Mr. Randi Misunderstands Patents

[Brown]

In the commentary of 27 July 2007, Mr. Randi wrote:

I just saw a typically naïve comment that repeats a canard about the requirements of our patent office:

Right, and in the US the USPTO, our patent office, actually requires people to send in physical working samples of supposed perpetual motion machines because they have historically received so many claims for patents to this system.

Wrong. The USPTO requires nothing of the kind. They’ll issue patents to just about anything, no matter how silly it is. For an example, see randi.org/jr/042602.html – do a search for “patent.”

The subject matter of the commentary to which Mr. Randi links is actually described in the April 19, 2002 edition of Swift, and it has nothing to do with perpetual motion. It is a silly idea, and a silly patent, yes; but it is not related to perpetual motion.

Silly patents... we'll come back to them later. Let's first deal with what Mr. Randi calls a "canard." Let's start with the act passed by Congress, 35 U.S.C. (United States Code) section 114:

The Director may require the applicant to furnish a model of convenient size to exhibit advantageously the several parts of his invention.

When the invention relates to a composition of matter, the Director may require the applicant to furnish specimens or ingredients for the purpose of inspection or experiment.

So the Patent Office can (i.e., it is legally empowered to) ask for working models.

Does the Patent Office ask for working models in most cases? For that, we turn to the administrative rules adopted by the USPTO, 37 C.F.R. (Code of Federal Regulations) section 1.91, which provide:

(a) A model or exhibit will not be admitted as part of the record of an application unless it:
...
(2) Is specifically required by the Office; or
(3) Is filed with a petition under this section including: (i) [a fee]; and (ii) An explanation of why entry of the model or exhibit in the file record is necessary to demonstrate patentability.

So the Patent Office generally does not require working models for most inventions, but there are exceptions.

Anybody want to hazard a guess as to what one of those exceptions might be? Here's an official link to the web site of the United States Patent and Trademark Office, specifically the Manual of Patent Examining Procedure. These are the rules that the Examiners follow, and they include quotations from both the U.S.C. and the C.F.R. passages that I quoted. MPEP 608.03 provides:

Models or exhibits are generally not admitted as part of an application or patent unless the requirements of 37 CFR 1.91 are satisfied. With the exception of cases involving perpetual motion, a model is not ordinarily required by the Office to demonstrate the operability of a device. If operability of a device is questioned, the applicant must establish it to the satisfaction of the examiner, but he or she may choose his or her own way of so doing. (emphasis mine)

There it is, in black and white. Feel free to check it out.

This is no canard. The person who pointed it out is not naive, but well-informed.

Mr. Randi is wrong, and he might even owe somebody an apology.

By the way, here's what The Economist recently had to say on the point:

Nowadays the United States Patent and Trademark Office (USPTO) refuses to review a perpetual-motion machine unless it is accompanied by a working prototype.

Now, having said that, the USPTO has apparently been duped into issuing a patent or two for a perpetual motion machine. It is rare, but it seems to have happened. Take a look at this patent, entitled "Motionless electromagnetic generator." Some physicists have said that this thing is tantamount to a perpetual motion machine, although the patent itself denies such a thing:

Thus, an electromagnetic generator operating in accordance with the present invention should be considered not as a perpetual motion machine, but rather as a system in which flux radiated from a permanent magnet is converted into electricity, which is used both to power the apparatus and to power an external load. This is analogous to a system including a nuclear reactor, in which a number of fuel rods radiate energy which is used to keep the chain reaction going and to heat water for the generation of electricity to drive external loads.

It should be noted that some patents refer to "perpetual motion" in the conventional sense, namely, that something is always moving (like the tides or the wind or the orbits of the planets or a hyperactive child). In other words, some patents do not use the term "perpetual motion" to describe any purported violation of the laws of thermodynamics.

