Supreme Court Rejects Software Patents On Performing Generic Functions; Pretends That Lots Of Other Software Must Be Patentable
from the a-step-in-the-right-direction dept
For a few years now we've been covering a key software patent case, Alice v. CLS Bank, which was another chance to show that pure software patents shouldn't be granted. As you may recall, four years ago, the Supreme Court got to tackle the question of software patents in the Bilski case, but chose to punt instead, rejecting that particular patent, and arguing that the specific test that everyone relied on shouldn't be the only test -- but otherwise leaving a lot of confusion in its wake. It did help dump a few software patents, but left the wider question pretty open.We had hoped that when the Supreme Court agreed to hear the Alice case it wouldn't miss another chance to actually add some clarity to what is and what is not patentable. It seemed like the perfect opportunity. As you may recall, the original appeals court (CAFC) ruling was a complete disaster, with 135 pages of different opinions -- with only one single paragraph having a majority view, rejecting the specific patent. But no one could agree on why or the larger questions.
It was as if CAFC were practically begging the Supreme Court to provide clarity and guidance.
Unfortunately, the Supreme Court didn't really do that. It technically "upheld" the CAFC ruling (that one paragraph) rejecting the patent (which basically covered a computerized escrow service) as unpatentable subject matter. It further makes clear that merely taking an abstract idea and doing it "on a computer" doesn't make it patentable. That's all good... But, while three Justices (Sotomayor, Ginsburg and Breyer) hoped the court would go further and basically say that business methods weren't patentable at all, the rest simply wouldn't go along with that, saying that "many computer-implemented claims are formally addressed to patent-eligible subject matter," but never giving any examples.
Instead, it notes that you can't get a patent if each step claimed in the patent "does no more than require a generic computer to perform generic computer functions." Except, uh, many people will point out that's all that software does. That's basically how software works, but the Justices don't seem to recognize that. So, it's a bit of a conundrum. The court says many software patents are perfectly good because they apply to patent-eligible subject matter, but that if the claims do no more than require a generic computer to perform generic computer functions, it's not patentable. You could read that to mean that basically most software patents are no longer allowed, but... that's going to involve an awful lot of wasted litigation to teach a bunch of courts, including the Supreme Court, that basically all software involves generic computer functions.
Part of the problem is that, like many non-technical people, many of the Justice seems to think that software is a lot more than it really is. They seem to think that there's some magic in software that goes beyond just a bunch of instructions for a computer to follow. So, now they're saying that just taking some ideas and telling a computer to follow instructions to implement that idea is not patentable... but they still argue that there's plenty of software that is patentable. So it's... still really punting on the issue, in part because the Justices don't seem to understand software.
The court relies a lot on two other big recent rulings which we've covered -- the Mayo ruling that rejected medical diagnostic patents and the Myriad Genetics ruling that struck down gene patents, but doesn't quite go as far with software and business method patents. Instead, it sorta half rejects software patents, kinda, without going as far as it needs to go. As some folks are pointing out, the language used in the ruling is "going to tie folks in knots" as they try to figure out what it means.
In the long run, this may be a very important ruling. It's easy to read this ruling to basically reject a very large number of software patents. But, because of the unfortunately all-too-common nature of the Supreme Court semi-punting on clear decisions on this particular issue, it's not entirely clear where this ends up, meaning that there's going to be a lot of patent litigation citing this ruling, with both sides seeking to tap dance around the language choices. And that just means a few years down the road, it's quite likely that we'll be back here again, with the Supreme Court asked to decide, once again, whether or not software and business methods are really patentable.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: business model patents, cafc, generic functions, patentable subject matter, patents, software patents, supreme court
Companies: alice, cls bank
Reader Comments
Subscribe: RSS
View by: Time | Thread
See ya, One Click patent. You had your 15 minutes of fame.
/tongue in cheek
[ link to this | view in chronology ]
Still good to have
[ link to this | view in chronology ]
Re: Still good to have
[ link to this | view in chronology ]
Re: Re: Still good to have
They are unclear and leave room open. They don't say specifically that a computer with attached screen can or can't ever be patentable based on that behavior/screen contents. But they leave some options open, including referencing Diehr (which used computer logic in an "industrial" machine). I think they might accept new algorithms/apps to some extent.
What they did do in part was to eliminated for near certainty the taking of descriptions that don't impress anyone (obvious to a PHOSITA etc) and adding a computer into the process via computer programming.
BTW, I think being non-obvious to a person having ordinary skill in the art is an extremely low standard that would almost surely not result in advancing the progress as anticipated in the Constitution. I hope SCOTUS addresses this issue more carefully some day.
[ link to this | view in chronology ]
Re: Still good to have
One side bases their claim on software being unpatentable because it's just giving generic instructions to a generic computer. The other side claims their patent is valid because the patent office wouldn't have granted it if it were invalid, and cites the SCOTUS decision because their ideas were original and creative, not generic.
And under this ruling, given that SCOTUS made two mutually exclusive arguments therein, BOTH sides of the future court case would be 100% correct about what the law says.
The problem with punting is that sometimes it gets returned, and you have to run it in anyway -- which you should have done all along.
[ link to this | view in chronology ]
[ link to this | view in chronology ]
Re:
I agree.
While it may not kill software patents completely, this seems to kill the idea of taking generic analog processes and filing a patent by just adding "on a computer."
Perhaps it's a minor distinction, but I'm not going to deny the possibility that there are just some things that CAN'T be done outside of software.
[ link to this | view in chronology ]
Generic...
I read that to mean that generic software running on a generic computer is not protected but software created for non-generic, for example, specifically created hardware for machines with very customized demands, would be protected. I think the ruling is fine in dismissing most things I run in windows and protecting software written for a factories embedded systems.
