The U.S. patent system will undergo its most substantial overhaul in decades if a bill introduced on Wednesday becomes law.
A fix for a broken patent system?
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Only inventions with a WORKING prototype can be patented… just having an idea and being the first to claim it shouldn’t mean a damn thing unless you can also get it to work.
I dunno if I agree with that. If its the first with a prototype then whoever has the most money will have the most patents. (more money = more workers = less time)
That’s, sure, the best news of the month! Fixing the chaotic North American patent system’s a priority for all the industry!
Does this rule invalidate Prior Art, and also award patent-based monopoly privileges to companies who may admittedly not have invented the device in question themselves?
I mean, if Joe J Average invents a better mousetrap in 2005, but doesnt file a patent for it, and MegaCorp Ltd., after finding details of J. Averages work in the data they snoop off his PC (legally due to EULA on their software), and file a patent on it, then they get granted a patent on a device that they:
a) played no part in inventing
b) knew that prior art existed for
And yet they would seemingly still be granted the patent, leaving Mr. Average zero recourse, and even if he could prove he invented it first – since first to file trumps first to invent – would make no difference.
Isnt this massively unjust?
I don’t think this is actually part of the bill. That would be ridiculously and profoundly stupid.
The idea I got when I RTFAed looks like an improvement. If you are interested I suggest you read it. I for one already sent an email to my representative.
Does this rule invalidate Prior Art, and also award patent-based monopoly privileges to companies who may admittedly not have invented the device in question themselves?
The article doesn’t address prior art specifically, but the wording of the article could be interpreted as invalidating prior art. I hope it’s just a poor choice of words in the article. The ability to challenge a patent would seem to leave room for prior art.
Agree this is a big step backward. It would likely INCREASE the number of bogus patents filed because the new law would establish that a patent trumps at least certains kinds of prior art. Large corporations with the wherewithal to conduct regular patent sweeps of their projects would be at an enormous advantage over smaller shops, except for the IP boutiques of course.
My one-word proposal for patent reform: change the word “non-obvious” in the current law to “highly non-obvious”.
So a person should be able to patent an idea regardless of whether or not it can later be produced? Should Jules Verne have been able to patent time travel or space flight just because he thought of the idea first?
Working implementations are the only things that should be patentable.
get rid of them. It is a damn lie that innovation depends on such “protection.”
“Working implementations are the only things that should be patentable.”
Now transfer that to the digital realm, and let the fun begin.
Don’t forget that they changed the law so you can now “own” life forms. I dunno, this country is just getting stranger and stranger. I’m about ready to bail out…
…but then again the UK is getting really strange too…
You need to look back in history, at one time people did have to submit working prototypes to the patent office, they often made scale miniatures. If you ever see Antiques roadshow on PBS you will see such things come up all the time.
But this simply isn’t very practical, it cost an inventor a ton of money, and the patent offices gets buried in things. Also, what if I say came up with an invention for say, crushing the earth into a cube instead of it’s current sphere shape. For me to deliver a protoytpe would be extremely hard, and definitely not fit in the mail slot of the patent office. How would I get a patent?
You still can’t patent an idea. You still have to submit drawings and concepts showing how something works. You can’t just say “I have a idea for a Planet Cubifier, give me a patent” you still have to show how to do it and so forth.
This doesn’t mean things don’t slip through sometimes. It is humans checking things afterall.
And once you get something very generalized to cover a good bit of what you have, it starts looking like an idea. You never make a patent to specific, otherwise it’s easy for someone to bypass it and your out of luck. Thus the patent is pointless.
What should also be done is the USPTO should offer patents to be publically reviewed (ie, peer-reviewed, similar to scientific papers etc) for feedback for non-obviousness, cross-field similarities (ie, an inventor is merely mapping an existing invention to a new field of research, and is not uniquely new) and that this feedback (whether good or bad) is included in the final decision to award the patent.
At the moment, people can review patents and applications, however there is no means of providing feedback to the examiners easily to warn them about bogus patents. If 10,000+ eyes are all reviewing every single patent application as it comes in, the chances of bogus applications will rapidly drop. (I hope!)