Now to silly patents. Man, there are a lot of those. Before I get started with the really goofy patents, let me say that some silly ideas are certainly legitimately patentable. If someone invents a new kind of practical joke device (say, an iPhone look-alike that squirts water or makes a squeaky noise or something), that could be a patentable device. It's silly, and its not going to cure cancer or increase the mileage on your car, but it could be a source of amusement. If it's novel and non-obvious and useful (even if that use is to provoke amusement), it's patentable. There is no requirement that it be "non-silly." (By the way, the costs of examination are picked up by the applicant, not by the friendly neighborhood taxpayer. So if the applicant wants his silly idea examined, and he's willing to pony up the fees, then perhaps he ought to get what he paid for.)

One recipient of a possibly silly patent might be our bud, Penn Jillette. See his patent here.

In this thread, I discuss some common misconceptions about patents, such as the difference between an issued patent and a patent application. In this thread, I corrected many other misunderstandings about patents, including some from Mr. Randi. This post argues that even if Mr. Randi is wrong in some respects, he has a legitimate point in other respects:

First, some (apparently) pseudoscientific or near-pseudoscientific patents occasionally get through [such as the one I mentioned above].
...
Second, some patents seem to get issued for previously invented things. A classic example of this may be US Pat. 6,368,227, which is discussed in Mr. Randi's commentary of April 19, 2002.
...
Third, some downright loony patent applications get filed. Some of these applications come from crackpots, some come from the mentally ill, some come from honest inventors who don't know any better. ... These patent applications are often published as required by law. The fact that they are published and appear on the USPTO web site does not mean that the Patent Office thinks they have any merit. (Chances are, the examiners laugh their butts off when they see such applications.) Rarely, very rarely, do patent applications like this get anywhere.

Fourth, the fact that a patent issues does not mean that the invention actually works, or that it works as well as the inventor says that it does.
...
And finally, many patent holders falsely convey the impression that, because the invention has "earned" a US patent, that the invention works and has a governmental endorsement. Such an impression is almost always false. Unfortunately, it is not the Patent Office's job to regulate advertising to see whether assertions made to the public about patents are honest. That job is performed through other governmental channels.

Now, to wind up this little essay, let's have some fun!

You want goofy patents? I got goofy patents for you! Take a look at "They Invented What?" for some funny patents and patent applications. Or Google "wacky patents" and see what comes up. You won't see many perpetual motion machines, but you may have a laugh or two.

There are a lot of howlers there, but take a look at a patent application (NOT an issued patent!!) for Godly Powers. And here's one that ought to make Mr. Randi's hair turn white. Oh, wait, he must have seen it already. Here it is: Bi-digital O-ring test for imaging and diagnosis of internal organs of a patient: Basically, you use your fingers as a CAT scan! Click the "Images" button at the top of the screen (you will need a TIFF viewer) to see the pictures. This is an issued patent!

 

[slyjoe]

Thanks for finding this Brown - I was too lazy to look it up, but I had previously heard about the patent office requiring models for PM machines. I only encounter patents in terms of our company's patent committee; once we approve a patent, the lawyers handle the dealings with the patent office.

 

[Horatius]

Something that needs to be considered when dealing with the question of models being provided to the Patent Office. I notice the wording of the statute you quote is similar to our Canadian law:

The Director may require the applicant to furnish a model of convenient size to exhibit advantageously the several parts of his invention.

Thing thing to note, that we've dealt with here in Canada, is that it specifies a model may be required, but not a working model.

That seems kind of silly at first glance, but consider the problems of providing a working model of something that, by its nature, must be very large, or very expensive. It might be impossible to provide a working model of such a device on a scale that it could be conveniently examined by the office. This gives them the wiggle room to get out of this bind.

This is reinforced by the office practices you quote:

Models or exhibits are generally not admitted as part of an application or patent unless the requirements of 37 CFR 1.91 are satisfied. With the exception of cases involving perpetual motion, a model is not ordinarily required by the Office to demonstrate the operability of a device. If operability of a device is questioned, the applicant must establish it to the satisfaction of the examiner, but he or she may choose his or her own way of so doing.

