the USPTO in over a decade.

      I think that “more clearly drawn” is accurate as far as it goes (it is more clearly drawn than the first CTNF), but you are right that this is not really saying much. The first CTFR is usually your second office action, and it is still pretty sketchy at that stage. The issues of real contention do not usually come well into focus until after the first RCE.

    3. 7.4

      “But it satisfies the examiners’ union, so all is well.”

      It also satisfies management. Every USPTO administration for over a decade has chosen cost and speed to the detriment of quality. Examiners are now paid 18% less than they were in 2010.

      1. 7.4.1

        Examiners are now paid 18% less than they were in 2010.

        That seems to scream for a little substantive backing.

        1. 7.4.1.1

          “That seems to scream for a little substantive backing.”

          We’re apparently getting underpaid by alot. Supposedly there was even some sort of pay adjustment agreed to like 20 years ago that never went into effect (because supposedly they had agreed only to implement it if it was justified by not being able to hire people). Fact is, if they ever have super hard time hiring and keeping people it’ll be because they’re only wanting to keep people that want to make a “show effort” rather than a quality effort as quality effort takes way too long per case. It won’t be because of 10-20k dollars.

          1. 7.4.1.1.1

            The “feeling” of getting underpaid (even if “by a lot”), is a feeling whereas “we are making 18% less than a certain time ago” is an objective assertion, and one that lends itself to factual analysis.

            Ben’s statement is incredulous.

            1. 7.4.1.1.1.1

              I know there’s no sense responding to the him, but for anyone else who thought I was exagerating

              link to opm.gov

              link to opm.gov

              link to data.bls.gov

              Take a GS14-1.

              2023 pay 2023 Dollars:132878
              2010 Pay 2010 Dollars:112647
              2010 Pay 2023 Dollars:157706

              (132878-157706)/132878*100%=-18.7%

              1. 7.4.1.1.1.1.1

                First attempt censored. Second attempt lost the links. The examiner special pay rate table is 0576 if anyone wants to find the links themselves.

              2. 7.4.1.1.1.1.2

                I know there’s no sense responding to the him, but

                While your response was ham-fisted (more than two hyperlinks does tend to hint one of the filters), the fact of the matter is that my call for substantiation was very much on point.

                Let that sink in.

              3. 7.4.1.1.1.1.3

                That being said, try a straight forward listing of the numbers, and then check your math.

                Your ‘batch’ listing appears to have an “extra” 2010 dollar listing of $112,647.

                1. “try a straight forward listing of the numbers”

                  PEOPLE LOOK AT POST 7.4.1.1.1.1.3

                  anon was confused by a list of three values, two in 2023 dollars and one in 2010 dollars. It wasn’t straight forward enough for him! This guy supposedly has an engineering degree!

                  THAT IS TOO FUNNY

                2. You flubbed the post and want to blame me for not understanding your flub…?

                  Why have three numbers for two years? It is NOT clear (at all) which number does not belong — and it does not take an engineering degree to figure out that choosing one to remove at random would not be appropriate.

                  But this type of “logic” from you does make it believable that you are an examiner.

                3. “You flubbed the post and want to blame me for not understanding your flub…?

                  Why have three numbers for two years? It is NOT clear (at all) which number does not belong —”

                  LOL

                  HE STILL DOESN’T GET IT GUYS!

                  “Why have three numbers for two years”

                  AHAHAHAHAHA

                4. As the saying goes, there are none so blind as those who will not see, and no one is more practiced at closing his eyes as anon. Suffice it to say, the rest of us can follow your math, but if you are hoping that anon is going to own up to his error here, you are going to be waiting a long time.

                5. This IS too funny from Greg:

                  and no one is more practiced at closing his eyes as anon.

                  As Greg has boasted that he cannot even see my posts.

                  That’s not a log in Greg’s eye – that’s the entire forrest.

                  As for Ben, he still wants to laugh at others for his error, even when his error is pointed out nicely (or perhaps simply for more nicely than he deserves).

