Join 3,418 readers in helping fund MetaFilter (Hide)


Just because you used a computer doesn't make your idea new.
June 19, 2014 10:32 AM   Subscribe

We hold that the claims at issue are drawn to the abstract idea of intermediated settlement, and that merely requiring generic computer implementation fails to transform that abstract idea into a patent-eligible invention.
...
We must first determine whether the claims at issue are directed to a patent-ineligible concept. We conclude that they are.
The United States Supreme Court has ruled 9-0 [pdf], invalidating many but by no means all software patents, in Alice v CLS Bank.

Analysis from the usual suspects:

End Soft Patents
Patently O
IPWatchdog
posted by atbash (57 comments total) 22 users marked this as a favorite

 
It sounds like this will invalidate a lot of the stupider software patents (the ones that are essentially "let's do X, but on a computer!") It will be interesting to see if the ones that are more complex (such as the ones that say "Let's do this complicated algorithm, but on a computer") pass muster or not.

I'm going to guess that the big winners in this case will be the people developing UIs, since in many cases that's what UI patents boil down to.
posted by Mitrovarr at 10:43 AM on June 19 [1 favorite]


*fist pump*
posted by Jpfed at 10:44 AM on June 19 [3 favorites]


Wow. 9-0? Good.
posted by zarq at 10:47 AM on June 19


Wow. 9-0? Good.

If you read the opinion (it's only 21 pages and you can skip a fair amount and still get the gist of it), it's actually 6 in the majority, 3 concurring, and none dissenting. Thomas wrote the opinion, which surprised me some. The concurring opinion's difference is quite small.
posted by atbash at 10:51 AM on June 19 [2 favorites]


This was a precedent ruling. They had clear cases Bliski, Benson, Parker they could use and seems like no-one on the court was spoling for a fight over a more expansive attack on software patents or a more comprehensive defence of them. A good row back of the worst excesses.
posted by Another Fine Product From The Nonsense Factory at 10:52 AM on June 19 [1 favorite]


[nelson.jpg]

This has to be one of those cases where the plaintiffs really regret litigating this far. If they'd stopped after the district or appeals court they would have lost the patent. But now they've lost an entire class of patents, potentially their entire portfolio.

Not that I'll be shedding any tears over this ruling. This should have happened a long time ago.
posted by sbutler at 10:55 AM on June 19 [7 favorites]


Tell me tell me tell me that this bankrupts Nathan Myhrvold. Please.
posted by Diablevert at 10:58 AM on June 19 [14 favorites]


[nelson.jpg]? I don't get it.
posted by Jasper Fnorde at 10:59 AM on June 19 [8 favorites]


Does this mean that Android can now let content bounce when you scroll past it?
posted by gsteff at 11:00 AM on June 19 [2 favorites]


Nelson is the Simpon's character who always say "ha ha"
posted by RustyBrooks at 11:01 AM on June 19 [1 favorite]


Unanimous? Wow.

*mic drop*

Wow, this is a pleasant surprise, I must say.
posted by absalom at 11:03 AM on June 19


Unanimous and Thomas woke up to write it? Damn, you just got sonned, Alice. Go check Bob!
posted by klangklangston at 11:06 AM on June 19 [6 favorites]


[nelson.jpg]? I don't get it.

In his spare time, Lord Nelson was a big-time patent troll.
posted by briank at 11:10 AM on June 19 [18 favorites]


Also, lol: Gene Quinn is a Patent Attorney and the founder of the popular blog IPWatchdog.com, which has for three of the last four years (i.e., 2010, 2012 and 2103)"

AWARDS FROM THE FUTURE HOW CAN YOU ARGUE WITH THAT
posted by klangklangston at 11:10 AM on June 19 [5 favorites]


and Thomas woke up to write it?

I kind of assume some clerk actually does that?
posted by atbash at 11:11 AM on June 19


It sounds like the ruling could be used to as the basis for invalidating all software patents or just those that are excessively vague or abstract. It really depends on what lower courts do with the ruling. Is that right?
posted by Cash4Lead at 11:23 AM on June 19


Wow, posted and commented here long before slashdot or reddit, gett'n to be the hotbed of nerdism news on the blue.
posted by sammyo at 11:25 AM on June 19 [1 favorite]


So will we be getting a one click shopping cart for donations here soon?
posted by sammyo at 11:27 AM on June 19 [2 favorites]


James Grimmelmann (University of Maryland professor of intellectual property law) wrote on Twitter:
1. Very tentative, but I read CLS Bank to eliminate "do X using software" patents but not patents on new software techniques.