This was a point I made in my talk at TAM. By its nature, an invention really has no limit on how complex it may be, and so we cannot put artificial limits on what an inventor may do to establish the validity of his invention. Again, it's this flexibility that makes it possible to scam the office, but this flexibility is also needed by legitimate inventors, so we can't really get rid of it.


Then there's also the practical question - even if they have the power under law to require a model, how often do they actually do that? I know in Canadian practice it is incredibly rare, and I don't see why US practice would be that different.

And then there's cases where the "perpetual motion" nature is very carefully concealed. I saw one case, during the last training session I taught, dealing with what was essentially a perpetual motion car, and I had a really hard time convincing other examiners that that's what it was, because the inventor had written his patent application so well (that is, sneakily).

 

[Brown]

This was a point I made in my talk at TAM. By its nature, an invention really has no limit on how complex it may be, and so we cannot put artificial limits on what an inventor may do to establish the validity of his invention. Again, it's this flexibility that makes it possible to scam the office, but this flexibility is also needed by legitimate inventors, so we can't really get rid of it.

Then there's also the practical question - even if they have the power under law to require a model, how often do they actually do that? I know in Canadian practice it is incredibly rare, and I don't see why US practice would be that different.

And then there's cases where the "perpetual motion" nature is very carefully concealed. I saw one case, during the last training session I taught, dealing with what was essentially a perpetual motion car, and I had a really hard time convincing other examiners that that's what it was, because the inventor had written his patent application so well (that is, sneakily).

I can't speak as to the practices of the Canadian patent office, so I'll stick to the US counterpart; but it's a good bet that Canadian practice is similar.

Let's get one thing right off the table right now: There are a lot of problems facing the USPTO these days, but handing out patents willy-nilly for perpetual motion machines is not one of them. (If you want to know what some of the real problems are, skip to the bottom of this post.) If Mr. Randi thinks this is what's going on, then he desperately needs to educate himself.

Patent examiners have a lot of work on their plates. If an application crosses an examiner's desk that looks like a perpetual motion machine, the easiest thing to do might be to tell the applicant, "Before I'll do anything, you must bring me your device"--the USPTO offices are in a suburb of Washington DC--"and demonstrate its operability." The applicant must at least travel to Washington DC at his own expense, and make some sort of model that "operates."

This will deter the vast majority of the cranks.

If the crank is foolish enough to admit that he's seeking a patent for a perpetual motion machine, the examiner has another tool at his disposal. Quoting one again from the Manual of Patent Examining Procedure, MPEP 2701.01:

II. WHOLLY INOPERATIVE INVENTIONS; "INCREDIBLE" UTILITY

An invention that is "inoperative" (i.e., it does not operate to produce the results claimed by the patent applicant) is not a "useful" invention in the meaning of the patent law [and therefore not entitled to a patent].
...
Situations where an invention is found to be "inoperative" and therefore lacking in utility are rare, and rejections maintained solely on this ground by a Federal court even rarer. In many of these cases, the utility asserted by the applicant was thought to be "incredible in the light of the knowledge of the art, or factually misleading" when initially considered by the Office. In re Citron, 325 F.2d 248, 253, 139 USPQ 516, 520 (CCPA 1963). Other cases suggest that on initial evaluation, the Office considered the asserted utility to be inconsistent with known scientific principles or "speculative at best" as to whether attributes of the invention necessary to impart the asserted utility were actually present in the invention. In re Sichert, 566 F.2d 1154, 196 USPQ 209 (CCPA 1977). However cast, the underlying finding by the court in these cases was that, based on the factual record of the case, it was clear that the invention could not and did not work as the inventor claimed it did. Indeed, the use of many labels to describe a single problem (e.g., a false assertion regarding utility) has led to some of the confusion that exists today with regard to a rejection based on the "utility" requirement. Examples of such cases include: ... a perpetual motion machine (Newman v. Quigg, 877 F.2d 1575, 11 USPQ2d 1340 (Fed. Cir. 1989)).... These examples are fact specific and should not be applied as a per se rule. Thus, in view of the rare nature of such cases, Office personnel should not label an asserted utility "incredible," "speculative" or otherwise unless it is clear that a rejection based on "lack of utility" is proper. (italics and underscoring in original, boldface added)

In the case of an invention overtly claimed to be a perpetual motion machine, there must be a factual record that it does not work. (In typical patent practice, the "factual record" is, shall we say, sparse--it's nowhere near as detailed as a court record.) If the device doesn't work, the examiner can slap it with a utility rejection and be fairly satisfied that case law supports such a rejection.

