Relax liability for infringers

Problem

Patent liability is strict liability. This means that all a patent holder must do is show that a defendant practices its invention, not that the defendant had any intention to infringe, or even any knowledge that the patent existed. The law, in its current form, allows trolls to target anyone who allegedly infringes.

Solution

Congress should pass a law letting alleged infringers avoid liability when they can show that they independently arrived at the invention they practice. If inventors can’t understand a patent, then they, as a matter of logic, cannot copy it. Likewise, if inventors don’t know about a patent, they cannot copy it. An innocent infringer defense would protect those inventors and, consequently, shut down a major revenue stream for the trolls.

Sergey Babkin
United States
Posted Sep 11, 2012 11:20 am
  • Software developer / engineer
  • Internet user
A horrible idea. It will make the patents useless because it will be impossible to prove that someone had seen a patent or an implementation based on it before they used it. On the other hand, the patents must not cover the obvious solutions. If given the problem, any engineer with reasonable experience in that area comes with a solution in an hour, this solution must not be patentable.
up
444 thumbs up.
Mike Boucher
Dakota Legal Software, Inc.
Lafayette
CO
United States
Posted Sep 5, 2012 12:47 pm
  • Patent owner
  • Software developer / engineer
  • Internet user
  • Victim of patent troll
The proposal is to let infringers avoid liability if they did not infringe on the inventor's control of the right to copy. Not surprisingly, this is something called copyright. Copyright is different from patent, as it should be.

A side note: a lot of these proposals seem to have been drafted by well-meaning people with no understanding of patent law. Not understanding that you are proposing a copyright system to replace the patent system is but one example. Getting a patent lawyer involved early in your process can only help. We resume our regularly-scheduled programming...

Despite apparently proposing a large change, the proposal to allow "infringers avoid liability when they can show that they independently arrived at the invention" actually changing a few words and phrases to get to exactly the same place. To see why this is so, understand the law of copyright from which this proposal seems to have originated.

To show copying, a plaintiff must show that a defendant had access to the information and that there is substantial similarity between plaintiff's protected work and defendant's infringing work. Showing substantial similarity is simply the analysis of infringement that already takes place. The new element is a demonstration that defendant had access to the copied information. Plaintiff has presumably filed an application for a patent on the invention with USPTO and had them publish it on their public web site, which is well-known by many and indexed by Google for everyone else, which is sufficient to show access.

The new element that you propose is satisfied by the existence of the patent, which must already be demonstrated to show standing, and the old element is already litigated under the present system. So your effort to convert the patent system into a copyright system is a nullity. The elements of copying are already established by facts that must already be proved to establish infringement. You may have a different idea of copying with different elements than are found in copyright law, in which case you are proposing something different. However, if so, you are probably proposing something defective because you are not taking advantage of centuries of copyright jurisprudence to determine whether there was or was not independent copying. Either way, this is a poor idea.
up
444 thumbs up.
Benjamin Fritz
United States
Posted Aug 24, 2012 8:24 pm
  • Software developer / engineer
  • Internet user
In the software world, just keeping a history of your code and the changes to it (read: version control), should be enough in many cases to show that you came up with the idea yourself or gradually evolved your software naturally. More complicated things would no doubt need dated design documents or mathematical derivations or whatever. I see no reason why it would be difficult to show that you came up with something without copying it. It's the same sort of information I understand is needed to defend a patent's validity in court.
up
444 thumbs up.
Chris B
San Jose
CA
United States
Posted Mar 1, 2013 7:19 pm
  • Patent owner
Independent invention in patent law is asinine. What's the motivation for disclosing your invention in a patent if some other yahoo can just say,"oh yea, I totally came up with that too..but I did it independently, so you can't touch me"

I surprised such an organization as smart as the EFF has proposed this moronic idea.
up
418 thumbs up.
Stephan Kinsella
Kinsella Law Group
Houston
TX
United States
Posted Feb 6, 2013 11:35 am
  • Lawyer
  • Academic
  • Venture capitalist / investor
  • Internet user
This defense would be good, and could be made workable, despite some concerns expressed by other commentators; essentially the patentee should have to prove copying, as copyright holders have to do in copyright suits. One way the defendant can prove they did not copy is to use a "clean room" approach as is done in software design to avoid copyright infringement.

