Magic Language in Patent Applications

by Dennis Crouch

The Federal Circuit handed down a mixed decision in Chewy, Inc. v. International Business Machines Corp., 2022-1756 (Fed. Cir. Mar. 5, 2024) ChewyvIBM. The district court had ruled against the patentee (IBM) — finding one patent ineligible and the other not infringed.  On appeal, the Federal Circuit largely affirmed, but found one claim that passes through the pre-trial gauntlet.  The patents at issue were IBM’s U.S. Patent Nos. 7,072,849 and 7,076,443, relating to improvements in web-based advertising.  On remand, a jury may need to decide whether claim 12 of the ‘849 patent is valid and infringed.

The first half of the post focuses on eligibility and is fairly standard.  The second half of the post is what all patent prosecutors need to read because it delves into “magic language” – binding statements – in describing the invention.

Let’s start with eligibility of the ‘443 patent. The Federal Circuit affirmed the district court’s ruling that asserted claims 13, 15, 16, and 17 are ineligible under 35 U.S.C. § 101 as directed to the abstract idea of “identifying advertisements based on search results.” According to the court, the claims recite the concept “without any specificity as to how this is accomplished.”

Lets look at the claims here for a moment.  Claims 15-17 of the ‘443 patent recite a method for providing advertisements related to search result items. The method requires several steps. First, the search result items are matched with related advertisements, and the search result items that have matched advertisements are designated in some way. However, at this initial step, the claim does not identify how the matching or designation occurs. Second, for each designated search result item, a corresponding graphical user interface (GUI) is provided to allow the user to select that search result item.  This appears to be likely a pop-up screen, but again the claim does not recite how this GUI might be provided.  Third, when the user selects one of the designated search result items via its GUI, the related advertisements for that search result item are retrieved. Finally, the retrieved advertisements are formatted and displayed to the user. Claim 16 adds a further step of assigning an identifier to the user when the user submits the initial search query. This identifier allows the system to track the user’s actions (e.g., search queries and search result selections) during that search session. Claim 17 specifies that the related advertisements are advertisements for related products.

15. A method for providing related advertisements for search result items from a search of an information repository, comprising:

matching said search result items to said related advertisements;

designating each of said search result items that have said related advertisements matched therewith;

providing a corresponding graphical user interface for each of said search result items so designated for subsequent user selection;

searching and retrieving said related advertisements for one of said search result items when said corresponding graphical user interface is selected by a user;

and, formatting and displaying said related advertisements upon selection.

16. The method of claim 15 further comprising assigning an identifier for said user when said user submits a query to said information repository.

17. The method of claim 15 wherein said related advertisements comprise related product advertisements.

At Alice step two, the court rejected IBM’s arguments that the claims contain inventive concepts. For instance, the claimed “information repository” for correlating ads with search results is merely a generic database performing well-known functions. Considering the elements as an ordered combination also did not provide an inventive concept, as the claims “do not recite any specific implementation of the abstract concept” The eligibility ruling was therefore affirmed.

= = =

On the ‘849 patent, the Federal Circuit affirmed the SDNY grant of summary judgment of non-infringement of claims 1, 2, 14, and 18, but reversed as to claim 12.

The ‘849 patent claims are directed to methods for presenting advertising content to users in an interactive service, such as a website. As described in the specification, the invention aim to avoid delays in presenting the requested application data to the user by “pre-fetching” the advertising content and storing it at the user’s computer before the user requests the associated application page. However, the claims themselves do not expressly require pre-fetching. And, importantly for this case, CHEWY’s accused process apparently does not pre-fetch.

Claim 1, which is representative of claims 2, 14, and 18, recites “selectively storing advertising objects at a store established at the reception system.” The Federal Circuit construed this limitation to require pre-fetching.  The court explained that:

The written description consistently describes the invention as including pre-fetching of advertising objects. . . . In light of these repeated descriptions of the present invention, a skilled artisan would understand the claimed invention requires pre-fetching of advertising objects.