Brushing off the cranks is one thing. As Horatius points out, there is still the possibility that someone will deliberately obfuscate things to fool the examiner into issuing the patent.

By the way, fooling the examiner is not allowed. If you fool an examiner, or it you even try to do so, your patent is unenforceable in court.

A more serious problem than perpetual motion is that of bad science, in which an examiner is intimidated by what appears to be a very advanced scientific explanation that is in reality gobbledygook. Or applications that recite scientific findings that are at best dubious. Take a look at US Patent 7,104,948:

The bracelet, in accordance with the present invention, that irradiates anion and far infrared rays was allowed to be worn by 50 patients with either the problem of blood circulation such as stitching in the shoulder or whole body or the problem of suffering from neuralgia in the wrist was targeted.

The result was that all the patients wearing the bracelet felt the aches steadily decrease and, after about 15 days, the aches nearly disappeared.

As described above, the bracelet that irradiates anion and far infrared rays, in accordance with the present invention, has an elegant appearance due to its gold plate and can be used as a bracelet as an accessory, and it does not give any inconvenience to the human body since it is attached to the body.

The bracelet in accordance with the present invention can irradiate anion and far infrared rays more effectively through static electricity occurred at the friction between bracelet and skin as well as sunlight, and by regulating cellular ion balance, the electric charge inside and outside the skin gets normal, and controls complex biorhythm such as promoting blood circulation and metabolism, besides activates body cells, lowers blood pressure to increase regenerate tissues, suppressing aging phenomena.

If you're an examiner, you might read this and say it sounds like BS. But according to the principles of MPEP 2701.01, "Sounds like BS to me" is not a proper ground for rejection. Besides, maybe the damn bracelet really does irradiate anion and far infrared rays, which might be useful for matters other than suppressing aging or some such puffed-up benefit.

I mentioned that there are some real problems facing the US Patent and Trademark Office. None of these involves perpetual motion, although bad science is involved tangentially. There are three major "shocks" to the US system, coming from all three branches of government.

From the Legislative Branch: The Congress is considering sweeping patent reform legislation. Some of the proposals make sense and some are so stupid that it's astonishing that they would ever go anywhere. Chances are that patent reform from the Legislative Branch will not go anywhere soon, but it is not dead yet, and folks who work in the patent field are reeling under what might be radical and foolish changes.

From the Judicial Branch: Three months ago, the United States Supreme Court issued a decision in the case of KSR Int'l Co. v. Teleflex Inc.. This case increased the flexibility of rejecting patent applications and voiding issued patents on grounds of obviousness. Incredibly, there are early indications that the USPTO is using this decision as a basis for providing shoddier examinations, even though the Supreme Court's opinion itself criticized (even insulted) the USPTO for providing shoddy examinations.

From the Executive Branch: The USPTO has just adopted a sweeping set of new rules, expected to take effect in two months, that will make it much harder to get a patent. These rules will affect all patent applications, not just those of the cranks, and the impact of these rules could be enormous. The purpose of these new rules, ostensibly, is to make patent examination easier and patents stronger. These might be noble purposes IF THE APPLICANTS WERE ASSURED THAT, IN EXCHANGE FOR ALL THIS EXTRA EFFORT, THEY WOULD BE GUARANTEED A THOROUGH, PROFESSIONAL EXAMINATION. Unfortunately, the rules are pretty one sided: the applicants have more obligations put on them, but they have no guarantee that the examiners will do a better job. Before these rules were adopted, the number one complaint was not that unworthy patents were being issued, but rather that deserving patents were being rejected because the examiners weren't doing their jobs. What the USPTO has done is comparable to a store taking notice that there are a lot of angry customers at its complaint department, so it "solves" the problem by reducing the hours that the complaint department is open. Voila! Fewer complaints!

