US Judge strikes out COMPUTER/HUMAN LOVE patent An American judge has delivered a comprehensive smack-down to a patent troll, invalidating a patent that tried to cover the practise of matchmaking simply because a computer was involved. In this judgement, District Judge Denise Cote decided that matchmaking, claimed by a company called Lumen View Technology, simply isn't …
“Merely directing a computer to perform a function does not transform the computer into a specialized computer. Such a principle would lead to the absurd result of allowing the patenting the computerized use of even the most basic abstract ideas,”
Now how about applying that logic to all software patents?
Steve, in the US "software" alone cannot be patented. However, software that is an intrinsic part of some hardware - like the software which is part of an engine management system system or the way the image on a mobile phone screen is displayed as a user fingers it. The EU also permits this type of software patenting.
Where the US differs from most of the rest of the world is that it permits software to be patented that is part of a "business process". Then you begin to get the contentious patents appearing such as the one-click patent issued to Amazon for buying products on a web site. If a patent lawyer is successfully able to argue that some software supports a business process then the patent is likely to be granted.
Clearly Judge Cote averred with respect to matchmaking. And good for her. It's not often that common sense prevails in such cases.
"Such a principle would lead to the absurd result of allowing the patenting the computerized use of even the most basic abstract ideas,” Judge Cote added."
I agree, absurd. But no more absurd than the idea of patenting the 'shopping cart'. Sure that was overturned but not before millions in settlements and, at any rate, Amazon still have their 'one-click' patent, which is little better. Most telling is the case of CLS Bank suing to invalidate Alice Corp's patents relating to settlement risk. Essentially, It was, just like this:
Existing concept + computers = magical, new, patentable concept. (Apparently)
The court invalidated the patent but the interesting part is in the conflicting opinions given by the justices. This from 4 judges:
". . .a computer programmed to perform a specific function is a new machine with individualized circuitry created and used by the operation of the software."
And:
". . . a general purpose computer in effect becomes a special purpose computer once it is programmed to perform particular functions pursuant to instructions from program software," on the theory that software "effectively rewires a computer."
Hmmmm . . .
I suppose the question is, if the very notion of such a patent as this is absurd then why are they being granted in the first place? The 'man on the street' could tell you it's not an overly strong case and judges seem to agree so why is there a disconnect with those issuing these patents?
Let's ignore the fact the patent is for software. The patent is far too broad to be defensible. Most successfully defended/enforced patents have an extremely narrow focus. It is a terrible misunderstanding of the value of a patent that leads to these overly broad things. Had the patent narrowly defined a specific 'thing' to protect it would probably not have been overturned.
A patent for an entire system is very rare. A patent for part of a system is normal and that's what people don't understand about prior art. In this example they were basically trying to define any computer based characteristics matching as their innovation. That's just dumb and has prior art all over it. If however their patent was for using a computer to find appropriate matches based on say, systolic blood pressure, that probably could get a defensible patent. The previous existence of computer based matching isn't patentable, but individual aspects of computer based matchmaking are patentable.
This post has been deleted by its author
The core problem (such as I understand it, which is imperfectly) is the prevalence of the opinion, as penned by Rader J, which I quoted, above, essentially stating that a general-purpose computer becomes a new, specialised, machine when paired with software.
This opinion then works with the "Machine-or-transformation" test of patent eligibility, and would lead one to believe that such a computer would qualify as a "particular machine or apparatus".
Thus, a business process can effectively be made eligible for protection by patent by writing software to carry this out.
And this is pretty much what Rader says in his opinion:
"All that is required is that the claims tie an otherwise abstract idea to a specific way of doing something with a computer, or a specific computer for doing something."
The problem, as I see it, is that it should be the "specific computer for doing something" that should be eligible for the patent - if anything is - and not the abstract idea. Such an idea does not magically become less abstract because it is implemented any more than gravity becomes becomes patentable because someone builds a see-saw..
And it's here that these things should be self-defeating. If a computer coupled with software does not become a "particular machine" for the purpose of the "machine-or-transformation" test, then the patent application/defence is greatly weakened. If, on the other hand, a computer + software does "create(s) a new machine", as Rader asserts, strengthening the claim for eligibility, then surely reason would hold that any change to the software would create an entirely new machine.
I.e., a different software package, even though designed to do the same thing, would create an entirely "new machine" and thus avoid infringing the patent.
As you can see if you read through the opinions, this issue is a long way from resolved but it seems judges are at least looking closely at this grey-area now, which is, tentatively, a good thing.
So, just like Apple, "Existing Concept + Smartphone = magical, new patentable concept (apparently)"
So using the same logic, Rounded Corners + Smartphone = completely UNPATENTABLE CONCEPT
(Because people have been "rounding the corners off objects" for centuries before Apple)
"True, but this ruling is more important than that. A judge stating that "Old idea + computer != patentable idea" could be a good first step for fixing many similar patent cases, and perhaps even to fix the USPTO*."
Perhaps the end of "Slide to Unlock"? The number of stupid patents coving the obvious + computer is growing.
all* computer programs are "computer implementations" of relatively simple abstract ideas
Really? You seem to have a pretty broad definition of "relatively simple". Relative to what?
I'm curious to know what's so simple about, say, graph spectral sparsification. Or what "relatively simple abstract idea" is embodied by, oh, the Linux kernel.
