There are so many conclusive arguments against software patents that the result is overdetermined. Nevertheless, lining the pockets of the patent bar is likely to overcome any quality of argument as long as computer programmers have even less Washington influence than illegal alien terrorists.
I'll give a few arguments, but first let's consider yours. Black boxes are never patentable. The concept of patents is that you reveal exactly how your process works and you get a limited monopoly on the totally new and non-obvious bits of your process for the public benefit. It always matters how the process works inside. That's the theory anyway; the patent bar loves to rob the public by cheating the bargain.
1. Computer programmers are overwhelmingly against software patents. For the first time in my life I see educated people advocating the abolition of the entire patent system, even at the cost of life saving drugs, just to get patents out of computer software (John Siracusa, to cite a public personality, but there are many, many more who agree). Programmers hate patents in their industry by something like ten or twenty to one.
2. Software patents are harmful to innovation. They created almost all the modern patent trolling crisis. They shutdown startups and innovative projects and block open source. They promote vendor lock-in and empower giants over small companies with new ideas.
3. Software is math and both math and mental processes are SCOTUS identified ineligible subject matter.
4. The quality of software patents we see is uniformly bad. None of Apple's, Motorola's, Oracle's, Samsung's, or Microsoft's recent patents in litigation or threats thereof have contributed anything to the public. The billion dollar patents from Eolas and NTP were pure garbage. None of the troll patents we hear about all the time ever contributed anything to our art or even constitute inventions at all. Bilsky, Ultramercial, and Alice are all ludicrously abusive. Whole teams of programmers could work for decades searching without finding a single plausibly valid and useful patent in their field.
5. When we bought our computers, ever since the first general purpose computers in the 1940s, we have done so expecting to run programs on them. Using a machine for its expected and customary use is not subject to any patents beyond the patents on the machine itself. Running any program at all is the expected use of a computer, at least for programmers who write their own programs, so any program running on a computer is automatically in the prior art. In fact, we already have a list of every program our computers are expected to run and it includes every program any computer can run. We haven't printed the list out (not enough paper), but we know exactly how to produce any part of the list for your pleasure on demand.
Now the patent bar wants to tell us that programs we expected to run, that are the conventional use of our machines, and that have been on our list since the 1940s are their brilliant inventions and they are going to prohibit us from running them. No thanks.
" lining the pockets of the patent bar is likely to overcome any quality of argument as long as computer programmers have even less Washington influence than illegal alien terrorists."
1. The patent bar[1] itself has little to no influence in Washington. Nobody cares what they think. Congress cares about what the companies some of these folks work for think, but AIPLA, for example, has never had any real lobbying impact. They've complained about every rule change, every bill, every everything that has ever "hurt" their members. As far as i'm aware, there has never been any changes that have resulted from their comments.
2. The influence of computer programmers in Washington is mostly due to the fact that they spend time posting here instead of doing anything about anything. This is an entirely self inflicted wound, and the sooner programmers stop convincing themselves this isn't true, the better off they will be. Watching this happen without fail since 1997 is one of the reasons i became a lawyer.
I would wager a large amount of money if even 1% of the local user groups of programmers of various sorts that still exist, each appointed a representative, and had that person go to their local congressperson/senator, and said "I represent a concerned group of local small business people and programmers, and software patents are destroying our ability to make a living and causing us to lose jobs", that software patents would have been a solved problems years ago.
[1] I'm also not sure why you keep saying "the patent bar".
The patent bar itself is not of one mind, and there is no real, single organized patent bar that does anything, because the patent bar are those attorneys and agents licensed to practice before the PTO, and since being a lawyer is not a requirement, they come from a variety of walks of life, not just a bunch of attorneys. The actual patent bar (IE as run by the PTO) offers literally no opinions on anything, and certainly does not lobby, just like the bar of the 9th circuit court of appeals doesn't lobby.
There are a number of outside organized collections of patent attorneys, like AIPLA, but as mentioned, none are really "the patent bar". There is no single mind among these folks, either, Most of the inside and outside litigators I have met in the corporate realm rail against software patents. So even "the patent bar" is not consistent here, it's more like you have the criminal defense lawyers vs the prosecutors, rather than a single "criminal law bar".
> ... and said "I represent a concerned group of local small business people and programmers, and software patents are destroying our ability to make a living and causing us to lose jobs", that software patents would have been a solved problems years ago.
That is very true, but my very strong hunch is that this hasn't happened because software patents are not really a problem -- at least not anymore. As I've said elsewhere, the vast majority of programmers barely even know what a patent is, and tons more consider them a good thing. Those are also overwhelmingly the people that don't spend time posting on places like this.
