Slashdot Mirror
Judges Debate Patents and If New Software Makes a Computer a "New Machine"
First time accepted submitter ectoman writes "A third party steps into a financial transaction to make sure all parties exchange funds at the same time and as expected. Can you patent this process? What if the third party is a computer? Rob Tiller, vice president and general counsel for Red Hat, details a recent court ruling on this very matter—one that has critical implications for the future of software patents, and one that divided the judges involved. Tiller writes that: 'The judges mostly agreed that the idea of managing settlement risk with a third party was abstract such that by itself it could not be patented. They differed, though, on whether using a general purpose computer for managing settlement risk meant that the patents avoided invalidity based on abstraction.' Interestingly, some judges suggested that a computer becomes a 'new machine' every time it loads different software."
1.Make it so that every time you load a new piece of software you invalidate the license of every other piece of software in an endless recursive loop.
2.????
3.Profit
4.goto 1
If we can slap 'on a computer' onto an old idea to patent it, can we slap on another 'on a computer' to patent it for virtual machines?
As much as I hate trivial patents in any field, I must admit that when I wipe windows off a pc and put debian on it, it sure feels like a new machine...
---- MISSING MISCELLANEOUS DATA SEGMENT --- [sigdash] trolololol
The computer is a new machine many times every picosecond as the data in the registers change, the data in ram changes, the data on the hard drive changes, the data flowing through network interfaces changes, etc...
Since the computer is *new* billions, if not trillions of times a second, then software doesn't make it unique.
Yet again, the clueless making decisions on things they cannot comprehend.
Is that they do not have to understand the technical stuff. Patent Judges are the reason for the problem. If they were really qualified, they would not have a low-paying federal job.
You are thinking of a Patent Examiner, not a "judge". These are two very different governmental positions.
I know I'll be flamed for this because everyone here is against software patents, but...
The distinction between a physical machine and a physical machine running software is somewhat pointless. Almost any software can be converted into a physical machine. Using a hardware H.264 encoder/decoder or a hardware crypto card is really no different from doing those things in software. If you want to make a distinction between physical objects and software for patent purposes, the logic part of computers needs to be considered software despite the fact that it's been manifested as a physical object. But general purpose computers running software aren't any different than purpose-built computers executing the same logic in hardware.
As long as they run a different OS we can get around the patent? Time to add some ideas to the public domain.
Someone want to make a website that scrapes through software patents on http://www.google.com/patents and adds ' on a linux based computer' to the end of them, and then reposts them as public domain?
What the judges need to understand is that fundamentally, people are general purpose machines. Without any training at all, people are not capable of solving many problems. However, people can solve all sorts of problems based on what you teach them. Sure, you aren't going to be able to get a bushman in the Serengeti to be able to build the Apollo but, the potential is there given enough training. If that same bushman were to go to school and eventually graduate from college, they would be just as capable as anybody else. So how is receiving training through college or a technical school any different than a computer receiving a new set of instructions on how to solve a particular problem?
Computers only run algorithms (which aren't supposed to be patentable). They follow a set of instructions step by step and can't do anything that you and I can't do with a pencil and piece of paper. (They just can do it a lot faster.)
Of course, the beauty of computers is that they only do what you tell them. When they don't do what you want, it is still your fault since you didn't provide the instructions you thought you did. ;-)
They are, of course, right. An algorithm is conceptually the same as a description of a machine for executing that algorithm, whether you draw logic gates to execute your algorithm, or solder together transistors, or write Python.
As a mathematician working on algorithms right now, I say they are right.
However.
Algorithms shouldn't be patentable, no matter how they are represented. But that's a political decision, not a technical one.
They are both the same in functionality and somewhat in implementation, only hardware is faster, more parallel but limited in functionality. Whereas software is slower, less parallel but has vastly more functionality than hardware.
35 USC 101 requires that a claimed invention be directed to patent eligible subject matter: a process, machine, article of manufacture, or composition of matter (it also requires that the invention is useful). The courts have decided that these categories are very broad, but don't include "abstract ideas" (though that term is never defined), laws of nature, or natural phenomena.
