Hacker News new | past | comments | ask | show | jobs | submit login
Abstract software patents struck down by Supreme Court (eff.org)
280 points by chaqke on June 19, 2014 | hide | past | web | favorite | 59 comments

>We will work to ensure the Federal Circuit—the court that hears all patent appeals—diligently applies the new filter on patentability.

Doesn't the Federal Circuit have something of a history on cases such as this, where they essentially say 'fuck it, we're not listening to the SCOTUS', and continue to rule as they please? What's to stop them from just doing business as usual in defiance of the Supreme Court?

Chief Justice Roberts will send Justices Kennedy and Ginsburg around to make an example of one of the Federal Circuit justices.

Kind of like how the Mob do business but will gavels instead of baseball bats.

"Ya know, that's a really nice bench ya got there. It'd be a pity if something were to happen to it..."

PS: I think the premise of the Supreme Court justices enforcing their judgments, Mafia enforcer-style, would make a great graphic novel. I can see Ginsburg going all Tommy DeVito on some Federal Circuit judge's ass and burying the body in the Rose Garden.

One of my favorite Onion videos to date is "Supreme Court Rules Death Penalty 'Totally Badass'." Complete with court sketches and everything.


There should be a threat of Supreme review and it should be random. And the Supreme reviewers should have the option of firing the judge.

NOBODY expects a Supreme Court Review! Our chief weapon is surprise...surprise and fear...fear and surprise.... Our two weapons are fear and surprise...and an encyclopedic knowledge of the law.... Our three weapons are fear, surprise, and an encyclopedic knowledge of the law...and an almost fanatical devotion to the Constitution.... Our four...no... Amongst our weapons.... Amongst our weaponry...are such elements as fear, surprise.... I'll come in again.

or Sotomayor all Robert Rodriguez style

> What's to stop them from just doing business as usual in defiance of the Supreme Court?

The Supreme Court itself, and, ultimately -- assuming they care -- Congress.

But, in any case, this "new filter on patentability" isn't new with this decision, really -- its a fairly direct application of previous USSC decisions in similar cases -- and isn't contrary even to the CAFC en banc decision in the same case, which came to the same substantive conclusion, with the dominant (though plurality) decision applying broadly similar logic applying the same precedent.

So while the "out of control CAFC defying the Supreme Court on patent issues" story might have some relevance, it really doesn't on this case.

(There's always the problem that what a larger en banc panel might do is different than what the smaller panels of the same court that hear appeals first would do, but that's not a particular problem of the CAFC but of the way circuit courts work in general. If you picked random three-judge panels of the Supreme Court to hear appeals to that court first and had to petition separately by a hearing of the full court, well, I'd expect a panel composed of Thomas, Alito, and Scalia (or, alternatively, Sotomayer, Kagan, and Breyer) would tend to return decisions on lots of issues that wouldn't look like the decisions the full court would return.

Sure, but then it would probably get appealed...

Yup. Just wrote about this, actually: http://www.sdtimes.com/content/article.aspx?ArticleID=71372&...

The Federal Circuit court has been, essentially, ignoring the supreme court, and has had just about every decision its made on patents in the last 2 years overturned upon appeal to the supreme. I talked to the EFF and they couldn't explain why the circuit was so fucked up. No one's really sure, but they clearly don't listen to the supreme court.

just about every decision its made on patents in the last 2 years overturned upon appeal to the supreme

The Federal Circuit sees 1000 patent cases a year. The Supreme Court has amped up it usual zero or one patent cases reviewed per year to five this past session. And those Supreme decisions have been largely 9-0 against the Federal Circuit. But about 99.5% of CAFC decisions utterly escape Supreme review.

Remember that WildTangent is still on the books as good CAFC precedent holding the exact opposite of Alice. The CAFC can pick whichever it prefers.

The Federal Circuit court has been, essentially, ignoring the supreme court

This is true. It will continue to do so. It will continue to be a corrupt juggernaut focused on bleeding our industry for grifters and trolls and the patent bar.

I talked to the EFF and they couldn't explain why the circuit was so fucked up

It's simple. Courts with general geographic jurisdiction get disinterested generalist judges who try to do a good job being fair and applying the law. Even if they get bad judges, the others on the circuit balance them out because they don't all belong to the same interest. Specialist courts with exclusive subject matter jurisdiction get judges beholden to special interests that promote abuse and parasitism for their own benefit.

Hey - thank you for this comment, it makes a murky situation a bit clearer for me.

My next question is, of course, what's the solution? I can't tell from your comment (being unversed in this landscape) who the 'specialist courts' you reference are -- the Federal Circuit themselves? Some subset of FC?

Or is there some mechanism that assigns these kinds of IP/patent cases to FC, or to that subset of FC, which might be tweaked?