Another indirect piece of evidence is the current uproar in Congress about patent trolls: A handful of trolls (that I'm aware of) -- like the WiFi guys, the scan-to-email guys, the vehicle-tracking guys -- send out demand letters to mom & pop businesses, and suddenly Congress is up in arms. It's possible, but I find it unlikely that, if there are so many companies involved in software patent litigation, there are not even a few that would reach out to their representatives.
Talking about he influence of the patent bar is utterly ridiculous. The plaintiffs bar is way bigger and has more clout, but that doesn't stop the Supreme Court from paring back the scope of class actions.
Only in the HN bubble is the argument against patents "conclusive." What Congress sees is people in the Internet tech sector complaining about patents, then people at a Oracle and Microsoft and Apple defending them, along with engineers at dozens of key American companies in aerospace, medicine, agriculture, etc.
The Supreme Court is required to give effect to Congressional intent in the patent act. The language of the patent act days that nearly any subject matter is potentially patentable. Unlike commenters on HN, they must approach software patents in a way that gives effect to that principle.
Black boxes are never patentable. The concept of patents is that you reveal exactly how your process works and you get a limited monopoly on the totally new and non-obvious bits of your process for the public benefit. It always matters how the process works inside.
I shouldn't have used the term "black box". Assume it's a pair of crystal clear boxes, and inside one is a beautiful, novel arrangement of gears and camshafts which I have painstakingly documented in patent A, and inside the other is a run-of-the-mill laptop, running software that I have painstakingly documented in patent B.
Both devices perform the same function. It's something no device has ever done before, and has substantial practical uses. Let's say it can accurately predict when your in-laws will next visit.
What makes patent A valid while patent B is "harmful to innovation"? Or are they both "software patents"?
I'll let alone the fact that your definition of "software patents" is flawed. Let me address your other points.
> There are so many conclusive arguments against software patents that the result is overdetermined.
Or, also known as "being in an echo chamber".
> Black boxes are never patentable.
The concepts of a Person of Ordinary Skill and Doctrine of Equivalents is not "black boxes". Are you saying all the so-called "bad" patents you've seen were things you couldn't implement by reading them?
>1. Computer programmers are overwhelmingly against software patents.
Echo chamber. The vast majority of developers barely know what a patent is. Many others think they're a point of pride, and that's where all these software patents are coming from. The rest are just really vocal.
> For the first time in my life I see educated people advocating the abolition of the entire patent system, even at the cost of life saving drugs, just to get patents out of computer software
"Educated" does not mean "well-informed", and it does not mean they know anything about patents. Economists are still having trouble quantifying the benefits and harms of patents, let alone software patents, so I'm guessing your "educated people" are basing their opinions on personal biases rather than evidence.
> Programmers hate patents in their industry by something like ten or twenty to one.
Wow, this echo chamber seems to have reached resonance. Here's a statistic closer to the truth: 99% of people complaining about patents don't know jack about how patents work and simply regurgitate what they tech media tells them, and tech media is not only clueless, it's deliberately misleading.
> 2. Software patents are harmful to innovation. They created almost all the modern patent trolling crisis. They shutdown startups and innovative projects and block open source.
Studies [1, 2], albeit with only publicly available data, have found no empirical evidence of a "modern patent trolling crisis", and I have heard of more small companies being ripped off by big guys (see i4i) than being shut down by patent trolls. On the other hand, I can list some studies that find "software" patents are no worse than other types of patents [3, 4, 5], and can actually be pretty useful for startups. [6, 7].
> 3. Software is math and both math and mental processes are SCOTUS identified ineligible subject matter.
"Machines are made of metals which occur in nature, which are SCOTUS identified ineligible subject matter."
> 4. The quality of software patents we see is uniformly bad...
I'm genuinely curious: how do you judge the quality of a patent?
> 5. When we bought our computers, ever since the first general purpose computers in the 1940s, we have done so expecting to run programs on them. Using a machine for its expected and customary use is not subject to any patents beyond the patents on the machine itself.
"When we first mined metals, we have done so expecting to make things out of them. Using a metal for its expected and customary use is not subject to patents beyond the patents on the metal itself."
There are plenty of those, including, say, IP Watchdog. Yes, some of us have been known to hang out in very pro-patent echo chambers as well.
The concepts of a Person of Ordinary Skill and Doctrine of Equivalents is not "black boxes".
Complete non sequitur. A black box is a device where the function is known, but not its workings. If something is a black box, it's hard to see how it would satisfy the 'enablement' standard. Well, unless it's black box software, then it's just fine! Judges apparently think that you just tell the computer what to do and it requires no undue experimentation. So you just need someone to come up with the brilliant idea to patent software that simulates the human brain and achieves sentience (or at least passes the Turing test) and everything thereafter is just a matter of typing it in once this disclosed. Yes really, see [1].