                  Say “La Vie”

                6. “if you are hoping that anon is going to own up to his error here”

                  I am almost hurt that you think I could be so naive. My purpose in my latter responses was to draw attention to his post so that others could enjoy.

                7. Ben and Greg playing patticake…

                  And for Greg’s hidden link, he should turn into the Zeihan podcast series that goes into extensive depth as to why the US manufacturing is doing what it is doing.

                8. Ben: “ My purpose in my latter responses was to draw attention to his post so that others could enjoy.”

                  Amusing but I’ve seen better (worse). Thank you regardless.

            2. 7.4.1.1.1.2

              “The “feeling” of getting underpaid ”

              Almost everyone in the whole world is underpaid, especially when compared to actual value of money in the 60’s, even 80’s. Objectively. And even with the cost of many things coming down and cheap chinese plastic goods everywhere and cheap indian/viet textiles.

              1. 7.4.1.1.1.2.1

                Ben would rather quip at imagined errors of others than realize his own mistakes.

          2. 7.4.1.1.2

            “We’re apparently getting underpaid by alot.”

            You’re free to seek greener pastures at any time.

            But we all know you won’t.

            1. 7.4.1.1.2.1

              It would help if there were a little bit of honesty and clarity about what the PTO is meant to do.

              The PTO was not chartered for the benefit of the inventors. Rather, the PTO is there to protect the public. That is why it is possible for there to be Art. III standing in cases like Diamond [PTO director] v. Diehr [inventor]. The PTO does not exist to further the inventor’s interests, so that is why there can be real adversity between the PTO and the inventor.

              So, what is the PTO meant to do for the public? The PTO is there to prevent claims from going to grant that would deprive the public of its right to practice the prior art. But this side of the eschaton, any operation is going to have some rate of imperfection. The closer you want to get to perfection (you will never get all the way there), the more it will cost.

              How much is it worth to achieve that little bit closer to perfection? To my mind, the cost/error point that we presently occupy seems fairly sound to me. We successfully weed out people trying to patent fire, or the transistor, or aspirin for fairly modest costs. The closer the claim is to the border between the genuinely new and the prior art, however, the more resource intensive it becomes to make that distinction accurately and fairly.

              The examination system, therefore, does not necessarily weed out people who are trying to patent the technology that is just barely on the wrong side of the border between new and old. Precisely because that is expensive, however, we can leave it up to private actors to police that border (in litigation and IPRs/PGRs). They have more at stake in those disputes, so it makes more sense for them to bear the costs of that discernment, rather than to socialize those costs during the examination stage.

              1. 7.4.1.1.2.1.1

                This:

                The PTO was not chartered for the benefit of the inventors. Rather, the PTO is there to protect the public.

                Is a critical misunderstanding – no doubt colored by Greg’s interaction with the Deny Deny Deny mindset of the Office.

                The plain truth is much more universally cooperative.

                Examiners are tasked with examining applications to meet the law. Those that pass are to be allowed.

                This function IS for both the public AND inventors, just as the Quid Pro Quo is an EVEN deal, benefitting the public and inventors.

                This notion of “protect the public” directly links to the inordinate Catcher of the Rye syndrome in which the Office deems that it is FAR better to err by saying “no” to a proper application than saying “yes” to a flawed one.

                1. The PTO is there for the benefit of “the public”. End of story.

                  The patents clause is there for the benefit of the public. The PTO is there to implement the patent clause.

                  The public benefits when patentable inventions receive the grant of a patent. The public loses, when inventions that are not eligible/patentable receive the grant of a patent.

                  Thus, the PTO is doomed to be in perpetuity a gatekeeper, sorting sheep from goats. When it discharges its statutory duty conscientiously and competently, it is destined to offend those who fondly suppose that they are the proud owner of a sheep that turns out to be a goat.

                  Labelling goats as sheep does the public a disservice. IPR’s are a valiant attempt, belatedly to label inventions more accurately. The public (represented in Congress) should strive to make the IPR system more efficient and effective. Is that not what we are currently observing?