During oral arguments, the judges repeatedly commented on how much the patentee had focused on pre-fetching.

Judge Stoll: Doesn’t your patent specification talk about how important it is? I mean, the way I read your patent specification, it says that it’s really important that you be able to pre-send down some advertisements. And the reason why you want to do that is so that your advertisements and your content aren’t being sent at the same time, creating a bandwidth problem.

Judge Moore: My big problem with your argument is really column three, right? And column three is called summary of the invention. So this is not a preferred embodiment. It’s the summary of the invention. And at column three, starting at line five-ish, you say briefly the method for presenting advertising in accordance with this invention. That’s like magic language. You know, the present invention is in accordance with this invention. That’s language that binds you, whether you like it or not. And then it goes through that paragraph and explains that advertising is prefetched. And so that is what bothers me the most is in the summary of your invention. Judge Stoll pointed out quite rightly the objects of your invention and how you’re basically showing the disadvantages of not prefetching advertising in the objects of your invention. Now, in the summary of your invention, you use this invention is language in the context of advertising being prefetched.

The court noted that during prosecution, the patentee stated that “selectively storing advertising objects” means “advertising objects are pre-fetched from the network and cached at the reception system in anticipation of being called for presentation” This prosecution history further supported the district court’s construction requiring “pre-fetching.”

In the end, the court did not find disavowal, but rather simply that the intrinsic evidence supports a particular interpretation of the claim language. “In light of the intrinsic evidence, we conclude ‘selectively storing advertising’ objects requires pre-fetching.”

But, the court contrasted claim 12 which does not recite the “selectively storing” or “pre-fetching” limitation. Instead, claim 12 focuses on targeted advertising by requiring the establishment of “characterizations” for respective users based on compiled user data. These characterizations, which can be based on the user’s interaction history and demographics, are then used to selectively supply advertisements to the user.

The important point here is that the patentee’s characterization of the invention requiring pre-fetching did not carry over into claim 12 because that claim did not have a textual hook within its claim language to associate with pre-fetching. While the specification and prosecution history emphasized the importance of pre-fetching advertising objects, claim 12 focused instead on the separate concept of establishing user-specific characterizations for ad targeting. Thus, the Federal Circuit found a genuine factual dispute regarding infringement of claim 12, as it lacked the pre-fetching limitation that doomed IBM’s infringement case on the other asserted claims.

50 thoughts on “Magic Language in Patent Applications

  1. 10

    Please Pardon Potential re(P)eat…

    Your comment is awaiting moderation.

    March 12, 2024 at 9:49 am

    “complete conception” – please explain how you are attempting to put this into the scenario of claims.

    1. 10.1

      … goes to Paul’s comment at 1.2

  2. 9

    Mindful of that old saw “There are no stupid questions” I ask (from outside the jurisdiction) the following: How much difference to the outcome in this specific case would it have made, if the problematic sub-title “Summary of the Invention” had instead read simply “Summary”?

    I have read only Dennis’s write-up and not the Decision in full but from the write-up I get the impression that it would have made not a blind bit of difference to the outcome.

    Magic words? Really?

    1. 9.1

      Please Pardon Potential re(P)eat…

      Your comment is awaiting moderation.

      March 13, 2024 at 7:47 am

      The saw of, “There are no [s t u p 1 d] questions” carries the notion that any such questions will actually drink from the well of wisdom that such questions may lead to.

  3. 8

    Think of all the shells for Ukraine that could have been bought with the expensive time & energy of the various courts and teams of lawyers to arrive back at the starting point for the establishment of “characterizations” for respective users based on compiled user data. These characterizations, which can be based on the user’s interaction history and demographics, are then used to selectively supply advertisements to the user

    But this level of g rift is toddler stuff compared to the creativity of the real masters.

    link to prospect.org

    1. 8.1

      Everybody knows that “characterizations” are ancient history. “Dynamic user identification objects” is where it’s at, baby.