Studies [1, 2], albeit with only publicly available data, have found no empirical evidence of a "modern patent trolling crisis"
They apparently haven't read the newspaper, either, wherein we find that every major tech company is apparently a thief, in spite of the fact that nobody actually reads patents (we're all warned not to, treble damages and all that). Interestingly, widespread reinvention of a patent is not considered empirical proof of obviousness, rather it proves that the idea was really valuable and the patent holder deserves lots of money for discovering that, say, cell phone computers can do email just as well as other computers. Also, we'd best not look at the steadily increasing number of patent cases, nor all of the goings on in Marshall, TX.
"Machines are made of metals which occur in nature, which are SCOTUS identified ineligible subject matter."
You're confusing "contains" with "is" here. I've also seen patent lawyers try to do a reductio ad absurdam saying that software is hardware (read some of the Patently O comments...). Apparently they don't realize that software is information and that numbers are no different than any other form of information. And, as with your broken logic, we can say that hardware contains information [software], but it's quite ridiculous to say that hardware is information.
I'm genuinely curious: how do you judge the quality of a patent?
By how much of it is actually new. If someone comes up with a new computing device, some idiot will try to patent having it do email, web surfing and everything else we already know that computers can do.
When we first mined metals, we have done so expecting to make things out of them. Using a metal for its expected and customary use is not subject to patents beyond the patents on the metal itself."
We can actually enumerate all of the programs that can go on a particular computer. It's the set of integers from zero to the largest number that would fit in memory (most of these are not useful, however). I don't believe you can do that with all things that could be made of metal, moreover the metal's properties are altered in certain combinations (e.g. amalgams) whereas the computer's properties are not altered by the information on it--the computer is still performing an instruction loop over the data it contains.
[1] This is a real howler for programmers, found quoted on Patently O: Fonar Corp. v. General Electric Co., 107 F.3d 1543, 1549, 41 USPQ2d 1801, 1805 (Fed. Cir. 1997) (“As a general rule, where software constitutes part of a best mode of carrying out an invention, description of such a best mode is satisfied by a disclosure of the functions of the software. This is because, normally, writing code for such software is within the skill of the art, not requiring undue experimentation, once its functions have been disclosed. * * * Thus, flow charts or source code listings are not a requirement for adequately disclosing the functions of software.”).
Glad to finally see someone who looks at the other side of things as well! I generally avoid IPWatchDog for the same reason reasons as Groklaw: too shrill, snarky and narrow-minded.
> Judges apparently think that you just tell the computer what to do and it requires no undue experimentation.
Yes, Fonar v GE is a ridiculous one when generalized, and I do wholeheartedly agree that the "enablement" requirement is very weak currently and a ton of patents don't meet it in my eyes. Take the PageRank patent, for one, often cited as an example of a good patent. It is very lacking in implementation details, so much so that a blogger set out to implement it and ended up with a bunch of posts complaining about all the undue experimentation he had to do.
But take the average case: how much undue experimentation is required for your average patent? How much undue fiddling would be required to implement, say, Twitter's pull-to-refresh, or the iPhone rubber banding patent, or the Google doodle patent, or the MS FAT patent? I think my point still stands: how many patents couldn't you implement just by reading them? (Also, keep in mind triviality of implementation is very different from non-obviousness.)
> They apparently haven't read the newspaper...
Like I said, tech media is not only clueless, it's deliberately misleading. Newspapers are merely anecdata, and worse, potentially biased anecdata [1]. Why rely on that when we have studies that tell you the data they looked at so you can find the flaws in them rather than accepting unsubstantiated, rageview-bait at face value?
> Also, we'd best not look at the steadily increasing number of patent cases, nor all of the goings on in Marshall, TX.
Actually, studies [2, 3, 4] did look at it and found no real increase. [2, 3] found that the AIA joinder rules, where previously a single case could have multiple defendants now must be individual cases, caused an inflation in lawsuits with no statistically significant increase in the number of litigants involved. [4] actually finds EDT is better than others with respect to plaintiff win-rates.
> Interestingly, widespread reinvention of a patent is not considered empirical proof of obviousness
A few points here, since this turns up often:
1. Reinvention many years later is not proof of obviousness. It simply means an idea has permeated widely enough to become obvious to others. A big reason the way US (and other) patent offices do examination, where they must support a rejection with enough prior art references, is to avoid hindsight bias.
2. There is very little "widespread" re-invention going on; what is happening is one product/project/company re-invents something(s) and that gets widely used. (Think Android, or the iOS SDK.)
3. Simultaneous independent invention could be considered proof of obviousness... Or it could be proof that two or more really smart people were working on the same problem at the same time, which has happened in "patent races". But at least at the PTO it happened so rarely that they changed to first-to-file.
