It isn’t because of the patent that I can’t reveal it. It is because of the nature of my contract. I can’t identify it in any context, including the context of patents. I was just answering the question about what I had patents for. I could have mentioned only the two, but if later my confidentiality were to be lifted, and I were to happen to mention the item, someone will inevitably point to this thread and yell, “You said you had patents for two things, and now you say you have three with some story about confidentiality. You’re just backpeddaling… and blah blah blah blah blah.” Stupid, I know. But I’ve run into that kind of thing before.
35 USC 101 says that a patented invention must have utility.
35 USC 112 says that it must enable a person of ordinary skill in the art to make and use the invention.
Way back when, you had to submit a working model for all patents, but that would be totally unworkable in the modern world. Nowadays, the USPTO only demands working models for things like perpetual motion machines (that is, the demand for a working model is basically an emphatic way to back up the 101 rejection that you get for disclosing an invention without utility). For chemicals, occasionally they will demand evidence that it works, but usually they take your word for it. The fight over that issue is more likely to take place in the courts when you try to enforce your patent.
Your disclosed utility can be pretty minor (although they’ve gotten a little better about it for chemicals - it used to be that if it was a solid, you claimed it was an abrasive, and if it was a liquid, you claimed it was a lubricant. If you weren’t sure, you claimed both). I don’t think that, “it makes people feel like they’re taking care of their health” would be an adequate utility, but I wouldn’t be shocked if that slipped through the USPTO, either.
The issuance of a patent definitely doesn’t mean that the product actually does what is claimed on its web site (or wherever). The web site is not part of the patent application.
I knew a guy who had a patent application rejected on the basis that his invention wouldn’t work. The rejection was incorrect - the guy took a model with him to the patent office in Washington, D.C. and proved that it worked.
The patent was for a new type of motorcycle compression release with a check valve to prevent dirt from being sucked into the cylinder on the down stroke. Like most compression releases it was controlled by a lever mounted on the handlebar. The release itself was mounted in the cylinder head. The release would open when you pushed down on the top of it, and a spring would close it when you stopped pushing. A cable between the handlebar lever and the compression release would transmit the force to open the release: the wire core of the cable was connected to the immovable base of the release so that when the lever pulled on the cable, it would cause the sheath of the cable to straighten out and push down on the movable top of the release.
The rejection from the patent office claimed that this cable mechanism wouldn’t work - that for the compression release valve to open, the engine would have to come up out of the frame.
When the inventor showed his model to a patent examiner, the examiner was apparently fascinated that the pulling action on the core of the cable was translated to a pushing action by the cable sheath. He showed it to another examiner who was an expert on cables, and that guy had never seen this, either. They tried to convince the inventor to apply for a patent for this application. He had to tell them that the idea wasn’t new - that all motorcycle compression releases worked this way.
Based on the current state of affairs with patents, I would say the answer to most of those questions is “no.”
Little if any verification. Does not have to be unique to be granted (they just wait for court challenges to sort it out). Possibly needs to make sense to someone, most of the crap I read about makes sense but is in no way novel and typically has lots of prior art.
So, will the PO automatically reject applications for devices that promise “free energy”? I seem to recall that a fair number of those worthless 'gas saving" devices for cars are actually pateneted. and what about wide ranging patents? like the use (by Microsoft) of “Windows” (which Apple calls “Icons”); are these rejected?
This can’t be right. I deal almost daily with banks and investment companies that have purchased patents for the sole purpose of suing someone for royalties as soon as it hits the market, but have no intention of making anything themselves.
Someone - the inventor - has to have figured out how to make the invention work (which is not the same thing as building a working one). Once they apply for a patent, they can sell the application (or later the issued patent). Whoever buys it then has the full rights, and can sue others for infringement. But the owner and the inventor(s) don’t have to be the same person.
Yes. The classic example of an unpatentable claim is a perpetual motion machine. If it’s ridiculous on its face, it will be rejected.
What do the patents actually claim that they do? The claims in the patents aren’t necessarily the same as the claims in the advertising.
This is not a patent issue. It’s a trademark issue. It happens to be handled by the same agency, the U.S. Patent and Trademark Office, but under a completely different set of legal principles.
For the purposes of a Microsoft claim over the term “Windows,” among the key questions are:
(1) Is the term “windows” a generic term for the purposes to which Microsoft puts it? (I don’t think so. The generic term is “operating system.” Apple calls its operating system the “Mac OS.”)
(2) Has Microsoft established a pattern of use (“use in commerce”) demonstrating that it is using the term “Windows” as a source identifier? (Yeah, I think so.)
(3) Does the allegedly infringing use (you’ll have to propose a hypothetical here) create a likelihood of confusion among consumers?
A patent holder is allowed to sit on his or her rights and not commercialize the invention. However, the patent is not valid unless the patentee actually knows how to implement it.
