The problem, you see, is their length. Seventeen years of monopoly is an eternity in Internet time. Instead, software patents should only be valid for seven years.
I'd argue even seven years is an eternity in Internet time.
First, the author can't decide if patents last for 17 years or 15 years. The correct answer (in the US) is neither: since 1995, US patents have enjoyed protection for a maximum of 20 years from the filing date. How long this protection lasts from the issue date depends on how long the application process took, and how much extra time is tacked on due to PTO delays, and whether or not all of the maintenance fees are paid.
Second, the author gives no justification for the 7-year period. Why are 7-year terms better than 2-year terms or 10-year terms? And how will any of them get rid of software patent trolls or fix abuses of continuation-in-part applications?
The article consists of nothing but unsupported assertion and incorrect information.
I was vague about the lenght of the protection because it depends on when you filed your patent. It's beside the point, the point is that current protection is very very long, roughly 20 years.
You are correct, I have no proof that 7 years is the right number. I'd be happy for people to debate if it should be 6 or 10 or whatever. But I believe it's a simple enough fix.
Believe me, I have been through enough patent issues, I strongly wish the system would be fixed. But most proposals so far are jist plain impractical. I'd love for patent examiners to really understand the technology and only grant truly innovative patents. But it's almost impossible. I'd like for lawsuits to not take so long and not be so arbitrary, but it would require a full reform of the legal system. Not realistic.
s/20/7/a sounds much easier to me, and achieves a lot. Not perfect, but a decent tradeoff. I'm an engineer, I do tradeoffs.
Fluff. The author suggests reducing the duration of software patents to seven years as a solution to current problems. There are no other suggested corrections. This does nothing to address the "One-Click" class of software patents. Further, it doesn't even make a very strong case for why seven year software patents would actually solve problems. I'm pretty sure many software patent trolls sue within seven years. In addition, the author seems to be under the bizarre assumption that Chinese companies actually respect US patents.
Yes, there are other suggested improvements to the patent office, such as rejecting obvious patents. Except you can't easily fix that, whereas substituting 20 by 7 is something that can be done by your average lawmaker.
Regarding China: you must have misinterpreted what I was saying. I pointed to China as an example of what would happen if we dropped patent protection completely. Indeed, China doesn't care much about patents currently, and it's not a world I want to live in.
One problem with having a software patent have a validity of seven years is that it often takes seven years or in some cases even more to get a patent through the USPTO.
If a patent's validity is up before it is issued, it is pointless.
17 years is long for anything now. Heck, 7 years is long for almost everything, and anything that invested in their product relies on die-hard secrecy rather than throwing patents around. IE, cars. Software in particular moves faster than most because it has zero manufacturing time, so yes, it should have a shorter limit as long as patents exist. It's very clearly a special case.
Early-industrial-age, retooling factories took a while. Now, many things are made to be modular or simply replaced outright for cheap. Things do not take as long to develop and produce now as they did when the patent office was new.
What may have legitimately been protection to begin with hasn't kept up with the world, and has become a hindrance. Just like anything else that doesn't change. Stagnation is death.
Everyone here is negative, but I think the general idea is reasonable (though not new). I don't know that the length should be 7 years, but something significantly shorter than ~17 seems like it could help fix a lot of the current problem.
This would definitely change the cost-benefit analysis on filing patents. Companies definitely take into consideration the costs of employee time, legal fees, filing fees, etc. when considering whether to pursue a patent. If the expected benefit is suddenly less than half of what it was, then decisions start to change. If some portion of patents can't even make it through the system in 7 years (or whatever), then that makes the expected benefit of software patents even lower.
Patents are an incentive to invent something new, with the knowledge that you alone will be able to profit from that thing for some period of time.
The idea is that this incentive will benefit society with these inventions more than it will cost society in reduced competition.
Looking at the software market today, I think it's pretty clear that the incentive that patents create for software innovations is not needed and that the cost that they create greatly outweighs the benefits. I can not think of any software innovations that would not have been made without the incentive of monopoly protection.
A work of creative fiction can also be innovative, but does this mean that books should be patentable? Whilst it may benefit large companies who can afford to spend money on lawyers, patenting software is as foolish and harmful to the industry as patenting books. One thing's for sure and that's that software patents do not benefit startups or independent software contractors.
Reducing the length of applicability of software patents does not reduce their harm. Seven years in a fast moving business like software may as well be eternity.
12 comments
[ 2.7 ms ] story [ 34.8 ms ] threadI'd argue even seven years is an eternity in Internet time.
First, the author can't decide if patents last for 17 years or 15 years. The correct answer (in the US) is neither: since 1995, US patents have enjoyed protection for a maximum of 20 years from the filing date. How long this protection lasts from the issue date depends on how long the application process took, and how much extra time is tacked on due to PTO delays, and whether or not all of the maintenance fees are paid.
Second, the author gives no justification for the 7-year period. Why are 7-year terms better than 2-year terms or 10-year terms? And how will any of them get rid of software patent trolls or fix abuses of continuation-in-part applications?
The article consists of nothing but unsupported assertion and incorrect information.
You are correct, I have no proof that 7 years is the right number. I'd be happy for people to debate if it should be 6 or 10 or whatever. But I believe it's a simple enough fix.
Believe me, I have been through enough patent issues, I strongly wish the system would be fixed. But most proposals so far are jist plain impractical. I'd love for patent examiners to really understand the technology and only grant truly innovative patents. But it's almost impossible. I'd like for lawsuits to not take so long and not be so arbitrary, but it would require a full reform of the legal system. Not realistic.
s/20/7/a sounds much easier to me, and achieves a lot. Not perfect, but a decent tradeoff. I'm an engineer, I do tradeoffs.
Yes, there are other suggested improvements to the patent office, such as rejecting obvious patents. Except you can't easily fix that, whereas substituting 20 by 7 is something that can be done by your average lawmaker.
Regarding China: you must have misinterpreted what I was saying. I pointed to China as an example of what would happen if we dropped patent protection completely. Indeed, China doesn't care much about patents currently, and it's not a world I want to live in.
If a patent's validity is up before it is issued, it is pointless.
Early-industrial-age, retooling factories took a while. Now, many things are made to be modular or simply replaced outright for cheap. Things do not take as long to develop and produce now as they did when the patent office was new.
What may have legitimately been protection to begin with hasn't kept up with the world, and has become a hindrance. Just like anything else that doesn't change. Stagnation is death.
This would definitely change the cost-benefit analysis on filing patents. Companies definitely take into consideration the costs of employee time, legal fees, filing fees, etc. when considering whether to pursue a patent. If the expected benefit is suddenly less than half of what it was, then decisions start to change. If some portion of patents can't even make it through the system in 7 years (or whatever), then that makes the expected benefit of software patents even lower.
The idea is that this incentive will benefit society with these inventions more than it will cost society in reduced competition.
Looking at the software market today, I think it's pretty clear that the incentive that patents create for software innovations is not needed and that the cost that they create greatly outweighs the benefits. I can not think of any software innovations that would not have been made without the incentive of monopoly protection.
Reducing the length of applicability of software patents does not reduce their harm. Seven years in a fast moving business like software may as well be eternity.