                2. As usual, MaxDrei is 100% wrong.

                  It’s as if the notion of Quid Pro Quo just does not register, but hey, he will keep on being “provocative.”

              2. 7.4.1.1.2.1.3

                You seem to think that “protect the public” means “deny patents”. The PTO’s job is to issue patents on patentable inventions, because the failure to do so imposes a cost on society, just as the issuance patents that shouldn’t be issued imposes a cost.

                This, to me, is one of the biggest losses of the Dudas era: experienced examiners who understood what it meant to work with an applicant to identify what’s patentable in the application.

                1. You seem to think that “protect the public” means “deny patents”.

                  My apologies if my earlier post left you with that impression. That was not the meaning that I intended to convey. I meant to convey that the PTO protects the public by denying clearly invalid claims.

                  The whole thrust of my discussion of trade-offs between cost and perfection was to convey that it is fine if the PTO lets some of the harder-to-spot errors slip through. I definitely do not mean to convey that it is somehow better to hold back a valid claim than to let an invalid claim through.

                  The PTO examination process should function as a coarse filter to stop the progress of the really obviously invalid claims (claims to fire, to aspirin, to the transistor, etc). We can let the courts and the IPRs system correct the harder-to-spot errors on the back end (where the correction is funded by those who have the biggest stake in the matter).

                2. Greg continues to miss…

                  I meant to convey that the PTO protects the public by denying clearly invalid claims.

                  The (over) emphasis by Greg is a clear sign of “Catcher in the Rye” syndrome, whereas the simple and direct mutual view of Quid Pro Quo covers the situation in which examination under the law simply dictates that the offered quo is not sufficient for the desired Quid — without invoking the Salinger mindset/psychosis.

            2. 7.4.1.1.2.2

              “But we all know you won’t.”

              I actually have looked around a bit. But in general “greener pastures” aren’t all that much of a thing. Keep hearing about everyone making a gagillion dollars here and there, but the only positions I’m seeing are paying around the same, sometimes a bit less, and they all claim paying more is basically highway robbery. And, it might actually be. I’m like top 9% of salaries in the US already iirc, though it depends on what study or who you’re asking, it could be top 8% or way down at top 13%.

      2. 7.4.2

        Wow, 18% less. When I was an examiner, school teachers in the area were making $80k a year while examiners only made $50k to $60k. The examiners union is largely toothless. When I was an examiner, I lived in section 8 welfare housing and couldn’t afford a car. I suspect the high rejection rate in final OAs has a lot to due with turn over, i.e. examiners quitting the PTO. I suspect new examiners don’t do thorough examination and the primaries have to clean up the mess. Thanks Ben for the actually meaningful input from the trenches rather than political rants.

      3. 7.4.3

        Lulz.

        Stream a PPAC meeting and listen to all of the career PTO officials tell you how they are doing record breaking outstanding quality work at every level. “The Master Review Form surveys indicate that 98.57% of 103 rejections contain a motivation to combine!!!!”

        Ok.

        More lulz.

    4. 7.5

      “But it satisfies the examiners’ union, so all is well.”

      I lelzed. Has nothing to do with the muh useless union.

      “I am seeing a definite decline in the quality of OAs.”

      I support that. As the workload of cases is never ending and has basically reached an initial breaking point for many examiners. I’ve got old heads reporting that they have more and more “impossible” cases. And I myself have the same. That is, cases that are impossible to examine in anywhere near the time allotted to a good degree of quality. Some additional people will just start doing what mgmt has always actually wanted to pay for (half-assing it). By paying for I mean in terms of hours spent and allotted per case.