Here are some unproven designs that have been patented:
6,960,975 - Space vehicle propelled by the pressure of inflationary vacuum state - an antigravity spaceship
6,341,372 Universal machine translator of arbitrary languages - Star Trek-style universal language translator
6,891,310 Neutrino light to photon light converting matrix - neutrino radio and Star Trek-style long range sensors
Here’s a Wikipedia section on patents granted to what were effectively perpetual motion devices: Perpetual Motion: Patents
Here’s what I would consider non-novel:
4,022,227 Method of concealing partial baldness - a comb-over hair cut!
If it’s true that these claimed inventions are either not enabled or not novel, they will be invalidated if challenged in court. Of course, that first requires that the patent holder try to enforce the patent.
However, I’d be hesitant to jump to conclusions. “A comb over hair cut” is not necessarily the same thing as “this particular method of combing over hair.” I note in particular that the patent claims:
Do you know for sure that this isn’t novel? Do you have access to prior art? I couldn’t say without looking at the details of the patent and the previously known ways of combing over hair.
(This is not to say that I don’t think it’s stupid to allow a patent on a hair style. But that’s a different issue.)
Actually, with an “Improbability Drive”, you can.
No - the inventor and the patent attorney have to write a claim which makes a plausible argument that the invention will work. And utility covers a lot of ground. It can be wildly inefficent, but still useful.
If someone really tries to build the invention, and discovers that it does not work, and adds to it to make it work, he might be able to get a patent on the changes. Question: would he have any success in trying to invalidate the original patent on the grounds that it doesn’t work?
You are mixing up the term “Windows” and the technology of creating a windows based operating system. This involves, among other things, the methods of storing the image in a window when it is not being displayed, how to overlap, focus, etc. I know that Bell Labs had at least a few patterns on this - from a Blit terminal - and I am sure there are many others. The lawsuits about windowing systems had nothing to do with the term Windows itself, which is indeed a trademark issue.
I’m not sure I quite grasp the “no.” Voyager’s and ENugent’s statements don’t strike me as obviously contradictory. And, barring some subtlety that might be escaping me at the moment, both statements strike me as being true.
Remember, utility and enablement are two different issues. Utility answers the question “is it useful?” Enablement answers the question “does this patent tell me enough so that I can make it work?”
Going back to the example of the combover – Yes, it has utility in that it helps people cover their bald spots. Yes, it’s enabled, because it tells you how to do it.
This sounds right also, but again, not entirely contradictory. The exact claims of the first patent are non-enabled and therefore not patentable. The second patent, if it is enabled, is a valid patent. Strictly speaking, the two patents aren’t the same, but I think that’s essentially what ENugent meant.
I believe the answer is “yes.” That’s a claim of invalidity on the basis of non-enablement. Second inventor goes to court asking that the patent be declared invalid because it’s not enabled.
Ah, I see. I can’t say I know enough about the technology of operating systems to try to respond to the question in detail, but going back to the general question –
There’s nothing in patent law that says a patent can’t be “wide ranging.” The patent owner is going to try to get as wide an interpretation of it as possible. The patent examiner is going to look at the prior art and say – You have to narrow the language so that it doesn’t encompass things that are already out there.
It’s quite common for the patentee to grugingly narrow the claims until the patent is awarded and then, when bringing suit against an alleged infringer, to then argue for as broad an interpretation as possible. Courts quite frequently slap these attempts down, saying “Well, you disclaimed all this stuff when you were trying to get your patent (the ‘patent prosecution’ process), so you can’t claim it now.”
But in the IT field this is an ongoing issue. IT developers frequently complain that it’s impossible to look at (1) an IT patent and (2) the product they’ve developed, and from that figure out whether the patent owner is going to try to say that the product infringes the patent. The IT industry, very generally speaking, is coming to the conclusion that the patent system is interfering with business and they’re pursuing legislative fixes to help them out. They’re being opposed primarily by the pharmaceutical industry, which has a very different business model and thus a very different view of patents.
I worked in the US Patent & Trademark Office (USPTO) examining patents. No, your invention does not have to work, BUT (a) it cannot defy laws of nature. If you do claim to have a perpetual motion machine, the Patent Examiner will ask you to bring in a working model for the USPTO to examine, typically involving dissection which has revealed concealed batteries in the past. if you try this, it may be considered fraud since you have to sign key papers about the validity of your invention. (i.e.: don’t claim you’ve invented warp drive unless you can put your money where your mouth is.)
(b) Also, your patent’s claims must be enabling. Enabling is a legal term meaning that an ordinary person in the arts (i.e.: a non-expert in your field of endeavor) would be able to make and use the invention as claimed. For example, one cannot claim a foot-powered car that can go up to 60mph with the body of the application simply describing a Flinstone’s car because this is not adequate to explain how one is to make such a car.
If your claims are not enabling, this is a cause of a patent rejection which the applicant would have one chance for written rebuttal to overcome. If the applicant is still unhappy with the outcome, s/he can file for appeal within the USPTO…which takes a long time thereby eating into the life of your patent.
So, think carefully about what your invention can and cannot do. - Jinx
Which is pretty much what I said about utility in my post before the one you quoted.