The CAFC is the sole court in the USA that hears patent appeals. [0] All patent appeals go to the CAFC.

The CAFC has limited jurisdiction. It mostly hears patent cases, but also hears federal employee claims and some foreign trade cases. All judges on the court are equal but they have different backgrounds. Judges Dyk and Prost write the most rational and evenhanded decisions on the court and have little background in patents and technology policy. Prost was a labor lawyer and Dyk worked on First Amendment cases before being appointed. The worst judges worked for big science organizations in government and industry, but they write as if they were completely ignorant of science in order to grind down innovators in favor of government granted privileges.

[0] This link is worth reading: http://arstechnica.com/tech-policy/2012/09/how-a-rogue-appea...

> Remember that WildTangent is still on the books as good CAFC precedent holding the exact opposite of Alice.

There is no such thing as "good precedent" from a circuit court that contradicts Supreme Court precedent on the same issue. (Leaving aside the question of whether there really is a direct conflict between WildTangent and Alice given the procedural differences in the cases.)

Precedent is a social convention; there is no rule that judges have to follow precedent.

And the CAFC chooses which precedents it likes to follow however it likes. Supreme Court precedents have no special power over the CAFC. Just look at how Benson is treated by State Street Bank or Alappat.

Like all law, yes, case law is a social convention, but there most certainly is within US case law, quite specific rules requiring judges to file precedent and defining which precedent they must follow (binding precedent) and which they are not required to follow.

This is what I was getting at. It's great if the Supreme Court overturns this stuff, however if the Federal Circuit goes on ignoring it, and every case has to be appealed to the Supreme Court thereafter, then the impact is considerably dampened.

And, I don't think there isn't anything anyone can do about it. Is there even a precedent for a lower court blatantly ignoring the decisions of a higher one, at this level? Is there anything that can be done about it short of an act of Congress (which we can safely assume would never ever happen)?

Couple thoughts. First, Congress could swear in Obama's nominees. I haven't checked but a few years ago, basically he'd nominated hundreds of judges, and congress was refusing to approve any of them. This happened with Bush, too, so it's a bi-partisan thing.

It also means that, if you have a patent case, you must be ready to go to the supreme court. That's a shit ton of money...

> This is what I was getting at. It's great if the Supreme Court overturns this stuff, however if the Federal Circuit goes on ignoring it

The Supreme Court affirmed the Federal Circuit in this case.

So, insofar as the precedent on this issue which the Supreme Court applied in this case, its pretty hard to say that the Federal Circuit was ignoring it.

Is there even a precedent for a lower court blatantly ignoring the decisions of a higher one, at this level?

Jim Crow era Southern courts and practices like lynching had a lot in common with the CAFC's software patent jurisprudence.

Honey badger of a Federal Circuit we've got there.

Yep, the Federal Circuit court has a way of interpreting around SCOTUS decisions if they don't like them. The Circuit court sees a lot of Patent cases and they do whatever they can to avoid reversing past decisions. There are a lot of second and third order consequences to this judgement. Hopefully it will stick.

> the Supreme Court reaffirmed that merely adding “a generic computer to perform generic computer functions” does not make an otherwise abstract idea patentable. This statement (and the opinion itself) makes clear that an abstract idea along with a computer doing what a computer normally does is not something our patent system was designed to protect.

This opinion gives me hope that reform is possible & that the U.S. government still works with reason.

Now the question is are patents still relevant in today's world of internet, rapidly accelerating innovation, and evidence that free/open source research reduces costs?

Each generation of CPU design or gas turbine design or wireless baseband design, drug design, or nuclear power plant design still takes more R&D money than the last. And nobody was working on a competitive open source LTE baseband last I checked.

"Accelerating innovation" is something which, if it exists, touches only a few areas of engineering. In most areas of engineering, innovation has gotten more expensive, not less. Very few things are amenable to a distributed group of hackers working on it. Jet engines are still best designed by a roomful of PhD's. And those cost lots of money.

We can argue about how necessary patents ever were, but I don't think they're any less necessary today.

The question, in my opinion, isn't whether patents are still relevant. Its whether we can draw lines better to leave out less useful ones while keeping that exception from swallowing the patent system. E.g. if a power control module for an LTE radio is patentable if implemented in an ASIC, should it become unpatentable when you implement it in a DSP? If not: how do you allow those patents while invalidating Amazon's One Click shopping patent? To a certain extent, its tempting to just say: "its too hard to draw the lines, its not worth the trouble." That's easy to say if you're not Qualcomm and don't have a dozen companies who'd love to use your work for free.

a power control module for an LTE radio is patentable

It's worth drawing a line between "a particular solution to the problem of power control on LTE radios" and "any method of doing power control on LTE radios". Far too many software patents seem to cover any solution to the same problem, not just a particular implementation.