These are the major concerns in patents in the US today. In view of these developments, there is no serious concern about silly patents for perpetual motion machines.

 

[LTC8K6]

Randi may have meant free energy devices such as the MEG.

 

[Brown]

Just as a quick follow-up, the three shocks to the US patent environment that I mentioned are not the only things making waves. Recent decisions from the Federal Circuit (the court one step below the Supreme Court) have held what appear to be perfectly legitimate patents invalid based upon, shall we say, surprising grounds. Waiting in the wings is a potential Supreme Court case that may cut back upon what is considered patentable (and it will likely have no effect upon bad science patents, although there is a chance that, in a perverse way, it may make them easier to obtain).

Also, if you do a search for issued patents for perpetual motion machines, you will be hard-pressed to find many. But if you do a search for patent applications for perpetual motion machines, you find some goodies. Remember, these are not written by, nor endorsed in any way by, the US Patent and Trademark Office, and are published on the USPTO web site only because the law requires it.

Energy Generation Device (by all means, look at the accompanying pictures: the applicant had them done up really nice and not on the cheap!):

[0003] This invention pertains to the field of perpetual motion machines. These are machines that require less energy to run that is outputted by the machine.
...
[0009] This invention pertains to a machine containing many moveable parts that outputs more energy than it takes to make it operate.

Here's another: Centrifugal motor (CM) (the pictures here look more like what you would expect from such an applicant):

[0003] The turbine's available power greatly exceeds that needed to keep the CM running i.e. to accelerate the necessary volumes of fluid and keep the cylindrical container spinning at operating speed. Such a "perpetual motion" machine was previously thought impossible; how could a machine actually produce energy? Energy is the ability to do work i.e. apply force over a distance. When Kinetic Energy (the energy of movement) is converted into Potential Energy (stored energy) and back again into KE, ordinarily there is at least some loss of energy due to friction. But the CM actually increases both the force and distance elements of the energy/work equation in two stages. In the first stage centrifugal force (artificial gravity) pressurizes fluid and produces a hybridform of energy i.e. Potential Energy in Motion. In the second stage, the PE of the spinning pressurized fluid is reconverted into KE as it drives a turbine. Using mechanical advantage in this re-conversion process, the force imparted to the turbine is increased many fold. With this force magnification, as is the case with use of mechanical advantage generally, the distance over which the turbine moves is correspondingly reduced; but only relative to the movement of the pressurization container. Since the actual or true movement of the turbine includes this relative movement plus the movement of the pressurized fluid at the periphery of the container (motion imparted in the first stage) the distance element of the energy equation is also increased; greater force plus greater distance equals greater energy!

Take a look at Torque generator, Fluid mechanics of inertia, or acceleration mechanics, or simplified calculus (there are lots of equations in this masterpiece, which hardly seems "simplified"), Perpetual motion comptrollers & energy molecule splitters (love that title!), Mechanical energy multiplier with long lasting accelerator cum brake (the language barrier is evident here).

There's also this nugget: Perpetual motion energy, which boldly claims (in a completely wrong format):

What I claim is:

1. A perpetual motion energy circuit for providing continuous or perpetual motion energy comprising:

2. Two or more electric batteries for providing electric current to a, Electric Wire Distributor Connector.

3. A Electric Wire Distributor Connector, for transmitting electric current and amperes to a electric motor.

4. A electric motor for providing power, and rotating a electric generator.

5. A generator for transmitting electric current to two or more batteries to charge said batteries.

6. A above said, Electric Wire Distributor Connector, for recycling one hundred percent or more energy back to above said batteries to recharge said batteries, which is perpetual motion energy.