The problem with this defense, however, is that patentees can still sue competitors who do actually copy them. There is nothing wrong with emulation; it is part of the free market process. The entire patent system should be abolished. Still, limiting patent rights by the proposed defense would be a step in the right direction, as it would reduce the harm patents do to the economy. Basically any reform that makes patent rights weaker is good. I discuss other reforms at http://c4sif.org/2012/03/2011/03/11451/how-to-improve-patent-copyright-and-trademark-law/

Though this might do a bit of damage to my career, as I'm a patent attorney.
up
415 thumbs up.
Peiti Ying
Auburn
WA
United States
Posted Jun 22, 2012 7:50 am
  • Software developer / engineer
  • Internet user
Some people are saying that you can't prove that someone arrived at the patented idea independently. Are we forgetting about:
http://www.lockergnome.com/news/2012/05/08/why-oracle-might-doom-software-as-we-know-it/
".. Google did not copy the Java source code over for use in Android and call it a day. In fact, a grand total of nine lines out of 15 million (that is, 0.000006%) were examined to be copied..."
Just like writing a book, we all have our own programming styles. We all write blocks and sections of code differently (short of the REALLY remedial stuff like a while statement or if x=1 statement). Some might say that someone who intentionally wants to infringe can just change a few variable names and call it done, but let's face facts. If someone did that in a book, we'd all know. And if someone went out of their way to change the coding syntax of entire blocks of code for the majority of the code to try to avoid infringement, then they're pretty much doing the work of writing from scratch. I don't know if getting rid of software patents completely is a good idea yet. But the current system needs to change. I personally think that changing the current system is less difficult a fight than trying to get it abolished altogether. One step at a time.
up
283 thumbs up.
Lucas Meyer
United States
Posted Jun 20, 2012 10:40 pm
  • Software developer / engineer
I agree with the previous comments that more than just a statement of innocence is needed, but consider:
1. Patents are only supposed to be granted for something "novel" and "non-obvious"; if two or more companies independently arrive at the same idea, then it's probably a good sign that whatever got patented is reasonably obvious.
2. Right now, the patent system works using a "first-to-file" method. The first person to patent something gets the patent. This means that if Company A has spent five years of R&D to create something, and Company B files for a patent on the same concept 3 days before Company A does, then Company A is screwed. In a scenario like this, it would be pretty easy for Company A to provide a whole lot of evidence that they independently invented the concept.
3. There's also no requirement, under the current laws, to demonstrate that you've actually invented something that you've patented. Anyone else remember the whole fiasco with Rambus and JEDEC? That was an instance where a single company patented a bunch of stuff that was created through cooperation with multiple companies. That sort of trolling behavior needs to be stopped.
up
280 thumbs up.
Austin D
CA
United States
Posted Jun 22, 2012 11:46 pm
  • Software developer / engineer
Too difficult to prove. Will be abused by clever lawyers.
up
268 thumbs up.
Andrew Fong
Berkeley Law
Berkeley
CA
United States
Posted Jun 21, 2012 5:00 am
  • Academic
  • Software developer / engineer
  • Internet user
(1) A jury (or judge if both parties waive their right to a jury trial) determines whether the defendant independently arrived at an invention. This is how the system works right now in determining whether a defendant "willfully" infringed a patent.

(2) The burden to show independent invention falls on the defendant. Worse-case scenario: The independent invention defense does nothing because no defendant can prove independence. But at least we don't accidentally destroy the whole patent system in the process (assuming we're worried about that). And legitimate patent holders won't have to bear the cost of disproving this defense.

(3) Allow the defendant to show independence by demonstrating that they could not have relied on a publication or disclosure by the patent holder. One of the key requirements of a patent is that that the patent disclose enough information to enable someone like the defendant to reproduce the patented invention. If the defendant can prove no access to a disclosure, then logically, the invention was independent. For example, the defendant shows that he came up with the infringing product on a vacation with no Internet access, proven by forensic analysis of journal entries he made on his computer.

(4) Allow the patent holder to rebut an independent invention through tracing. For example, suppose the defendant shows that he came up with the idea not from the patent, but from conversations he had with his college physics professor, who had never seen the patent. The patent holder could rebut this by pointing the physics professor regularly collaborated with another professor, who had seen the patent.

