The patent system in general and the USPTO in particular is a very strange amalgam. As a holder of multiple patents (both hardware and method/soft), I've found the process to be extremely odd.
First you have to phrase whatever your technology involves in a legal vernacular that is weirdly constraining and rarely maps well onto what the thing in question actually does. You always have the sense that you're playing a game of legal-mumbo-jumbo performance art more than actually describing an invention.
Second, you then have to interact with patent examiners. Some are great, many not so much. One I encountered was possibly the dumbest person I have ever met. Thick as two short planks with two extra short planks nailed on.
Third, the process just seems to go on and on. The examiner will make an objection or point to prior art. These actions may have little basis in logic or common sense, but they do seem to move pieces of paper from one box to another. Meetings occur with the examiner and/or their supervisor where your jaw will drop at how little anyone involved understands the technology being claimed.
You can cut through a lot of this with good legal counsel, but that is really just contributing to the gamesmanship of the whole process.
And, in the end, a patent is a pretty meaningless thing. A portfolio of defended patents might be an "Intellectual Property" asset (what an awful term). But so many people have to jump through the process just as a defensive act so they have a sliver of reassurance that using the novel thing they came up with will not lead to immediate cease-and-desist injunctions from IP-fortified competitors. What a collective waste of human creativity.
> You can cut through a lot of this with good legal counsel, but that is really just contributing to the gamesmanship of the whole process.
True and true. In some ways perhaps that’s the point. It is a game of sorts, it is business competition; businesses temporarily monopolizing business on inventions in return for disclosing them. And it makes sense that legal protections for businesses are better & easier for those businesses when they spend money on lawyers, if a little pessimistic. Unfortunately the system is somewhat motivated to try to cut out people who can’t afford lawyers.
You mentioned bad/dumb patent examiners. I’ve never worked with them and only worked with patent lawyers, and I have to say I’ve been constantly impressed with their ability to absorb, understand, and describe highly technical work in legalese (and yes the language is sometimes performative and can sound funny to engineers). Teams creating these inventions take months and years to make something patentable, and when I think about how the lawyer’s job is to understand these new inventions in a short amount of time, over and over, week after week, I feel like they have a hard job and I’m routinely surprised at how well and how quickly they understand it.
>> I’ve been constantly impressed with their ability to absorb, understand, and describe highly technical work in legalese (and yes the language is sometimes performative and can sound funny to engineers).
This is one of the saddest aspects: so much high-value effort and skill towards an end that, in the whole, I view as a massive drain and retardant on human development.
- Documenting and classifying inventions is valuable
- people choose to become a patent clerk/lawyer because of the stability. It enables them to take care of family, pursue hobbies, etc (see Einstein).
- the alternative for some is being an engineer but for others it might be a librarian
Patents are close to completely useless as a form of documentation. It's not valuable.
You may be thinking about old patents where people wrote real engineering information with real details on them that excluded non-working alternatives. Patents are not like that anymore.
The latter points are valid, but for the first - software engineers are usually best advised not to spend time looking at the patent database. For a long time (is it still true?) there were triple damages for knowingly infringing a parent, and in any case the function of the patent database as a publication of ideas is extremely small in software. I have literally not heard of anyone locating something they need to implement by looking at it.
A good lesson I learned from one of my patent attorneys was: "Never try to write claim language yourself." You're much better off writing it in straightforward language that you understand yourself, and let the lawyers translate it into good claims, which you can then review.
Of course this assumes access to lawyers who are familiar enough with the subject matter. But if you don't have that, then you're screwed, and your patents will be weak.
In my submissions to the patent lawyers, I have a section such as "proposed claims" and I write them in conversational language. I explain why I think each claim is important. This is also useful when other people such as managers have to be involved in deciding whether an invention is worth patenting or not.
> You always have the sense that you're playing a game of legal-mumbo-jumbo performance art more than actually describing an invention.
My first patent was filed a few weeks ago, with consultations with a patent attorney. This is pretty much the highlight of my takeaway as well.
It helps to have a good understanding of the English dictionary and thesaurus to be able to negotiate wording. But yes, it's still mumbo-jumbo. Inventions might be "obvious" to someone else in the trade, but it's hard to say if it's going to be obvious if it isn't described using that trade's mumbo-jumbo instead of Legal mumbo-jumbo.
> But so many people have to jump through the process just as a defensive act so they have a sliver of reassurance that using the novel thing they came up with will not lead to immediate cease-and-desist injunctions from IP-fortified competitors.