If the claimed invention passes that low threshold for 101, 35 USC 102 requires that the invention must be new or novel. That's a higher bar, but not a huge one - if I'm the first person to make a red car and claim that in a patent application, that's new, even if blue cars existed.
If the claimed invention passes that threshold, then 35 USC 103 requires that the invention must be nonobvious. The red car is obvious if blue cars existed, even if no one has ever made a red car before.
So, for example, if some piece of software causes a computer to paint the screen in red paisley and that's never been done before, it's new... but it's obvious and still patent ineligible.
The problem is when these get confused or conflated into a single requirement, because obviousness and novelty require evidence, while subject matter eligibility does not. And so, you get cases like CLS or Bilski where the judges want to invalidate the patent because it's stupidly obvious, but they have no evidence on the record... so they declare it an abstract idea and invalid. In particular, here, the judges started carving out everything from the patent claim that made it non-abstract, declaring it irrelevant, until the only thing left was abstract. The outcome may be the right one, but it's for the wrong reason - it's like finding a murderer guilty because you hate his face. Maybe he was actually the murderer, but you're finding him guilty for the wrong reason.
Someone needs to slap the cowards in Congress to clarify this w.r.t. a "general-purpose machine".
Merely shoving some algorithm that is done by hand onto a computer is nothing novel. This isn't to say a particularly clever and non-intuitive software implementation couldn't be patented. But just doing it in software is not novel; it is obvious.
Software is indeed a virtual -- and specialized -- machine -- that is the idea behind computers as "generel-purpose machines". But not in the legal sense driving patents: novel and non-obvious innovation.
I keep recommending these rules:
1. If it's already being done in the real world, doing it on a computer is not patentable per se.
2. Doing a simulation of a real-world item is similarly not patentable per se.
3. Doing something wirelessly formerly done over a network, or remotely formerly done locally, or on a lil' phone or tablet or tricorder, is also not patentable per se.
4. This is not to say particularly clever implementations (the "machine" part of "virtual machine") could not be patented.
There, follow those rules, cowardly Congress, and you protect patentable innovation while eviscerating a ton of current patent problems.
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
Reading this story makes me feel like a new man. Those bills? You'll have to track down the old guy if you want them paid.
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
I completely disagree that loading a computer with new software makes it a new machine. The machine "part" of the computer is the hardware not the software, the software is the operator. So reloading a computer with new software will change the operational purpose of the machine but not the machine itself. To put this another way, if I have a conveyor belt that just moves objects around, like a computer moves data around, then it doesn't matter who is operating the machine it will always be model xxy, I might change the operator many times in the life of the machine but that doesn't change the machine itself.
Now lets be clear that I'm not talking about a FPGA! In a FPGA if you "reload" the software you will change the machine because in essence the "software" for a FPGA is like the internal gear system, you can configure it in 1000 different ways to do different tasks. One day it can be a conveyor build and the next it will be a bottle cap remover. So before someone comes in here and blows a gasket at me, I fully accept the fact that a FPGA is very different then a Computer.
So then, since the Jacquard loom used a different set of cards to 'program' each different cloth pattern, it qualifies for millions of separate patents.
We're talking about technology almost 300 years old. I'm thinking some of those judges are not only technically inept, but are in fact Luddites in black robes.
But, does teaching your secretary a new way of taking notes make her a new secretary?
If so, does she have a new 3 month probation?
A different program DOES make it a different machine, and I'll explain why that must be true. That's not enough to defend an overly general patent, though.
However, it seems to me that's the wrong question to ask in this case. It seems that someone wants to patent the concept of "a machine to do X" rather than patenting one SPECIFIC machine that does X. You can't patent "a machine to take people from place to place". You'd have to invent and patent some specific new transportation machine. So the patent would be invalid on that ground. The patentors claim that "a computer machine to do X" is specific enough to be patentable, and most people disagree - you've got to patent some specific new invention, and "on a computer" isn't new, or specific.