(5) Allow the fact-finder to consider the degree of infringement in assessing damages -- e.g. this invention was only 25% partially copied from patent so the defendant only pays 25% as much. This also more accurately reflects reality. For example, suppose there's a patent for "a method to make cats stop shedding." The defendant has never seen the patent, but frequently makes cat jokes with a friend who has. One of those cat jokes inspires the defendant to make something that is very similar to the patented invention. Independent or not? I think you could argue it either way, which is why being to assign partial fault is useful. The percentages won't be precise, but this isn't without precedent. We already allow juries to assign relative fault in negligence cases, so why not for patents?

Under this implementation, the independent invention defense would be expensive since the defense would be tasked by gathering tons of evidence about exactly how a particular innovation came about. But it would still be better than the status quo, and if you could force a patent troll to reimburse these costs upon winning a suit, that would deter non-meritorious claims.
up
268 thumbs up.
Andrew Fong
Berkeley Law
Berkeley
CA
United States
Posted Jun 21, 2012 4:47 am
  • Academic
  • Software developer / engineer
  • Internet user
(1) A jury (or judge if both parties waive their right to a jury trial) determines whether the defendant independently arrived at an invention. This is how the system works right now in determining whether a defendant "willfully" infringed a patent.

(2) The burden to show independent invention falls on the defendant. Worse-case scenario: The independent invention defense does nothing because no defendant can prove independence. But at least we don't accidentally destroy the whole patent system in the process (assuming we're worried about that). And legitimate patent holders won't have to bear the cost of disproving this defense.

(3) Allow the defendant to show independence by demonstrating that they could not have relied on a publication or disclosure by the patent holder. One of the key requirements of a patent is that that the patent disclose enough information to enable someone like the defendant to reproduce the patented invention. If the defendant can prove no access to a disclosure, then logically, the invention was independent. For example, the defendant shows that he came up with the infringing product on a vacation with no Internet access, proven by forensic analysis of journal entries he made on his computer.

(4) Allow the patent holder to rebut an independent invention through tracing. For example, suppose the defendant shows that he came up with the idea not from the patent, but from conversations he had with his college physics professor, who had never seen the patent. The patent holder could rebut this by pointing the physics professor regularly collaborated with another professor, who had seen the patent.

(5) Allow the fact-finder to consider the degree of infringement in assessing damages -- e.g. this invention was only 25% partially copied from patent so the defendant only pays 25% as much. This also more accurately reflects reality. For example, suppose there's a patent for "a method to make cats stop shedding." The defendant has never seen the patent, but frequently makes cat jokes with a friend who has. One of those cat jokes inspires the defendant to make something that is very similar to the patented invention. Independent or not? I think you could argue it either way, which is why being to assign partial fault is useful. The percentages won't be precise, but this isn't without precedent. We already allow juries to assign relative fault in negligence cases, so why not for patents?

Under this implementation, the independent invention defense would be expensive since the defense would be tasked by gathering tons of evidence about exactly how a particular innovation came about. But it would still be better than the status quo, and if you could force a patent troll to reimburse these costs upon winning a suit, that would deter non-meritorious claims.
up
268 thumbs up.

Pages

Buster Bo
Sweden
Posted Jul 26, 2012 5:51 am
  • Software developer / engineer
  • Internet user
"35 U.S.C. § 103 describes the condition of patentability referred to as non-obviousness. This provides that a patentable invention must not have been obvious to a "person having ordinary skill in the art" in view of the appropriate prior art." Quote: http://en.wikipedia.org/wiki/Title_35_of_the_United_States_Code

If more than one inventor comes up with the same invention, isn't it kind of an obvious solution then?
The main problem is that it is too easy to get patents granted on obvious solutions or even on prior-art. Causes to that could be granting officers of USPTO or patent lawyers who lacks the "ordinary skill in the art", or that they get paid by the number of applications they grant, or they get paid on the side by the mega-company to grant trivialities.

Millions of programmers solve problems every day (that's what programmer do) in the most obvious way they can think of. Yet they infringe on dozens of patents for every program they write, without knowing it. Nothing new can be made without paying gigantic license fees. The time it would take to check all your solutions against patent databases and then applying for licenses or doing workarounds would be multiple times longer than it takes to develop the software.

Software patents results in:
*Worse program code - Due to unnecessary workarounds of obvious solutions.
*Longer development times - Due to searching for infringed patents, licensing bureaucracy or inventing in-optimal, non-obvious workarounds.
*More expensive software - Due to licenses, lawyers and longer development time.