Also, making the onus of defending a patent and harder penalties for bogus (ie, extortionist) patents should be on the applicant so they don’t try to blackmail people or companies for trivial ideas they slipped through. Patent awards should also be shortened to 5-10 years tops on computer related patents, 25+ years is too long in the field.
Little guy invents something. Big guy comes along and copies it and files for a patent. Little Guy if he does not file an “appeal” within a short time frame loses all his rights to his work.
There is no prior art.
There is no concept of “original invention”.
These new changes make a mockery of a system that DOES protect the little guy far more than the Microsoft-Oracle IP Theft Patent System.
Joke. Total Joke. Just get rid of software patents and use copyrights to protect what’s needed. How hard is that to grok?
Davw
” What should also be done is the USPTO should offer patents to be publically reviewed (ie, peer-reviewed, similar to scientific papers etc) for feedback for non-obviousness, cross-field similarities (ie, an inventor is merely mapping an existing invention to a new field of research, and is not uniquely new) and that this feedback (whether good or bad) is included in the final decision to award the patent.
At the moment, people can review patents and applications, however there is no means of providing feedback to the examiners easily to warn them about bogus patents. If 10,000+ eyes are all reviewing every single patent application as it comes in, the chances of bogus applications will rapidly drop. (I hope!) ”
I think you don’t realize how the patent office works, it’s not like you submit something and a random someone grabs it and goes to see if it’s good or not.
Everything goes through a sort, and this sort is based on subject and results in every patent application for similar things getting reviewed by the same guy or group of guys, so say ever patent for a fuel injector has gone through the same person, so they know what is already there very well, and are deeply researched in everything that has been done and currently available things. Most patent office people are engineers and have deep backgrounds in the patents they review. So your first paragraph is already done. Furthermore if you wanted a outside review board, it will be impossible to get impartial ness. Since everyone will basicly try to block everything since someone on that panel is bound to have a interest in something. And your not going to get a panel member to be part of the same obligations as a patent office person (they can never patent anything and have to show no connections to things they review).
This also leads into the second part, you aren’t going to have a panel of eyes who isn’t going to have people shooting it down. With a thousand eyes a few of them our bound to through up flags cause they don’t want it patented. Furthermore like opensource software using this idea, even if it’s possible for 1000s of eyes to see it, you still have to get willing eyes. Everyone in Opensource assumes people are looking through the code cause one could, in reality even important parts of the linux kernel probably rarely get looked at and then by very few people.
I think in general people think there is a lot more bogus patents going through then there really is. Most of which involve software, which needs patents the same as anything else, but it’s far harder to deal with then the thing being patented isn’t even a material object in production form. Also I think the big thing is even patents on software lots of people hate, they aren’t really bogus. People just hate them from doing things they want to do even when they are so massively simple. And being that software is so cheap and easy to re-create it’s even harder, but thats no reason to stop a patent. If your a mechanical engineer and go to make something, you have to check all sorts of things about the simplest compents, like how you assembled some fasteners, or a basic gear train, cause someone has a patent on it. And it’s nto a real issue, you might be able to just call up the patent holder and they will tell you sure no problem go for it, or you just have to pay a tiny fee, and you move on, not a problem. It’s only a problem if your trying to give stuff away for free, which only really comes up with software, and then thats not a issue of the patent office or holder or patents in general, thats the software makers issue who wants to give it away for free instead of paying the fee to the patent holder like everyone else has done for a long long time.
A patent doesn’t stop you, infact most the time you can just pay the holder and they gladly will since thats why they patented it in the first place, and you move on. Companies make large amounts of money just doing patents and licensing them. Such as IBM, licensing IP is one of the main sources of income for IBM. Its also important to remember how much a patent cost. It comes in around $20,000 for a 20 year run of a patent. This sums up a couple parts of it since you don’t just get it for 20 years (or whatever the current max is), it goes in steps and you have to renew. But before you even get the patent you have at least 7 grand into the process.