      1. 7.5.1

        I think 6 is right because there was an IG or GAO report that examination was a fraud back about 10 years ago since examiners were not given near enough time to do the job. Fundamentally, the PTO’s self funded monopoly factor is not serving the public by issuing bad patents that screw up commerce and clog the court system since the PTO is only interested in collecting the maintenance fees by issuing as many patents as possible. (Examination is not a priority.) I think the PTO should be funded from the general budget and not self funding since this creates a giant slush fund for the executive branch in violation of Congress’s tax and spending authority. I can’t talk about the abuses of the slush fund but it is real.

        1. 7.5.1.1

          Would be nice to find a link to that report from a decade ago.

          This of course confirms my own decade’s long call for any “patent quality” focus to be more appropriately focused on ‘patent examination quality.’

          Reminder: “Just say No” is NOT examination quality.

          I also note that my prior remarks as to the utter banality of the Examiner’s Union is again front and center.

          How does the bulk of examiners put up with this ongoing representation mess (is it a perpetual excuse for whining)…?

        2. 7.5.1.2

          Changing the funding mechanism of the PTO will have zero impact on patent examination.

  9. 6

    However, by the time the final rejection stage is reached, these issues are usually addressed, and the lines of difference are more clearly drawn.
    This comment is more aspirational than factual. A final office action almost immediately follows a first action on the merits. The lines of difference, in the vast majority of instances, have just started to come into view.

    The uppermost (orange dashed) line on the chart represents applications examined by the eCommerce art units (specifically the 3620s, 3680s, and 3690s).
    FYI, just because an invention was classified in the eCommerce art units does not mean that the invention is directed to an improvement to commerce. In most instances that I’m aware of, the improvement is to computer technology that is used for e-commerce. Unfortunately, most examiners in TC3600 presume that if something was classified to their art unit, then it is, by default, not patent eligible.

    The majority of cases that receive a final rejection are at least three years removed from their effective filing date, and so more recent filings might be doing a better job.
    One would hope that 9 years after Alice that patent attorneys might have learned something. However, despite those 9 years the rejection rate in 3600 is 70% and the rejection rate of AI-related inventions is increasing — not decreasing.

    That rejection rate says a lot about the clarity of the law that attorneys and their clients still cannot figure out how to draft claims that don’t get snagged by 35 USC 101.

    1. 6.1

      Woah there WT. You aver that the ongoing high rate of 101 rejection is because, 9 yrs after Alice, those seeking patent rights still cannot figure out how to write 101-compliant claims.

      I would suggest otherwise.

      I would suggest that they continue to push at the eligibility envelope even while they know that their efforts are likely to fail. This repetitive banging of heads against a brick wall is because the prospect of success, however remote, is perceived as worth the effort. Put another way, anybody in need of investment refreshment can’t afford to do anything else but push that envelope.

      Why is that, when the courts are not enforcing such futile claims?

      Perhaps because a duly issued patent, with its high presumption of validity, even now continues to impress patentee’s investors and, indeed, has huge intimidatory effect on the threatened party’s investors, sufficient to extract lorry loads of tribute rather than get dragged into, and smeared by, near-endless litigation?

      1. 6.1.1

        MaxDrei – how long have you Fricken been on these US patent boards, ever proclaiming that you really do not know US prosecution whenever pressed on any particular point, and YET, venture forth such an uninformed opinion as you do here….

        Unglaublich.

      2. 6.1.2

        I would suggest that they continue to push at the eligibility envelope even while they know that their efforts are likely to fail.
        Instead of speculating (i.e., “I would suggest”) perhaps you should ask some real US patents attorneys whether or not they are trying “to push at the eligibility envelope.” From my perspective, few US patent attorneys want to deal with 101 rejections. We don’t want to be anywhere near the line between patent eligible and patent ineligible.

        This repetitive banging of heads against a brick wall is because the prospect of success, however remote, is perceived as worth the effort.
        Again, TOTAL SPECULATION on your part. Neither I nor my clients are interested in banging our heads up against the wall. They do not have an infinite budget of cash or patience. My clients want quick prosecution — not drawn out prosecution.

