Great article. "Innovation" may not cease with the loss of patents, but investment certainly will. Most of my clients are startup companies raising Angel and Venture capital. The first thing they are asked by investors is "Do you have a patent?" If there is no patent, then there is no investment. The same is true for inventors licensing their innovations to larger companies. If there is no patent, there is no licensing.
I'm enjoying seeing the ideas generated by my provocative post. Most, not all, of the comments lean in the direction of reforming the patent system through reducing the number and reach of patents, and/or the lifetime of the patents. I'm also in this camp. Patent claims are so broad, that it is easy to see why there are accidental infringements.
Another way to look at this could be this: Instead of taking the current system and justifying each change from that system, go to a "zero-based budgeting" approach. Imagine that there were no patent laws, just copyright, trademark, and trade secret law. Then add patent law where there are clear cases that doesn't work, and the new laws don't come with considerable cost themselves. I still think drug patents may fall into this category. In any case, we would probably end up with a much different and more streamlined system than we have today.
Just as you often see one innovative movie then a long sequence of imitators (think of vampire movies, for instance) so you have some truly innovative technical work then a lot of subsequent work in the field by competitors essentially doing 'patent engineering' ie doing fairly obvious things to get around the original patent whilst enabling a directly competitive product to exist without paying a license fee - think of the smart phone market - there are only so many ways to produce almost identical touch screen devices!
I am not sure that this use of patents is positive for anyone, other than lawyers.
A patent should really be used for two purposes:
1) to enable the originator to recoup their R&D costs for innovation by giving them a legally protected headstart - because there are very few innovations that are hard enough to reverse engineer that a bigger or cheaper competitor could not catch up quickly
2) to educate the world (and primarily competitors) on how things are actually done - therefore spurring innovation AND ensuring that the knowledge is a matter of record
The combination of these is of course what drives 'patent engineering' - if you can see how the originator did it and see the holes in his patent then you can still eat his lunch.
There does need to be a more thorough block on being able to patent the obvious, particularly obvious variations on someone else's patent. But then there needs to be an expectation that the originator will license the patent to competitors for a reasonable fee.
Loking at recent stories here, for instance, there seem to be at least two almot identical MEMS solutions for mobile phone clock generation, driven by a common demand and using (at least in part) common innovations to achieve features. Do these all have patents, and if so do they overlap? Where does the fault lie?
You raise a number of interesting points. The "accidental infringement" concept is an interesting idea but how would the "accidental" part be proven)? Would everyone infringe until they were caught and then negotiate a license or develop a work-around? What about monopoly laws? Should it be legal to refuse to license a technology at a rational price with the intention of driving a competitor out of business? Meanwhile, the patent office maintenance fees do have a tendency to cause companies to drop patents that are not paying their way. Finally, there is the issue of patents before their time. If you invent something 12 years before the world is ready, is it fair to require it be commercially launched within 4 years?
- patent owners must be commercially utilizing their patent to keep it. (within 4 years of patent application) - software and business process should not be patentable (ie 1 click) - accidental infringement be not initially punishable (ie stop it, then you have the choice to stop using/licence or litigate, but no backdated punitive action) - the bar for novelty to receive a patent be way higher than it is now. - patent application / renewal fees be based on number of patents held eg x = v * n(n+1)/2 where v = the base patent cost. Lets say v = $100 1 Patent = $100/annum to maintain 2 Patents = $300/annum to maintain 10 Patents = $5.5k/annum to maintain 100 Patents = $500k/annum to maintain 1000 Patents = $50M/annum to maintain 10k Patents = $5Billion/annum to maintain
This would force very large companies / trolls to only keep their best patents and toss out the dross polluting the patent system. You would need rules to stop companies spawning sub related companies to get around the intent.
I know of one small company who were on the wrong end of a patent infringement case, where a large corporation had bought a very poor patent, but had the money to defend it. They spent about $2m defending themselves and eventually settled, as they couldn't rely on a jury of ordinary people to understand the issues. In this case, the corporation made it clear that they wanted them out of the market as they wouldn't allow them to licence 'the technology' at anything other than the crippling fee of 30% of sales revenue. In the end the corporation agreed to a settlement as all of their development budget for this type of product was being spent on lawyers.
In this case, the patent should have been rejected by the patent office, as the first thing that the patent stated was that it was a combination of two well-known products so should have failed the KSR v Teleflex test.
In the USA, it seems that the patent office doesn't have the time to properly examine patents, so they are awarded and and then everyone sorts things out in the courts. The only beneficiaries are hugly expensive patent lawyers.
In Europe, they are much more strict about requiring a novel step, and they would not grant the same patent.
A major new patent bill, passed in a 117-4 vote by New Zealand's Parliament after five years of debate, has banned software patents.
The relevant clause of the patent bill actually states that a computer program is "not an invention." Some have suggested that was a way to get around the wording of the TRIPS intellectual property treaty, which requires patents to be "available for any inventions, whether products or processes, in all fields of technology."
Processes will still be patentable if the computer program is merely a way of implementing a patentable process. But patent claims that cover computer programs "as such" will not be allowed.
It seems there will be some leeway for computer programs directly tied to improved hardware. The bill includes the example of a better washing machine. Even if the improvements are implemented with a computer program, "the actual contribution is a new and improved way of operating a washing machine that gets clothes cleaner and uses less electricity," so a patent could be awarded.
One Member of Parliament who was deeply involved in the debate, Clare Curran, quoted several heads of software firms complaining about how the patenting process allowed "obvious things" to get patented and that "in general software patents are counter-productive." Curran quoted one developer as saying, "It's near impossible for software to be developed without breaching some of the hundreds of thousands of patents granted around the world for obvious work."
"These are the heavyweights of the new economy in software development," said Curran. "These are the people that needed to be listened to, and thankfully, they were."
The head of New Zealand's Institute of IT Professionals, Paul Matthews, celebrated the passage of the bill, tweeting: "Software patents are now disallowed in New Zealand. #Historic #Awesome #Yay." Matthews also noted that the new law will only apply to new patent applications, so existing software patents will continue to stand in New Zealand.
It's an open question whether other countries will take up New Zealand's example. An outright ban on software patents in the US seems unlikely given the large corporations—in tech and other sectors—that would oppose such a move. However, this year has been one of unprecedented concern about "patent trolls," with six bills introduced in Congress addressing that issue in the last several months.
i have a patent pending. it's a very simple idea... once it's on the market anyone will be able to copy it, fabricate it, and even improve on it. BUT nobody thought of it before, at least nobody has manufactured it commercially before. shouldn't i be able to profit from my invention?
Having been criticised by various government instrumentalities like the tax office and government R&D support departments because I don't hold patents for certain defence projects ... which I regard as 'secret' ... does create something of a challenge, and highlights the blind following of the patent system by certain officials.
On one side, I have the Australian Defence Export Control Office advices that restrict the export of 4 of my technologies, and then the tax department that infers that the technology does not exist because I will not detail who did what in the defence system sub assemblies. By lining up the required detailed invoices for any month, it will clearly explain how it is done .... secrecy, patents and the tax department is an interesting mix. And the tax department claims their compliance to 'commercial confidentiality' has a higher status than national defence secrecy.
Ah yes ... have patents run their course .... nah, they are a good starting point for the book!
Blog Doing Math in FPGAs Tom Burke 20 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...