And if someone won’t let you use their patented way, you just have to get creative and find a work around. What software is going through right now is like the late 1800s when the same issue came up surrounding internal combustion engines. Otto and Diesel locked up most the patents, and rightfully so since they put together the best way of doing it. This caused others to have to find ways to make engines not using ideas already patented, so people got very creative and a lot of interesting ideas came out of it. One person was Atikinson who came up with a different type of engine that was more efficient then the otto engine. It never took off, but now 100+ years later, his engine is making a comeback in hybrids such as the prius and escape. If he hadn’t been blocked by otto, that engine may have never been developed (key word is may, just because someone or something didn’t happen doesn’t mean it wouldn’t have happened anyways somewhere by someone sometime).
In a few more years, many software patents will begin to expire, and people will do whatever with them, but in the mean time many ways of doing things will have been thought up. The timeframe for innovation in software is short, one could probably already state that the number of fresh ideas is dwindling.
” Little guy invents something. Big guy comes along and copies it and files for a patent. Little Guy if he does not file an “appeal” within a short time frame loses all his rights to his work. ”
This isn’t really an issue. If a little guy invents something, he can go patent it, if he doesn’t well tough luck for him.
Secondly the only way the big guy can steal it is if the little guy showed it off. If you invent something, then show it off without patenting it, and someone tries to patent it, thats your problem, you shouldn’t have done that. But still if you showed it off they can’t patent it since there is prior are.
If you want to make sure no one patents it, and your not going to, you could probably very easily send a note to the patent office with your idea and just say “heres an idea, it’s out there, don’t let anyone patent it”
If you do want to patent it, and don’t want someone else to do the same, don’t show it off (no posting on slashdot), and go file a patent. You can also have your invention put on hold if you will at the patent office, this shows you have file, and stops others from copying you, but gives you a window to find investors without fear of someone stealing your idea.
I’m afraid I have to agree with Goldstein here. At first glance, this appears to be an attempt to shift the power in the system from independent inventors to large corporations.
I understand that Microsoft wants to eliminate patent shops, but many provisions in the proposed legislation do more than that. The simplest way to eliminate EOLAS-type lawsuits would be to demand that the invention be implemented within five years of being filed and raise the bar so that only very or highly non-obvious applications are accepted.
Clearly, Microsoft doesn’t want to raise the bar because that would invalidate many of their own patents. What Microsoft wants to do is make patents the exclusive domain of well established corporations and eliminate any threats that startups might pose to them.
Read it carefully. There are many ways to solve these problems, but these provisions are specifically aimed at decreasing the power of independent inventors.
Proposal: Lowering the penalties for willful infringement.
Result: Encourages willful copying without paying inventor.
Proposal: First to file rather than first to invent
Result: With this provision, Microsoft and/or IBM could make regular passes of SourceForge and file patents on all innovative work there. On hearing of any new product from a startup, they could patent the techniques used in that product immediately. Note that even if they didn’t receive the patent, there would be no penalties. In fact, they would benefit because by filing first they eliminated the possibility that the original inventor could receive a patent.
Proposal: Allowing judges rather than the patent office to review challenges,
Result: I’m not sure what this means because the details aren’t spelled out. In the best case, it is a needed fix because it catches mistakes made by the patent office when a bad patent slips through. What worries me is the choice of forum. In the worst case, it is shifting challenges from the patent office to a courtroom. The same thing could be accomplished without moving it to the courtroom by simply funding the patent office. The purpose of moving it to the courtroom – which is more expensive for the government – could be to allow large corporations to crush startups who cannot afford the legal costs to survive a challenge. Or it could be to make it prohibitively expensive for the open source community to challenge bogus patents by Microsoft. Either way, the specification of forum is of some concern. The patent office may not be effective, but its best aspect is that small inventors can afford it.
There is a need for patent reform, but the brief description of this legislation sends up warning flags. Raise the bar on what can be patented and fund the patent office so it can spend more time reviewing. Allow patents to be challenged if the inventors haven’t creating a prototype within five years of filing. These and several other approaches would be far simpler and more effective at solving the current problems than Microsoft’s proposals.
— N
“And once you get something very generalized to cover a good bit of what you have, it starts looking like an idea. You never make a patent to specific, otherwise it’s easy for someone to bypass it and your out of luck. Thus the patent is pointless.”
And that is part of the problem… instead of granting a patent for an implementation, they are granting patents for things so broad that there could probably be 100 unique ways to implement them in some cases.
