Software Patents Compared to Hard Patents 134
Arie writes "The Slate discusses the obvious differences between patenting an algorithm and a drug. The article introduces the Fence test, which basically says that if you can physically protect your property, you have a case patenting it. In addition, it claims that the burden on a programmer identifying whether he is infringing on a patent or not involves excessive research burden, essentially to the inherent lack of physical boundaries. Obviously the article starts off with mentioning the patent dispute between RIM and NTL."
Re:QED (Score:5, Interesting)
Its a good example of applying an obvious concept to a new application. Every TV set ever made uses an XOR to add and erase the HSync signal in the VSync, but that isn't drawing a line, so it doesn't count as prior art. Thank you, USPTO.
Re:Investment, risk, compensation (Score:5, Interesting)
Well said, sir.
I think the fence analogy is rather confusing, however. (No offense to Jefferson, of course.) I prefer to draw the line between ideas and implimentations. Without this distinction, there would be no such thing as a competing product.
For example, let's say you think of a novel new way to wake people up in the morning; something that's never been marketed before, like pouring water on your head. So you take this idea and distill it down into an implimentation called the AlarmWaker3000, which you patent. Now your good friend down the street has this same idea, and creates himself the WakeUp Machine, which he patents. But because the AlarmWaker3000 used a bucket and a piece of string tied to the hands of a clock, and the WakeUp Machine uses a hose crimped by a weight that gets lighter over time, neither one of you are infringing on each other's patents. You didn't patent "A method for waking someone up by splashing their face with cold water," you patented the "AlarmWaker3000" and the "WakeUp Machine".
The question of infringement on a patent usually boils down to the similarity between the two devices. Two companies can build, and patent, two different motherboard designs with exactly the same functionality: onboard video, sound, ethernet, whatever. But as long as the circuits are dissimilar enough (which is up to a judge to decide) then the implimentations are different and they don't infringe.
The problem with software patents is that the the line between ideas and implimentations has been blurred. Amazon patents their code for purchasing things with a single mouse click, once all your information is on file and you are logged in. Suddenly, anyone creating a shopping system that allows registered customers who are logged in to purchase things with a single mouse click is infringing on the patent, even if their implimentation is different. These companies are trying to use the patent system to enforce artifical monopolies on ideas instead of implimentations, and effectively cut out the competition. The truly troubling part is that the Patent Office, and lots of patent courts settling disputes around the country, seem to be going along with it.
Maybe the fault lies with an overworked Patent Office staff; maybe it lies with a culture that irrationally rewards new technologies. I just don't know.
Re:Investment, risk, compensation (Score:4, Interesting)
Actually, I don't like the fence analogy either and it wasn't 'exactly' the model Jefferson used.
The true test is whether the research can be transmitted by speech/print.
k=1/2mv^2, E=mc^2, "Now is the winter of our discontent. .
If you can recite the research than it falls under those laws intended to protect speech, not things. As you say, ideas as opposed to implimentations.
It is, unfortunately, computers that have blurred the line between the two with the decision that since the physical device impliments the idea that it is somehow itself the device.
This is the sort of bullcrap we get when people cannot differentiate between the physical and logical levels. At its extreme I have seen plans for a piece of furniture that has no patent protection on it try to apply EULA copyright restrictions to your ablility to impliment the plans and claim a license fee for every instance of the furniture you build.
Oh, yeah, and a website whose legal page actually begins with "By viewing you agree to these policies . .
It's gone totally crazy out there.
KFG
Physical patents are by far more evil (Score:4, Interesting)
This is becuase software patents are about controlling information, but hardware patnets are about physical controll. To controll information often requires BS and deception, but to controll physically often requires physical coercion and violence.
Just as the false property of slavery was destined to end in a violent civil war as those who "owned" slaves lost controll, physical patents brought to their logical conclusion will result in the violent death of billions as society enters into the replication age and physical creation becomes more and more imposible to controll for the sake of monopolizing profit. In some ways we are already seeing a warm up. Millions in africa die of AIDS because access to generics are forbidden, millions of elderly are strongly pressured into using drugs which may lock in markets, but have all sorts of strange hidden side effects because the natural alternatives can't be patented. Safety devices on cars don't get installed causing 1000s of unneded deaths per year because other auto companies own the patent and won't let it be used. Billions and billions of incompatable parts and pieces to appliances that have no need to be incompatable except for patents. (and billions of uneeded enviromental waste because of it). The examples go on and on...
Fencing (Score:2, Interesting)
While TFA uses the term "fence" to denote clear, definable boundaries, there's another interpretation that helps explain the essence of patent law:
The underlying purpose of the patent laws is to encourage the sharing of useful ideas. In order to get a patent you must publicly disclose how your invention works, and the "best mode" for using it (i.e. the best way to take advantage of its benefits).
As you can imagine, normally these are just the kind of disclosures that a profit-seeking inventor hates to make. The patent laws are thus designed to encourage more social, knowledge-sharing behavior on the part of even these self-serving entities. Note that the law is NOT about "protecting" or "rewarding" inventors per se, except as a means to this greater end.