7. Above said batteries do not have to be recharged by any other source.

8. Above said circuitry for powering electric cars or vehicles.

9. Above circuitry for providing electricity, and heat energy for residential and commercial buildings.

10. Above said circuitry for producing perpetual motion energy.

What makes this especially funny is that this inventor has a previous patent, in which he asserted that his invention enabled perpetual motion. That patent is this one: Electric wire distributor connector, in which he stated this incredible result:

The present invention relates to electric wiring, and to equipment for use in distributing electric current in one or more directions. More particularly, the present invention relates to an electric wire distributor connector for receiving and distributing electric current. More importantly the electric wire distributor connector can be modified to be used with any type of electric wiring. Another use is that it can also provide the means for the electric car to run continuously, by allowing the batteries to recharge themselves, and also run an electric motor, which runs a generator that recharges the batteries. This allows the electric car to run continuously, or in other words perpetually without being recharged by any outside source.

The claims in the patent, which are the legally operative part of the document, relate to an electric wire distributor connector, and make no mention of or allusion to perpetual motion. The connector, even if it didn't give you something for nothing, still could be a potentially useful configuration. Unfortunately, obtaining this patent may have encouraged this individual to open up his wallet and try for more patents.

Anyway, the point here is that there are plenty of perpetual motion patent applications that are pending, but there are nowhere near as many issued patents that allow claims to the concept.

It is always possible that an inventor, striving to find perpetual motion and thinking he's found it, may actually invent a better valve, or a more frictionless bearing, or a lower impedance connector, or a more robust ball-bearing conveyor. In such cases, he might get a patent on what he actually did invent, even if it doesn't really produce perpetual motion.

And by the way, even if his patent application has a load of pseudoscientific crap mixed in with the legitimate science, the law generally requires that any issued patent include all of the explanation filed in support of it, pseudoscientific crap and all. There's no editing allowed to chop out the pseudoscience, so it is likely that many patents have pseudoscientific pronouncements in them. Those pronouncements do not mean, however, that all of the document is pseudoscientific, and that the Patent Office is authorized to dismiss the application summarily. The time to become concerned is when the Patent Office allows claims for pseudoscientific crap.

 

[Mojo]

Models or exhibits are generally not admitted as part of an application or patent unless the requirements of 37 CFR 1.91 are satisfied. With the exception of cases involving perpetual motion, a model is not ordinarily required by the Office to demonstrate the operability of a device. If operability of a device is questioned, the applicant must establish it to the satisfaction of the examiner, but he or she may choose his or her own way of so doing. (emphasis mine)

There it is, in black and white.

Well, Steorn did say that they couldn't patent their invention. In view of the recent fiasco at Spitalfields, I guess this explains it.

 

[Horatius]

Brushing off the cranks is one thing. As Horatius points out, there is still the possibility that someone will deliberately obfuscate things to fool the examiner into issuing the patent.

By the way, fooling the examiner is not allowed. If you fool an examiner, or it you even try to do so, your patent is unenforceable in court.

I pretty much agree with what you've said in this post, and your later one. In fact, you highlight a lot of the points I made in my talk at TAM 5. As above, the cranks are pretty easy to dissuade, it's the people who deliberately set out to acquire a bad patent that are hardest to deal with. And they don't really care if their patent is unenforceable, as they never expect to take it to court. They instead use the prestige of having a patent to mislead investors or consumers.

As to the question of how many "perpetual motion" patents are out there, the relative number is pretty small, but that still allows for a disturbingly large number of such patents. In the 10 years I've worked as an examiner, the US has issued abut 2 million patents. Even if only a fraction of a percent involve perpetual motion, you're still talking about potentially hundreds of such patents. There will be a lot more applications than patents, but there's still a lot of issued patents.

But as Brown says, there are much bigger issues that the patent profession needs to addressed first. There's not much we as skeptics can do to fix those problems, but we can do something to raise awareness of the bad patents issue, so that people don't get conned into investing in shady technologies, which is really what these "inventors" are looking for. That's what I see my role here as, and that's why I submitted my talk to TAM last year - I wanted to give other skeptics some insights in to these questions, and some tools they could use to address them.

 

[Horatius]

Well, Steorn did say that they couldn't patent their invention. In view of the recent fiasco at Spitalfields, I guess this explains it.