2. To expand on that idea, a patent on a new technique for memory defragmentation would, I think, pass muster.
If so, that would be a great improvement over the previous status quo.
posted by mbrubeck at 11:30 AM on June 19


In my extremely cursory reading of a couple blog posts, am I correct in suspecting that this mostly affects process patents (eg, shopping cart) and UI patents (eg, pull-to-refresh), but leaves alone things like algorithm patents (eg, search engine rankings)?
posted by ardgedee at 11:33 AM on June 19


If they'd stopped after the district or appeals court they would have lost the patent. But now they've lost an entire class of patents, potentially their entire portfolio.

From what I can tell, in this case the plaintiff (Alice) is basically a patent troll set up with a very small number of closely-related patents and a mandate to use them offensively to extort obtain licensing fees.

They either win, or die. I don't think there was an opportunity for them to lose partially; the patents are too closely related. Given that, who else goes down with them if they lose isn't their problem.

In fact, it might be better to scorch the earth if they're going to lose, rather than lose only their patent; if they lose only their patent, then they lose individually. If they escalate to the point of having an entire class of patents invalidated, then their loss is also their competitors' loss.
posted by Kadin2048 at 11:37 AM on June 19 [1 favorite]


I'm seeing a variety of reactions from the wonks. Not sure if there's consensus yet on what this means.

For a good lay overview I liked the Ars Technica article.
posted by wemayfreeze at 11:38 AM on June 19 [1 favorite]


ardgedee I'm curious to see a post that argues this invalidates UI patents. Link?
posted by wemayfreeze at 11:40 AM on June 19


and Thomas woke up to write it?

Thomas is a super-sharp jurist with a coherent judicial philosophy. Pretending otherwise is racist, and any liberal who doesn't take his move from Malcolm X to originalism seriously is, well, not serious. He's conservative, and there's good reason to reject that conservatism. But don't pretend he's dumb or lazy; you sound racist.
posted by anotherpanacea at 11:46 AM on June 19 [13 favorites]


I'm also doubtful that this ruling would affect most UI patents.
posted by gyc at 11:47 AM on June 19


It may be a bit premature to celebrate. My understanding of this ruling is that the class of patents that are now invalid are those of the form "we do [pre-existing thing] on a computer!"

So a one-click patent that just claims to patent the process of buying stuff with a computer would be invalid, but a one-click patent that describes the process of conducting a transaction over HTTP would probably still stand.

If my understanding is correct, this is certainly bad for patent trolls and good for everyone else. But it still leaves the door open to patent trolls - they just need to be a bit more descriptive in their patents. The net result is that they might need to file ten different patents to cover implementations of a concept, rather than one broad patent.

IANAL.
posted by b1tr0t at 11:51 AM on June 19


Pretending otherwise is racist

I get the feeling that klangklangston was probably alluding to Thomas' notorious tendency to all-but-sleep-through oral arguments because he thinks they're superfluous to the decision-making process. But feel free to persist in a persecution fantasy.
posted by fifthrider at 12:01 PM on June 19 [24 favorites]


Thomas is a super-sharp jurist with a coherent judicial philosophy. Pretending otherwise is racist.

Whoa, I think that's going a little far. There are plenty of reasons not to like Thomas that have nothing to do with his race. He's (probably) a sexual harasser, he doesn't recuse himself from cases that maybe he should (maybe!), he doesn't ask any questions at oral argument, he's very conservative in an "originalist when I agree with what the founders said" kind of way.

You may disagree with one or all of those reasons (I've started to come around to his side about not asking questions), but "and Thomas woke up to write it" is not a racist statement, nor is a dislike him necessarily based on race at all.
posted by sparklemotion at 12:07 PM on June 19 [30 favorites]


endorfireworks.gif
posted by drezdn at 12:07 PM on June 19 [3 favorites]


And back on topic: This decision annoys me because it doesn't really say anything. Invalidity here hinges on the Alice patent being, at its core, an abstract idea.

But what is an abstract idea? I feel like using a photovoltaic substance to convert sunlight into electricity sounds pretty abstract to me, but no one would claim that a new type of solar panel is not patent eligible subject matter.
posted by sparklemotion at 12:15 PM on June 19 [1 favorite]


Yeah a vote for moderation in responses to klangklangston's comment. As a super lay observer of SCOTUS I was also under the impression at one point that Thomas didn't do much in general due to his much talked–about silence during oral arguments. Certainly we can't remove race from the equation (because USA) but an ignorance of Thomas's activities on the court in my mind doesn't equate to racism.
posted by wemayfreeze at 12:18 PM on June 19 [1 favorite]


Does this invalidate that guy who's trying to claim podcasts are the same as his idea to mail cassettes of news programs to people?
posted by lumpenprole at 12:20 PM on June 19 [1 favorite]


> ardgedee I'm curious to see a post that argues this invalidates UI patents. Link?