        Perhaps because a duly issued patent, with its high presumption of validity, even now continues to impress patentee’s investors and, indeed, has huge intimidatory effect on the threatened party’s investors, sufficient to extract lorry loads of tribute rather than get dragged into, and smeared by, near-endless litigation?
        You really haven’t been paying attention the last 10 years, have you? There is no high presumption of validity. It is a farce that doesn’t exist in reality. And very few are threatened by patent lawsuits these days. Patents get killed under 101 at via IPRs ALL THE TIME. We’ve got patents to garage door openers, cameras, and prop shafts being declared as patent ineligible. Very few are afraid of patents these days. Moreover, very few are “extract[ing] lorry loads of tribute” on these.

        Seriously. Wake up. It isn’t 2005 anymore.

        I had you on ignore for the longest time, and now I regret taking you off. Your opinions are terribly uniformed. I hope you don’t talk to real attorneys like this IRL.

        1. 6.1.2.1

          From my perspective, few US patent attorneys want to deal with 101 rejections. We don’t want to be anywhere near the line between patent eligible and patent ineligible.

          This is, in my experience, true as far as it goes, but there is more to be said. Yes, the patent attorney may not want to push the boundary, but often our clients do. At that point, our desire to stay away from the boundary does not much matter.

          That does not mean that we have learned nothing in the years since Mayo/Alice. It does not even mean that our clients have learned nothing (although sometimes that is true). Rather, it mostly means that our clients have a business objective that is served by maintaining (at least for a time) a public position that they are entitled to a claim that is objectively hard to achieve under Mayo/Alice.

          1. 6.1.2.1.1

            Thank you, Dozens. You surely must be right. Once again, somebody takes time to explain to me something that is obvious to everybody else here. I am most appreciative.

            1. 6.1.2.1.1.1

              The comment at 6.1.2.1.1 is in the wrong place. My reply to Dozens 6.1.2.1 is also thanks, that you confirm what I was suspecting and that you express yourself on the point more clearly than I did.

        2. 6.1.2.2

          WT, I post here with the intention of getting informative answers. I am grateful for your having taken the time to answer so informatively.

          When I talk to real correspondent attorneys, it is under my real name. My demeanour then is quite different, of course. Here, under a pseudonym, I indulge myself, being provocative. The downside of it is that some reactions are not informative at all.

          Feel free to take any measures you like, to keep yourself from ever again getting provoked.

          You say that patents are being 101-slaughtered all the time by IPR. But I had thought that the only IPR’s being filed are by those who feel threatened by a patent duly issued by the USPTO. Are these people imagining things? What is it that I’m not appreciating here?

          1. 6.1.2.2.1

            You say that patents are being 101-slaughtered all the time by IPR. But I had thought that the only IPR’s being filed are by those who feel threatened by a patent duly issued by the USPTO. Are these people imagining things? What is it that I’m not appreciating here?

            I know that Wt wrote that “[p]atents get killed under 101 at via IPRs ALL THE TIME,” (emphasis added), but I feel fairly certain that he means “101 or IPRs.” One is not allowed to litigate 101 issues in an IPR (only 102/103). Probably, he typed “ot” (T being right next to the R on a QWERTY keyboard) and autocorrect took that for “at.”

            1. 6.1.2.2.1.2

              My thanks for this explanation, Dozens, has appeared at 6.1.2.1.1.

              Sorry for the mis-placement.

            2. 6.1.2.2.1.3

              I feel fairly certain that he means …
              there should have been a “the District Court or” following the “at” so as to read “get killed under 101 at the District Court or via IPRs.” Not sure where it went. I appreciate you picking up on my intent.

          2. 6.1.2.2.2

            When I talk to real correspondent attorneys, it is under my real name. My demeanour then is quite different, of course. Here, under a pseudonym, I indulge myself, being provocative. The downside of it is that some reactions are not informative at all.

            No.

            The downside is that you are simply awful at being ‘provocative,’ and your comments (nearly always repeating a same B$ view) are merely you throwing
            C
            R
            A
            P
            against the wall.

    2. 6.2

      One would hope that 9 years after Alice that patent attorneys might have learned something. However, despite those 9 years the rejection rate in 3600 is 70% and the rejection rate of AI-related inventions is increasing — not decreasing.

      That rejection rate says a lot about the clarity of the law that attorneys and their clients still cannot figure out how to draft claims that don’t get snagged by 35 USC 101.

      I carry no brief for Alice, but I do not think that your conclusion really follows from the data. It is true that the line for e-commerce is still way up, but the line for data processing is only increasingly lately. It shoots up post Alice, then turns on a dime in 2018 and declines most of the way back to its pre-Alice steady state, then turns again in 2020 and starts rocketing up in 2022.

      Well, what happened in 2018? What happened in 2022? I think that we all know what happened in each of those years. Michelle Lee held over from the Obama administration to the Trump administration for several months, so Pres. Trump did not end up appointing a PTO director until late 2017, and Dir. Iancu did not take office until early 2018. Sure enough, that is exactly the point at which the direction of the slope inverted. Similarly, 2022 was when Dir. Vidal took over.

      In other words, it looks like the knowledge and skill of the patent bar has a lot less to do with the rate of U.S. eligibility rejections than the partisan valence of the administration in power and the personal identity of the PTO director. The data say less about either the clarity of the law or the knowledge and skill of the patent bar than they say about the nature of the political coalitions who stand behind each of the two major parties in our country.

      1. 6.2.1

        In other words, it looks like the knowledge and skill of the patent bar has a lot less to do with the rate of U.S. eligibility rejections than the partisan valence of the administration in power and the personal identity of the PTO director.
        You make some valid points, but they do not impact the validity of my statement.

        Let me restate it this way. 35 USC 112(b) is known standard. While it got some reworking awhile back, the USPTO still uses their old standard (not Nautilus). Notably, because it is known, patent attorneys know what they need to do to overcome it. While a first office action may reveals some typos or inadvertent antecedent basis issues, by the time the action goes final, I would wager that 95% of those actions do not include 112(b) rejections. This is a characteristic of a known and well-understood standard.

        In contrast, 9 years after Alice we are still seeing substantially higher rates of rejections based upon 35 USC 101 than 35 USC 112(b). The applications getting rejected today were filed 4 to 6 years ago. The attorneys had plenty of time to react to Alice. So why are we still getting high levels of 101 rejections? The reason is simple — the law is about as clear as effluent from a pig farm.

        1. 6.2.1.1

          9 years after Alice we are still seeing substantially higher rates of rejections based upon 35 USC 101 than 35 USC 112(b)… So why are we still getting high levels of 101 rejections? The reason is simple — the law is about as clear as effluent from a pig farm.

          I used to think that Alice was unclear, but that criticism no longer appears accurate. Nine years of case law from the CAFC yields the depressingly clear result that the patentee loses. Nothing unclear about this, even it is unsound as a matter of public policy, or unfaithful to the statutory text.

          I gather, however, that you mean to say that after nine years the Alice standard is less clear than other standards that the SCOTUS has articulated around that time or since then in the patent context. I can certainly agree with that. I think that the wretched implications of Alice are fairly clear by now, but I can agree that they are not as clear as those of Nautilus or Octane Fitness* (etc.).

          * Which, ironically enough, should be less clear, given the amorphous “totality of circumstances” test that was thrown at us.

          1. 6.2.1.1.1

            I used to think that Alice was unclear, but that criticism no longer appears accurate. Nine years of case law from the CAFC yields the depressingly clear result that the patentee loses.
            Let’s unpack that statement for a second. You said that the “patentee” loses. To be a “patentee” necessarily requires that a patent has to have been issued and my guess is that the majority of non-expired patents have now been examined under Alice. This means that the Federal Circuit, in many instances, are now invalidating patents that the USPTO granted under Alice. As such, the agency empowered to apply the patent laws cannot apply 101 correctly. Is that not indicative of an unclear law?