> You're confusing "contains" with "is" here.
No, you're confusing both, the nature of software and the inventions being claimed. The software is what makes a machine do useful things, and those useful things are the inventions that are patentable.
> I've also seen patent lawyers try to do a reductio ad absurdam saying that software is hardware (read some of the Patently O comments…).
What they mean is, it's "equivalent", which is perfectly accurate. Also, please explain how software can exist without hardware. What's in your mind is steps to implement or recreate software, but it cannot exist without hardware outside your mind. Just like any other invention.
> Apparently they don't realize that software is information ...
How is "Software is information" different from "machines are metal"? Doesn't seem to be my logic that's broken.
> By how much of it is actually new.
Sure, but how do you determine that a patent covers something actually new?
> We can actually enumerate all of the programs that can go on a particular computer. It's the set of integers from zero to the largest number that would fit in memory (most of these are not useful, however).
I don't see the point, but I can trivially debunk this:
1) The order in which you "interpret" or execute those bytes gives you a completely different program. So if you can fit in 1GB in a high-end smartphone, the number of all possible programs approaches factorial(1e9). That's a number about 8 billion digits long, if Google serves correctly.
2) Now consider that you fill that memory up with programs that generate other programs, or variations of themselves. Something like, say, Conway's game of life. I cannot begin to calculate how many different programs that could generate.
These are absolutely humongous numbers, no different from the number of ways physical objects can be arranged. Now consider that only an infinitesimally small part of them are actually useful. I find that an argument to the patentability of software than against it.
> ... whereas the computer's properties are not altered by the information on it--the computer is still performing an instruction loop over the data it contains.
The computer's properties are not altered?!? It goes from being a heap of semiconductors to actually doing something! How is that different from a machine going from a heap of components to an useful implement?
And each program makes it do something different so each program changes the purposes of the computer, and so becomes a different tool! A simple thought exercise: you're on Youtube. Someone asks you what you're doing. Is your answer:
a) "I'm using the computer"
b) "I'm watching videos"
Interesting isn't it? Now repeat the exercise with games, programming, document editing. And then explain to me how running different software on the same machine does not change its function.
Yes, Fonar v GE is a ridiculous one when generalized, and I do wholeheartedly agree that the "enablement" requirement is very weak currently and a ton of patents don't meet it in my eyes. Take the PageRank patent, for one, often cited as an example of a good patent. It is very lacking in implementation details, so much so that a blogger set out to implement it and ended up with a bunch of posts complaining about all the undue experimentation he had to do.
Glad to hear it.
[4] actually finds EDT is better than others with respect to plaintiff win-rates.
I might be missing something, but I don't see it taking the really high settlement rate into account. If you include those as mostly losses, well yes, Northern California still sucks as well and the overall win rate sucks.
There is very little "widespread" re-invention going on
Using SSL with RC4, scan to email, pick any "we own the internet" patent, really.
The order in which you "interpret" or execute those bytes gives you a completely different program.
No, the hardware performs an instruction loop no matter what is in memory or the registers (unless you break it).
Now consider that you fill that memory up with programs that generate other programs, or variations of themselves. Something like, say, Conway's game of life. I cannot begin to calculate how many different programs that could generate.
None of which fit into the device (you've already used up all memory, remember?). It's also a finite number. You're thinking of a Turing Machine with infinite memory (wish we had one, but we don't). There are functions with finite values that are not computable incidentally (the Busy Beaver function, which incidentally is related to your idea, is one such function). Yes, there are more programs than fit into memory for any finite amount. Sadly, we do not have infinite memory. Yes, you can increase the number of possible programs by adding memory (this should be no surprise to anyone who has ever used a computer and needed more space to install X). Every program (equivalently, every mathematical statement) can be identified with a number. Even this text is nothing more than a very long number. I believe I already mentioned Godel's work.
no different from the number of ways physical objects can be arranged.
That's infinite, though (as far as physics knows). Anything you can fit into memory will be finite (though it can be extremely large, as you've noticed). One is exhaustible and enumerable, the other is not. The computer is designed to let you put any value whatsoever into its memory. We do not have a general atom-arranger that can make any construct out of matter. Though how I wish we did have one of those from Star Trek! ... I'm sure you'd then have people patent making X with the synthesizer... for every X that already exists. Which is about where we are with the computer, no? That's why we think that new synthesizers (computers) should get patents but not every trivial, already existing X that you can imagine putting on there. At least when it's a new X we can respect it a bit, but when you're doing old X + computer, it gets silly fast, because many, many people can figure out how to do old X + computer ... but the lawyers think it patentable for some reason.