Last I heard, they wanted to patent the individual parts of their device, in so far as that's possible. As Brown mentioned with the Electric wire distributor connector example, even if the part doesn't do exactly what they say it does, it still might have some other utility.

The problem with Steorn is, I haven't seen any evidence that they've actually applied for those patents. The one published application they have is a part they're said isn't a part of their over unity device. If they filed just before they started this whole brouhaha, I expect the applications should be published no later than September or October of this year, so we'll have a better idea in a couple of months.

 

[Mojo]

The one published application they have is a part they're said isn't a part of their over unity device.

Although it looked as if it might be for something like a PM device, because it was (as far as I recall) some sort of moveable shielding for magnets that used very little energy to move.

 

[bjb]

In Randi's defense, it appears the USPTO does grant patents to inventions that are intended to be fraudulent:

http://www.uspto.gov/web/offices/pac/mpep/documents/0700_706_03_a.htm

A rejection under 35 U.S.C. 101 for lack of utility should not be based on grounds that the invention is frivolous, fraudulent or against public policy.

and

…Congress never intended that the patent laws should displace the police powers of the States...we find no basis in section 101 to hold that inventions can be ruled unpatentable for lack of utility simply because they have the capacity to fool some members of the public

Randi may be correct about the USPTO granting patents to obviously fraudulent devices, but they must do so because they may not reject an invention simply because it is fraudulent. Randi tries very hard to inform the public about frauds so I understand his frustration. At best, he can point out fraudulent devices as they recieve patents, but there's not much he can do to stop this from happening.

 

[Horatius]

Randi may be correct about the USPTO granting patents to obviously fraudulent devices, but they must do so because they may not reject an invention simply because it is fraudulent. Randi tries very hard to inform the public about frauds so I understand his frustration. At best, he can point out fraudulent devices as they recieve patents, but there's not much he can do to stop this from happening.

I'd be careful about reading too much into that judgment. From what I've read, that decision dealt with the question of whether or not imitating another thing constitutes an invention. See a discussion of that here. It would seem the drink dispenser in question does do something real (dispenses drinks), but it also creates an impression that it does it in a different manner than what is actually used. So there is a some deception involved, and you could even argue that the use of such a device would be a fraud against the customer, but the device still ultimately has a real functionality.

A perpetual motion device, or some other fraudulent thing like the earlier mentioned bracelets, wouldn't have that underlying functionality, even if they managed to fool someone into thinking they were doing something. I think that would make a difference in how the law is applied.


But that's just my opinion, I could be wrong

 

[Brown]

In the US, patent examiners are supposed to perform their duties as outlined by the law. They don't monitor deceptive ads, or false promotions, or dirty tricks against investors, or mail scams, or criminal activity in general. Those are the bailiwicks of other governmental bodies.

The basic criteria the patent examiners have to apply are: useful, novel, nonobvious. There are other things they check as well (definiteness, double-patenting, etc.), but utility, novelty and nonobviousness are the big three. And those criteria are statutory, i.e., adopted by Congress. (Outside the US, many countries use a term like "having an inventive step" rather than "nonobvious.")

I've written about utility before. In the case of an electrical or mechanical invention, an absence of utility is almost never asserted. It is generally recognized that certain items have no "utility" in a patentability sense, even though they might be useful to nefarious folks. "A bomb that maximizes random civilian casualties during a terrorist operation" is the classic law-school example of a device that lacks all social utility, and therefore could not be awarded a patent.

Utility could also be lacking in an invention that does not work at all ... like a perpetual motion machine. But the law generally does not require inventors to build prototypes before getting a patent, and the USPTO is not a product-testing agency (there are other agencies that perform that function). So from the examiner's point of view, the examiner has to make a judgment about whether the idea might work. In the case of a pseudoscientific application that is flooded with drawings and equations and double-talk and legitimate science, the examiner might be deceived into thinking that a problem thought to be impossible has been solved--or at least dented--by some crackpot working out of his basement.

Perpetual motion is a special case. Makers of such devices are batting .000. No such device has ever worked, and vast work in the field of thermodynamics strongly suggests that no such device ever will. So if the inventor claims a perpetual motion device, such a device almost certainly has no utility, because it doesn't work, and all current knowledge says it cannot work. (As I mentioned earlier, the inventor still might have a legitimate claim to a component of his device, like a novel hinge. Even if the hinge doesn't eliminate all friction, it still might be useful to, say, hang a door.)

Whether such devices are "fraudulent" or not is not really in issue. Some crackpots actually have fooled themselves into thinking they've found the Holy Grail, and from a legal standpoint, they might lack the mental component of fraud, an intent to deceive others. Also, the issue is not really whether the purpose of the claimed invention is to create a false impression. There are thousands of patented inventions that do that: a toothpaste package that contains as much toothpaste as a competitor's package, but looks bigger; a magic trick apparatus; a challenging puzzle that appears easy to solve; a new kind of natural-looking glass eye or prosthesis; and so on.

 

[Brown]

Here's a deceptive US patent: Method and means for creating anti-gravity illusion. The whole invention is devoted to creating a false impression for purposes of amusement. (This link is to a different web site, for which you need a pdf viewer, not a TIFF viewer, to see the drawings. Look at FIG. 6 for an overall illustration of what the invention allows the user to do.)

The patent is notable because of the first named inventor. Yes, it's him. In the patent, he identifies himself as: "a professional entertainer [who] has incorporated dance steps in his recorded video performances...."

 

[Brown]

Cross-reference to a related thread in the Science, Mathematics, Medicine, and Technology sub-forum.

 

[T'ai Chi]

Randi is welcome to try and change the way the PTO operates rather than endless complaining.

 

[hoemaco]

I agree that patent officials can't check for the later uses / ads of the devices, that is, they only see a working drink dispenser, and don't know that it'll be sold as using some bioenergy to make you live for 200 yrs (just guessed, don't know the real story).

There are actually countries (eg. Hungary, and afaik most nearby countries?) where law (back from 19th century!) requires the patent office to check for the scientific soundness and applicatibility of the application. I'm not good at reading laws (that'd be a happy moment when and if only 1 country started writing laws in any 'human' language), but this is what I know: The officers analyze the claim (they try to assign it to someone who has expertise in that field, so they employ people with education and experience), and determines if the device can or can't work. If he thinks it can't, he has to prove it / but afaik if he can't prove it, he can ask the inventor to prove it, so they won't give a patent just because they couldn't prove it wrong. Also, they have to check some bigger countries databases (eg US, UK, Germany, France) to see if it's been already patented.

I believe it's a good system, though very taxing on the patent office, and would need some more restricted pre-filters esp in a big country like US. (I think they do reject PMs very quickly here), so the practical applicability is in question. Also, the whole process takes 3 years (yep), though you get the patent retroactively.

Some trick the frauds use is that they request a trade mark or similar thing (as that's the only thing they can get), for example someone I met had a method of astrology which she name 'genodynamics' (and actually stated it's not astrology and despised other astrologists, typical), and then claimed that she had a patent for this. When I asked the patent guys, they told me she only applied to protect the name of the method. Also, she told me she managed to have the method (which is also called astro-psychology, for better understanding ourselves) to be available for teacher's upgrade course, and the district council pays for it... (I hope this wasn't true).

 

[Horatius]

Randi is welcome to try and change the way the PTO operates rather than endless complaining.

And since his complaints about patents are being read by employees of the US and Canadian Patent Offices, his complaints might very well change them.

He's certainly having more effect than you.

 

[Horatius]

There's another comment on patents today. Seems we're not alone in trying to educate Randi

 

[Brown]

Mr. Randi wrote:

I stand corrected and contrite.

This single remark separates him from the vast number of persons who promote pseudoscience. Mr. Randi is willing to be shown the facts, and is willing to change his views (no matter how deeply held) if the evidence supports such a change.

I cannot recall a pseudoscientist ever saying something similar.