Comment by rayiner | Hacker News Reader

Comment by rayiner | Hacker News Reader

This is a very good opinion, tightly argued based on the precedent. And frankly, unsurprising given the relevant precedent (Diehr, Benson, Mayo).

The opinion has three parts:

1) The Court identifies the abstract idea contained in the patent as the general concept of intermediated settlement. (Slip. Op. at 9) The key takeaway here is the discussion about whether an "abstract idea" has to be on the level of a "law of nature" in order to be excluded. The Court says that an abstract idea need not reduce to some fundamental truth that has always existed. It is sufficient for it to be a "fundamental economic practice" or "longstanding commercial practice," like the concept of "risk hedging" in Bilski. (Slip. Op. at 10)

2) The Court looks to see if the patent adds anything more to the abstract idea, and concludes that all it does is describe how to implement the idea in a general computer, in the generic and conventional way. (Slip. Op. at 12, 15) The Court analogizes to Mayo, in which it decided that the claimed method amounted to no more than instructing doctors as to how to use a well-known process for measuring metabolite levels to diagnose their patients. (Slip. Op. at 11) The Court states that the claimed method simply describes how to instruct a computer to perform the abstract idea of intermediating settlement. The Court makes a point of noting that the computer implementation described here is "wholly generic" and that "[i]n short, each step does no more than require a generic computer to perform generic computer functions." (Slip. Op. at 13, 15)

3) The Court concludes that the "system" claims are no different in substance from the method claims: "But what petitioner characterizes as specific hardware—a 'data processing system' with a 'communications controller' and 'data storage unit,' for example, see App. 954, 958, 1257—is purely functional and generic. Nearly every computer will include a 'communications controller' and 'data storage unit' capable of performing the basic calculation, storage, and transmission functions required by the method claims." (Slip. Op. at 16)

Haven't finished reading it, but...

> The Court says that an abstract idea need not reduce to some fundamental truth that has always existed. It is sufficient for it to be a "fundamental economic practice" or "longstanding commercial practice," like the concept of "risk hedging" in Bilski. (Slip. Op. at 10)

If your quote is accurate, and they applied the same logic as Bilski, this sounds like they're conflating patent eligibility (101) and patentability (102, 103 etc.) again. To understand the difference between "patent-eligible" and "patentable", if you invent a car colored red [1], it's patent-eligible because a car is an articles of manufacture i.e. it satisfies 101. But it's not patentable because red cars have been around forever, i.e. it fails 102.

For those of you following along at home, 101, 102 and 103, etc. are bars that patent claims must clear to be valid. 101 is "patent eligibility", i.e. it excludes things like laws of nature and "abstract" ideas. 102 is novelty, i.e. it excludes things found in the prior art. 103 is non-obviousness, i.e. it excludes things that are obvious (note: not "trivial") combinations of existing prior art. These are long-standing principles for judging patents, each providing a distinct criteria for judging and invalidating a patent, and this arrangement has a long, well-known and mostly consistent history behind it.

Now essentially, the SC is saying "prior art" can sometimes make something an "abstract idea". You don't have to be a patent lawyer to see that this does not really make sense.

This patent is obviously invalid, but we already have the tool for invalidating it, and that's section 102: a "longstanding commercial practice" or "fundamental economic practice" clearly falls under the umbrella of prior art -- things we already knew.

By mixing 102 analysis with 101 analysis, they're just "muddying the water" even more. I'm guessing this is going to make the PTO's and Federal Circuit's job even harder and more inconsistent.

1. Whether a "car" is an abstract concept is not relevant to the analogy.

I can't tell if you are heavily drinking the kool-aid, or trolling, but ...

You say each has a "well-known and mostly consistent history behind it"

None of these have a consistent history, being subject to the whims of changing panels and ideological en-banc fights every couple years when a new judge gets appointed.

You act as if they are well thought out, well applied doctrines, when they are completely hit-or-miss bullshit that has caused the MPEP to explode due to the inconsistencies.

"I'm guessing this is going to make the PTO's and Federal Circuit's job even harder and more inconsistent."

The federal circuit needs no help becoming more inconsistent, they do a great job on their own.

This is more true to the realities of this issue than anything else in these comments. We're still stuck with an "I know it when I see it rule" that depends on what the examiner/judge had for breakfast.

> I can't tell if you are heavily drinking the kool-aid, or trolling, but ...

No, just armchair-lawyering :-) My experience in this field, while more extensive than most non-attorneys, has been as a hobbyist. What I lack in hands-on experience, however, I try to make up by reading empirical studies.

I would agree that the standards of 101, 102, etc. have not been evenly applied, to say the least. In fact, that's why the Federal Circuit was formed. There are studies evaluating to what extent they have been successful in harmonizing patent law. To your point, there is even a study on how consistent (or not) the CAFC is. (Will try to dig them up when I get a chance.)