The computer's properties are not altered?!? It goes from being a heap of semiconductors to actually doing something! How is that different from a machine going from a heap of components to an useful implement?
It goes from executing an instruction loop on no data... executing an instruction loop on data. Yes, the instruction loop was a very fundamental, useful idea, precisely because it can compute any algorithm given appropriate input. The computer is unlike any other invention precisely because we found such a general mathematical construct to put behind it. But the data is just a large number. Yes, printing a few GB of data as a decimal number would run you out of paper, but there's absolutely no reason we couldn't do it.
Comments
There are so many conclusive arguments against software patents that the result is overdetermined. Nevertheless, lining the pockets of the patent bar is likely to overcome any quality of argument as long as computer programmers have even less Washington influence than illegal alien terrorists.
I'll give a few arguments, but first let's consider yours. Black boxes are never patentable. The concept of patents is that you reveal exactly how your process works and you get a limited monopoly on the totally new and non-obvious bits of your process for the public benefit. It always matters how the process works inside. That's the theory anyway; the patent bar loves to rob the public by cheating the bargain.
1. Computer programmers are overwhelmingly against software patents. For the first time in my life I see educated people advocating the abolition of the entire patent system, even at the cost of life saving drugs, just to get patents out of computer software (John Siracusa, to cite a public personality, but there are many, many more who agree). Programmers hate patents in their industry by something like ten or twenty to one.
2. Software patents are harmful to innovation. They created almost all the modern patent trolling crisis. They shutdown startups and innovative projects and block open source. They promote vendor lock-in and empower giants over small companies with new ideas.
3. Software is math and both math and mental processes are SCOTUS identified ineligible subject matter.
4. The quality of software patents we see is uniformly bad. None of Apple's, Motorola's, Oracle's, Samsung's, or Microsoft's recent patents in litigation or threats thereof have contributed anything to the public. The billion dollar patents from Eolas and NTP were pure garbage. None of the troll patents we hear about all the time ever contributed anything to our art or even constitute inventions at all. Bilsky, Ultramercial, and Alice are all ludicrously abusive. Whole teams of programmers could work for decades searching without finding a single plausibly valid and useful patent in their field.
5. When we bought our computers, ever since the first general purpose computers in the 1940s, we have done so expecting to run programs on them. Using a machine for its expected and customary use is not subject to any patents beyond the patents on the machine itself. Running any program at all is the expected use of a computer, at least for programmers who write their own programs, so any program running on a computer is automatically in the prior art. In fact, we already have a list of every program our computers are expected to run and it includes every program any computer can run. We haven't printed the list out (not enough paper), but we know exactly how to produce any part of the list for your pleasure on demand.
Now the patent bar wants to tell us that programs we expected to run, that are the conventional use of our machines, and that have been on our list since the 1940s are their brilliant inventions and they are going to prohibit us from running them. No thanks.
" lining the pockets of the patent bar is likely to overcome any quality of argument as long as computer programmers have even less Washington influence than illegal alien terrorists."
1. The patent bar[1] itself has little to no influence in Washington. Nobody cares what they think. Congress cares about what the companies some of these folks work for think, but AIPLA, for example, has never had any real lobbying impact. They've complained about every rule change, every bill, every everything that has ever "hurt" their members. As far as i'm aware, there has never been any changes that have resulted from their comments.
2. The influence of computer programmers in Washington is mostly due to the fact that they spend time posting here instead of doing anything about anything. This is an entirely self inflicted wound, and the sooner programmers stop convincing themselves this isn't true, the better off they will be. Watching this happen without fail since 1997 is one of the reasons i became a lawyer.
I would wager a large amount of money if even 1% of the local user groups of programmers of various sorts that still exist, each appointed a representative, and had that person go to their local congressperson/senator, and said "I represent a concerned group of local small business people and programmers, and software patents are destroying our ability to make a living and causing us to lose jobs", that software patents would have been a solved problems years ago.
[1] I'm also not sure why you keep saying "the patent bar".
The patent bar itself is not of one mind, and there is no real, single organized patent bar that does anything, because the patent bar are those attorneys and agents licensed to practice before the PTO, and since being a lawyer is not a requirement, they come from a variety of walks of life, not just a bunch of attorneys. The actual patent bar (IE as run by the PTO) offers literally no opinions on anything, and certainly does not lobby, just like the bar of the 9th circuit court of appeals doesn't lobby.
There are a number of outside organized collections of patent attorneys, like AIPLA, but as mentioned, none are really "the patent bar". There is no single mind among these folks, either, Most of the inside and outside litigators I have met in the corporate realm rail against software patents. So even "the patent bar" is not consistent here, it's more like you have the criminal defense lawyers vs the prosecutors, rather than a single "criminal law bar".