(Note that the software in the DSP also gets the benefits of copyright protection)

I'm sceptical that patents are an essential part of the ASIC business either. Hardware designs would still be covered by copyright law, so a competitor would still have to reimplement it from scratch. Which is actually more work than doing so for software because hardware languages use a lower level of abstraction, and the design needs to be tested much more thoroughly because you can't fix bugs post-hoc. Also, giant SoCs present a natural barrier to entry to new companies because a new company needs to have all the blocks, not just one or two. There may be a dozen companies trying to make LTE basebands today, but that situation is unsustainable and we are already seeing exits. In short: ASIC innovation would be profitable even without patents. Also, we seem the same kind of behavior in the hw sector: Qualcomm was found to be abusing its patents: https://en.wikipedia.org/wiki/Qualcomm_Inc._v._Broadcom_Corp.

Jet engines are still best designed by a roomful of PhD's. And those cost lots of money.

Possibly the wrong example for your argument :) Jet engines can now be designed by a team at SpaceX rather than at NASA, and the team can 3D-print them. That seems like reduced-cost innovation compared to decades past.


(PS - I definitely don't disagree with your general point, but it might be becoming harder to know which areas need patents and which don't.)

SpaceX is very neat, but I'm not sure they represent a change in the pace of design. Scaled Composites has been operating with a lean team for decades now. And arguably, Scaled Composites has been more willing to tread unproven ground with novel technology. SpaceX has been sticking more to proven concepts, which mitigates one of the big risks of this sort of R&D.

I'm not quite sure what difference you're pointing out. Why does is matter if the PhDs are at SpaceX or NASA?

Yes, I'm aware they aren't delegating the rocket design to the marketers. Smaller team, faster to market, cheaper. The discussion is around accelerating innovation and SpaceX perfectly demonstrates this, despite "rocket engines" being the earlier example given for how it's not true.

This is about the same thing that's already on the front page with 70 comments, including a great one by grellas. I'd go there if you're interested in this: https://news.ycombinator.com/item?id=7916160

That post was just a link to the actual Supreme Court decision. This article explains what it's about in language that's easier for people who aren't legal scholars to understand, so I think it serves a useful purpose.

> This article explains what it's about in language that's easier for people who aren't legal scholars to understand

I really strongly disagree with this. Supreme Court decisions are usually quite readable and will give a you a nuanced overview of the legal background and justifications for the ruling. They're usually very accessible and not nearly as crammed with legal jargon as you might guess. I try to read as many important decisions as I can and I have no legal training at all. I only occasionally have to look up a term. I would say they're easier, on average, for me to read than a CS paper. And after reading a few Scotus papers, you'll have a much deeper understanding of how our legal system works, and of course a great deal more about the individual decisions. You'll even read news articles about a decision and say, "That's not what the decision was about at all!"

Supreme Court rulings are, of course, quite long, and there's nothing wrong with a summary article like this one. But my point is that you shouldn't fear the legalese.

> Supreme Court rulings are, of course, quite long

Interestingly, judges such as Richard Posner think recent opinions are too long.

http://www.law360.com/articles/455840/posner-says-justices-s... (Ignore the linkbait title; the article really is quite good.)

I would add that the court's opinions are typically prefaced with an excellent summary, which can be a solid alternative or supplement to an article.

> This article explains what it's about in language that's easier for people who aren't legal scholars to understand

More accurately, this press release puts the spin of an interest group popular on HN on the decision; it is not a neutral explanation of the decision.

But now I'm looking at it on a mobile device. That's totally novel.

This is great ammo for all the jack-wagons I encounter in daily life who counter my position on patents with the genius position, "What if someone came and just took your house, or your car away. That's your property and you wouldn't like it, would you?"

while the decision is pretty tightly tailored, it seems like it's SCOTUS' way of opening the door for subsequent arguments regarding pure software patents. take google's page rank patent[1]. one could argue that assigning scores to, say, academic papers based on references cited and then doing that repeatedly is an "abstract idea" and google's recitation of "apply[ing] it" using a computer is not enough to enter into patent eligible subject matter. if you read/listen to the oral argument, the counter argument that doing so on a massive scale is impracticable was flatly rejected. indeed, a million monkeys counting the citations of scholarly pubs could probably perform that method on a decent sized corpus. is the method therefore inherently abstract? will be interesting to see if Alice alters the analysis (albeit slightly) for these types of questions and whether we'll see a pure software issue (without the negative financial/business method clouds in the background) sooner rather than later.

[1] http://www.google.com/patents/US6285999

What exactly is an abstract idea? Is it possible for an idea to be concrete?

From looking at examples, "abstract" seems to mean trivial. By trivial, I don't mean obvious or silly, I mean small; i.e. an idea that can be reduced to one or two sentences, like "e-commerce shopping cart."

Otherwise it's pretty hard to distinguish ideas about software from ideas about hardware and explain why one is more patentable than the other.

