by Dennis Crouch
Many patent applications are not fully reduced-to-practice by the time the patent application is filed. Although reduction-to-practice is a required element of invention, the Courts and Patent Office have long permitted the filing of a patent application to constructively satisfy the RTP requirement. Still, the application must fully satisfy the disclosure requirements of 35 U.S.C. 112(a), namely written description, enablement, and best mode.
A patent specification will typically include a series of examples, embodiments, use-cases and/or experimental results. In this post, I’ll refer to these collectively as “examples.” One way of categorizing the examples is to divide between actual “working examples” and predicted “prophetic examples.” As you might guess, working examples are typically more compelling and indicative that the invention will actually work as claimed. But, prophetic examples also have their use and are absolutely permissible. Although I have not measured this, I expect that the vast majority of patentees rely on prophetic examples to some extent in order to expand the scope and depth of their disclosure. Thus, even if the patentee has reduced the invention to practice and explained the RTP in a working example, the specification may also provide a set of prophetic examples with differing arrangements, elements, and outcomes. Because of their differences, the courts and USPTO have called on patent applicants to take care in distinguishing between the two in their patent applications.
My definition of prophetic example is broad: I’ll pause here to recognize that my definition of prophetic examples is broad. Traditionally, many have focused the “prophetic example” doctrines only on reports of experimental results that are typically found in chemistry and biology related patents. In her Prophetic Patents article, Professor Janet Freilich primarily limited her analysis to “experiments that report protocols that were not actually conducted and describe results that are made up, or prophesized.” Using that definition and an automated search, Freilich found about 50% of chem/bio patents include prophetic examples. The MPEP has a broader definition that more closely parallels mine: “A prophetic example describes an embodiment of the invention based on predicted results rather than work actually conducted or results actually achieved.” MPEP 2164.02. Basically, if you are describing something that has not been actually reduced to practice, it is a prophetic example.
The USPTO has published a new notice to patent applicants with the same reminder cautioning applicants to distinguish between working and prophetic examples:
The USPTO is reminding patent applicants of their duty to ensure that patent applications are written in a manner that clearly distinguishes prophetic examples with predicted experimental results from working examples with actual experimental results. Prophetic examples, also called paper examples, are typically used in a patent application to describe reasonably expected future or anticipated results. Prophetic examples describe experiments that have not in fact been performed. Rather, they are presented in a manner that forecasts simulated or predicted results. In contrast, working examples correspond to work performed or experiments conducted that yielded actual results.
Notice: Properly Presenting Prophetic and Working Examples in a Patent Application, PTO-P-2021-0020 (Fed. Reg. July 1, 2021) (“Notice”). The notice indicates that examiners generally do not question disclosed test results unless there is a reasonable basis to do so. But, prophetic examples that fail a reasonableness test will be rejected for insufficient disclosure (enablement and/or written description).
In addition, applications that include prophetic examples should make clear whenever the example is prophetic. Otherwise, the application may be deemed to include a misleading statement in violation of the applicant’s “duty of candor and good faith in dealing with the Office.” 37 C.F.R. 1.56.
QUESTION for PATENT DRAFTERS: When you describe various embodiments in your patent application, do you tell the PTO which ones are working-examples (actually reduced to practice) and which ones are prophetic?
— Dennis Crouch (@patentlyo) June 30, 2021
One way to disclose prophetic examples is to expressly label them as “Prophetic Examples.” “It is a best practice to label examples as prophetic or otherwise separate them from working examples to avoid ambiguities.” Notice. The PTO has also deemed it appropriate to use future-tense language in describing prophetic examples and past tense for working examples. “When drafting a patent application, care must be taken to ensure the proper tense is employed to describe experiments and test results so readers can readily distinguish between actual results and predicted results.”
The USPTO Notice focuses on prophetic examples stemming from experimental results, and thus remains to be seen whether the requirements will be practically extended to situations involving prophetic embodiments and use-cases presented without any experimental results. One approach for patent applicants is to write everything as if it is a prophetic example.