Claims do not “argue” anything. Claims describe the metes and bounds of the invention, very much like the boundary description describes the metes and bounds of a piece of physical property. You look at all of the limitations described in the claims, and if the accused device/method has all of them, then it infringes the claims (barring a bizarre doctrine called the “reverse doctrine of equivalents” that may or may not actually exist any more). If it doesn’t quite have all of them literally, but it has something else that uses the same function in the same way to achieve the same result, it may infringe under the doctrine of equivalents.
The specification must describe the invention in enough detail to enable one of ordinary skill in the art to make and use the invention. In some cases, the claims (which are part of the specification) may themselves give enough detail to enable the invention - this is more likely in the software arts than, say, chemical.
I’ve been writing patent applications for ten years. How about you?
I was addressing your point that the inventor had to figure out how to make the invention work, which was the question. The level of detail that I, as an inventor, tell the patent attorney and others involved in writing the claims is not enough to prove that the invention would actually work, especially since most of mine have been software. I’ll agree that the patent does not contain an argument that the invention works, in any strict sense of argument. That someone would be able to reproduce the invention does not guarantee that it would work.
I just looked at the claims and documentation for my last patent, and the description of the algorithm - ooops - “program storage readable by a machine, tangibly embodying a program of instructions executable by the machine to performs a method …” would not allow a person skilled in the art to verify the claim.
I assure you, I’ve read enough patents to never wish to write one. But I can’t see how you, or an examiner, can expect to verify that the invention “works.” And I understand that works in this context has nothing to do with being useful in a production environment, say. I did ask if the demonstration that the patent does not work, based on the invention being made by someone following the patent, is grounds for invalidation.
Software is a funny animal for patenting, in that to a large extent, if you can describe it, you can do it. But you don’t have to “prove” it works - you have to describe it well enough that someone can build it without undue experimentation. In close cases, the tendency in the past (it’s significantly less so today) has been to go ahead and grant the patent, and let the patentee sort it out in litigation if the patent is ever enforced.
I’m not entirely sure what you mean by “verify the claim”. If you mean that one of ordinary skill in the art can’t program something that includes every limitation in your claims without undue experimentation, then yes, you have a problem with your patent. For your sake, I hope the patent owner is not attempting to assert it, and that you’re never deposed on it (and that the accused infringer doesn’t find this post).
If an invention obviously can’t work, based on casual inspection, the examiner will reject it on that basis (e.g., a perpetual motion machine). If there’s a serious question, he may ask for proof that it works. But a sworn statement by an inventor that he built it and it works may be enough “proof”. If the patent is litigated, there will be much, much more effort put into showing whether or not it satisfies the statutory standards of enablement, written description, and utility. It’s more efficient to do that at the stage where people are actually fighting about it than to have the USPTO test every application. (The vast majority of patents are never litigated. Virtually all important patents are litigated.)
Again, I’m having trouble following what you’re asking here with “based on the invention being made by someone following the patent”. Answering the question I think you’re asking, yes, if it can be shown that the patent does not enable one of ordinary skill in the art to make and use the invention without undue experimentation, then the patent can be invalidated. This generally comes up in the context of patent litigation. Note that some experimentation is ok, as are minor errors in the specification that one of skill in the art would recognize once he started working on it.
(If that wasn’t the question, could you restate it?)
And this, I think, is the answer to the OP, and to my last question, which you did interpret correctly.
The claims are correct, because in this case I did implement the technique before patenting it. In my understanding, “skilled in the state of the art” does not mean a master, but someone reasonably competent in the given field. Would there be an issue if to accomplish the claim, a programming trick is required? The trick itself might exist in the literature and be neither a claim nor patentable. In this case anyone doubting the claim could be convinced by a demonstration.
Thanks. That is the answer I was looking for. If I understand correctly, unless the patent violates known physical laws, or is somehow internally inconsistent, then it is assumed to work based on the statement of the inventor, righrt? Do you know what percent of patents are litigated based on assertions that it does not work, as opposed to undiscovered prior art or inadequate claims, etc.? I don’t remember reading about any. I assume the specific incorrect claim would be invalidated, or would the entire patent be?
Usually. Unless the examiner gets a bug up his butt for some unknown reason. Sometimes he just plain doesn’t believe it, or sometimes he disputes the meaning of experimental data provided in the patent or in a later supporting declaration.
Validity is on a claim-by-claim basis, although an argument would probably also be made that the entire patent is unenforceable on the basis of fraud against the patent office. (That’s a tough claim to prove, though - you have to show material misrepresentation/omission and bad intent.)
Almost all patent suits include counterclaims that the patent is invalid because it is not enabled or fails written description, and often it is also claimed that it lacks utility. Those may or may not be the basis for decision, but it’s certainly an argument that’s made every day. It’s somewhat less common in the software arena, because in most cases if you can describe it, you can program it. It’s extremely common in chemical cases.
Patently-O is a nice blog for summarizing appellate patent opinions. Here’s a few writeups of cases decided on enablement/utility grounds.
Elster Electricity v. Schlumberger
In re Dash and Keefe
In re Fisher
(The last two concern cases where the USPTO denied a patent on enablement and/or utility grounds.)