Having a patent is not a defense to an allegation of infringement and if that's what your legal counsel told you, then you should seek new counsel.
As a holder of multiple patents (some thru employers, and several, now lapsed, that I tried to get personally) the entire system is a joke. As a few people have said here, it seems like it is designed entirely to obfuscate the process, make money for lawyers, and entrench the value of the existing portfolios of large patent holders.
In the process of doing some research on patents a while back I came across this article [0] which really resonated with me.
Part of the problem is that there are no more easy inventions left. The time when you could get a dozen new inventions just sticking a pair of electrodes into a vacuum chamber in your garage is long over.
“If you disclose your invention publicly and do not file an application within one year, you could be barred from receiving a patent on that exact invention,”
I am not a layer not patent attorney, but I have extensive experience as a former asset manager of a IP portfolio of an international stock market traded information company and inventor/co-inventor on multiple US and non-US patents and patent applications.
This sentence is poor advice, since the one-year rule exists only in the U.S., whereas other countries have stricter rules. So it is recommended to always keep things secret (meaning you can only tell people under NDA about it) until a patent application is filed. You will normally want to be able to exercise your patent rights (a government-granted, time-limited, territorial monopoly) in multiple jurisdictions, not just the U.S., as your competitors will likely sit not just in the U.S. (if you file only in one country, people can work around your patent by using the method disclosed therein in another, unprotected jurisdiction instead, which is perfectly legal).
Another piece of advice, namely to file a provisional application first, is only advisable if you are not quite ready to file the full application, yet you are afraid someone else may file first. In most circumstances it may be smarter to file a full application as the first action; the reason is that no new material matter may be added to the provisional application, and in the course of writing and "fine tuning" the language of the final application, one often wishes to include additional matter. To sum it up, a provisional freezes the scope of the application, while not providing you an earlier "priority date".
You need to understand IP well to play this game; most people don't. It's also a game with large capital requirements ($100k/lifetime of a patent, including extension fees, foreign filings and translations) and very long time lines (double decades rather than weeks). But if executed carefully, a single transaction (case of patent litigation or licensing) may bring an IP owner more $$$ than they may earn with their normal operational business.
Unless you have a business that can easily support these costs AND a plan to make extra money from the exclusivity that patent award gives you then I don’t see how this makes financial sense for any small entrepreneur.
“If you disclose your invention publicly and do not file an application within one year, you could be barred from receiving a patent on that exact invention,”
Wasn't this rule retired? I vaguely remember something in this regard.
"It's also a game with large capital requirements "
True. And many BS patents from big corporations are getting granted because their lawyers fight day and night to have it granted. And many solid patents from single inventors get rejected. The patent examiner is just lazy, likely does not even read your applications and rejects it on grounds that make it obvious that he has neither, read nor understood your applications (dont ask :-) ). While this can be fixed, it leaves your skeptical about the whole system.
There should be a legally accepted way to publish something, free of cost, that demonstrates a new invention, that prevents other people patenting it, and can allow the person who published it to claim the credit for doing so.
I'm not sure what's so difficult about this concept in the age of verifiable chains of communications and transactions.
There is: Publishing it on a blog or in a scientific publication does that. At trial, your blog can be used to invalidate any patent claims that it predates.
That is supported by the article, but I would find it hard to imagine that an inventor who is trying to assert their rights against a much more well-resourced competitor would want things to go to trial. The article is pretty clear that filing first is what matters, not posting first, and first to file provides much, much, stronger evidence of your claim than a social media post.
Well there’s literally nothing you can do about that when every party has a right to a trial. How are conflicts supposed to be resolved if one party doesn’t want to go to court?
I'm confused because my understanding is that "first to file" rather than "first to invent" is now basically universal.
My assumption was that tagline meant what to said, that it didn't matter when you create or publish something, that the government was open to giving a monopoly to whomever jumps through the legal hoops first.
...which has been a major load of salt I've carried around with me.
When you make a public disclosure (this includes things like "releasing a product that uses that method", fwiw), that starts a countdown clock [1]. When that clock reaches 0, that disclosure is now prior art, and anyone who files for a patent after that point is unable to get a patent for that method. Sufficiently old public disclosure invalidates any patent, independent of any filing system for patents.
First-to-file and first-to-invent are instead what happens if you get two different people who try to file for a patent before that point. In a first-to-invent system, there's an adjudication process that's going to look at all of the internal records to figure out who "invented" the process first (which, in the absence of evidence, is going to default to the first person to file anyways). In a first-to-file system, the first person to get their paperwork in the door is going to get the patent.
This doesn't mean that an unscrupulous person gets to look for public disclosures and rush their own patent applications to beat out the original inventor. You still need to demonstrate independent conception of the idea (although this would likely only be demonstrated in the context of an actual trial).
[1] I don't remember it's length, but I want to say it's like a year.
> This doesn't mean that an unscrupulous person gets to look for public disclosures and rush their own patent applications to beat out the original inventor. You still need to demonstrate independent conception of the idea (although this would likely only be demonstrated in the context of an actual trial).
IPRs and PGRs can also be used for this, and are cheaper than doing it at a full-blown trial.
These rules only apply when two people with the same invention both file a patent. If you publish something on Monday and someone files a patent (really - a provisional) for it on Tuesday, you are prior art to their patent and can be used to make it "obvious."
First to file determines who gets the patent, if anyone. Anything out in the public as verifiable prior art can invalidate all such patents, giving you (and your competitors) the freedom to operate.
A simple social media post can work as prior art, especially if it has details or some kind of diagram. But, that only limits their ability to patent it. Your own ability to monetise it is not protected by that. For independent inventors, licensing a patent is far more practical and lucrative than actually manufacturing something or releasing complex software packages. If you are raising VC money, having any kind of patent can increase your chances by a substantial factor.
Does licensing from smaller actors happens much in practice? It feels to me as there is a huge asymmetry of the ability of corps either infringing the patent directly, or circumventing it vs an individuals budget for infringement discovery and lawyers. I wonder even if bigger players like universities have much success in licensing their IP.
This has always existed. You disclose the idea. Things that are public cannot be patented.
Put your great idea up on arxiv and it becomes prior work which the patent office can find. Or publish it in some conference. Etc. It gets officially timestamped and you're all done. You don't need any of this useless verifiable chain of communication nonsense. The world figured out how to do this centuries ago in far simpler ways.
Whether the patent office finds your prior art or not, that's a whole other question. They may still grant a patent no matter how obviously it overlaps with prior art. But then you can use your prior art when they come to try and shake you down, if it comes down to that.
It’s easy to fake if you are part of YC. You think they would lose 1B+ potential revenue if the only thing they have to do is to change one database record that nobody except them is monitoring ?
it's pretty hard to unpublish something on the internet though. How many system do you think are consuming HN content? how many consume those? You can't un-ring a bell, so a simple social post is almost always discoverable in the future.
It took 4 years for me to get any communication from the USPTO after I filed. I had a patent lawyer prepare the application and transmit it electronically. There was a confirmation code of receipt and that was it. Nothing for 4 years and then an office action saying no. Ok, fine. One of the reasons offered was that there was a Chinese patent for something similar, owned by Bytedance. I read their patent and it is a huge word salad of a vast array of concepts, with very little application. When it comes to technology, I believe the whole America patent system is dominated by corporations and foreign players. Despite the new marketing of USPTO there is no way an independent inventor has a chance in this system.
Because patent specialists are experts at defining the widest possible scope for their idea, when I have reason to read a patent I find it a confusing word mess of noise, that does not have any meaningful interpretation until a case is argued in court, at which point the word salad is brought into the appropriate interpreted focus.
I am not convinced an independent inventor is able to ever file a patent that can not be countered from another patent that had no intention to cover the inventor's use case when written, it's just a useful tool to protect the filer in any and all scenarios.
Does anyone - except patent specialists who make their living from this - see value in the patent process at all any more? We're a long, long way from the original intent of the process these days, it seems.
Hi, I work at USPTO, the following statements and opinions are my own and represent me only:
1) While many do indeed file for the broadest patent possible, in my (limited) experience so far this makes an application much easier to reject. The application that I've seen so far that was the most difficult to reject was made difficult by extremely specific claims.
2) The best argument I've ever heard in favor of the USPTO is that it allows small inventors to hedge market competition risk. Imagine you are an AI startup, developing a novel model architecture for a domain-specific application. You file a patent for the method, and once your product begins selling and Microsoft creates a fast-follow product you can sue. Microsoft will have much more to gain from buying your company at that point than giving up the market to some plucky dorks with a Patent, and your IPO is just the cost of doing business for them.
3) To the extent that "it is very difficult to get a man to believe something his paycheck depends on him denying", I am aware of a personal incentive to argue for the utility of the USPTO. Idk what else to say besides offer me a job on your engineering team and I'll quit immediately. (edit: actually I do know what else to say: If you don't care about patents don't apply for one. Fully optional endeavor. If you do care about patents, then I am here to help.)
I remember learning about the pros and cons of patents vs trade secrets a few years ago- it's a lot less cut and dried than that. With trade secrets, you just have to make a reasonable effort to keep them secret, and then if it gets out you can sue for theft (or some such). As long as there's a secret sauce you can keep out of the public eye it's nearly functionally equivalent.
> The application that I've seen so far that was the most difficult to reject was made difficult by extremely specific claims.
Tradeoffs: the narrower the patent, the less useful it is, because it's that much easier for your competition to just tweak something inconsequential.
> once your product begins selling and Microsoft creates a fast-follow product you can sue
Not that easy - you need to be able to afford a patent law fight. Big companies like MS might just take it anyway and risk the lawsuit, because they can afford to drag it out longer than you can stay solvent.
On the other hand, if you go the trade secret route, Microsoft can't fast-follow because you didn't publish the whole thing.
Having filed multiple patents, it does not leave a good taste in your mouth (esp software patents). You know you are bullshitting but unfortunately it needs to be done because others will do it and deny you to use the invention.
Fully agree. Over 20 years ago, I did the same job as the Patent Engineer (I had a slightly different title, but same role) at a big law firm. Half of the patents I filed were BS software patents for a name brand tech company, and the other half were for actually novel and cool inventions for things like super-efficient distillation/chilling systems for coffee and satellite antennas for moving vehicles.
Getting that job was almost as easy as getting a software job in 2019. I had a masters in EE and a few years of SWE experience. I sent out 5 applications to top local law firms, got 3 interviews, and one offer over the course of a month. Apparently, the bulk of people in that role were post-docs in bio/chem who weren't excited about managing student loan debt while being underpaid lab people. (The EE/CS contingent was 5 out of 30 in the role, and I was the only one with actual SWE experience, so it gave the me and the firm creditability with inventors)
My salary matched my SWE salary, I had my own office + secretary, and the firm paid overtime and law school tuition. They also charged $250/hour for my time. I also only lasted about year in that role since it was very isolating and not collaborative.
The coolest part of that whole experience came a couple years later when I went to dinner with my wife, her close friend, and the friend's serious boyfriend. In the "what do you do" chit chat, he mentioned working with someone on a new way to brew and chill coffee quickly that they were close to bringing to market.
When I asked if it works like [describes invention exactly] his mind was absolutely blown away. As it turns out, I wrote their patent.
This assertion needs a lawsuit and deep pockets. Most startups will not be able to do this.
Infact the patent holder can file a lawsuit against a startup. Now even if the startup can prove that the patent is invalid, it has to spend precious money on the lawsuit.
>"If you disclose your invention publicly and do not file an application within one year, you could be barred from receiving a patent on that exact invention."
Does this mean if I disclose an invention, that nobody can patent it after a year?
This is my explicit reason for writing about BitGrid out in the open, since 2004.[1]
If it turns out to be useful, that's awesome. If not, it's been a fun mud ball to polish.[2]
> Does this mean if I disclose an invention, that nobody can patent it after a year?
Others can. But during their application who is going to tell that there is a prior art? Patent offices only do basic checks on existing patents and not on the wider internet.
Once the others get their patent, you can appeal [1] it.
The patent system in general and the USPTO in particular is a very strange amalgam. As a holder of multiple patents (both hardware and method/soft), I've found the process to be extremely odd.
First you have to phrase whatever your technology involves in a legal vernacular that is weirdly constraining and rarely maps well onto what the thing in question actually does. You always have the sense that you're playing a game of legal-mumbo-jumbo performance art more than actually describing an invention.
Second, you then have to interact with patent examiners. Some are great, many not so much. One I encountered was possibly the dumbest person I have ever met. Thick as two short planks with two extra short planks nailed on.
Third, the process just seems to go on and on. The examiner will make an objection or point to prior art. These actions may have little basis in logic or common sense, but they do seem to move pieces of paper from one box to another. Meetings occur with the examiner and/or their supervisor where your jaw will drop at how little anyone involved understands the technology being claimed.
You can cut through a lot of this with good legal counsel, but that is really just contributing to the gamesmanship of the whole process.
And, in the end, a patent is a pretty meaningless thing. A portfolio of defended patents might be an "Intellectual Property" asset (what an awful term). But so many people have to jump through the process just as a defensive act so they have a sliver of reassurance that using the novel thing they came up with will not lead to immediate cease-and-desist injunctions from IP-fortified competitors. What a collective waste of human creativity.
> You can cut through a lot of this with good legal counsel, but that is really just contributing to the gamesmanship of the whole process.
True and true. In some ways perhaps that’s the point. It is a game of sorts, it is business competition; businesses temporarily monopolizing business on inventions in return for disclosing them. And it makes sense that legal protections for businesses are better & easier for those businesses when they spend money on lawyers, if a little pessimistic. Unfortunately the system is somewhat motivated to try to cut out people who can’t afford lawyers.
You mentioned bad/dumb patent examiners. I’ve never worked with them and only worked with patent lawyers, and I have to say I’ve been constantly impressed with their ability to absorb, understand, and describe highly technical work in legalese (and yes the language is sometimes performative and can sound funny to engineers). Teams creating these inventions take months and years to make something patentable, and when I think about how the lawyer’s job is to understand these new inventions in a short amount of time, over and over, week after week, I feel like they have a hard job and I’m routinely surprised at how well and how quickly they understand it.
>> I’ve been constantly impressed with their ability to absorb, understand, and describe highly technical work in legalese (and yes the language is sometimes performative and can sound funny to engineers).
This is one of the saddest aspects: so much high-value effort and skill towards an end that, in the whole, I view as a massive drain and retardant on human development.
- Documenting and classifying inventions is valuable - people choose to become a patent clerk/lawyer because of the stability. It enables them to take care of family, pursue hobbies, etc (see Einstein). - the alternative for some is being an engineer but for others it might be a librarian
Patents are close to completely useless as a form of documentation. It's not valuable.
You may be thinking about old patents where people wrote real engineering information with real details on them that excluded non-working alternatives. Patents are not like that anymore.
The latter points are valid, but for the first - software engineers are usually best advised not to spend time looking at the patent database. For a long time (is it still true?) there were triple damages for knowingly infringing a parent, and in any case the function of the patent database as a publication of ideas is extremely small in software. I have literally not heard of anyone locating something they need to implement by looking at it.
A good lesson I learned from one of my patent attorneys was: "Never try to write claim language yourself." You're much better off writing it in straightforward language that you understand yourself, and let the lawyers translate it into good claims, which you can then review.
Of course this assumes access to lawyers who are familiar enough with the subject matter. But if you don't have that, then you're screwed, and your patents will be weak.
In my submissions to the patent lawyers, I have a section such as "proposed claims" and I write them in conversational language. I explain why I think each claim is important. This is also useful when other people such as managers have to be involved in deciding whether an invention is worth patenting or not.
> You always have the sense that you're playing a game of legal-mumbo-jumbo performance art more than actually describing an invention.
My first patent was filed a few weeks ago, with consultations with a patent attorney. This is pretty much the highlight of my takeaway as well.
It helps to have a good understanding of the English dictionary and thesaurus to be able to negotiate wording. But yes, it's still mumbo-jumbo. Inventions might be "obvious" to someone else in the trade, but it's hard to say if it's going to be obvious if it isn't described using that trade's mumbo-jumbo instead of Legal mumbo-jumbo.
No matter the situation, in America the lawyers always win.
> But so many people have to jump through the process just as a defensive act so they have a sliver of reassurance that using the novel thing they came up with will not lead to immediate cease-and-desist injunctions from IP-fortified competitors.
Having a patent is not a defense to an allegation of infringement and if that's what your legal counsel told you, then you should seek new counsel.
It's just a way to get a seat at the negotiation table. If they sue you, you could potentially countersue them for patent infringement.
Yes, it's all a bunch of bullshit, but that is what you get when you have a flawed notion of how inventing works.
As a holder of multiple patents (some thru employers, and several, now lapsed, that I tried to get personally) the entire system is a joke. As a few people have said here, it seems like it is designed entirely to obfuscate the process, make money for lawyers, and entrench the value of the existing portfolios of large patent holders.
In the process of doing some research on patents a while back I came across this article [0] which really resonated with me.
[0]: “Against intellectual property” — Brian Martin, https://documents.uow.edu.au/~/bmartin/pubs/95psa.html
Part of the problem is that there are no more easy inventions left. The time when you could get a dozen new inventions just sticking a pair of electrodes into a vacuum chamber in your garage is long over.
“If you disclose your invention publicly and do not file an application within one year, you could be barred from receiving a patent on that exact invention,”
I am not a layer not patent attorney, but I have extensive experience as a former asset manager of a IP portfolio of an international stock market traded information company and inventor/co-inventor on multiple US and non-US patents and patent applications.
This sentence is poor advice, since the one-year rule exists only in the U.S., whereas other countries have stricter rules. So it is recommended to always keep things secret (meaning you can only tell people under NDA about it) until a patent application is filed. You will normally want to be able to exercise your patent rights (a government-granted, time-limited, territorial monopoly) in multiple jurisdictions, not just the U.S., as your competitors will likely sit not just in the U.S. (if you file only in one country, people can work around your patent by using the method disclosed therein in another, unprotected jurisdiction instead, which is perfectly legal).
Another piece of advice, namely to file a provisional application first, is only advisable if you are not quite ready to file the full application, yet you are afraid someone else may file first. In most circumstances it may be smarter to file a full application as the first action; the reason is that no new material matter may be added to the provisional application, and in the course of writing and "fine tuning" the language of the final application, one often wishes to include additional matter. To sum it up, a provisional freezes the scope of the application, while not providing you an earlier "priority date".
You need to understand IP well to play this game; most people don't. It's also a game with large capital requirements ($100k/lifetime of a patent, including extension fees, foreign filings and translations) and very long time lines (double decades rather than weeks). But if executed carefully, a single transaction (case of patent litigation or licensing) may bring an IP owner more $$$ than they may earn with their normal operational business.
Unless you have a business that can easily support these costs AND a plan to make extra money from the exclusivity that patent award gives you then I don’t see how this makes financial sense for any small entrepreneur.
“If you disclose your invention publicly and do not file an application within one year, you could be barred from receiving a patent on that exact invention,”
Wasn't this rule retired? I vaguely remember something in this regard.
"It's also a game with large capital requirements "
True. And many BS patents from big corporations are getting granted because their lawyers fight day and night to have it granted. And many solid patents from single inventors get rejected. The patent examiner is just lazy, likely does not even read your applications and rejects it on grounds that make it obvious that he has neither, read nor understood your applications (dont ask :-) ). While this can be fixed, it leaves your skeptical about the whole system.
I, too, am not a lawyer, but ended up overseeing the IP portfolio of a publicly traded EU company. Small world...
I comment in geneal support of this advice.
We have started to submit a presentation that would be used for an examiner interview as our provisional application.
There should be a legally accepted way to publish something, free of cost, that demonstrates a new invention, that prevents other people patenting it, and can allow the person who published it to claim the credit for doing so.
I'm not sure what's so difficult about this concept in the age of verifiable chains of communications and transactions.
There is: Publishing it on a blog or in a scientific publication does that. At trial, your blog can be used to invalidate any patent claims that it predates.
That is supported by the article, but I would find it hard to imagine that an inventor who is trying to assert their rights against a much more well-resourced competitor would want things to go to trial. The article is pretty clear that filing first is what matters, not posting first, and first to file provides much, much, stronger evidence of your claim than a social media post.
You don't need to go to trial to invalidate a patent. It happens frequently at the motion to dismiss or summary judgment phases.
you still need the resources to go through the courts though. This closes out the option for almost all.
Well there’s literally nothing you can do about that when every party has a right to a trial. How are conflicts supposed to be resolved if one party doesn’t want to go to court?
This is correct. And it is called ensuring 'Freedom to operate' - FTO.
I'm confused because my understanding is that "first to file" rather than "first to invent" is now basically universal.
My assumption was that tagline meant what to said, that it didn't matter when you create or publish something, that the government was open to giving a monopoly to whomever jumps through the legal hoops first.
...which has been a major load of salt I've carried around with me.
When you make a public disclosure (this includes things like "releasing a product that uses that method", fwiw), that starts a countdown clock [1]. When that clock reaches 0, that disclosure is now prior art, and anyone who files for a patent after that point is unable to get a patent for that method. Sufficiently old public disclosure invalidates any patent, independent of any filing system for patents.
First-to-file and first-to-invent are instead what happens if you get two different people who try to file for a patent before that point. In a first-to-invent system, there's an adjudication process that's going to look at all of the internal records to figure out who "invented" the process first (which, in the absence of evidence, is going to default to the first person to file anyways). In a first-to-file system, the first person to get their paperwork in the door is going to get the patent.
This doesn't mean that an unscrupulous person gets to look for public disclosures and rush their own patent applications to beat out the original inventor. You still need to demonstrate independent conception of the idea (although this would likely only be demonstrated in the context of an actual trial).
[1] I don't remember it's length, but I want to say it's like a year.
> This doesn't mean that an unscrupulous person gets to look for public disclosures and rush their own patent applications to beat out the original inventor. You still need to demonstrate independent conception of the idea (although this would likely only be demonstrated in the context of an actual trial).
IPRs and PGRs can also be used for this, and are cheaper than doing it at a full-blown trial.
> "first to file" rather than "first to invent"
These rules only apply when two people with the same invention both file a patent. If you publish something on Monday and someone files a patent (really - a provisional) for it on Tuesday, you are prior art to their patent and can be used to make it "obvious."
First to file determines who gets the patent, if anyone. Anything out in the public as verifiable prior art can invalidate all such patents, giving you (and your competitors) the freedom to operate.
A simple social media post can work as prior art, especially if it has details or some kind of diagram. But, that only limits their ability to patent it. Your own ability to monetise it is not protected by that. For independent inventors, licensing a patent is far more practical and lucrative than actually manufacturing something or releasing complex software packages. If you are raising VC money, having any kind of patent can increase your chances by a substantial factor.
Does licensing from smaller actors happens much in practice? It feels to me as there is a huge asymmetry of the ability of corps either infringing the patent directly, or circumventing it vs an individuals budget for infringement discovery and lawyers. I wonder even if bigger players like universities have much success in licensing their IP.
Yes, it does, but it often needs at least an implied threat of a lawsuit or some extra deliverables that help implementation of the technology.
Patents also turn "I got crushed by Google" stories into legal battles done by patent trolls that often have a decent payoff.
I guess if you have a patent and a big corp infringes it, then a patent troll would be more than willing to buy it from you.
This has always existed. You disclose the idea. Things that are public cannot be patented.
Put your great idea up on arxiv and it becomes prior work which the patent office can find. Or publish it in some conference. Etc. It gets officially timestamped and you're all done. You don't need any of this useless verifiable chain of communication nonsense. The world figured out how to do this centuries ago in far simpler ways.
Whether the patent office finds your prior art or not, that's a whole other question. They may still grant a patent no matter how obviously it overlaps with prior art. But then you can use your prior art when they come to try and shake you down, if it comes down to that.
Couldn’t you post it on hacker news or X? That would have a pretty strong timestamp.
HN timestamps get rewritten for second chance pool submissions and possibly others.
It’s easy to fake if you are part of YC. You think they would lose 1B+ potential revenue if the only thing they have to do is to change one database record that nobody except them is monitoring ?
it's pretty hard to unpublish something on the internet though. How many system do you think are consuming HN content? how many consume those? You can't un-ring a bell, so a simple social post is almost always discoverable in the future.
They're not the only ones monitoring their own database.
It took 4 years for me to get any communication from the USPTO after I filed. I had a patent lawyer prepare the application and transmit it electronically. There was a confirmation code of receipt and that was it. Nothing for 4 years and then an office action saying no. Ok, fine. One of the reasons offered was that there was a Chinese patent for something similar, owned by Bytedance. I read their patent and it is a huge word salad of a vast array of concepts, with very little application. When it comes to technology, I believe the whole America patent system is dominated by corporations and foreign players. Despite the new marketing of USPTO there is no way an independent inventor has a chance in this system.
Because patent specialists are experts at defining the widest possible scope for their idea, when I have reason to read a patent I find it a confusing word mess of noise, that does not have any meaningful interpretation until a case is argued in court, at which point the word salad is brought into the appropriate interpreted focus.
I am not convinced an independent inventor is able to ever file a patent that can not be countered from another patent that had no intention to cover the inventor's use case when written, it's just a useful tool to protect the filer in any and all scenarios.
Does anyone - except patent specialists who make their living from this - see value in the patent process at all any more? We're a long, long way from the original intent of the process these days, it seems.
Hi, I work at USPTO, the following statements and opinions are my own and represent me only:
1) While many do indeed file for the broadest patent possible, in my (limited) experience so far this makes an application much easier to reject. The application that I've seen so far that was the most difficult to reject was made difficult by extremely specific claims.
2) The best argument I've ever heard in favor of the USPTO is that it allows small inventors to hedge market competition risk. Imagine you are an AI startup, developing a novel model architecture for a domain-specific application. You file a patent for the method, and once your product begins selling and Microsoft creates a fast-follow product you can sue. Microsoft will have much more to gain from buying your company at that point than giving up the market to some plucky dorks with a Patent, and your IPO is just the cost of doing business for them.
3) To the extent that "it is very difficult to get a man to believe something his paycheck depends on him denying", I am aware of a personal incentive to argue for the utility of the USPTO. Idk what else to say besides offer me a job on your engineering team and I'll quit immediately. (edit: actually I do know what else to say: If you don't care about patents don't apply for one. Fully optional endeavor. If you do care about patents, then I am here to help.)
I remember learning about the pros and cons of patents vs trade secrets a few years ago- it's a lot less cut and dried than that. With trade secrets, you just have to make a reasonable effort to keep them secret, and then if it gets out you can sue for theft (or some such). As long as there's a secret sauce you can keep out of the public eye it's nearly functionally equivalent.
> The application that I've seen so far that was the most difficult to reject was made difficult by extremely specific claims.
Tradeoffs: the narrower the patent, the less useful it is, because it's that much easier for your competition to just tweak something inconsequential.
> once your product begins selling and Microsoft creates a fast-follow product you can sue
Not that easy - you need to be able to afford a patent law fight. Big companies like MS might just take it anyway and risk the lawsuit, because they can afford to drag it out longer than you can stay solvent.
On the other hand, if you go the trade secret route, Microsoft can't fast-follow because you didn't publish the whole thing.
Having filed multiple patents, it does not leave a good taste in your mouth (esp software patents). You know you are bullshitting but unfortunately it needs to be done because others will do it and deny you to use the invention.
Fully agree. Over 20 years ago, I did the same job as the Patent Engineer (I had a slightly different title, but same role) at a big law firm. Half of the patents I filed were BS software patents for a name brand tech company, and the other half were for actually novel and cool inventions for things like super-efficient distillation/chilling systems for coffee and satellite antennas for moving vehicles.
Getting that job was almost as easy as getting a software job in 2019. I had a masters in EE and a few years of SWE experience. I sent out 5 applications to top local law firms, got 3 interviews, and one offer over the course of a month. Apparently, the bulk of people in that role were post-docs in bio/chem who weren't excited about managing student loan debt while being underpaid lab people. (The EE/CS contingent was 5 out of 30 in the role, and I was the only one with actual SWE experience, so it gave the me and the firm creditability with inventors)
My salary matched my SWE salary, I had my own office + secretary, and the firm paid overtime and law school tuition. They also charged $250/hour for my time. I also only lasted about year in that role since it was very isolating and not collaborative.
The coolest part of that whole experience came a couple years later when I went to dinner with my wife, her close friend, and the friend's serious boyfriend. In the "what do you do" chit chat, he mentioned working with someone on a new way to brew and chill coffee quickly that they were close to bringing to market.
When I asked if it works like [describes invention exactly] his mind was absolutely blown away. As it turns out, I wrote their patent.
> it does not leave a good taste in your mouth (esp software patents)
The very _concept_ of software patents should leave a bad taste in anybody's mouth.
If your invention is a prior art, then it is a defense to infringement. The asserted patent is invalid.
This assertion needs a lawsuit and deep pockets. Most startups will not be able to do this.
Infact the patent holder can file a lawsuit against a startup. Now even if the startup can prove that the patent is invalid, it has to spend precious money on the lawsuit.
>"If you disclose your invention publicly and do not file an application within one year, you could be barred from receiving a patent on that exact invention."
Does this mean if I disclose an invention, that nobody can patent it after a year?
This is my explicit reason for writing about BitGrid out in the open, since 2004.[1]
If it turns out to be useful, that's awesome. If not, it's been a fun mud ball to polish.[2]
[1] https://bitgrid.blogspot.com/2004/
[2] https://en.wikipedia.org/wiki/Dorodango
> Does this mean if I disclose an invention, that nobody can patent it after a year?
Others can. But during their application who is going to tell that there is a prior art? Patent offices only do basic checks on existing patents and not on the wider internet.
Once the others get their patent, you can appeal [1] it.
[1] I am not sure its the right word
> appeal [1] I am not sure its the right word
You can challenge the patent at the USPTO to instigate an in partes review.
You can also raise a defense of 'the patent is invalid' if the patent owner tries to sue you.
"you'll regret it" is basically the right advice. Unless you're carrying at 10 figure legal budget and a portfolio of thousands of patents.