Prolepsis: "Those are collections of many simple ideas" isn't a persuasive rejoinder, since the act of combining simple ideas in particular ways adds information entropy.
There's an enormous misunderstanding among the general public of what patents are. A patent is absolutely, in no way proof that your idea/thing is original. It is proof that the reviewing patent agent could not find, either in prior art on file with the USPTO or in his approved list of sources and his knowledge of the industry, an idea/thing that does what your idea/thing does in the same manner.
A patent allows you to legally pursue someone infringing on your patent, but you still have to defend it and show that the patent is truly novel and unique. It doesn't mean that you as the patent holder are automatically in the right. The validity of the patent will be determined in court by a judge.
All that to say, the USPTO is in no way, nor has it ever been the arbiter of uniqueness. That role has always been the role of the court. The job of the USPTO is to compare your idea with ideas inside a narrow data set and ensure its uniqueness there, not its uniqueness globally or even in another industry. If your idea is found to be unique within those confines you are awarded a patent.
It is then up to you to sue those who you believe to be infringing. It is up to the defendant to prove your idea was not worthy of a patent and a judge to weigh both sides and decide who is right. A patent is the key that allows you to start the process of enforcing the patent, but the patent itself must first be validated.
TL;DR - you are correct, but that doesn't mean it's not a chronically broken system.
True, but such legal defences cost money. A lot of money. Even if a defendant wins a legal proceeding, that is rarely the end as there are often appeals, further sapping time and money. That is why only about 5% of such cases ever get to trial.
A patent does indeed allow "you to legally pursue someone infringing on your patent", as you say. The unfortunate part is that, by-and-large, that is all that is required to start pursing someone.
Simply owning a patent allows you to send threatening letters demanding settlement or cessation, which is the way most patent actions seem to proceed. A patent may not be truly valid until ruled on by a court but the problem is that such a ruling does not have to occur before sending demands to all and sundry.
In essence, the job - and expense (greatly inflated) - of rigorously examining a patent claim is left to the defendant, rather than the body tasked with assessing the application in the first place (the USPTO) or the party asserting that their invention is both unique and eligible for protection by a patent (the patent applicant).
For someone being targeted by a patent claim, the first step is to engage specialists to review the patent and assess its validity. Then they must do an infringement assessment, analysing the allegedly infringing device/software/concept/shape/colour/facial-expression to see if it does actually infringe upon the patent.
Depending on the patent and product, that process could well cost over $10K* - not an insignificant sum for a small business or start-up. If you don't intend to capitulate straight-away, that process is necessary as without it you could be found to be 'wilfully infringing', vastly inflating any potential settlement.
It's worth noting that, as there is no legal framework around making such demands, the patent holder doesn't need to specify what action of the product infringes or which claim in the patent is being infringed. That only needs to happen once litigation has been started and the alleged infringer is already knee-deep in legal bills.
So a patent holder can, at minimal cost, and with no legal requirements, send your company a letter demanding licensing payment for some alleged infringement. To even assess the validity of the threat, you have to fork-over ~$10K.
I accept that the current state is that court is where patents are truly validated, but that arrangement is broken as it allows, and has lead to, a great proliferation of entities extorting money from anyone and everyone. The small companies just can't afford to take it to the courts and the larger companies that could afford to test the validity of a claim in court overwhelmingly appear to choose not to because someone has calculated that it is not financially worth it and that settling is cheap. 95% of claims end without the patent ever being tested in court.
So, while patents must be tested in court to properly determine their validity, that, in practice seldom happens and the result is that patents are largely assumed to be valid and treated as if they are.
To be clear, I'm not suggesting you actually agree with the current system.
If I remember correctly you hold one or more industrial/engineering process patents and so are far better versed in the system than many others here* and your input is therefore of somewhat greater value (and interest) than my own!
For my part, I believe that patents are important and that they do provide valuable incentive for people to invent. I also believe they provide a very strong disincentive as well, as anyone trying to bring a new idea to market will likely find they infringe upon multiple process and/or technology patents, meaning they will be subject to legal manoeuvrings, either in setting up licensing agreements, at the civil end, to cease-and-desist demands, through to aggressive legal action.
It's my assessment, as an external observer, that the patent system, in its current state, provides benefits to larger companies to invest in research while adding barriers to smaller groups or individuals developing new products and technologies. Without the protections of a broad patent portfolio, such smaller entities are at risk of being litigated into oblivion by those larger companies who wish to either exploit 'easy' targets or destroy competitors.
Of course, patents and the legal processes for suing against infringement can and do protect individuals and start-ups from having their inventions stolen by competitors and provide compensation and reward for their work. It's just that the way things have become, there are innumerable patents out there that should not be worth the proverbial paper they're printed on but yet are used to extract millions in settlements and damages, and in doing so, act as a disincentive to innovation; the opposite aim of the patent system.
* - With the exception of @jake, who likely continues to make a handsome living from software patents licensed to Google and firmware patents licensed to Intel and ARM.
This post has been deleted by its author
It seems to me that a key plank of the fight against patent trolls needs to be laid at the Patent Office. Currently it has no incentve to stop issuing questionable patents, no doubt reasoning that it minimises its own workload and that interested parties will just fight it out in court. Maybe if it was made to fund the defendant's legal costs it would concentrate minds a bit better.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