I am genuinely shocked that nearly 76% of those answering would merely lump them together (as of there were no actual difference).
Any of those care to comment on the apparent disregard for the duty of candor that may be implicated with such “lumping together”…?
How I wonder how Malcolm Mooney feels on this point.
Still no takers on the ethical implications…?
It’s too bad this blog does not have an ethics section.
😉
The comments below are just so nasty and illustrate the anti-patent judicial activists game.
I stated what actually goes on and how it actually works. That is reality. I’ve been doing this for over 20 years and this is reality.
It is your job to figure out what reality is. The game of starting with all these patents are from trolls and all the disclosure are terrible illustrates there is no interest in understanding what reality is but rather an interest in beating down the patent system.
Williamson is a good example of 112 games. Williamson is technically wrong. It is the closest I’ve seen to a criminal taking from judges beyond local corruption. The patent application is interrupted through the eyes of a person of ordinary skill in the art. A person of ordinary skill in the art of computer science who is familiar with algorithms does not need a full algorithm and a diagram. They only need an explanation of the changes to a known existing algorithm. That is merely a fact, which with the likes of Taranto has little meaning and with the anti-patent judicial activists below has less than little meaning.
But guess what? Try as you will reality is still reality. The Sun doesn’t orbit the Earth. What I said below is simply reality.
Slashdot Reader’s comments regarding 6611289 are a good example of this.
The game for the propaganda machine that is being played is to pretend that a patent application is supposed to disclose the experimental results. They do not have to. The game is to pretend that a patent application has to be understandable to the average joe on the street. It does not have to be. The patent application is supposed to enable a person of ordinary skill to practice the invention. That is the test not whether it is a good primer on the invention and the underlying technology.
Does anyone remember US Patent 6611289:
An improved digital camera comprising:
a first and a second image sensor closely positioned with respect to a common plane, said second image sensor sensitive to a full region of visible color spectrum;
two lenses, each being mounted in front of one of said two image sensors …
Was this invention ‘reduced to practice’ when the patent was filed? There were a lot of comments here about it being a ‘physical’ and ‘ not abstract’ invention, but I find it hard to believe that anyone actually had a 4 lens camera with a matching digital module in their hands…
Why do you find that hard to believe? I used to work with one of the largest manufacturers of cameras. I’d image that it would take very little time in the lab to set something like that up.
In fact, there are very few patent applications I’ve written where the inventor didn’t make a prototype and often of the time the prototype became a product.
This is just more anti-patent nonsense that patent application are disconnected from the real-world and it is only capable of being made an argument to incredibly ignorant judges that know very little about actual inventing, product development, science, or even patent law.
I read the entire patent and there is nothing in the document that indicates that a physical camera ever existed. If I reduce an invention to practice but no one can tell from the patent, does it still count? Is there a secret stash of extra evidence in the patent office?
“If I reduce an invention to practice but no one can tell from the patent, does it still count?”
That actually sounds like an argument in favor of the notion of constructive reduction to practice: it doesn’t matter because the subject matter is so predictable.
Thank you! Constructive Reduction to Practice was the answer to my question. (The phrase appears 21 times in MPEP so it must be a good one!)
“The phrase appears 21 times in MPEP so it must be a good one!”
I think you may have a future in the USPTO’s artificial intelligence based automation department.
It actually sounds like someone who has no clue about patents.
+1
I just don’t know how people just lie, lie, and lie, and they have no shame about it. And, they don’t lose their credibility with others.
People who lie repeatedly (on this blog and elsewhere) definitely lose their credibility. I gather from watching the phenomenon (here and elsewhere) that many just do not much care. They hold their credibility rather cheap.
N.W. you need look no farther than Trump, many members of Congress, and their voters. The answer to your general question is they don’t read or listen to anything that might interfere with preconceptions. But “truth will out” in the long run in most cases.