—
And just producing a diagram does nothing to illustrate that what you are submitting actually works at all… it’s just a drawing in many cased. I can draw, maybe I should start dreaming up things that I have no clue how to actually implement… but as long as I have a drawing with some radio waves coming out of something and a few paragraph explaining what it’s “supposed” to do, then I can get my own patent granted. Give me a break; producing a working implementation doesn’t mean that it has to be a final product ready for sale tomorrow, it could be parts from radio shack or something made out of wood or prototype software written that at least demonstrates that your invention actually works and isn’t just you trying to lock up an idea that you really have no clue how to actually bring into existence.
More like how to scam people further.
“Furthermore if you wanted a outside review board, it will be impossible to get impartialness. Since everyone will basicly try to block everything since someone on that panel is bound to have a interest in something.”
I don’t care if people on the external panel are not impartial. I don’t care if a half of them specifically WANTS a patent to be invalidated. All I care of is what FACTS they come up with. If they find facts that invalidate a patent application, then it doesn’t matter if the person who discovered these facts was impartial. THIS APPLICATION HAS TO BE REJECTED.
But of course, many people will be scared of this solution, because this makes it much harder to extort money.
The proposed legislation is merely an attempt to prevent an inevitable backlash against the broken US patent system which (in a just world) would see Microsoft’s precious software patents invalidated. By focusing on blatant opportunists, Microsoft and friends are attempting to marginalise the opposition to software patents by highlighting the most obvious abusers of the system (who just happen to be costing them millions of dollars) and extinguishing them, but the injustice of software patents will remain. Just like certain gestures from the patent lobbyists at Nokia, this is another example of giving a little with one hand whilst taking as much as possible with the other.
Microsoft has filed TONS and TONS of patents for things it NEVER invented.
“First to file” gives Microsoft legitimate patents on this stolen IP.
Because “first to invent” (the backbone of a good patent system) is being canned, Microsoft moves into the patent driver’s seat and will have effectively stolen thousands of patents worth of inventions.
This “reform” is legitimizing theft. It is nothing more than this and will only help Microsoft and a few other large software companies.
You can say goodbye even faster to the US software industry if this bill passes.
The rationale: MS and a few other big corporations were not happy, so USPTO and the governent had to do something to the patent system
Somehow I get the feelings that these “fixes” are planned mostly just to make those big corporations happy. But I guess that is no surprise, the USA being such an antidemocratic plutocracy and oligarchy as it is… (sorry, but that’s the way I see it)
What about protecting true innovation, competition and freedom of enterprise? What about prior art, individual programmers, open source and non-commercial software?
Patent systems exist to promote innovation. The governments responsible for them should ensure that patent holders bear the cost of their mischevous lawsuits.
Also, if somebody independently invents something covered by a patent, that patent should be overturned, rather than that person being made guilty of patent infringement.
@Brad
> the patent offices gets buried in things.
They wouldn’t if they rejected enough patents to discourage some bad applications.
Is this going to affect linux in any shape or form?
I think there is some wilful misinterpretation of first-to-file going on here. FTF is the basis of patent systems almost anywhere else and if this reform is also intended to bring it in line with systems elsewhere then this is what it will work like:-
If you think up an idea, you may file for patent protection. Big corp down the road may file for a patent on this same idea. The patent, if awarded will go to the first person to file, not on your wealth.
Say you don’t file and keep silent. Big crop thinks up the same idea and files, they get the patent. Which is what is intended – that an idea becomes available to society since you didn’t reveal it when you thought of it, society did not benefit and rightly rewards someone else for revealing it.
Say you have this idea and you want society to have it. Then “publish” it is some way – i.e. make your idea public.
For scientists, merely speaking about it in a conference without a prior filing constitutes publication. When you do this, the idea becomes prior art and no one else can file for it either. This is again how it is intended to work – society rewards those with ideas to release them with a limited monopoly, it does not reward others for taking this publically revealed idea and running with it. So big crop cannot hoover sourceforge and patent it – sourceforge constitutes publication.
What if you had this idea and talked about it in a small circle and big crop hears of it and files. Here is where you get problems. Generally, you should not talk about an idea unless you are prepared to file or publish. If you want to talk to big corp about it, there are legal docs you have to exchange beforehand to protect the confidentiality and rights on what is being discussed.
The first-to-invent system is fraught with difficulties. A patent may be awarded to big corp. After it expires, small guy fighting the priority of big corp’s patent may finally win his case and a fresh patent gets awarded to him/her. Society ends up paying royalties twice over. That’s what happened with the laser. Bell labs got the patent (and working prototype). Small inventor with idea (but never any bloody working implementation) fights them forever and finally wins on first-to-invent. Patent reissued. Having “invented” an algorithm before, I think a practical implementation of an idea is very much more difficult than just thinking up of it and patents ought to require reduction to practice as is required elsewhere. Also,here can be doubts as to when you invent but there is no uncertainty as to when you file. Let’s not have a system that merely keeps lawyers employed 🙂
I think the USPTO is trying to bring the US patent system inline with the rest of the world. It is quite possible that they may be insane, the current US system is, and intending to have first to file without having a prior art exclusion, but I doubt it.
> Say you don’t file and keep silent. Big crop thinks up the
> same idea and files, they get the patent. Which is what is
> intended – that an idea becomes available to society since
> you didn’t reveal it when you thought of it, society did not
> benefit and rightly rewards someone else for revealing it.
Okay, suppose I develop something “new” called “on-click shopping”. I don’t file either because I don’t think it’s really that much of an innovation or I just don’t have the money. Now Amazon comes along and patents “on-click shopping”. With “first to file”, my web site violates “their” patent and has to be taken down, even if I can prove I was up before Amazon was even incorporated.
It always bugged be about the claim “idea becomes available to society”. Does a patent *really* do that?
Seriously, how many of you go out and actively look in the patent database for new ideas?
(crickets chirp)
Okay, how many of you go out and actively to see if you violate patents so you can *work around the patent*?
(several lawyers from big corps raise there hands. crickets chirp in the small business part of the room)
How many of you develop without looking at the patent database and hope that somehow you’re not violating some patent somewhere and that the patent holder doesn’t come after you.
(99% of developers, assuming they’re even aware of the patent issue)
Okay, how does that help society again?
This isn’t good news at all and does nothing to “fix” the patent system, it just takes away rights from the little guy and tells him it’s for his own good. It’s very much in the spirit of the way things are being done in this country these days.
In Australia we currently don’t recognise prior art. You keep something secret until you lodge a patent application. Once your patent is approved you publicise it. If you’re really smart you sell your patent to a big corporation.
One of the richest men in Australia sold his engine technology patents to a major corporation. They were never commercialised but he invested the proceeds into commercial real estate and is now worth around US$700 million.
In reverse order…
“You keep something secret until you lodge a patent application. Once your patent is approved you publicise it.”
“Which is what is intended – that an idea becomes available to society since you didn’t reveal it when you thought of it, society did not benefit and rightly rewards someone else for revealing it.”
The problem with software patents is that they “corrupt the commons”. Software development is very “fluid” and ideas-intensive – if you want to develop anything more than “hello world”, there’s going to be lots of algorithms, techniques and best practices in that software, many of which are freely exchanged and published. Having patents as some kind of reward mechanism for publication may seem like a good idea, but given the number of ideas and concepts in any non-trivial piece of software, it’s easy to inadvertently re-discover or to incorporate some means of doing things that someone claims to have patented.
This is why software is very different from industrial inventions. For example, if you want to make a better water filtration plant, investigating patents, licensing “intellectual property” and improving on it is probably one of the better (or more effective) ways to go about doing so – you can’t trivially obtain the blueprints for some company’s filtration equipment; ordering the stuff and reverse-engineering it would probably cost a lot of time and money; and no-one is likely to be spilling the beans about that company’s technology without clear indications that the information is proprietary. I would also argue that in many industrial disciplines above a certain level of specificity (ie. compare filtration equipment to the underlying scientific processes which would be fairly general), it would be quite unlikely that people stumble across a patented solution to such a particular industrial problem.
In contrast, software is easily distributable (often with source code) and stems from mathematics – traditionally a completely open discipline with no artificial “intellectual property” constraints. Software patent lobbyists want to pervert the discipline by imposing a completely inappropriate economic/social model that will produce damaging, artificial monopolies with high margins for the patent business. Thus, a discussion about the merits of patents in general really doesn’t address the issues related to the original article (or the motivations of the protagonists).
As soon as the new patent system is in place, established parties will find a way to subvert it and make it even worse than the current one.
One one hand: making the patent date be the time of the patent application, instead of the presumed time of the invention: seems good.
On the other hand: opposition requests mean that almost ALL software patents will be challenged.
In fact, it is strategic for a Microsoft or an IBM to challenge each and every patent that is issued. Sooner or later, the new patent system won’t be able to grant any patents (which perhaps is an unintended good outcome).
NO SOFTWARE PATENTS.
Not for the US, Europe, or the world.
The article didn’t mention it, but there is already pressure from Big Corp to lengthen the period of validity of their patents. In recent trade negotiations, the US has been proposing bumping it up from 20 years to 25. This is the same pattern we’ve seen with copyrights. Someday in the future, we’ll have patents forever. Even though Microsoft copied things like tcp/ip and email from Unix, wanna bet that they won’t patent this (and much more) and have a patent that remains valid as long as Microsoft exists?
Everything goes through a sort… Most patent office people are engineers and have deep backgrounds in the patents they review.
I’m sure they are, but when I see some software related patents get through which covered obvious techniques in a field I’m personally interested in, I have my doubts that they’re all that effective on their own. In general, my experience is that working a vacuum is not as effective as everyone seeing what you’re doing.
I could pull out some simple patents on things like emulation, Copy-On-Write, Non-Uniform Memory Access which have been patented (and awarded) that anyone familiar with the field would have classed as being obvious. Email me for details if you’re interested.
Furthermore if you wanted a outside review board, it will be impossible to get impartial ness.
True, I didn’t say we should replace the examiners with a free for all board, I just said that people should be able to chip in their say on a particular application and that the examiner should have a look at it. The final say still comes down to the examiner[s].
With a thousand eyes a few of them our bound to through up flags cause they don’t want it patented.
Of course. However, there’s a difference between “this idea is obvious” and “damn, why didn’t I think of that!?” I have no problems with people getting rewarded for radically new and innovative (read: never been done before) inventions, but I do have issues with people who aren’t willing to do the work and patent any old thing as they can.
I think in general people think there is a lot more bogus patents going through then there really is.
I’m sorry, but my readings in software show there are a lot of bogus (ie, obvious) patents being awarded that I personally wouldn’t have let through if I’d been the patent examiner. Hardware is a different story, I’m not much into that, but it’s probably less of a problem. Again, I can send some examples.
… be able to just call up the patent holder and they will tell you sure no problem go for it, or you just have to pay a tiny fee, and you move on, not a problem.
Well, I’d like to say “it’s a tiny fee”, but in software, you see small fish trying to go for big fish all the time. Think Eolas (vs MS). RamBus (vs JEDEC). I just read yesterday someone was suing MS for some trivial thing and wanted several million dollars.
And if someone won’t let you use their patented way, you just have to get creative and find a work around.
I’m sorry to say, in software the avenues for “working around it” are very very difficult to find and use. For example, if you want to use a patented DCT transform, there is only so many ways you can do it. You’re free to find a better transform, but that’s pretty hard to do. Hardware is a different story however.
In a few more years, many software patents will begin to expire
I hope so. There’s still a lot that have 15+ years to go however. Things have cleaned up since 2000 however, but there’s still a lot out there just waiting to be used (or abused).
The timeframe for innovation in software is short, one could probably already state that the number of fresh ideas is dwindling.
Therefore software patents should expire after 5-10 years, max. Not 25 years. Otherwise, someone will come up with a unique idea from existing ideas, then get sued for patent infringement, even if the original patent was 20+ years ago (the equivalent of 100-200 years in industrial design terms, at times). It’s merely the case of people capitalizing on the good work of other people with the minimal of work on their behalf, in some cases.
Some of these software patents cover ideas that could be used for market manipulation or market price fixing schemes. If the implementer sells the capability with the intent or knowing to defraud consumers, then whistle-blowers could be deprived of the right to a jury trial should a patent be enforced in front of a panel of not-so-impartial judges likely to side with a corporate entity.