There is nothing good with software patents. Delete!
up
465 thumbs up, including yours.
John Allred
United States
Posted Jun 24, 2012 8:39 am
  • Academic
  • Internet user
I like this idea.

I believe that if a defendant can prove that they were not aware of patents, or had arrived at the idea BEFORE they knew of a patent, and after they had made appropriate inquiries at the patent office.

Additionally, if a patent can be proven to be too vague, it makes sense that there should be an acknowledgement of "No Willful infringement". This sort of defense should be perfect for discouraging the litigation on vague patents which trolls like to use and abuse.
up
267 thumbs up.
Al L
Thailand
Posted Jun 22, 2012 11:51 pm
  • Software developer / engineer
  • Venture capitalist / investor
"Arrives independently" - How do you determine that? Sounds to me like the biggest loophole ever to avoid liability for patent infringement. Absurd.
up
267 thumbs up.
Austin D
CA
United States
Posted Jun 22, 2012 11:46 pm
  • Software developer / engineer
Too difficult to prove. Will be abused by clever lawyers.
up
268 thumbs up.
Peiti Ying
Auburn
WA
United States
Posted Jun 22, 2012 7:50 am
  • Software developer / engineer
  • Internet user
Some people are saying that you can't prove that someone arrived at the patented idea independently. Are we forgetting about:
http://www.lockergnome.com/news/2012/05/08/why-oracle-might-doom-software-as-we-know-it/
".. Google did not copy the Java source code over for use in Android and call it a day. In fact, a grand total of nine lines out of 15 million (that is, 0.000006%) were examined to be copied..."
Just like writing a book, we all have our own programming styles. We all write blocks and sections of code differently (short of the REALLY remedial stuff like a while statement or if x=1 statement). Some might say that someone who intentionally wants to infringe can just change a few variable names and call it done, but let's face facts. If someone did that in a book, we'd all know. And if someone went out of their way to change the coding syntax of entire blocks of code for the majority of the code to try to avoid infringement, then they're pretty much doing the work of writing from scratch. I don't know if getting rid of software patents completely is a good idea yet. But the current system needs to change. I personally think that changing the current system is less difficult a fight than trying to get it abolished altogether. One step at a time.
up
283 thumbs up.
Ruediger Jungbeck
RSJ Software GmbH
Germering
Germany
Posted Jun 21, 2012 8:02 am
  • Software developer / engineer
A patents protects the idea independent of the way you arrive at it. (You could infringe on a patent (but not on a copyright) without ever knowing about it)

So what you really want, is to end software patents (which is a very good idea for this and many other reasons).

Relaxing reliability might be better than the current situation, but proving how you arrived at a specific implementation (without knowing someone else did the same), might be practically impossible (especially as patents are public records anybody can see).
up
266 thumbs up.
Andrew Fong
Berkeley Law
Berkeley
CA
United States
Posted Jun 21, 2012 5:00 am
  • Academic
  • Software developer / engineer
  • Internet user
(1) A jury (or judge if both parties waive their right to a jury trial) determines whether the defendant independently arrived at an invention. This is how the system works right now in determining whether a defendant "willfully" infringed a patent.

(2) The burden to show independent invention falls on the defendant. Worse-case scenario: The independent invention defense does nothing because no defendant can prove independence. But at least we don't accidentally destroy the whole patent system in the process (assuming we're worried about that). And legitimate patent holders won't have to bear the cost of disproving this defense.

(3) Allow the defendant to show independence by demonstrating that they could not have relied on a publication or disclosure by the patent holder. One of the key requirements of a patent is that that the patent disclose enough information to enable someone like the defendant to reproduce the patented invention. If the defendant can prove no access to a disclosure, then logically, the invention was independent. For example, the defendant shows that he came up with the infringing product on a vacation with no Internet access, proven by forensic analysis of journal entries he made on his computer.

(4) Allow the patent holder to rebut an independent invention through tracing. For example, suppose the defendant shows that he came up with the idea not from the patent, but from conversations he had with his college physics professor, who had never seen the patent. The patent holder could rebut this by pointing the physics professor regularly collaborated with another professor, who had seen the patent.

(5) Allow the fact-finder to consider the degree of infringement in assessing damages -- e.g. this invention was only 25% partially copied from patent so the defendant only pays 25% as much. This also more accurately reflects reality. For example, suppose there's a patent for "a method to make cats stop shedding." The defendant has never seen the patent, but frequently makes cat jokes with a friend who has. One of those cat jokes inspires the defendant to make something that is very similar to the patented invention. Independent or not? I think you could argue it either way, which is why being to assign partial fault is useful. The percentages won't be precise, but this isn't without precedent. We already allow juries to assign relative fault in negligence cases, so why not for patents?