As to the question in the summary, ANY electronic machine* can be implemented is software, and any software machine can be easily rendered as a hardware
machine. Converting between hardware circuits and software instructions is a trivial bit of arithmetic, so the exact same machine can be flashed into memory and
and we'd call it software, or it can be flashed into a gate array and we'd cal it hardware. Therefore, if any electronic device is a device, a software device must also be a device. (Because the exact same functions can be flashed as hardware or software, it's still a device or not either way.)
Gate arrays show that one cannot really distinguish between hardware and software programs. With the original gate arrays, the factory would take some stock
electronic logic parts and add layers of metal to implement a specific product. This isn't much different from soldering components together. It's definitely hardware. The cool thing was, the metal connections would be added based on a precise written description of of how the device should behave - a PROGRAM. In the next generation, the parts started out with fused connections between ALL of the parts and the factory would blow most of the fuses, leaving only the desired connections. Again, the list of which 10,000 fuses to blow was generated from written code, code that is pretty much a software program, a software definition of the hardware. Still, hardware produced according to a software program. In the third generation, the CUSTOMER could blow or reset fuses to reconfigure the electronics in the field, "re-wiring" the hardware. That generation is the FPGA, field PROGRAMMABLE gate array. Hmm, the customer can load different programs, that sure sounds like software programs. Those software programs are implemented as physical connections, so that sure sounds like hardware. It's both software and hardware, showing that in the end, codes that define how a machine operates really are both hardware and software - they can be implemented as either or as both at the same time.
So yes, software can control the other parts of the machine to act differently, making it a different machine. But only specific, newly invented machines are supposed to be patentable, regardless of whether the newly invented machine is rendered as hardware or as software on any given day.
If loading software turns a general-purpose computer into a new machine, the patent for that "new machine" should include all the internal details to make it work, particlularly the source code. Then if somebody else implements the same concept with different source code and different algorithms, they made a different machine so it's not infringing.
---------
There is inferior bacteria on the interior of your posterior.
Uh, lets see. Computers are pretty much a practical implementation of a Universal Turing Machine which Turing proved could compute any computable problem. Therefore a computer is, in effect, a universal machine. Since changing the software on what is already a universal machine doesn't really change the fact that it is still a universal machine then it seems that no software for any computer should be patentable.
I am a scientist. I call into question the assumption that, "Patents are Beneficial for Society as a Whole". This is an untested hypothesis. Until we have PROOF that patents are beneficial for our society, they should not be allowed. What if patents are irreparably harming the innovation of mankind?! You have NO PROOF to show whether they are or are not harmful or helpful. There is only conjecture and assumptions. Prove they are beneficial, until then abolish them.
If I hand you a gun and tell you to put it to your head and pull the trigger, It's beneficial for you. Would you just assume I'm telling the truth? A rational being would examine the gun first to ensure it was not actually harmful... Some would test fire it to ensure it was not harmful. The most rational would require proof that it was beneficial before even touching the damn thing.
We're slowly blowing our culture's brains out with the legal weapons called patent and copyright. I'm a scientist, so if you think otherwise, then I have two words for you: PROVE IT. As it stands my statement that patents are harmful is just as valid a hypothesis as that they are beneficial. It's unconscionably reckless to continue subjecting the world's economy and culture to this potential danger.
Wait, I say there's no proof either way, but there is evidence that patents are not required at all. Neither the automotive or fashion industries are allowed copyright or patents for their designs yet look how innovative they are in design; Indeed often design is their core selling point. That's two real world data points indicating patents are not necessary. I've read many responses that say, "Without patents businesses wouldn't fund innovation." These statements are equally bogus because this statement: "Without patents ______," is bogus. We don't the-hell know what it would be like until we run the damn test. The potential risk associated with not performing the experiment should be enough for any rational being to give pause.
Abolish patents. It's the only rational thing to do. The hypothesis must be tested. Only then will the argument for or against them have any leg to stand on. We can re-institute any laws we want once we're sure they're not harmful. If the scientific method is good enough to develop things that could be patentable, then why turn a blind eye and ignore the fact that no one has applied the scientific method to the patent system itself?