An abstract idea isn't physical. You can patent a machine, or a drug. You can't patent a method for dividing two numbers. However, a machine for dividing two numbers—assuming it's novel, not obvious, etc.—is fair game.

Ok, but what's actually protected is the idea behind the machine, right? A different machine that employed the patented idea would infringe the patent.

Maybe what is meant by "abstract idea" is really "an idea in the abstract," i.e. an idea that is not used to make a machine can't be patented; it's the employment of the idea that is patentable.

You're getting confused by the word "idea." You patent a machine, not an idea. If you insist on saying that you patent the "idea" for a machine, then what you can't patent is an "idea about an idea." Ideas about machines are good. Ideas about ideas are no good. Make sense?

I think I understand patents as well as any average non-lawyer. What I'm trying to do is make sense of the language used to describe patents. It's clear that people can't agree on what they are, because the Supreme Court keeps needing to step in and refine the definition. The language seems not to make sense, which would help to explain why people can't agree on what they are.

Ok, you patent a machine. But that statement is vague and ambiguous at best. If I patent a machine that I built, then how could the patent apply to a different machine that I didn't build? Yet, that's the whole point: they apply to classes of machines, including machines that haven't even been built at the time that the patent is granted. The class would seem to be inherently something abstract.

Correct, ideas are abstract. Mathematics (including computer science) is abstract. If you invent a new math, or a new piece of software, it is abstract and thus non patentable.

If you invent a new piece of hardware, it's concrete. The ideas embodied are abstract. If you came up with hardware that works differently in a concrete way (e.g. you figured out how to use carbon in place of silicon, opening the door to entirely new categories of manufacturing) then it would be potentially patentable. Your methods would be concrete methods involving substances which are not abstract.


I would argue (playing devil's advocate in favor of software patents) that a new /piece of/ software is decidedly not abstract, it is concrete and it has an abstract part that is (in the US today) patentable.

What's not patentable is "old idea, now on a computer for the first time" -- covering any implementation of "old idea" that nobody thought to put on a computer before, "but on a computer." That's clearly abstract. It's not an implementation. There are no implementation-specific details to put in the patent as claims (they would have to stand up to the obviousness test either individually or as a group). With nothing new except "on a computer" or "on a mobile device" it's not novel or specific and therefore not patentable.

I'm actually not going to argue the point "now if you HAD a specific implementation" because I'm not sure I actually believe in patents (or more specifically software patents) at all, but I do believe that even without a patent: if you have some source code that your competition doesn't, and you made it... and they acquire it somehow without getting a license, there are already copyright laws that protect you (so long as you can establish your own original ownership of that work) if they should decide to rip it off wholesale.

My favorite part about this is that it shows how idiotic USPTO is for approving such patents in the first place. Hopefully, this will make them reconsider whether they should be approving other types of patents, too, before the Supreme Court puts them in a bad light again, but I'm not holding my breath.

There are no ramifications for the USPTO.

In addition, approving a patent takes almost no work while not approving one takes significantly longer and can have more bouts of back and forth.

The USPTO right now operates on a very loose definition of 'patentability' simply because it is more convenient to do so and the course will sort it out eventually anyways.. and there's absolutely no downside for them other than their own conscious.

They do not care.

They can't care. They don't set policy, Congress does. They just act out Congress' laws.

"Admittedly, the Supreme Court did not offer the clearest guidance on when a patent claims merely an abstract idea, but it did offer guidance that should help to invalidate some of the more egregious software patents out there." Where can i find this guidance?

Despite EFF syndrome, patents are a tool, they are not evil, patent trolls are evil, patents are just tools.

This was always going to be the case eventually (a task that was already performed, this just invalidates adding "on a computer").

It's much harder for the law (I would say impossible) to actually judge stuff that is obvious. For example wavelet patents in certain video codecs are absurd - improving fourier transforms in a very very obvious way - but it won't be obvious to a judge or a jury and certainly isn't covered by this supreme court decision.

The more specialised people become in certain fields the more 'obvious' discoveries become. The law has no means to understand what rights it's protecting, and for how long they should be protected, which means it will never police the patent's system effectively, no matter what this judgement says.

Given how fast software patents and technology change and that the law can't judge them effectively I would suggest that instead of all this mess can we not just have software patents that last a maximum of 5 years.

Some tools empower evil more than good or even just accidentally do more harm than good more often than not.

Land mines for example.

Absurdly oversimplified.

Patents are a tool that was being abused so badly that the Supreme Court has been taking multiple cases just to try to set things straight.

The abuse has been evil.

EFF didn't say patents were evil. I can do that too:

Despite inane anonymous commenting syndrome, you are a person, you are not evil, you are just a person.

Registration is open for Startup School 2019. Classes start July 22nd.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact