Patent Specifications Continue to Rise in Size

by Dennis Crouch

In 2008, I reported on a trend that I termed the “Rising Size and Complexity of the Patent Document.”  Since then, the size of patent specifications have continued to grow. Patent specifications continue to grow — another 10% on average since 2008.

The chart below shows the average number of characters in the body of utility patent applications grouped by application filing year and includes data from all applications filed during the given year and published since 2005.

PatentlyO138.

As the second chart suggests, shorter applications (those in the first quartile) are less likely to issue as patents within the four years of the publication date. To build this chart, I calculated the specification size for each application published 2005–2007 and also determined whether a resulting patent had issued.

PatentlyO139

41 thoughts on “Patent Specifications Continue to Rise in Size

  1. Ned, have you ever tried in PTO proceedings the line “You allowed that one so, now, you have to allow this one”. It doesn’t work. I say to you that you cannot cite one solitary claim allowed by the EPO in Prom and tell me it marks the definitive line on vlalidity underv the EPC. Perhaps the allowance was anomalous. How about that? Could that be? Not even EPO Examining Divisions are infallible in ex parte proceedings, are they?

    The more correlation claims get opposed at the EPO, the better the EPO will get at the extremely delicate and difficult task of sorting the sheepy claims from the goaty claims.

  2. Max, the EPO apparently allowed a patent to the Harvard mouse as a composition of matter. If you can do that with a mouse, what can you do with a plant?

    On software, OK.

    On information, I saw from Paul Cole’s thread what the EPO will allow in the EPO analog of Prometheus style claims. It seems to me that a patent on information as the claims actualy do not require any use of the information for infringement. The measurement itself is old. The information it produces is old in kind. The only thing new are the correlations and they can be published in books. It is not as if one measured, as in Abele, a CAT SCAN signal, processed it using new mathematics, and then displayed the result in a vastly improved display. The claims simpy require a measurement, which is old, and no processing whatsoever by anyone to do anything.

  3. Happy reading. I think you will find Activis rewarding. Watch out for Robin Jacob’s image of a national Supreme Court ship captain’s duty to his jurisprudential ship, when he perceives that his (EPO) commodore is steering the entire European flotilla on to the rocks.

    And in T_1099/06 look out for para of the Reasons, on page 8 of the print. I quote: “In the legal system established under the EPC there is no principle of binding case-law”.

    Do you see now?

  4. I confess to not having read the materials offered.

    I will do so and see if your seemingly contradictory stance is alleviated.

  5. Newbe, did you read me on the other thread on April 24 at 10:07am? Did you click on the link and read the Hal Wegner note, the Activis decision and the T Decision?

    It seems not. If those materials haven’t helped you, I’m at a loss what more I can do to convince you.

  6. Good morning MaxDrei,

    I find it odd that you left our conversation in a huff, and then return to the very same conversation (with your very same stance) on this thread.

    Apparently, you do have the time and energy to repeat yourself, but not the time or energy to actually explain the differences (and similarities) that I pointed out. Without your version of SCOTUS (EBA) and its Binding Precedent (which is very much in place), you do not have legal certainty at all. Why you shield your eyes from this very simple observation is beyond my ken.

    Your position on legal certainty is thus left unconvincing.

    You appear to be just another person who wants their turn atop the soapbox, to have their say, and to move on without any dialogue or understanding.

  7. There are more than a few glaring errors of logic and fact in your post that need to be corrected. Two of which are:

    1. There is no point of novelty test in the USA.

    2. Prometheus never claimed information was the invention.

  8. Ned I read what you say but find that it does not accord with what I observe. In Europe, getting past the PTO brings no great presumption of validity, and so you should not equate what SCOTUS says about a Prom claim with what a less than fully concentrated EPO Examiner of the Prom claims let through to issue.

    On software I don’t understand your remarks. Software in general fails the EPO’s obviousness test because the only sort of non-obviousness that establishes patentability is “technical” non-obviousness. Some software does get through, when it is convincingly presented as the non-obvious solution to a technical problem.

    Under the EPC, Art 52, plant varieties are not eligible. Biology is not eligible. There is a wealth of caselaw on what biotech is patentable in Europe, and what is not.

    In Europe, effectively, the Supreme Court on patent validity is the Enlarged Board of Appeal at the EPO. It is a specialist patent court and it does a good job on patent validity. In Europe, we know what is eligible and what is not, and how to analyse whether the subject matter of any claim is obvious. I doubt anybody in Europe wants to swap its EBA jurisprudence for your SCOTUS jurisprudence.

    Of course Binding Precedent is normally essential to bring legal certainty. But in the 40 country EPC we have a special situation. The EPO’s Boards of Appeal handle >1000 appeals every year. After 30 years, they have developed a consistent body of law which, through survival of the fittest lines of logic, and the very absence of Binding Precedent, has brought a high level of legal certainty to issues of patent validity in all 40 countries.

    Readers please note: my defence of Europe is only on the narrow issue of legal certainty about which patent claims are ultimately valid and which are not. We all know that there is a lot of legitimate criticism that can be made, about how patent disputes are litigated in Europe.

  9. Max, I hear what you say, but there is something lost in translation when you allow Prometheus-type claims where the only invention claimed is a correlation, and that is pure information, information that can be published in a book for pity’s sake.

    You say you cannot patent information, but you do when the only novelty in the claim is the information, not the use of information, and that information can be published. Do you really make the spread of knowledge an infringement?

    You say you cannot patent software, but you do, when the only novelty in the claim is the software and the recital of the computer is simply incidental — that it execute the software.

    I presume you allow patents on the improved asexually reproduced plant or animal. But once that plant or animal is exposed to nature, it spreads naturally and become part of nature. How can one patent the environment?

    There is a large gap between what you claim to do and what you actually do.

    I am not saying that the US is pure as the driven snow. But I think we do have the advantage of a Supreme Court that actually tears away the veil of form and looks to the substance of claimed subject matter. I wonder at times how Europe can continue to exist without a Supreme Court.

  10. Do really want to give some examiner another weapon to use against you during prosecution?

    True, they might never mention it but why even give them the option?

    I prefer to win battles before they begin.

  11. Is it truly a requirement if the Office will not examine to it and the courts will not review it?

  12. I have no idea what this article is about. My patent applications are always at least 11 inches long.

  13. “For the EPO, a patentable invention is the solution to an objective technical problem. But what problem? Look to the specification as filed, ”

    Since the EPO has no objective way to define technical no such thing as a patentable invention exists in the UK.

    Oh and where the EPO is going we won’t need wheels.

  14. The (US) requirement is “preferred,” not best, and all of the examples cannot be “preferred” as that makes none of the examples “preferred.”

    Actually all of the examples can be preferred with the understanding that the invention is not limited to just the examples in the specification.

    So one may state preferred embodiments for each example with the understanding there are other examples that could not be included because of time and space factors.

    And as long as “2165 The Best Mode Requirement – 2100 Patentability” exist there will be a best mode requirement in the USA.

    2165 The Best Mode Requirement

    A third requirement of the first paragraph of 35 U.S.C. 112 is that:

    The specification. . . shall set forth the best mode contemplated by the inventor of carrying out his invention.

    “The best mode requirement creates a statutory bargained-for-exchange by which a patentee obtains the right to exclude others from practicing the claimed invention for a certain time period, and the public receives knowledge of the preferred embodiments for practicing the claimed invention.” Eli Lilly & Co. v. Barr Laboratories Inc., 251 F.3d 955, 963, 58 USPQ2d 1865, 1874 (Fed. Cir. 2001).

    The best mode requirement is a safeguard against the desire on the part of some people to obtain patent protection without making a full disclosure as required by the statute. The requirement does not permit inventors to disclose only what they know to be their second-best embodiment, while retaining the best for themselves. In re Nelson, 280 F.2d 172, 126 USPQ 242 (CCPA 1960).

    Determining compliance with the best mode requirement requires a two-prong inquiry. First, it must be determined whether, at the time the application was filed, the inventor possessed a best mode for practicing the invention. This is a subjective inquiry which focuses on the inventor’s state of mind at the time of filing. Second, if the inventor did possess a best mode, it must be determined whether the written description disclosed the best mode such that a person skilled in the art could practice it. This is an objective inquiry, focusing on the scope of the claimed invention and the level of skill in the art. Eli Lilly & Co. v. Barr Laboratories Inc., 251 F.3d 955, 963, 58 USPQ2d 1865, 1874 (Fed. Cir. 2001).

    The failure to disclose a better method will not invalidate a patent if the inventor, at the time of filing the application, did not know of the better method OR did not appreciate that it was the best method. All applicants are required to disclose for the claimed subject matter the best mode contemplated by the inventor even though applicant may not have been the discoverer of that mode. Benger Labs. Ltd. v. R.K. Laros Co., 209 F. Supp. 639, 135 USPQ 11 (E.D. Pa. 1962).

  15. Thanks Leopold. I’m grateful. Your last para exposes the different pages that First to Invent USA and First to File ROW are on. For the EPO, a patentable invention is the solution to an objective technical problem. But what problem? Look to the specification as filed, and all will be revealed (if it was written outside the USA).

    Or not. As the case may be.

  16. Max, I’m not sure where the requirement to “list one as preferred” comes from – as far as I know that requirement doesn’t exist. I rarely use the word preferred in an application.

    Genius is generally correct, in my opinion. There’s not necessarily any need to talk about the relative worthiness of different embodiments of the invention at all. That said, I certainly would avoid saying anything remotely resembling “They are all pretty much the same.” If a variant has a particular advantage, by all means point it out. But that advantage may be meaningless in some applications, and in the engineering world I come from many advantages come at the cost of something else. Paraphrasing Genius, “better is in the eye of the specific use,” and I probably don’t know all the specific uses for an invention when I write the specification.

    With regards to obviousness, the evidence for the “unexpected results” flowing from the claimed combination doesn’t have to come from the specification. Thus, with respect to the four corners of the application, “‘different’ suffices.”

  17. Thanks Genius but I’m bound to reply, that both of your statements strike me as dubious propositions.

    Anybody else care to enlighten me?

  18. Thanks but we are not yet on the same wavelength. So I’ll try again.

    That results are surprising or unexpected is worth something in the USA, for patentability, I gather from all the times I read “surprisingly” as the first word of one para or another in the specification of a patent application. But if in your drafting you have lumped together all your embodiments as much of a muchness, that doesn’t help you when the time comes, later, to assert that a selected one of them delivers a surprising or unexpectedly good technical effect (or at least one rthat is different and better than that of the known one). If you started with the message (whether explicit or implicit) that “They are all pretty much the same” that will hurt you when later you want to say “Examiner, the one I now claim is different and better than the one in your search reference”. You just contradicted yourself, didn’t you?

  19. The (US) requirement is “preferred,” not best, and all of the examples cannot be “preferred” as that makes none of the examples “preferred.”

    Just list one as preferred without denigrating the rest. If the preferred one gets knocked out, you are not in trouble as you have viable alternatives and you have met your (US) duty.

  20. Thanks. But does it? I was thinking more about obviousness.

    Suppose, at the outset, you present all the variants as equally effective, nothing to choose between them, all equivalent to each other, and then it turns out that the patentability of just one of them is prejudiced by the art. You might then want to press forward with a selected one, arguing that it is in fact more effective than the one that has turned out to be in the art. In those circumstances, what you wrote in the spec as drafted is not going to help you and, in fact, might well blow your chances, at least in Europe.

  21. “do I present them as each being as good as any of the others, or should I aim to include in my specification as many intermediate levels of generalisation as I can, all the way up from the pinpoint illustrated embodiment to the surrounding balloon of the widest independent claim?”

