The http://defendinnovation.org site is EFF's attempt to evangelize the idea that current patent system is broken and gather support for changes that would improve the system a lot.
Any lunatic apparently can start a petition on change.org and some did because EFF's (realistic) position isn't radical enough for his taste: he wants abolition of software patents.
Now, personally I would prefer abolition to the reform ideas proposed by EFF but this is not about what is better. EFF probably has to consider things like: does a given idea has a chance of being implemented by congress.
EFF is fighting the good fight. They try to improve the awful patent system and I have no doubt they do it in the best way they can.
Nothing stops the guy who created this petition to evangelize abolition of software patents. He can even start his own foundation for defending internet and civil liberties.
But attacking EFF in this way is harmful. He brings nothing positive to the table and can only derail EFF's efforts by instigating infighting among people who want the same thing: fix broken patent system.
Finally, one can support EFF's reform as well as patent abolition. We would be better off with either one.
Just don't attack EFF for trying to do something about patents.
By all means sign both the EFF petition and my petition. Still, EFF has never been the kind of organization that tries to cut deals inside the beltway. They've always been an organization that takes principled positions. Given that reputation, I think the implicit message of their campaign--that abolition is too radical even for EFF to endorse--undermines the argument for software patent abolition. I have no problem with them advocating incremental reforms, but they should also say that the ideal reform is eliminating software patents altogether.
There's one thing about software patents that I've come to realize: If a products sole competitive advantage over otherwise exchangeable goods is algorithmic... Is the world a better place if those insights are kept proprietary, or if they are made public knowledge via the patent process?
Software patents have been misused in many contexts, but for work that genuinely transformatively advances the state of the art, is protecting the right to have a viable business around this insight while making the mechanics of this knowledge public, is that a bad thing?
And just to repeat myself, broadly yes software patents have been awarded inappropriately time and again. But for those true insights, is a world better with a patented computational insight or a proprietary one?
For brilliant masterworks that would otherwise have gone unknown for decades, sure. But if I demand $10M for an invention that two vendors could have otherwise independently recreated for $1M and five years each, I have made that disclosure a net loss for society and the industry. The problem is that there's no market mechanism ensuring that licenses are available fairly at lower costs than would otherwise be spent on reinvention. And of course the avalanche of patents which merely disclose the obvious and claim the entire problem, yet can only be overturned individually at enormous risk and (uncompensated!) expense. And the 3x wilful-infringement doctrine over which practitioners are well advised never to even look at patents.
Since the replies seem to be thinking about patents in the context of large businesses, let me clarify the context that I am thinking of:
small businesses such as perhaps some iphone app developer that has spent some time developing an app/software that solves an "obvious problem" that no one has bothered to solve before, and that without patent protection would easily be reverse engineered for all the insights? Especially if those insights in fact also require advancing the state of the art in computer science?
I'm not talking about big corp X, i'm not talking about truly obvious ideas, I am ONLY talking about small business owners who are writing software that requires computer science research that advances the current state of the art, and which have no way of protecting themselves against copycat software makers (a la happens on the various smart phone app stores) unless patent protection is possible?
Can you name very many examples of cases where researchers have been searching for an algorithm and found it buried in an expired patent somewhere?
How common is it for some researcher to have an insight into an algorithm that no one else would be able to come up with independently (given 20 years) should the need arise?
If you read the material at defendinnovation.org, you'll see that they do not oppose patents on software. They call for changes to the patent system designed to make software patents work better. Which is fine, but not the same thing.
If you're not going to patent it, publish and/or otherwise disclose it ASAP, to establish priority. Otherwise, in many countries, you can be sued for using your own invention by someone who bothered to go to the patent office.
What some guy wants is for EFF to instead go for the kill (abolition of patent system).
Why does he feel that he should dictate what EFF does? I don't know.
Did he provide any reasoning as to why EFF should do what he wants as opposed to what they are actually doing? Not really.
Why he started a petition vs. writing a blog post describing his displeasure? Beats me.
To recap: EFF is doing something about broken patent system. They started http://defendinnovation.org and are trying to gather support for patent system reform and actually make this happen.
Some guy spent 5 minutes creating a petition asking EFF to drop what they're doing and do something else instead. Because he clearly knows betters.
It's not quite the same. For other patents, your invention needs to be some device or method for doing something. Now, this invention was obviously designed with math, but only the end result is patentable. That is, if you design some novel tire (for example), only the itself is protected--the math used to get the design in the first place is not protected. If somebody used the same math to come up with a different sort of tire, they would not have to license your patent.
With software patents, there is no device or method. You're taking some math and patenting it just because it's being used on a general computing device. The program itself is not what's protected by the patent--it's the actual math the program is isomorphic to that's protected. Somebody could write a completely new program using the same mathematical foundations and it would infringe on the software patent.
So yes, most patents are going to boil down to math one way or the other. The difference is that software patents protect the math directly where other patents only protect some specific device or invention.
EDIT: The original post seems to be deleted. Here it is so that my answer makes sense (I don't believe in deleting my own posts)
>Serious question, since I see this said a lot, but isn't that true of anything other than process patents? I'm hard-pressed to think of a patent in any other engineering discipline that at the end of the day doesn't boil down to math.
Here's how I look at it: If someone come up with a better car engine, they can patent it because it's a design and application of math. It's not like they patented the Carnot Cycle and precluded ANYONE from ever building ANY internal combustion engine. It's a specific design, not a general formula.
Patenting an algorithm, which is a purely mathematical construct, precludes any work done in the field. Patenting the design an ASIC that implements the algorithm would be the closest example I can think of to patenting an engine based on the Carnot cycle (and I'd be fine with a patent on your ASIC design).
To start, EFF doesn't support patents.
The http://defendinnovation.org site is EFF's attempt to evangelize the idea that current patent system is broken and gather support for changes that would improve the system a lot.
Any lunatic apparently can start a petition on change.org and some did because EFF's (realistic) position isn't radical enough for his taste: he wants abolition of software patents.
Now, personally I would prefer abolition to the reform ideas proposed by EFF but this is not about what is better. EFF probably has to consider things like: does a given idea has a chance of being implemented by congress.
EFF is fighting the good fight. They try to improve the awful patent system and I have no doubt they do it in the best way they can.
Nothing stops the guy who created this petition to evangelize abolition of software patents. He can even start his own foundation for defending internet and civil liberties.
But attacking EFF in this way is harmful. He brings nothing positive to the table and can only derail EFF's efforts by instigating infighting among people who want the same thing: fix broken patent system.
Finally, one can support EFF's reform as well as patent abolition. We would be better off with either one.
Just don't attack EFF for trying to do something about patents.