BTW, most of the many patent applications I wrote over the years were also based on actual physical engineering prototypes. And before the AIA gutted enforcement of the “best mode” disclosure requirement.
That certainly applies across BOTH sides of the aisle, Paul.
Right now I am prosecuting an application that captures a feature that 100’s of millions of people use everyday. Another patent application I prosecuted is a consumer product where 10’s of millions of them were made.
I am writing a patent application right now where the software is about 80 percent done and will become part of a desktop assistance for building applications and will either be used by half a billion people or those people view the results of someone else using it.
So, please STFU with these constant and continuing lies about the patent system.
Perhaps you could identify the particular “lie” you see in post 8? Surely you wouldn’t want to be seen just be randomly accusing people of wrongs.
Surely you wouldn’t want to be seen just be randomly accusing people of wrongs.
I think that this is an example of just not caring, as I mentioned in 8.2.1. Some folks just hold their own credibility rather cheap.
^^^ case at point.
(But what are the chances that Greg will recognize his own parsimoniousness?)
I think I made it clear what the lie is. See below. The propaganda is that patents are fantasy disclosures of non-products that are marginally enabled at best.
This is simply not true for the vast majority of patent applications and certainly for the large technology companies. In fact it has pretty much nothing to do with my practice. And, yet, the drums beat on and on about this as part of the propaganda movement to use 112 to further limit patents. Williamson was about as close to a criminal taking as I’ve seen from the CAFC. And the idea is to continue on with 112 to continue to limit patents.
>>This is just more anti-patent nonsense that patent application are disconnected from the real-world and it is only capable of being made an argument to incredibly ignorant judges that know very little about actual inventing, product development, science, or even patent law.
The propaganda is that patents are fantasy disclosures of non-products that are marginally enabled at best.
This is simply not true for the vast majority of patent applications and certainly for the large technology companies.
Honestly, how could anyone pretend to know this. We are on the 11 million series of patents right now. To opine about the sufficiency of enablement of even a simple majority of patents—to say nothing of the “vast majority”—one would need to be familiar with more than 5.5 million patents, their respective technology fields (at each of their filing dates), and their file wrappers.
You are not familiar with that many patents. Nor am I. Nor is any person alive today.
In other words, you could not possibly know what you are talking about, and yet you say it anyway. Like I said, some do not value their credibility all that highly…
Really Greg? The just question in wonderment the statements of the opponent rhetorical technique?
Let’s see I’ve been doing this for over 20 years. I’ve prosecuted thousands of patents and reviewed literally 10’s of thousands of patent applications. Probably getting close to 100,000 patent applications in my career. Do you know statistics? I’d say you don’t.
The sample size I am dealing with is statistically significant so I have reviewed enough patent applications to be able to have an opinion that is relevant and –guess what–statistics backs me up on that.
Greg has a point here in regards to any opining on direct knowledge of explicit counts of actual patents reviewed.
The counter of course is that it simply is not an unreasonable proposition (demand) that examiners in the patent office actually DO their job and that patent grants, having followed the law, DO carry the point expressed by Night Writer, and anyone ‘caring’ about their credibility would recognize that taking a contrary stand would place the onus of support on that person.
Yes, I am just stating the obvious – but Greg does have a tendency to miss the obvious (given his own obsess10n with his fixed views).
Hi Night Writer,
I expect you read many software patents as well as write them. Could you point out a couple of really good ones? We usually discuss the marginal ones here which went the Scotus. Do you have one or two which you think would and should be universally acclaimed on this site and by the courts?
SR,
You seem to (once again) have a hard time getting out of your own way when it comes to understanding that patents are legal documents and are NOT technical engineering documents.
I also invite you to understand both the presence AND level of the legal presumption of validity.
See 35 U.S. Code § 282
Slashdot Reader, how about you take a random sampling of 100 software patents. Pick an AU that is primarily software.
Post the 100 and then we can discuss.