Under this implementation, the independent invention defense would be expensive since the defense would be tasked by gathering tons of evidence about exactly how a particular innovation came about. But it would still be better than the status quo, and if you could force a patent troll to reimburse these costs upon winning a suit, that would deter non-meritorious claims.
up
268 thumbs up.
Andrew Fong
Berkeley Law
Berkeley
CA
United States
Posted Jun 21, 2012 4:47 am
  • Academic
  • Software developer / engineer
  • Internet user
(1) A jury (or judge if both parties waive their right to a jury trial) determines whether the defendant independently arrived at an invention. This is how the system works right now in determining whether a defendant "willfully" infringed a patent.

(2) The burden to show independent invention falls on the defendant. Worse-case scenario: The independent invention defense does nothing because no defendant can prove independence. But at least we don't accidentally destroy the whole patent system in the process (assuming we're worried about that). And legitimate patent holders won't have to bear the cost of disproving this defense.

(3) Allow the defendant to show independence by demonstrating that they could not have relied on a publication or disclosure by the patent holder. One of the key requirements of a patent is that that the patent disclose enough information to enable someone like the defendant to reproduce the patented invention. If the defendant can prove no access to a disclosure, then logically, the invention was independent. For example, the defendant shows that he came up with the infringing product on a vacation with no Internet access, proven by forensic analysis of journal entries he made on his computer.

(4) Allow the patent holder to rebut an independent invention through tracing. For example, suppose the defendant shows that he came up with the idea not from the patent, but from conversations he had with his college physics professor, who had never seen the patent. The patent holder could rebut this by pointing the physics professor regularly collaborated with another professor, who had seen the patent.

(5) Allow the fact-finder to consider the degree of infringement in assessing damages -- e.g. this invention was only 25% partially copied from patent so the defendant only pays 25% as much. This also more accurately reflects reality. For example, suppose there's a patent for "a method to make cats stop shedding." The defendant has never seen the patent, but frequently makes cat jokes with a friend who has. One of those cat jokes inspires the defendant to make something that is very similar to the patented invention. Independent or not? I think you could argue it either way, which is why being to assign partial fault is useful. The percentages won't be precise, but this isn't without precedent. We already allow juries to assign relative fault in negligence cases, so why not for patents?

Under this implementation, the independent invention defense would be expensive since the defense would be tasked by gathering tons of evidence about exactly how a particular innovation came about. But it would still be better than the status quo, and if you could force a patent troll to reimburse these costs upon winning a suit, that would deter non-meritorious claims.
up
268 thumbs up.
Lucas Meyer
United States
Posted Jun 20, 2012 10:40 pm
  • Software developer / engineer
I agree with the previous comments that more than just a statement of innocence is needed, but consider:
1. Patents are only supposed to be granted for something "novel" and "non-obvious"; if two or more companies independently arrive at the same idea, then it's probably a good sign that whatever got patented is reasonably obvious.
2. Right now, the patent system works using a "first-to-file" method. The first person to patent something gets the patent. This means that if Company A has spent five years of R&D to create something, and Company B files for a patent on the same concept 3 days before Company A does, then Company A is screwed. In a scenario like this, it would be pretty easy for Company A to provide a whole lot of evidence that they independently invented the concept.
3. There's also no requirement, under the current laws, to demonstrate that you've actually invented something that you've patented. Anyone else remember the whole fiasco with Rambus and JEDEC? That was an instance where a single company patented a bunch of stuff that was created through cooperation with multiple companies. That sort of trolling behavior needs to be stopped.
up
280 thumbs up.
Darron Black
Houston
TX
United States
Posted Jun 20, 2012 7:07 pm
  • Software developer / engineer
  • Internet user
How on EARTH would a patent holder succeed in court if the defense could just claim ignorance?

Sorry, this just makes the whole enterprise look naive. Nobody's going to take this seriously if you leave such huge holes in your proposed system. I wanted to sign this petition, but now I can't.

The key to reform is a WORKING software patent system or NONE AT ALL. This takes the current broken system and just throws the broken on the other side of the fence.
up
524 thumbs up, including yours.

Pages

First
Last
JavaScript license information