  4. 7

    Let’s have a laugh at the expense of the children in the comments below who think that the “real problem” with these worthless claims is the way they were construed by the baddies at the CAFC.

    The problem with these claims is that there is no invention. The claims were written by an imbecile, for an imbecile, and they never should have been filed much less rubber-stamped by the paint huffing Examiner at the PTO.

    Shame on everyone involved with this dreck.

    1. 7.1

      It appears the inventors and those involved in the prosecution understood what the invention was. And then someone decided to assert the patent against something that wasn’t the invention, requiring the litigators to make bad arguments. It happens quite often among people who see the purpose of patents as something other than protecting their own specific contribution to the art.

      1. 7.1.1

        This. The plain truth is that you can and should say all kinds of true things about “the invention” as such statements will help prevent the Examiner or others from being confused about what the invention is and what it isn’t.

        Of course, if there isn’t an invention to begin with, sure, just write a bunch of buzzwords into a word salad, kick up dust and pretend that nobody can see you doing that. Then sell it to someone even more dishonest and less intelligent. Then you are ready for the bigtime aka … Texas.

        LOL

        1. 7.1.1.1

          Comment #6 and its progeny is refreshing to read. Its authors will no doubt be condemned here as people who are anti-patent but to me the very opposite seems the case; they are amongst the ones who are striving to support and maintain a properly-functioning patent system.

          1. 7.1.1.1.1

            Sadly, this is only because of your bias and improper understanding of what “properly-functioning” means.

        2. 7.1.1.2

          However, in addition to the known problems of inadvertently narrowing claim scope, that is is overlooking the fact that so many applicants do not really know what their CLAIMABLE “invention” will even be when they file, because they do not normally do, or are paid to do, a really thorough prior art search before preparing the application.

          1. 7.1.1.2.1

            I should have thought, Paul, that it is a flat 100% of applicants who, on the day they file, don’t know how much scope they can validly claim. But that doesn’t stop them extrapolating from their reduced-to-practice embodiment all the way up the ladder of abstraction as far as a single independent claim and then descending the ladder all the way down to the narrowest worthwhile claim conceivable.

            How else shall one balance the interest of the inventor and that of the public, in a properly functioning patent system, other than to require this of every applicant for patent?

            1. 7.1.1.2.1.1

              Again – you appear to have an odd almost ivory-tower-like view of things.

              Sadly, this is only because of your bias and improper understanding of what “properly-functioning” means.

            2. 7.1.1.2.1.2

              Re: “…extrapolating from their reduced-to-practice embodiment all the way up the ladder of abstraction as far as a single independent claim and then descending the ladder all the way down to the narrowest worthwhile claim conceivable.”
              Not a bad description of what the changing claims in the prosecution file of too many applications looks like in going from from filing to issuance.
              But the main topic here is what, or what not, and how to express, what the Specification of the application should say about what might distinguish the “invention” from the prior art. And, of course a specification can Not be so easily changed.

              1. 7.1.1.2.1.2.1

                Nor can the prior art be known – even when an extremely diligent search has been conducted.

                Prof. Crouch ran a study (now a few years back) on how often Examiners came up with their own art, regardless of extensive IDS filings.

              2. 7.1.1.2.1.2.2

                Yes, the debate is about how much freedom an applicant (or patent owner) shall have, to announce to the world, following the post-filing coming to light of one or other prior publication, that “the invention” is not what I said it was in my patent application as filed but, rather, something different from that.

                There is a similar debate raging in Europe, about the degree to which surprising or unexpected effects not flagged up in the patent application as filed can help, post-filing, to evidence patentability over prior art not addressed in the specification as filed. It’s indeed a fine and delicate balance between giving the public reasonable legal certainty while, at the same time, giving the applicant a fair measure of protection for their contribution to the art.

                The way I see it, drafting is a higher level activity than litigating.

                1. that “the invention” is not what I said it was in my patent application as filed but, rather, something different from that.

                  How tone-deaf can you be?

  5. 6

    Just as an FYI bros the crypto bull appears to perhaps be kicking off a bit early. Hope you packed your bags.

  6. 5

    Talking about magic language, there’s people on here telling us cancel culture does not real n sheet. Meanwhile in real life we have actual profitable companies (making GOOD consulting money for their field of lolwriting consultants) employing people literally using it to extract their money.

    link to youtube.com

    1. 5.1

      What’s your point 6?

      Dontcha know that’s the new “center?”

      1. 5.1.1

        “What’s your point 6?”

        Lulz

        1. 5.1.1.1

          Let’s see what 6 has to say before proceeding to “Lulz

          1. 5.1.1.1.1

            That cancel culture does real n sheet, derpa.

            1. 5.1.1.1.1.1

              Hey – I was just drawing out your deeper views – I made no judgment on those views.

              You might have a bone to pick with Smelly, though.

  7. 4

    There is an ancient problem with PTO examinations of applications filed or amended with plural independent claims. That is, examiners missing the one claim that does not have the key limitation that is in all the others. It could be done deliberately by applicants to obtain unintended allowance of a broader claim, or to avoid a restriction and divisional filing requirement. This has been greatly reduced by excess claim fees and compact prosecution in modern times, but can still occur.

    1. 4.1

      Paul — did you use the phrase “present invention” during prosecution If so, where and for what purpose did you use it.

      1. 4.1.1

        The [not actually required] subtitle “Summary of the Invention” was and still is to be particularly avoided. It makes whatever words follows that subtitle look like they are an essential part of the invention, even if not claimed in all claims.
        If you say “the present invention is distinguished by …” or the like to distinguish prior art in prosecution, but did not so distinguish the prior art in all claims, it might also create a prosecution history [file wrapper] estoppel issue for those other claims. I.e., better not to say it unless you ARE willing to own it. Or, avoid it by just pointing out actual specific claim elements and their absence from the examiner-applied art? [Others here may have other suggestions.]
        Nobody is perfect, especially with so many patent applications being filed in a rush under time and/or financial pressures. In contrast to the almost unlimited billing times available to litigation defense counsel staff to studiously pick them apart.

        1. 4.1.1.1

          Best to just say what you mean. Summarize what you believe you invented, write clear claims, and be happy that you have a monopoly on what it is that you actually contributed to the art. The problems arise when you try to claim (or later claim you claimed) what you don’t actually own.

          1. 4.1.1.1.1

            Sorry no – patent profanity is real and is not as you try – something to blame on the applicant.

            1. 4.1.1.1.1.1

              Billy remembers intimately every one of the tens of thousands of patents that were mercilessly rendered invalid in the courts because someone was just too honest about the invention and its relationship to the prior art. Oh, the humanity!

              1. 4.1.1.1.1.1.1

                Your bizarro land comment lands firmly in the zone of meaninglessness.

  8. 3

    The ‘849 is a divisional of a 1989 CIP. The ‘443 was filed in 2000. My recollection is that use of “present invention” was pretty common until 2001.

    Good time for Dennis to do another no-no words/phrases article.

  9. 2

    My big problem with your argument is really column three, right? And column three is called summary of the invention. So this is not a preferred embodiment. It’s the summary of the invention. And at column three, starting at line five-ish, you say briefly the method for presenting advertising in accordance with this invention. That’s like magic language. You know, the present invention is in accordance with this invention. That’s language that binds you, whether you like it or not.
    Then why have claim language if the Federal Circuit is going to rely upon the Summary of the Invention? I thought “the name of the game is the claim”?

    The Summary of the Invention is a useless section in a patent application, and I can see this case as an excuse not to include it in future filings.

    As I have written in the past, the Federal Circuit of today (unlike the Federal Circuit of the past) is constantly looking for new land mines to plant for unwary applicants to stumble upon.