” how about you take a random sampling of 100 software patents”
How about he just asserts that his sample is independent, Hypocrite?
The irony…
“I find it hard to believe that anyone actually had a 4 lens camera with a matching digital module in their hands…”
I don’t see why this is so implausible. Perhaps you are conflating mass production of such a device with prototyping a rough test?
On the other hand, I would agree that the “invention” of claim 1 was likely not actually reduced to practice at the time of filing, as it appears likely that the two-sensor embodiment was not contemplated at the time of filing.
Well when I read the patent he never mentioned taking a picture or how it would look better, except to say:
Further, with images of the same imaging target from the multiple sensors, it is possible to enhance images, such as noise removal and color correction at 820. More importantly, a true color image with true resolutions is derived from the enhanced images at 830.
But that sounds like phosita stuff…
What happens in real life is that inventors at the large tech companies come up with inventions and then often build small prototypes to see if they make sense in a product.
Often when things are in a nascent form the inventors don’t have a lot of results or they want it filed very quickly or there is a very low budget.
Sometimes they don’t want to disclose their experimental results as they know that it will only encourage other to take their invention.
And there is always the Chinese who will copy any product pretty much.
Etc.
This reads like an excuse for why so many patents do have enablement problems, not an explanation that they do not.
No Greg. It is an explanation for why so many patents don’t include more meaningful experimental results.
I have to wonder (given Greg’s own point concerning any actual reviewing of the sheer number of patents out there), what does “many” mean to Greg?
Is Greg expecting perfect examination? How “few” does “many” have to drop to to be acceptable in Greg’s ‘real world ?’
… and how does Greg think that his bloviating affects his own ‘credibility’ here?
The thing that gets me about the commentators on this blog is that they just throw shxt on the wall to see what sticks.
You knock them down and they lose miserably and then they don’t respond. Then they post the same shxt in the next post.
It is clear that they have no intention of a real dialogue but merely the intent to post propaganda.
Again, in this thread and the other one I am fighting their propaganda that the average patent is junk and doesn’t correspond to a real product. This is not true.
[I]n this thread and the other one I am fighting their propaganda that the average patent is junk and doesn’t correspond to a real product. This is not true.
No one—on this thread or the other—is advancing a thesis about “the average patent.” That concept would be meaningless, and the thesis pointless, therefore, to defend. The actual thesis being actually advanced is that something well short of “the overwhelming majority” of patents emerge from the PTO fully enabled across their entire claim scope.
Of course, all of the ones that I prosecute comply with all statutory requirements. 😉
Go back about a decade or so ago, and my words to Prof, Crouch (during one of those ‘how can we have a better blog’ stints) were exactly like:
“You knock them down and they lose miserably and then they don’t respond. Then they post the same shxt in the next post.”
It only takes a few times of attempting to engage on the merits with that type of (perpetual) response, and snark becomes a justifiable ‘self-assistance’ mechanism.
If the other side is not going to bother being serious, I might as well enjoy a bit of John Maynard Keynes at their expense.
I have found that the notice and this blog post are a bit unclear and misleading. My understanding of the notice is not that you have to distinguish between working and prophetic examples when those examples are described, but that you need to distinguish between experimental results that have actually been achieved (from a working example) and experimental results that are simulated or expected (from a prophetic example).
Since experimental results are not required to be disclosed in a specification, and given the risk for inequitable conduct outlined in the notice, I tend to believe that an applicant should not include experimental results. I can understand the benefit of having that information in the specification to avoid later having to generate a declaration against an enablement rejection or obviousness rejection (to show unexpected results or criticality); however, that need at the time of drafting the application is speculative. The potential harm seems to outweigh the benefit. I will concede that my perspective is from one that works in the electrical arts that has significant predictability, and not from one that works in the chemical or life science arts that may be unpredictable.
+1
Boo! You require a Twitter login.
I get that 112 may be a way to make it harder for people that seem to ride the way and claim things that they haven’t reduced to practice and don’t plan on reducing it to practice but assert the patent.