> ... and said "I represent a concerned group of local small business people and programmers, and software patents are destroying our ability to make a living and causing us to lose jobs", that software patents would have been a solved problems years ago.
That is very true, but my very strong hunch is that this hasn't happened because software patents are not really a problem -- at least not anymore. As I've said elsewhere, the vast majority of programmers barely even know what a patent is, and tons more consider them a good thing. Those are also overwhelmingly the people that don't spend time posting on places like this.
Another indirect piece of evidence is the current uproar in Congress about patent trolls: A handful of trolls (that I'm aware of) -- like the WiFi guys, the scan-to-email guys, the vehicle-tracking guys -- send out demand letters to mom & pop businesses, and suddenly Congress is up in arms. It's possible, but I find it unlikely that, if there are so many companies involved in software patent litigation, there are not even a few that would reach out to their representatives.
Talking about he influence of the patent bar is utterly ridiculous. The plaintiffs bar is way bigger and has more clout, but that doesn't stop the Supreme Court from paring back the scope of class actions.
Only in the HN bubble is the argument against patents "conclusive." What Congress sees is people in the Internet tech sector complaining about patents, then people at a Oracle and Microsoft and Apple defending them, along with engineers at dozens of key American companies in aerospace, medicine, agriculture, etc.
The Supreme Court is required to give effect to Congressional intent in the patent act. The language of the patent act days that nearly any subject matter is potentially patentable. Unlike commenters on HN, they must approach software patents in a way that gives effect to that principle.
Black boxes are never patentable. The concept of patents is that you reveal exactly how your process works and you get a limited monopoly on the totally new and non-obvious bits of your process for the public benefit. It always matters how the process works inside.
I shouldn't have used the term "black box". Assume it's a pair of crystal clear boxes, and inside one is a beautiful, novel arrangement of gears and camshafts which I have painstakingly documented in patent A, and inside the other is a run-of-the-mill laptop, running software that I have painstakingly documented in patent B.
Both devices perform the same function. It's something no device has ever done before, and has substantial practical uses. Let's say it can accurately predict when your in-laws will next visit.
What makes patent A valid while patent B is "harmful to innovation"? Or are they both "software patents"?
Well said. It's a topic I find myself at a loss of words to describe to the uninformed, but you do so nicely here.
I'll let alone the fact that your definition of "software patents" is flawed. Let me address your other points.
> There are so many conclusive arguments against software patents that the result is overdetermined.
Or, also known as "being in an echo chamber".
> Black boxes are never patentable.
The concepts of a Person of Ordinary Skill and Doctrine of Equivalents is not "black boxes". Are you saying all the so-called "bad" patents you've seen were things you couldn't implement by reading them?
>1. Computer programmers are overwhelmingly against software patents.
Echo chamber. The vast majority of developers barely know what a patent is. Many others think they're a point of pride, and that's where all these software patents are coming from. The rest are just really vocal.
> For the first time in my life I see educated people advocating the abolition of the entire patent system, even at the cost of life saving drugs, just to get patents out of computer software
"Educated" does not mean "well-informed", and it does not mean they know anything about patents. Economists are still having trouble quantifying the benefits and harms of patents, let alone software patents, so I'm guessing your "educated people" are basing their opinions on personal biases rather than evidence.
> Programmers hate patents in their industry by something like ten or twenty to one.
Wow, this echo chamber seems to have reached resonance. Here's a statistic closer to the truth: 99% of people complaining about patents don't know jack about how patents work and simply regurgitate what they tech media tells them, and tech media is not only clueless, it's deliberately misleading.
> 2. Software patents are harmful to innovation. They created almost all the modern patent trolling crisis. They shutdown startups and innovative projects and block open source.
Studies [1, 2], albeit with only publicly available data, have found no empirical evidence of a "modern patent trolling crisis", and I have heard of more small companies being ripped off by big guys (see i4i) than being shut down by patent trolls. On the other hand, I can list some studies that find "software" patents are no worse than other types of patents [3, 4, 5], and can actually be pretty useful for startups. [6, 7].
> 3. Software is math and both math and mental processes are SCOTUS identified ineligible subject matter.
"Machines are made of metals which occur in nature, which are SCOTUS identified ineligible subject matter."
> 4. The quality of software patents we see is uniformly bad...
I'm genuinely curious: how do you judge the quality of a patent?
> 5. When we bought our computers, ever since the first general purpose computers in the 1940s, we have done so expecting to run programs on them. Using a machine for its expected and customary use is not subject to any patents beyond the patents on the machine itself.
"When we first mined metals, we have done so expecting to make things out of them. Using a metal for its expected and customary use is not subject to patents beyond the patents on the metal itself."
1. http://www.gao.gov/assets/660/657103.pdf
2. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1396319
3. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=970083
4. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=650921
5. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=999098
6. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=510103
7. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=989592
There are plenty of those, including, say, IP Watchdog. Yes, some of us have been known to hang out in very pro-patent echo chambers as well.
Complete non sequitur. A black box is a device where the function is known, but not its workings. If something is a black box, it's hard to see how it would satisfy the 'enablement' standard. Well, unless it's black box software, then it's just fine! Judges apparently think that you just tell the computer what to do and it requires no undue experimentation. So you just need someone to come up with the brilliant idea to patent software that simulates the human brain and achieves sentience (or at least passes the Turing test) and everything thereafter is just a matter of typing it in once this disclosed. Yes really, see [1].
They apparently haven't read the newspaper, either, wherein we find that every major tech company is apparently a thief, in spite of the fact that nobody actually reads patents (we're all warned not to, treble damages and all that). Interestingly, widespread reinvention of a patent is not considered empirical proof of obviousness, rather it proves that the idea was really valuable and the patent holder deserves lots of money for discovering that, say, cell phone computers can do email just as well as other computers. Also, we'd best not look at the steadily increasing number of patent cases, nor all of the goings on in Marshall, TX.
You're confusing "contains" with "is" here. I've also seen patent lawyers try to do a reductio ad absurdam saying that software is hardware (read some of the Patently O comments...). Apparently they don't realize that software is information and that numbers are no different than any other form of information. And, as with your broken logic, we can say that hardware contains information [software], but it's quite ridiculous to say that hardware is information.
By how much of it is actually new. If someone comes up with a new computing device, some idiot will try to patent having it do email, web surfing and everything else we already know that computers can do.
We can actually enumerate all of the programs that can go on a particular computer. It's the set of integers from zero to the largest number that would fit in memory (most of these are not useful, however). I don't believe you can do that with all things that could be made of metal, moreover the metal's properties are altered in certain combinations (e.g. amalgams) whereas the computer's properties are not altered by the information on it--the computer is still performing an instruction loop over the data it contains.
[1] This is a real howler for programmers, found quoted on Patently O: Fonar Corp. v. General Electric Co., 107 F.3d 1543, 1549, 41 USPQ2d 1801, 1805 (Fed. Cir. 1997) (“As a general rule, where software constitutes part of a best mode of carrying out an invention, description of such a best mode is satisfied by a disclosure of the functions of the software. This is because, normally, writing code for such software is within the skill of the art, not requiring undue experimentation, once its functions have been disclosed. * * * Thus, flow charts or source code listings are not a requirement for adequately disclosing the functions of software.”).
Glad to finally see someone who looks at the other side of things as well! I generally avoid IPWatchDog for the same reason reasons as Groklaw: too shrill, snarky and narrow-minded.
> Judges apparently think that you just tell the computer what to do and it requires no undue experimentation.
Yes, Fonar v GE is a ridiculous one when generalized, and I do wholeheartedly agree that the "enablement" requirement is very weak currently and a ton of patents don't meet it in my eyes. Take the PageRank patent, for one, often cited as an example of a good patent. It is very lacking in implementation details, so much so that a blogger set out to implement it and ended up with a bunch of posts complaining about all the undue experimentation he had to do.
But take the average case: how much undue experimentation is required for your average patent? How much undue fiddling would be required to implement, say, Twitter's pull-to-refresh, or the iPhone rubber banding patent, or the Google doodle patent, or the MS FAT patent? I think my point still stands: how many patents couldn't you implement just by reading them? (Also, keep in mind triviality of implementation is very different from non-obviousness.)
> They apparently haven't read the newspaper...
Like I said, tech media is not only clueless, it's deliberately misleading. Newspapers are merely anecdata, and worse, potentially biased anecdata [1]. Why rely on that when we have studies that tell you the data they looked at so you can find the flaws in them rather than accepting unsubstantiated, rageview-bait at face value?
> Also, we'd best not look at the steadily increasing number of patent cases, nor all of the goings on in Marshall, TX.
Actually, studies [2, 3, 4] did look at it and found no real increase. [2, 3] found that the AIA joinder rules, where previously a single case could have multiple defendants now must be individual cases, caused an inflation in lawsuits with no statistically significant increase in the number of litigants involved. [4] actually finds EDT is better than others with respect to plaintiff win-rates.
> Interestingly, widespread reinvention of a patent is not considered empirical proof of obviousness
A few points here, since this turns up often:
1. Reinvention many years later is not proof of obviousness. It simply means an idea has permeated widely enough to become obvious to others. A big reason the way US (and other) patent offices do examination, where they must support a rejection with enough prior art references, is to avoid hindsight bias.