However, what I'm trying to say is, regardless of how the statutes have been implemented, the theory is sound and rational: a patent has multiple, largely orthogonal, dimensions and it should be judged along those lines. To my engineering mind, anything that conflates those dimensions is an inelegant and problematic ruling.

> However, what I'm trying to say is, regardless of how the statutes have been implemented, the theory is sound and rational: a patent has multiple, largely orthogonal, dimensions and it should be judged along those lines. To my engineering mind, anything that conflates those dimensions is an inelegant and problematic ruling.

Yeah, the problem with that idea is that rulings aren't intended to implement abstract theories independent of the controlling enactments (the statutes and the Constitution). The theory that patents have "multiple, largely orthogonal dimensions", at least as you seem to identify those dimensions -- and particularly the idea that newness is a consideration only for Section 102 patentability, and not Section 101 patent eligibility -- is inconsistent with the actual text of Section 101, which explicitly requires for eligibility that the thing sought to be protected is (a) an invention that is (b) new and (c) useful.

Your aesthetic preference for orthogonal considerations where newness would be a considered only for Section 102 patentability is, of course, a legitimate aesthetic preference -- but its not what the statute says, and I think it would a more "problematic ruling" for the court to ignore the statute simply out of aesthetic preference for orthogonal categories with a certain configuration.

Replying since I can't edit the post... Some studies regarding the contributions, consistencies and inconsistencies of the CAFC.

1. "Promoting the Progress: Three Decades of Patent Jurisprudence in the Court of Appeals for the Federal Circuit", Damon C. Andrews" - http://law.missouri.edu/lawreview/files/2012/11/Andrews.pdf

2. "On the Causes of Unpredictability of Federal Circuit Decisions In Patent Cases", Paul M. Janicke - http://papers.ssrn.com/sol3/papers.cfm?abstract_id=960636

3. "How Federal Circuit Judges Vote in Patent Validity Cases", Mark Lemley and John R. Allison - http://papers.ssrn.com/sol3/papers.cfm?abstract_id=189512

Your point is, at a purely statutory level, fairly compelling. And it's essentially the point the plaintiffs made in this case: the "abstract principle" exception to subject matter eligibility should be construed narrowly, given the broad wording of 101 and the fact that it's a judicially-created exception.

But, invalidating this patent on 101 grounds versus 103 grounds is not the same. Subject matter eligibility is something that can be ruled on in a motion to dismiss, In re Bilski, 545 F.3d 943, 951 (Fed. Cir. 2008), while obviousness will generally need to wait after claim construction. It should be possible to knock out patents that claim a fundamental practice in any field as early as possible.

You can make other distinctions between "a fundamental practice" and "prior art," too. In the field of cooking, do we have to point out prior art for making a roux, or do we just accept that someone, perhaps many someones independently, did it sometime long ago, but it's so established now that it might as well be a fundamental law of the field?

Over time, the Supreme Court has been beefing up 101 as a bar to patent eligibility. Is the interpretation in CLS v. Alice the one that's most faithful to the text of the statute? Probably not. But interpretation in the name of controlling the flow of litigation in lower federal courts is one of the key roles of the Supreme Court.

I have a tangentially related question: has there been any thought given to raising patent validity standards around teachability? A big constitutional point of a patent is to teach someone "skilled in the art" how to replicated the subject matter of the patent. That's typically intepreted as "can you pay an expert witness to claim the patent teaches them to construct the claimed infringement," which is, of course, always true. But to me, this seems like fertile ground for the defense, and ought to be a negative test: can the defense find an expert witness to say that the patent is too vague, unspecific, or abstract to teach them to construct the claimed infringement.

Is this approach ever tried in these cases? It seems to me that the more convoluted and abstract the patent (which seems to be the point of many of these patents explicitly written to try and reach as broad as possible), the more juries will tend to identify with the expert saying "I can't make heads or tails of this thing" and invalidate bad patents as simply failing to adequately teach the construction of the infringement claim.

This would be a challenge to the validity of a patent on Section 112 grounds. However, there would not be an expert witness involved, as it's a question of law to be determined by a judge. See: http://scholar.google.com/scholar_case?case=1212705333151944... (Section B, starting on page 495). This does get used in practice, successfully.

Ah, I had not considered how this would affect downstream litigation procedures. Thanks for the explanation. I strongly believe we need more tools to dismiss frivolous litigation as early as possible, so in that light this seems more favorable.

> If your quote is accurate, and they applied the same logic as Bilski, this sounds like they're conflating patent eligibility (101) and patentability (102, 103 etc.) again. To understand the difference between "patent-eligible" and "patentable", if you invent a car colored red, it's patent-eligible because a car is an articles of manufacture

101 does not (and much more does not only) require that something relate to "an articles of manufacture" to be patent-eligible, it requires it to be an invention of a "new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof"

It is quite arguable that, assuming cars exist, a "red car" is, to the extent that it is new not useful, and to the extent that it is useful not new, and therefore patent ineligible.