    Depends on if the law requires a best mode of invention, doesn’t it? Regardless every embodiment should be presented as a preferred embodiment.

  22. What you mostly get below 20,000 words that throws off the curve are pro se applications, many of which are written in crayon, drafted with no room to amend the claims, prosecuted inexpertly,.

    “So says the expert that has never run a business, earned a patent, or invented anything in their life.

  23. 20,000 characters really isn’t that long, if you think about it.

    About ten pages of typewritten text, I estimate. And yes, that sounds quite close to the minimum length of a professionally-drafted application. Indeed, I’m surprised that even one-third of applications half as long manage to issue within four years…

  24. “abandoned after a rejection is made “final” because surely that was their last chance, or because RCEs cost money.”

    All perfectly legitimate reasons to abandon a case. Just because your clientlols don’t know any better because you “taught” them differently doesn’t mean anything.

  25. “Like you said, if they’re below that, there are numerous problems that will keep it from issuance.”

    Such as 101, 112, 102 or 103?

  26. It would be interesting to know how subject matter affects the “Likelihood of Patent Issuance Grouped by Specification Size.” I mean, maybe the data show nothing more interesting than an increase in biotech applications, which tend to be larger than mechanical and electrical/computer.

  27. You also get the direct foreign translations that have all kinds of support and enablement issues.

    I never have an app that’s less than 15 pages including claims, even the simplest. Like you said, if they’re below that, there are numerous problems that will keep it from issuance.

  28. Sure, there was an accurate observation and recording of the size of specifications and the chance of patent issuance within four years of publication. However, these factors alone are insufficient to conclude that smaller patent applications having less than 20 thousand words in the specification are less likely to issue as patents w/in the 4 years of publication date. The term “specification size” is vague and using it to infer likelihood of patentability of any patent application is likely to produce inaccurate results.

  29. As the second chart suggests, shorter applications (those in the first quartile) are less likely to issue as patents within the four years of the publication date.

    20,000 characters really isn’t that long, if you think about it. Most sophisticated applicants would have no trouble writing that much about even the simplest invention, and lots of agents are innately wordy and include lots of boilerplate.

    What you mostly get below 20,000 words that throws off the curve are pro se applications, many of which are written in crayon, drafted with no room to amend the claims, prosecuted inexpertly, and abandoned after a rejection is made “final” because surely that was their last chance, or because RCEs cost money.

  30. What is a typical application and how does a typical application differ from that of an average application?

    It’s roughly the difference between mean and median/mode.

    The average person has one ovary and one testicle. A typical person has either two of one or two of the other.

  31. What is a typical application and how does a typical application differ from that of an average application?

  32. I’d really like to see a similar chart for validity on patents challenged under 12, either on court or on reexam. What criteria? Reexam grant rate. Summary judgement survival rate. At least one claim invalid under 112. Percent claims invalid under 112. Etc.

  33. A cause looking for followers.

    All you have here is a natural reflection of the drop off of the smaller and less likely to be patented applications driven by the general economy and the tendency of the patent role to return to its “sport of kings” status. By and large, the typical application is not growing in size (and by looking at the plateau, such growth is not likely to improve patent ability).

  34. All these “different embodiments” I’m going to mention: in mech eng cases, do I present them as each being as good as any of the others, or should I aim to include in my specification as many intermediate levels of generalisation as I can, all the way up from the pinpoint illustrated embodiment to the surrounding balloon of the widest independent claim?

    With the likelhood that claim 1 is too wide, and the rising importance of a supporting “written description” for the amended narrower claim, I’m going to need these statements of intermediate levels of generalisation in from the outset, aren’t I?

    Is that another factor pushing specifications to greater length?

  35. It’s important for an applicant to have a long and detailed specification with as many different embodiments as can be thought of, these days.

    Especially with the rising use of BRI as a justification to invalidate claims in virtually any category and for almost any reason.

    One needs the ability to fight back with BRI in light of the specification!!!

Comments are closed.