I was guessing at what the ruling would affect. Since a lot of process concepts involve distinct UI (eg, one-click checkout), I was assuming there would be sufficient overlap to affect both aspects. But I don't know, which is why I asked.
posted by ardgedee at 12:22 PM on June 19 [1 favorite]


Well, this is gonna be interesting.
posted by stoneweaver at 12:24 PM on June 19


It certainly appears that the SCOTUS is hinting at going down the European route of requiring a technical effect.
posted by gyc at 12:29 PM on June 19


But what is an abstract idea? I feel like using a photovoltaic substance to convert sunlight into electricity sounds pretty abstract to me, but no one would claim that a new type of solar panel is not patent eligible subject matter.

Apparently, it's already a qualification for patents, intended to limit the applicability of patents to specific innovations. You can't (or shouldn't be able to) patent the concept of converting light to electricity, since that's a concept so broad that it includes coal-fired power-plants. You can patent specific processes for producing or improving photovoltaic materials.
posted by CBrachyrhynchos at 12:37 PM on June 19


For whatever it's worth I suspect the major software companies (Apple, Microsoft, Google, Amazon, Adobe...) are quietly happy about this, even if they aren't certain of the consequences yet either.

Whatever protections they gain for themselves through software patents, they are each having to spend hundreds of millions of dollars a year on the costs of constructing defensive patent portfolios, defending lawsuits from real competitors and patent trolls alike, on licensing existing patents, on managing the licenses they grant to others... This could stand to be an entire class of expenses that don't have to happen and are probably routinely costing them more money than they earn on their own patents.
posted by ardgedee at 12:40 PM on June 19


You wouldn't patent "using a photovoltaic substance to convert sunlight into electricity", and you're correct in thinking that's pretty abstract. You could, probably, patent using some specific substance, perhaps also arranged in a particular way, to convert sunlight to electricity; the "not abstract" part is the specificity of the substance and/or arrangement.

Obviously, people filing patents want those patents to be as broad as possible, so they are going to constantly push the boundaries of what the USPTO will allow. The PTO at various times has had very robust standards for patentability (cf. 19th c. patent models), but then the courts got involved and since then it's been pretty much "anything goes! patent all the things!" ... so it's nice to see the court rein that in.
posted by Kadin2048 at 12:44 PM on June 19 [2 favorites]


"But don't pretend he's dumb or lazy; you sound racist."

He barely participates in oral arguments (hence "wake up"), and he's so often off on his own even on concurrences that when he delivers the majority opinion, it tends to be unanimous and a slam dunk. For the last essentially unanimous opinion I remember on a similar issue, AMP v. Myriad, Thomas also delivered pwnage.
posted by klangklangston at 12:49 PM on June 19 [6 favorites]


Alex Tabarrok posits that this is the Supreme Court smacking down the Federal Circuit.
posted by bruceo at 12:55 PM on June 19


It certainly appears that the SCOTUS is hinting at going down the European route of requiring a technical effect.

I was curious so I tried to read the Wikipedia page that talks about technical effect in European patent law and it makes no sense to me. What is a technical effect, exactly? What would be an example of a software-related patent that might fail the technical effect test?
posted by enn at 1:00 PM on June 19 [1 favorite]


What's amazing is that it is 2014 and we are still squabbling over the basics here. Think of the many thousands of patents, millions in legal and patent agent fees, and repeated Supreme Court decisions and we're still at the point where the Court has to repeatedly remind the Federal Circuit (let alone the patent office) that there's some kind of basic standard for patentability.
posted by zachlipton at 3:29 PM on June 19


… 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.
posted by ChurchHatesTucker at 4:18 PM on June 19 [2 favorites]


"A generic computer [performing] generic computer functions" might not be a category which includes, say, a patent in which an otherwise generic computer acts on signal from an accelerometer. Looking just cursorily at the '479 patent, the most "esoteric" physical components of the system seem to be telephone modems connecting computers in a way that was decidedly common already on the patent's priority date in 1992, and an element which I assume Alice Corp. would have argued was not essential (in case someone said, "but we connect our computers with fibre channel not telephone modems").