    1. 2.1

      It gets better – the Office requests that “Summary of the Invention” expressly omit language that sounds like claims.

      1. 2.1.1

        Summary is suggested. 37 CFR 1.77: “The specification should include the following sections in order:” As for the claims, the MPEP states “The brief summary of invention should be consistent with the subject matter of the claims.”

        Regardless, I would expect my more sophisticated clients to stop requiring that section.

        1. 2.1.1.1

          Judge Moore: a turnip know more about patent law than that supposed expert.

      2. 2.1.2

        Oops – my bad, I was thinking of “The Abstract.”

        We’ve long ago move to shortening to just “Summary,” and including disclaimer that that section was by no means any limitation on the rest of the specification including claims.

      3. 2.1.3

        Based on my experience, that generally means (the use of the word “means” not intended here to invoke means-plus-function) the Summary shouldn’t include words like “comprises”. Otherwise, re-writing claims in prose is otherwise acceptable with “comprises” replaced by “includes”.

    2. 2.2

      The name of the game is the shame.

    3. 2.3

      And we see why there is so much conflict between Moore and Newman. Judge Newman would have jumped all over that.

    4. 2.4

      The issue was not the title, “Summary of the Invention,” the issue was the language “this invention.” IBM repeated “the invention” and “this invention” multiple times, pretty much nailing down specific limitations. When we prepare the “Summary of the Invention,” we indicate that “this disclosure includes.” In other words, we write inclusive words versus exclusive words. The moment you state that “this invention is…,” you are done.

      1. 2.4.1

        “ The moment you state that “this invention is…,” you are done.”

        As already pointed out a zillion times, you are only “done” if your only goal is a patent that covers more than what you described. It’s just as likely (or more likely) that a narrow description of “the invention” or its operation will PROTECT your patent from being invalidated as obvious or non-enabled. Sophisticated drafters are well aware of this. Mindless pond-feeding faces not so much.

        1. 2.4.1.1

          lol – it’s Opposite Day (yet again) for Malcolm and his penchant for exacting picture claims as the only “true” way of claiming.

  10. 1

    “According to the court, the claims recite the concept ‘without any specificity as to how this is accomplished.'”

    Then be honest and 112-trash these claims. Don’t dishonestly 101-trash them.

    (Note to IBM: Welcome to our world. Sure hope you’re investing some of your profits in restoring eligibility via Congress to all areas of innovation — including yours.)

    1. 1.1

      Then be honest and 112-trash these claims. Don’t dishonestly 101-trash them.
      They cannot honestly trash them under 112. Claims aren’t meant to enable the practice of the invention (i.e., “specificity as to how this is accomplished”). That is the role of the specification under 112(a). And if one wants to argue lack of enablement, there are a slew of factual findings that need to be considered before determining that the specification fails for lack of enablement. Consequently, the Federal Circuit cannot hand wave reject a specification for lack of specificity under 112 like they do under 101.

      The Supreme Court supposedly believes that “This Court has “more than once cautioned that courts ‘should not read into the patent laws limitations and conditions which the legislature has not expressed.’ ”” See Bilski (quoting Diamond v. Diehr). However, the Courts have shown time and time again that this is merely aspirational.

      1. 1.1.1

        This.

      2. 1.1.2

        Thanks Wandering. I do know of what you speak. Should have made clear that Chewy should / could have first raised 112; which the court then could have ruled on.

        But as you point out, that would require an actual, in-depth, fact-based, (expensive-for-Chewy) analysis.

        Easier to hand-wave.

    2. 1.2

      Is this also a semantics issue? Was it really claims that only “recite the concept,” or was it, more accurately, claims that only recite the desired result without claiming the “concept” in the patent law sense of a complete conception?

      1. 1.2.1

        “complete conception” – please explain how you are attempting to put this into the scenario of claims.

Leave a Reply

Your email address will not be published. Required fields are marked *

You can click here to Subscribe without commenting

Add a picture