Back to fences: Imagine that a company comes up with a new invention. The patent law gives them an alternative to "fencing it in" (by keeping it secret) and thereby preventing the public from learning about the discovery.
But note what this implies: if it's impossible to "fence in" the technology (i.e. prevent the spread of knowledge about how it works), then there's little reason for patent law to apply; The public would likely find out anyway, so why reward the inventor for disclosing? If knowledge is easily discoverable through some reverse engineering or simply by using a small amount of observation/experimentation, then the principled argument for patentability is weaker.
On the other hand, if an inventor could successfully use a new invention without the public ever learning how it worked (i.e. if it were possible to "build a fence" around the knowledge), then there's a good case to be made for offering the incentive for disclosure.
Just a different way to think about the "fence," that's all...
a plague (Score:3, Interesting)
The coming patent Apocalypse (Score:3, Interesting)
I stronly suspect that no professional programmer has not infringed on US software patents. I occasionally do a keyword search for "computer" on the USPTO to see the latest tech patents, and I'm always appalled by what comes up.
It's not that they even always violate the rules of the USPTO -- sure, for a lot of them, there's prior art. It's that they are *never* clever ideas that another person wouldn't immediately think of if presented with the same problem. This simply does not drive technical advancement.
What I'm really scared about is the upcoming patent Apocalypse.
For the past decade or so, the rate of granting tech patents has *vastly* accelerated. There are now a huge body of tech patents out there.
Thus far, we've only had a few problems with tech patents -- usually the ones that are getting long in the tooth, about to expire, and ones that the holders feel that they have to hurry up and make money on. Kind of like the GIF/LZW patent from Unisys.
Towards the end of a patent's life, whoever is holding the patent -- maybe an inventor, maybe a patent troll, maybe a company in financial trouble (a la SCO) knows that they are in a "use it or lose it" situation. It's just that there's about a fifteen year delay from the time that the patent is granted to the time that it gets really urgent to start litigating on it.
We've talked about patent reform, but no matter what happens, legislators will never, ever invalidate existing patents. To do so would produce business chaos -- business decisions were made based on the value of those patents, and there would be deep concern from companies if that IP value vanished.
Even if I turn on my computer tomorrow and discover that, wonderfully, Congress has officially banned the granting of software patents, there is still a decade's worth of glut of software patents out there.
What happens in another five years when software patent holders start warily eyeing the expiration date on their patents, wondering when they're going to make a return? Oh, sure, maybe IBM isn't going to go out and start suing people left and right, but they can easily sell their patents off to a patent troll. That way, they get a flat return on their patent and don't suffer any PR damage. Patent trolls don't give a damn about PR, because they aren't in a business where PR helps them in the least.
That's what worries me. And no matter how bad the situation gets, there isn't a whole lot that Congress can do. They can't reasonably do anything about *existing* patents. And there isn't much that the industry can do to work around the problem. Sure, they can ship software development jobs overseas to developers that aren't hamstrung by US software patents...but if you want to sell your finished product to the lucrative US software market, you *still* have to abide by the patents. This affects everyone, because just about every software development company out there depends on at least some US sales.
The problem isn't even just patent trolls. Given their recent exploits against Microsoft and similar folks, I'm pretty sure that IBM/MSFT/etc are more than happy to push for legislation that makes life miserable for trolls. But they sure as hell don't want to stop the stream of patents that *they* are acquiring. And there are *plenty* of bullshit patents going to both of those folks.
The really doubly frusterating thing is that if you're a researcher, a PhD that's gone into industry, almost always *have* to file for patents. It's a metric of "how well you're doing" in a field where it's *really* goddamn hard for your superiors to figure out how effective you are.
You've seen academics that have hundreds of papers with some authorship credit. Very, very few people have hundreds of papers worth of importa
Re:Investment, risk, compensation (Score:3, Interesting)
Copyright is all about protecting particular expressions of ideas. Patent law is all about actually protecting ideas.
You can't obtain a copyright in, for instance, the idea of a certain plot, or a certain genre of song, or any other such 'idea.' Copyrighted works must be fixed, and must be a particular expression of such ideas, for instance a particular story which uses a plot, or a particular song which may belong to a genre or pattern...
Patents are most emphatically not about protecting such particular expressions, or particular devices, as you state. Patents are about protecting the IDEA of the invention, and they always have been. Your example of the method of waking someone up with cold water could be a patentable invention, if it had never been done before, and was not obvious based on what had come before...
The particular devices are just what people don't want to patent; nor would it be useful to do so, since inventions are not like books - inventions can be used in millions of different forms, and still serve their purpose perfectly. The devices that embody inventions could be so varied that protection for only a particular one would be basically useless.
This is why the claims of a patent are not supposed to be limited by the embodiments described in the written description portion of the patent, except in special cases. The claims determine the invention, not the particular embodiments that the inventor describes in the specification.
For example, if I claim a chair, by claiming something like a sitting surface with at least 3 legs, etc., I am the inventor of THE CHAIR. It's an idea, not a particular type or style of chair. It's a completely new sort of thing. I will describe a couple of different ways to make chairs in the specification of the patent, to show how it could be accomplished, but these are not what is covered by the patent. What is covered by the patent is anything encompassed by the language of the claim.