On the other hand, that line of reasoning indicates that at some point the connection of an accelerometer to a generic computer plus some possibly-novel software would no longer be a patentable combination. I am comfortable with this consequence.
posted by jepler at 4:53 PM on June 19


"… 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 a bit of a weird misread, so far off that from another outlet I'd assume it was intentional. The missing part there would be that what is patentable would be describing SPECIFIC METHOD that the computer was utilizing. It's not enough to say that the computer should, say, check some numbers and run a formula, then update the numbers. It would have to describe what the formula is, how that formula works and why that's a technical innovation. It's a higher level of specificity.
posted by klangklangston at 5:27 PM on June 19 [1 favorite]


… 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.

I'd say that a most software beyond "echo 'Hello world!'" does more than just shuffle data from one register to the other. That is, we don't give patents just because the innovation in question was built in a machine shop, or synthesized in a test tube. We should give patents just because the claimed innovation runs on a Turing system either. If we go there, we end up with patent trolls like the guys who are suing cities because moving a document from office to office is such a radically new idea if the document is digitized in transit.
posted by CBrachyrhynchos at 5:44 PM on June 19


I'd say that a most software beyond "echo 'Hello world!'" does more than just shuffle data from one register to the other.

Such as…?

Seriously, we should be at the point where "… on a computer" is obvious by now.
posted by ChurchHatesTucker at 6:26 PM on June 19


Such as…?

Transforming that data in various novel and useful ways. Whether those transformations are worthy of patent protection is another matter altogether. To use an example cited by the decision, turning voltage from a thermocouple into information about the the process of curing rubber apparently is patent-worthy.
posted by CBrachyrhynchos at 8:39 PM on June 19


[A couple of comments deleted; please drop the Clarence Thomas derail.]
posted by taz at 11:51 PM on June 19


zarq: Wow. 9-0? Good.

atbash: If you read the opinion (it's only 21 pages and you can skip a fair amount and still get the gist of it), it's actually 6 in the majority, 3 concurring, and none dissenting. Thomas wrote the opinion, which surprised me some. The concurring opinion's difference is quite small.

So, what you're saying is: "9-0"?
posted by IAmBroom at 11:33 AM on June 20


A six member opinion with three concurrences isn't the same thing as a truly 9-0 opinion.
posted by ROU_Xenophobe at 12:26 PM on June 20


Yes, I get that, and yet, it still it effectively the same.

Pedantic shaved hairs aside, the notion that vague-assed software patents have some sort of bounds on their idiocy has been unanimously affirmed.
posted by IAmBroom at 12:48 PM on June 20


It really depends on the nature of the concurrence. They can be anywhere from "I have some minor differences" to "You people are morons who only made the correct decision by blind luck. "
posted by ROU_Xenophobe at 1:06 PM on June 20 [1 favorite]


Tell me tell me tell me that this bankrupts Nathan Myhrvold. Please.

While he is a patent troll, he isn't all evil. I'd point you at things like Modernist Cuisine for one (and his continued dedication to innovation in the culinary world), a novel approach for killing malaria-bearing mosquitoes with lasers, etc.
posted by feckless fecal fear mongering at 1:49 PM on June 20


It really depends on the nature of the concurrence. They can be anywhere from "I have some minor differences" to "You people are morons who only made the correct decision by blind luck. "

ROU_Xenophobe, that's nice in an academic sense, but in the situation we're actually discussing, "The concurring opinion's difference is quite small."

So, I concede the point: 6 + 0.999*3 is not equal to 9.

Except for all practical purposes.
posted by IAmBroom at 2:53 PM on June 20


So, what you're saying is: "9-0"?

I mean, yeah, the reason I said 9-0 in the post and explained the nuanced difference in a comment when somebody made a note of it is that it's nuance.
posted by atbash at 9:32 PM on June 20 [1 favorite]


Yeah, the "score" on a SCOTUS decision only refers to the votes for and against what the Court held. Concurrences still count toward's the Court's holding.

It's like how Brazil had a 3-1 game over Croatia. You don't mention in the score itself that Croatia's one point was actually an "own goal" by Brazil. That would only come up in the commentary on the game.

But, when a person points out that Croatia's one point was an own goal, they're not contradicting the 3-1 score. They're just providing more detail.
posted by Sticherbeast at 4:45 AM on June 21


« Older Clarity Campaign Labs invites you to use TargetSma...  |  Rainbow-Cake Recipe Inspires C... Newer »


This thread has been archived and is closed to new comments