Frankly, all the patents I’ve written over the last 5 years and maybe much longer are ones where the software was reduced to practice.
But, I will say that back 15 years ago the large corporations realized that one way to defeat the non-reduction to practice software patents was just stay ahead of them. There was an effort to write more patents and do more research and stop letting them get a year or two ahead of them. This was very positive.
For me, when drafting a detailed descrption in the chem/bio area, one refers to “Examples” of the claimed inventive concept whereas in engineering cases the detailed description is of different “embodiments” of the claimed inventive concept.
Common to both areas is the need to enable over the scope of the claim. Think of that as an inflated balloon, with the Examples/embodiments being data points within the balloon.
Then think of what material you get from the inventor, with which to draft the case. One prototype, perhaps, or one set of experimental data, on one particular molecule. Perhaps only one real data point, that is the inspiration for the balloon that you formulate around that data point. How big you make that balloon is down to your judgement, experience and your knowledge of the state of the art. Reach too far and the claim will one day crash. Reach not far enough and the competitors design around.
But then again, the inventor will have done so much exploration that she/he will know intuitively, despite not having rigorous data points, what else will “work” and what will more likely not work. That experience is what you rely on, to shape the balloon.
But once you have settled upon the size and shape of the balloon, you need to fill it with the “fluid” that will maintain its inflated state. That one data point isn’t enough. In the “predictable” engineering field, one will know what “works”. But also in chem/bio, these days, one will have a very reliable feeling what else is likely to work and what not. The unpredictability of chem/bio is steadily diminishing, isn’t it? Hence, the growth in patenting activity for inventions supported by the sort of detailed description which EPO jurisprudence calls a “One Way is Enough” disclosure to enable a claimed concept.
In short, I am very dubious about the proposition that you can sort worked from prophetic examples like you can sheep from goats. In a rainbow, it is an illusion that there are seven (or six, or five) separate colours. It is also an illusion that there is a binary distinction between worked and prophetic examples. All the lawyerly effort being put into drawing that rigid distinction is a waste of effort, a distraction, a plyground for lawyers and not helpful for industry.
Put even shorter, what does it mean, to “reduce to practice” that which we call “the invention”? Does the statute include a definition of that notion? Does the Statute even mention the term “reduce to practice”? If not, does everybody then have a different idea, what it means, and when any particular Example gets over the threshold of “reduced to practice”. I suspect most Examples in most detailed descriptions are somewhere in between 100% worked and 100% prophetic.
But I’m not a chem/bio drafter. All those reading this who say I’m wrong, call me out, do.
MaxDrei,
Do you have a whitebook (or other) link to show what you mean by, “supported by the sort of detailed description which EPO jurisprudence calls a “One Way is Enough” disclosure to enable a claimed concept.“…?
Sort of, anon.
Start with EPC, Rule 42(1)(e)….”describe in detail at least one way…..”. As The EPC Implementing Regulations expressly envisage that “one way” might be enough, there are by now innumerable cases where “one way” was found to be “enough”. And plenty of cases where just one way was not “enough”. You know, it’s a fact-based enquiry, so each case turns on its unique matrix of fact.
I would suggest, that your “matrix of fact” also is dependent on the particular art to which the innovation belongs.
Previously, you had seemed to want to look the other way for art units lacking “predictability” when it came to possession and/or enablement, positing that if prophetic examples were later shown to be false, then there might be an issue.
Here, you seem to be singing a different tune (your ‘distraction, a [playground] for lawyers and not helpful for industry’ comment).
So please clarify — is it ‘no big deal and not a concern for patent quality,’ or is it ‘bad for industry’ (and potentially, a sign that patents have been improvidently granted)?
To be fit for purpose (promoting the progress of useful arts) a patent system has to recognise the reality of life as an inventor being creative, and the reality of life as a manufacturer obliged to come up with a creative design around. In a First to File system, inventors are very well aware of the imperative, to file at the PTO without delay, lest some other inventor beats them to it. Publication of their enabled disclosures, 18 months later, promotes technical progress magnificently.
But to be fair to them, inventors need to know, with certainty, when it is still too early to file. The thing about FtF is that, if you jump the gun, file prematurely, you end up with nothing.
If you have just one Example or embodiment, is that enough to enable the full width of your claim? Can that one Example be prophetic or must it be based on pre-existing experimental data? These questions preoccupy attorneys at law in the USA but not the patent attorneys practising everywhere else in the world.
Everywhere else in the world, there are two professions of equal standing, namely, patent attorney and attorney at law. This debate about how “prophetic” any given Example is can run only in the USA. This is not a coincidence. For those looking in on the USA from the rest of the world, the US patent system is dysfunctional precisely because it is firmly in the grip of attorneys at law rather than patent attorneys who are not attorneys at law.
Please explicate what exactly this “structure” of two types of legal practitioner have to do with the immediate point at hand.
Still waiting…
Thanks Max.
I written so many applications that have been prosecuted before the EPO, the US, and the China patent offices that I don’t want to count.
There isn’t really much difference other than amendments where at the EPO and in China, the standard is higher. Other than that there isn’t much difference beyond the technical problem/technical solution.
Biotech space here:
Typically, we only write prophetic examples for planned future experiments, and the purpose is to more easily bring the data in during prosecution to support the claims if needed (e.g. obviousness hurdles). But, our applications also often include embodiments that are not exemplified in either prophetic or working examples, but are rather reasonable extrapolations. (E.g. claims for K+ salts in place of exemplified Na+ salts).
Re: “I expect that the vast majority of patentees rely on prophetic examples to some extent in order to expand the scope and depth of their disclosure.”
I doubt that, and it may depend on how broadly a “prophetic example” is defined. The term itself is rarely used outside of chemical or pharmaceutical patents, such as for alternative embodiments that have variations in formulations or processing which might also achieve the desired result. Alternative enablements are not the same as specification statements of possible alternative uses or applications.
“ The term itself is rarely used outside of chemical or pharmaceutical patents”
See post 1 below.
“The MPEP has a broader definition that more closely parallels mine: “A prophetic example describes an embodiment of the invention based on predicted results rather than work actually conducted or results actually achieved.” MPEP 2164.02. Basically, if you are describing something that has not been actually reduced to practice, it is a prophetic example.”
I’m imagining Mel Brooks in a suit turning to the camera and saying “It’s good to be a software applicant.”
It’s been awhile since Random popped up with his emphasis on the difference between the “predictable arts” and — as Paul points out above — those arts for whom prophetic items simply are not “predictable.”
Such a view directly clashes with your insinuation here.
And the collision does not stop there. It echoes through a host of other “anti’s” such as Malcolm Mooney, the Slashdot/TechDirt lemming crowd, and even to MaxDrei (on the non-computing arts side of the spectrum).
People may not “like” the fact that innovating in the computing arts is ‘easier,’ but this only tends to mirror the old Supreme Court view that was expressly repudiated by Congress (the ‘must be worthy by showing a Flash of Genius’ mindset).
“ the fact that innovating in the computing arts is ‘easier,’ ”
Ya gotta love it when the truth leaks out, even when it’s a hilarious understatement like this one. Just about nothing on earth is easier to ‘innovate’ and most of the things that are marginally more difficult are ineligible, for great reasons.
LOL – your obfuscation is noted.
There is nothing at all untowards with a truth that innovation in some areas is easier than innovation in other areas.
The patent act does not segragate innovation to be protected under a view that any such ‘is the hard.’
Quite in fact, this proposition is what generated the patent law of:
“Patentability shall not be negated by the manner in which the invention was made.”
See 35 USC 103.
Truth — clearly it is not what you think it to be.
(I am not suprised)
Comments are closed.