2. There is very little "widespread" re-invention going on; what is happening is one product/project/company re-invents something(s) and that gets widely used. (Think Android, or the iOS SDK.)
3. Simultaneous independent invention could be considered proof of obviousness... Or it could be proof that two or more really smart people were working on the same problem at the same time, which has happened in "patent races". But at least at the PTO it happened so rarely that they changed to first-to-file.
> You're confusing "contains" with "is" here.
No, you're confusing both, the nature of software and the inventions being claimed. The software is what makes a machine do useful things, and those useful things are the inventions that are patentable.
> I've also seen patent lawyers try to do a reductio ad absurdam saying that software is hardware (read some of the Patently O comments…).
What they mean is, it's "equivalent", which is perfectly accurate. Also, please explain how software can exist without hardware. What's in your mind is steps to implement or recreate software, but it cannot exist without hardware outside your mind. Just like any other invention.
> Apparently they don't realize that software is information ...
How is "Software is information" different from "machines are metal"? Doesn't seem to be my logic that's broken.
> By how much of it is actually new.
Sure, but how do you determine that a patent covers something actually new?
> We can actually enumerate all of the programs that can go on a particular computer. It's the set of integers from zero to the largest number that would fit in memory (most of these are not useful, however).
I don't see the point, but I can trivially debunk this:
1) The order in which you "interpret" or execute those bytes gives you a completely different program. So if you can fit in 1GB in a high-end smartphone, the number of all possible programs approaches factorial(1e9). That's a number about 8 billion digits long, if Google serves correctly.
2) Now consider that you fill that memory up with programs that generate other programs, or variations of themselves. Something like, say, Conway's game of life. I cannot begin to calculate how many different programs that could generate.
These are absolutely humongous numbers, no different from the number of ways physical objects can be arranged. Now consider that only an infinitesimally small part of them are actually useful. I find that an argument to the patentability of software than against it.
> ... whereas the computer's properties are not altered by the information on it--the computer is still performing an instruction loop over the data it contains.
The computer's properties are not altered?!? It goes from being a heap of semiconductors to actually doing something! How is that different from a machine going from a heap of components to an useful implement?
And each program makes it do something different so each program changes the purposes of the computer, and so becomes a different tool! A simple thought exercise: you're on Youtube. Someone asks you what you're doing. Is your answer:
a) "I'm using the computer"
b) "I'm watching videos"
Interesting isn't it? Now repeat the exercise with games, programming, document editing. And then explain to me how running different software on the same machine does not change its function.
1. http://www.paulgraham.com/submarine.html
2. http://www.gao.gov/assets/660/657103.pdf (same as [1] in post upstream)
3. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2346381
4. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1597919
Glad to hear it.
I might be missing something, but I don't see it taking the really high settlement rate into account. If you include those as mostly losses, well yes, Northern California still sucks as well and the overall win rate sucks.
Using SSL with RC4, scan to email, pick any "we own the internet" patent, really.
No, the hardware performs an instruction loop no matter what is in memory or the registers (unless you break it).
None of which fit into the device (you've already used up all memory, remember?). It's also a finite number. You're thinking of a Turing Machine with infinite memory (wish we had one, but we don't). There are functions with finite values that are not computable incidentally (the Busy Beaver function, which incidentally is related to your idea, is one such function). Yes, there are more programs than fit into memory for any finite amount. Sadly, we do not have infinite memory. Yes, you can increase the number of possible programs by adding memory (this should be no surprise to anyone who has ever used a computer and needed more space to install X). Every program (equivalently, every mathematical statement) can be identified with a number. Even this text is nothing more than a very long number. I believe I already mentioned Godel's work.
That's infinite, though (as far as physics knows). Anything you can fit into memory will be finite (though it can be extremely large, as you've noticed). One is exhaustible and enumerable, the other is not. The computer is designed to let you put any value whatsoever into its memory. We do not have a general atom-arranger that can make any construct out of matter. Though how I wish we did have one of those from Star Trek! ... I'm sure you'd then have people patent making X with the synthesizer... for every X that already exists. Which is about where we are with the computer, no? That's why we think that new synthesizers (computers) should get patents but not every trivial, already existing X that you can imagine putting on there. At least when it's a new X we can respect it a bit, but when you're doing old X + computer, it gets silly fast, because many, many people can figure out how to do old X + computer ... but the lawyers think it patentable for some reason.
It goes from executing an instruction loop on no data... executing an instruction loop on data. Yes, the instruction loop was a very fundamental, useful idea, precisely because it can compute any algorithm given appropriate input. The computer is unlike any other invention precisely because we found such a general mathematical construct to put behind it. But the data is just a large number. Yes, printing a few GB of data as a decimal number would run you out of paper, but there's absolutely no reason we couldn't do it.