[ link to this | view in chronology ]
Re: Generic...
[ link to this | view in chronology ]
Re: Re: Generic...
Groping their way towards distinguishing ASICs …from CPLDs and FPGAs …from x64 and ARM CPUs …from GPUs ?
Or groping their way towards distinguishing Verilog / VHDL from C ?
[ link to this | view in chronology ]
Re: Generic...
If the software is only the 'computer translation' of a process or algorithm that, in itself - outside a computer application, is not eligible for patenting, the 'with a computer' addition cannot make it patentable.
More importantly, anything that was already patented cannot be patented again because of simply adding "with a computer" to the claim!
With this I hope that the ruling - finally - understands that doing something 'on a computer' is not innovative in itself.
In other words:
As anything patentable, the software-patent-claim must be a truly innovative creation, not just a translation into software of something that is in itself not patentable.
[ link to this | view in chronology ]
Poorly written article
[ link to this | view in chronology ]
Re: Poorly written article
However, since I read your comment and it so clearly framed all sides and left nothing out, I am unable to add anything helpful to the conversation.
You have saved me a great deal of time.
Thank you again.
[ link to this | view in chronology ]
Re: Re: Poorly written article
This article promotes the position that software does no more than require a generic computer to perform generic computer functions. The reader is encouraged to read the opinion with this predisposition and dismiss nine Justices as technologically impaired. It totally ignores the discussion of how the patent at issue fails to fulfill the "add more" requirement
Then the article says this "They seem to think that there's some magic in software that goes beyond just a bunch of instructions for a computer to follow." This totally misses the "add more" discussion again. It also dismisses the analysis of what software actually does, the "intermediary" discussion.
[ link to this | view in chronology ]
Re: Re: Re: Poorly written article
[ link to this | view in chronology ]
Generic computing [was Re: Re: Re: Re: Poorly written article]
Going way off topic —but— you did ask.
“Intel Mates FPGA With Future Xeon Server Chip”, by Timothy Prickett Morgan, Enterprise Tech, June 18, 2014
The line between hardware and software has grown increasingly blurry over the past couple decades. No doubt that trend will continue.
But, for this decade, I think Intel is announcing a non-generic computing platform—and it'll run non-generic software.
[ link to this | view in chronology ]
Re: Generic computing [was Re: Re: Re: Re: Poorly written article]
"Intel is taking [hardware] seriously as a means of accelerating [software] and has crafted a hybrid [piece of hardware] that marries a [piece of hardware] to a [piece of hardware] and puts them in the same processor socket."
I don't see where the line is getting blurry.
[ link to this | view in chronology ]
Re: Re: Generic computing [was Re: Re: Re: Re: Poorly written article]
Do you know how to “program” an IBM Tabulator? Starting with the 1906 Hollerith Type I Tabulator? Going up through the 1949 IBM 407 Accounting Machine? Like ENIAC, the “program” is a plugboard configuration.
Now, in this century, have you ever used VHDL to configure an FPGA?
[ link to this | view in chronology ]
Re: Re: Re: Generic computing [was Re: Re: Re: Re: Poorly written article]
More on topic, it seems generally agreed that microchip designs are eligible for patents. Is anyone seriously proposing that FPGA configurations should be patentable? If so, how would that work?
[ link to this | view in chronology ]
Re: Re: Re: Re: Generic computing [was Re: Re: Re: Re: Poorly written article]
These days, of course, the simulation itself may run in a virtual machine, on virtual cores.
A Turing machine, of course, is a precisely-defined model of computation. If we are to be rigorous in using words and definitions, we should remember that the Turing class of machines consists of Turing-equivalents.
So—no: At the end of the day, Intel has announced another Turing-equivalent machine, but not an actual Turing machine.
Certainly not me. Rather, I'm pointing out that a distinction based on actual silicon, transistors, caps, all hardwired and soldered—versus simulated,virtualized and reconfigurable machines—is a distinction built on shifting sand in the middle of a stream.
Otoh, I don't really expect anyone who isn't intimate with the technology to really understand how a design may be moved from simulation to FPGA to ASIC.
[ link to this | view in chronology ]
Re: Re: Re: Poorly written article
Software doesn't do more than math and generic functions. True some specific hardware/software combinations for specific patented process may well be patent worthy. With the current set of hardware commonly available to all ... I have trouble even imagining one that would pass the test.
[ link to this | view in chronology ]
Re: Re: Re: Poorly written article
[ link to this | view in chronology ]
Re: Poorly written article
[ link to this | view in chronology ]
Bilski and Congress
No matter how small, I'm glad that SCOTUS did something - I've not read the decision yet but am guessing that creating precedent was done somewhat reluctantly.
Unfortunately this effectively rewards Congress for not doing what it is supposed to do; I guess SCOTUS hasn't done a lot of child-raising to know that's a bad idea.
[ link to this | view in chronology ]
This should be pretty obvious, but apparently...
Patent Laws are written by.... Congress.
The SCOTUS doesn't want to, and indeed can't, rule on the total mess that the US patent law has going on now.
We need updated laws for that to happen in any way other than slowly, painfully, specifically, and over lots of time.
Write to your congressperson... if you have one.
[ link to this | view in chronology ]
[ link to this | view in chronology ]
Re:
[ link to this | view in chronology ]
Re: Re:
/sarc
[ link to this | view in chronology ]
[ link to this | view in chronology ]
I have a pc that carries out non-generic computer functions
[ link to this | view in chronology ]
Rapido.
[ link to this | view in chronology ]
Oops
[ link to this | view in chronology ]
if only they excluded "connected to the internet"
Or maybe some inciteful attorney will one day point out thats all the internet is and so include all patents based on being co nected tk the internet as being equal to run on a generic computer.
[ link to this | view in chronology ]