I would also like to see a “cold room” exception. If you can show you invented the idea using completely independent thought, you don’t violate the patent and the patent is invalidated.
This seems to make a lot of sense at first glance. But how do you ever prove that you were unaware of the patented invention?
I propose a different solution. The burden of proof should be on the patent applicant to show nonobviousness, rather than the burden being on the PTO to show obviousness, and objective evidence should be required. The kinds of objective evidence that could be supplied to show nonobviousness have already been delineated by the Supreme court:
() commercial success
() long-felt but unsolved needs
() failure of others
For software, if you managed to get a peer-reviewed paper accepted, I think that could also count as evidence of nonobviousness -- just how strong would depend on the prestigiousness of the conference or journal in question. Or maybe you could point to a passage in a textbook or someone else's published paper to the effect that the problem you are solving had been open for some time.
>() commercial success () long-felt but unsolved needs () failure of others
I don't know how appropriate these are for software. The second one would almost always be impossible to show simply because of the recency and rate of change of the industry as a whole: How do you show "long" anything when the market segment has only existed for five years? Likewise failure of others; most of the problems haven't even existed long enough for anyone to have seriously tried and failed to solve them.
And then commercial success goes the other way. Because software is so complicated, it infringes a zillion patents by accident, so is all you need to show commercial success that Android or iOS is allegedly infringing and you're done because they're commercially successful? Or do you have to show that the patent was the main cause of the success, which will then go back to almost never being the case?
I suppose setting a nearly impossible standard that would result in a de facto abolition of software patents is one way to fix it, but I would much prefer just actually abolishing them. The alternative is to give the Federal Circuit yet another opportunity to read language broadly and expand the scope of patentability beyond what is reasonable.
> most of the problems haven't even existed long enough for anyone to have seriously tried and failed to solve them
Bingo. Currently, it's my impression, a lot of patents are being granted for completely uncreative solutions to novel problems. New technology (multi-touch screens, for example) creates new technical possibilities; along with those possibilities come a bunch of new problems. People are coming upon these problems, solving them quite straightforwardly, and then patenting the solution. It is exactly this I want to put a stop to, because I think it's this that is doing so much mischief.
Your second point is well taken. I think you'd have to show that the patent was a primary cause of the success -- which, as you say, is rarely the case. In any case, commercial success depends on a number of factors, and the courts would have to take that into account when weighing its probative value.
I don't think it's quite "de facto abolition", but yes, it is both my belief and my intention that this change in the rules would rule out most current software patents.
The few that remained would be for truly novel technology that would solve problems that were well known to be difficult. And those, I think, deserve to be patentable.
Can someone please clarify what it means to "endow a chair." I get from the article it has something to do with the EFF but no idea what it actually means.
It means that he has donated enough money to the EFF's endowment[1] to fund a position at the EFF in perpetuity. That is, it's a big enough chunk of money that it can sit in an investment account, and the interest can be used to pay for a person's salary and other costs.
Calling it a "chair" just means that Mark Cuban's name gets associated with the position, and the role comes with specific instructions. In this case, fighting stupid patents.
That isn't the case here. While it's a nice sum of money, Notch and Cuban agreed to give $250,000 (the news broke back on Dec 20th about this). That won't be enough to perpetually fund a position, but perhaps enough to hire two lawyers for a year.
You know who always wins? Lawyers are used in the litigation (for both parties). If you receive a letter written by another lawyer, you'll need to consult one in order to tell them to get lost. There are essentially no downsides to all the lawyers other than their clients no longer being able to pay.
40-50% of Congress and Senate are lawyers. I would be astonished to find them voting against their own profession's best interests.
This is an interesting claim, except for the fact that if you lose a lot, nobody hires you. Even if you are a corporate counsel, making bad decisions around patents will get you fired.
So no, lawyers don't "always win".
Nobody in senate/congress who is a lawyer gives a crap about "their profession's best interests", because it's not something that makes sense to do, politically or financially. They don't need to cater to lawyers to get jobs when they leave congress/senate, and lawyers contribute fairly equally to all parties.
Also, patent law is a very specialized area of law that requires separate testing and certification, it's almost a separate profession.
Lawyers are the perennial scapegoats for the wishes of the shitty clients they represent. Yes, plenty of them (particularly in the patent world) do it happily,, but behind every patent lawsuit is a crazy client who hired a lawyer to sue someone.
Lawyers grease the wheels of this activity. Consider what the effect of having loser pays on the system would have such as in Britain. That would seem to be a good path to go down and likely reduce the volume of legal activity.
Sometimes yes they are greasing, sometimes no, they aren't.
It varies depending on the type of firm.
Loser pays has its ups and downs. For example, you don't want loser pays when you are trying to fight small guy vs large guy, which is the case for most consumers.
Legislators should be required to have law degrees. Think about it: they write the laws. How is it a bad thing that they're equipped to know how they'll be interpreted?
It would be far better if congress is representative of the people of the country and be composed of doctors, engineers, teachers, taxi drivers, farmers, candlestick makers etc.
The legislators do not actually write the laws. They have staff as well as there being a process to help improve the text before legislating (eg publishing drafts). For some legislation (eg covering the financial industry) often the law that is passed just requires a government agency (eg SEC) to come up with the rules.
One of Michael Moore's films had a clip of a congressman explaining that they don't even read the legislation they vote on. It is a common claim that much of the legislation is written by lobbyists. And if you actually read any of it, it would obviously not really be what you would expect (short, legible).