            The fact that the Federal Circuit says, in most instances, if a colorable 101 argument can be made then the patent is invalid doesn’t take away from the fact that how that law is applied — not just by the Federal Circuit but by the USPTO — is also unclear.

            even it is unsound as a matter of public policy, or unfaithful to the statutory text.
            No disagreement there.

      2. 6.2.2

        Please pardon potential rePeat (filter)…

        Your comment is awaiting moderation.

        June 28, 2023 at 5:45 pm

        t”u”rns

        With Greg’s penchant for feeding himself Sprint Left propaganda, turning to Krugman is of no surprise.

    3. 6.3

      “In most instances that I’m aware of, the improvement is to computer technology that is used for e-commerce.”

      In most instances I’m aware of, attorneys arguing that an application classified in eCommerce contains an improvement to computer technology are wishcasting.

      “It is difficult to get a man to understand something, when his salary depends on his not understanding it.”

      1. 6.3.2

        In most instances I’m aware of, attorneys arguing that an application classified in eCommerce contains an improvement to computer technology are wishcasting.
        Unless you are examining in those art units, you are just speculating and/or working with sparse anecdotal evidence.

  10. 4

    Is it possible to show the rate of 101 rejections for [All Apps] – [E-Commerce + Data Processing]? Failing that, is it possible to mention what percentage of all apps is accounted by E-commerce and what percentage is accounted by data processing?

  11. 3

    Hmm … so … AI inventions not smart enough to hippity-hop over the eligibility morass?

    Is artificial intelligence perhaps … not so intelligent after all?

    1. 2.1

      The law is being applied again instead of ignored. Good.

      My longer comment about the predictably short lifetime of “drop this trendy buzzword into the claims” strategery is stuck in the filter.

  12. 1

    “The main focus of 2020s examination has now shifted predominantly to issues related to AI simulation and modeling.”

    Using a computer is … using a computer and it doesn’t matter what trendy buzzword you use if there is no “meat” on the bones . Tossing a trendy buzzword like “AI” into one’s spec and claims is a time-tested strategy that might increase your chances of obtaining an invalid patent for a couple years but won’t even get you that far once the PTO sees through the ploy (because it’s swamped with “do it using AI” j u n k).

    Anyway, the key point here is that — predictably — data processing dreck is being selectively hammered under 101, and it doesn’t matter what buzzword you use. You want to avoid an eligibility rejection? Invent something more than “use a computer to compare two things and make a decision based on the comparison”. And then claim it properly, using objective structural terms to distinguish the invention from the prior art. This works almost all the time, as Dennis’ chart here shows.

    Maybe Dennis can work on that chart which shows that progress in computing technology (and everything else!) has hit a brick wall in the US because not enough patents are being granted. Just kidding, Dennis. Hahaha.

    1. 1.1

      Same old same old…

      You want to avoid an eligibility rejection?

      Invent something more than [fill in trite strawman]…

      And then claim it properly, using objective structural terms to distinguish the invention from the prior art. [false solution – as repeatedly shown by my provided counter point that one COULD do this with an unrecognizeable and unexaminable “pits and groove” physical media that painstakingly uses the “hallowed” objective structural terms]

      Bottom line (as usual) for Malcolm: understand the innovation of the Fifth Kondratiev wave (even as that wave is now subsiding and making wat for the Sixth Kondratiev wave), and put your money where your mouth is and abstain from any technology that you would deny patent protection to.

    2. 1.2

      +1

      Although many prep/pros practitioners insist they can write and prosecute software-based innovation apps, precious few actually can.

      The typical practitioner will write the typical “use a computer to compare two things and make a decision based on the comparison” type claims, or some other approach that claims the “what” (i.e., what it does) of the innovation and not the “how” (i.e., how it works – algorithmic or other computational structural elements).

      They get the eligibility rejection and continue to argue the “what” back at the Examiner and then complain the Examiner is digging in their heels; no…the Examiner is doing their job. In a number of cases, the Examiner gives in and allows these claims to issue, and they predictably fall in litigation. Did “the patent system” fail? Yes and No. No – the practitioner failed and the Examiner failed. Yes – the courts did their job.

      1. 1.2.1

        +1 (Addy) and +1 (anon)

        Addy: Would you consider providing 2 or 3 of such “how it works” issued patents / claim sets? If you prefer, they don’t have to be your own.

        Would be helpful to many who visit here; including many prep/pros practitioners.

      2. 1.2.2

        >>Yes – the courts did their job.

        Total nonsense. The “how” is almost always in the applications and claims of the invalidated software patents. The ladders of abstraction are ignored by the courts. And the courts fabricate nonsense like in Williamson that a complete algorithm is needed when NO person skilled in the art would say that.

        Etc.

        1. 1.2.2.1

          Who cares if those skilled in the art do not think that a complete algorithm is necessary? Williamson (following Aristocrat Techs) is a statutory construction case regarding the proper construction of §112(f). What do those of ordinary skill know about construing statutes? They have no more insight into discerning the will of Congress than the CAFC judges have about how to program a computer to place a pop-up ad.

          1. 1.2.2.1.1

            Total nonsense Greg. The claims and specification are supposed to be viewed through the eyes of a person of ordinary skill in the art.

            Williamson is saying that a few lines describing a change to an algorithm or method known to POSITA is insufficient. That is 100 percent wrong and judicial activism to invalidate valid patents.

            The interpretation of the statute is supposed to be a process where the disclosure is viewed through the eyes of a POSITA or an any person skilled in the art.

            1. 1.2.2.1.1.1

              Looks like Greg needs another reminder to watch his comments in areas that he has already admitted that he knows nothing about.

              Confusing the issue, he A S Sumes that 112(f) is necessarily in play, while you correctly note that “how” in a specification handles the error of Addy (wanting it instead in the claims).

              1. 1.2.2.1.1.1.1

                … maybe he should spend less time with his hidden signaling of the likes of Drum…

            2. 1.2.2.1.1.2

              The interpretation of the statute is supposed to be a process where the disclosure is viewed through the eyes of a POSITA or an any person skilled in the art.

              The interpretation of a claim is supposed to be done through the eyes of the PHOSITA. The interpretation of a statute is not. While claim construction and statutory construction are both questions of law for the judge, that does not mean that they are the same question.

              If you think that Williamson (or Aristocrat Techs.) was principally a claim construction case, you need to go back and re-read. These were questions about the construction of the statute not the claims (why would the CAFC take the case en banc just to resolve a claim construction dispute?). The PHOSITA has no insight about how to construe a statute.

              1. 1.2.2.1.1.2.1

                The issue in Williamson is whether the few sentences describing an algorithm was sufficient to avoid 112(f). The interpretation of those few sentences is supposed to be through the eyes of a POSITA or any POSITA. What is sufficient to avoid 112(f) after the interpretation of those sentences by a POSITA is what the judge’s role is.

                1. The issue in Williamson is whether the few sentences describing an algorithm was sufficient to avoid 112(f).

                  No. That was not the issue. Go back and re-read Williamson. You cannot “avoid” §112(f) by inserting added detail in your spec.

                  The question in Williamson was whether a functional claim term invokes §112(f) in the first place. Once one determines that §112(f) is invoked, then the sufficiency of disclosure comes into play, but that was not the point on which the CAFC took en banc review. They took en banc review to determine whether §112(f) was invoked in the first place. That is a question of statutory construction, not claim construction.

                2. And the claims are supposed to be interpreted in view of the specification, which is supposed to be interpreted how a POSITA would view the specification.

                  And whether those sentences were sufficient after the 112(f) determination was supposed to be interpreted how a POSTIA would interpret them.

                  The judge runs the process but is supposed to interpret the claims and the sentences in terms of how a POSTIA would interpret them.

                  The fact that the CAFC did not illustrates they are unfit to serve.

Comments are closed.