The representatives use legislation as a fund raising mechanism (contributions are well documented). For example there were several taxes added temporarily to see if they worked well. They did, but congress doesn't make them permanent. Instead they wait to be paid each time to renew them temporarily again. (This started in the Reagan administration.)
See this excellent talk by Lessig: http://www.youtube.com/watch?v=Ik1AK56FtVc
The whole tax provisions as fund raising is covered at 7m30s. And the corruption is obvious at 25m20s.
> Legislators should be required to have law degrees.
Yes, and authors should be required to have English degrees before inflicting their opinions on the public, musicians should be required to have music degrees, philosophers should have philosophy degrees, and so forth. In short, people without degrees should have no right to take part in public affairs. Right?
If this system were in place, Albert Einstein would not have been allowed to submit his first physics papers for publication. And Abraham Lincoln, regarded as one of our most remarkable presidents, would have been barred from running for office (he had almost no formal education of any kind).
Possession of a law degree -- or any degree -- should not be a precondition to take part in public affairs. That's why we have a voting booth -- that's a much better competence filter than a sheepskin. But only if voters are educated and skeptical.
The difference is that I'm not required to read any particular author's books. Following the laws is more compulsory, and I think legislators should be held to a more rigorous standard.
> I think legislators should be held to a more rigorous standard.
There is a long democratic tradition (and politics is as much about tradition as it is about common sense) that says representatives should resemble those they represent. This is why there are two houses of congress -- the house of representatives (aptly named) to accommodate those who think representatives should be ordinary people, and the senate, to accommodate those who think representatives should be a cut above. The founding fathers also deliberately produced multiple branches of government to avoid too much power in too few hands.
The senate can't enact legislation without the cooperation of the "lower" house, and vice versa. Just as the founding fathers intended.
There is a term for the kind of government you're describing -- "elitism". If history teaches us anything, it is that people who have law degrees aren't better at representing the pubic's interest than those who don't.
There's some truth to that, and I don't pretend to have a solution to this difficult question. I remember from earlier posts that you are a lawyer yourself, so I don't need to give you examples of all the poorly drafted laws one comes across.
Do you think there's an ideal proportion of lawyers that should be involved in the legislative process?
He's wrong about AMD and Intel; it's not that they don't patent the stuff they use, they have had cross-licensing agreements for x86 for a long, long time.
AMD started producing Intel-compatible processors even before X86, as in 1975 they produced a reverse-engineered and unlicensed clone of 8080. Doing something like that would be unthinkable today. Also, AMD became a licensee of Intel's architecture after IBM told Intel that they require at least 2 sources for the chips, an agreement that Intel tried to cancel for 80360.
Back in those days these licensing agreements were more about copyright and trade-secrets rather than patents. And AMD had to license X86 from Intel because reverse-engineering was too time-consuming, not because they felt threatened. The patents arms race started around year 2000, when the bubble happened.
And it's not like Intel and AMD don't have a long history of litigation. They do.
Basically AMD had enough time to get big and innovate or buy valuable technology on its own. After AMD Athlon was released, the big and mighty Intel was in real jeopardy of being driven out of the market, as Athlons were not only cheaper, but better in every way (compared to its predecessor, the AMD K6, which was good and with extra goodies like 3D Now, but lacked a good coprocessor). Having to license AMD64 was probably a big embarrassment for Intel too, considering Intel was the one that tried to introduce a new architecture that was not backwards compatible with X86 and failed ... though it's interesting to think about the reasons why Intel tried a new and backwards incompatible processor and personally I think existing licensees of X86 had something to do with it.
Of course, today AMD seems to have lost their touch. For instance I'm typing this on a Thinkpad with an Ivy Bridge processor, that I bought because of Intel HD, the only graphics card with which you have absolutely no problem under Linux, as Intel's drivers are open-source. I was expecting AMD to do things like this after they bought ATI, but unfortunately they didn't. Such a shame.
I think the quickest way to do that would be to convince small country like Antigua for example to become a patent-free zone opened for infinite innovation.
I guess it's the times we live in, but the EFF has usually managed to seem a little more subdued than folks who would be fine with pushing a brand in return for a donation. After all it was never "Lotus 123 presents the hacker defense fund"
Obviously its been done in universities for some time, but for example the link in question discusses a chair that was permanently endowed via property upon his death. A bit different than covering 2-3 years of a borderline lobbyist for your business interests. I did a quick check of some similar organizations, this is the only named chair in the EFF, and EPIC, FSF, and the center for democracy and technology all don't have any.
I'm sure I'm rather biased though, as I sincerely doubt I'd have said anything if it was the Lawrence Lessig chair.
This seems to make a lot of sense at first glance. But how do you ever prove that you were unaware of the patented invention?
I propose a different solution. The burden of proof should be on the patent applicant to show nonobviousness, rather than the burden being on the PTO to show obviousness, and objective evidence should be required. The kinds of objective evidence that could be supplied to show nonobviousness have already been delineated by the Supreme court:
() commercial success () long-felt but unsolved needs () failure of others
For software, if you managed to get a peer-reviewed paper accepted, I think that could also count as evidence of nonobviousness -- just how strong would depend on the prestigiousness of the conference or journal in question. Or maybe you could point to a passage in a textbook or someone else's published paper to the effect that the problem you are solving had been open for some time.