So, no, while I agree that your analogy may be relevant, I don't think it illustrates that the Supreme Court is confusing patent eligibility with patentability. I think you are reading patent eligibility far too broadly.

> 102 is novelty

Its true that 102 has specific novelty requirements on top of the patent-eligibility requirements in 101, but that doesn't mean that things don't need to be "new and useful" under Section 101 before one even needs to consider whether they meet the specific novelty rules in Section 102.

You're right that the actual wordings of the statutes make them far less orthogonal. However, the way I've most frequently seen the statutes applied has been that 101 is applied broadly -- only used to exclude things like laws of nature -- whereas 102, 103 etc. have been used to judge the actual "quality" of a patent. To me, that also makes technical sense, but as you said downthread, that's more a personal aesthetic preference than anything grounded in the actual language on the books.

Edit: however, my original point regarding the opinion was, the SC did not just say this fails 101 because it's not new, but they went on to say that "long prevalent" practices (i.e. prior art) can make something "abstract". This, to me, simply makes no sense.

> Edit: however, my original point regarding the opinion was, the SC did not just say this fails 101 because it's not new, but they went on to say that "long prevalent" practices (i.e. prior art) can make something "abstract". This, to me, simply makes no sense.

In what I've seen of the Supreme Court jurisprudence on patent-eligibility, "abstract idea" seems to have (based on the way it is used, not any explicit statement) evolved into a sort of a term of art for things that fail both the "new and useful" and "process, machine, manufacture, or composition of matter" requirements of Section 101.

I would say that this is one of the ways in which case law gets unnecessarily and confusingly self-referential and loses its visible connection to the underlying enactments that it is interpreting, which makes it less accessible and avoidably opaque. But good luck getting people on the Supreme Court that see that as something important to minimize.

that's not at all what the SC is saying, what they are essentially saying is that business method patents are nonstatutory, with or without the computer in which executed. An iphone on the other hand is not a business method (though shopping with one is.

This is an opinion everyone can love because the term "abstract idea" means something different to each reader. I think that's intentional. They're shifting authority from juries to judges. The wording gives judges a lot of leeway to kill patents by applying the loosely defined "abstract idea" label and then looking for conventional steps that remain, an analysis that seems a lot like the obviousness analysis that used to be reserved for juries.

This is REALLY important. Juries were getting snowed. Look at the Apple-Samsung verdict.

Every time I see someone that brings up the Apple-Samsung verdict I know they don't what what that suite was about. It wasn't about patents; it was about design patents. A design patent is to protect ideas outside of what is covered by trade dress and is a distinctly different idea than a patent.

Yes it's a terrible name. Yes it's easy to get confused. Just never use Samsung/Apple when talking about patents you'll be fine.

The Apple-Samsung verdict was about design patents AND utility patents (United States Patent Nos. 7,469,381, 7,844,915, and 7,864,163).

Thanks. I agree this is a very valuable case, as so many horrible software patents boil down to "this common thing everyone knows, but inna fone!" Software people have understood this for a long time as a completely unacceptable mis-use of the patent system, and it's nice to get that translated to an SC decision.

To rant a sec, how many times will the federal circuit need to be slapped upside the head on these issues before they get it through their skulls that their whole permissive approach to patent law is just fundamentally unconstitutional, and they need to adapt their philosophy, or the SC will continue to apply the boot to them? Is this a case of there just being a handful of particular incorrigible bad actors who we just have to wait to die?

The fundamental issue is that Congress has stated patent eligibility in very broad terms. Where the Federal Circuit and the Supreme Court differ is that the former is reading it broadly, while the latter is trying to reinforce narrowing exceptions to the statutory text it created itself.

This is, fundamentally, Congress problem, and a problem with the breadth of the statutory text they've given the USPTO to work with.

> Where the Federal Circuit and the Supreme Court differ is that the former is reading it broadly, while the latter is trying to reinforce narrowing exceptions to the statutory text it created itself.

That would be a compelling story, except that the CAFC and Supreme Court came to the same conclusion on this case, both on the judgement (that the patent was invalid), and largely on the rationale (the plurality of 5 out of 10 judges in the en banc rehearing found that all the claims were patent-ineligible as the Supreme Court found, and a majority of 7 found that at least some of the claims were patent-ineligible.)

I know, I meant more generally. I think the original panel opinion, vacated en banc, which required it to be "manifestly evident" that the idea is abstract is an example of reading the statute broadly.

how many times will the federal circuit need to be slapped upside the head on these issues

The Federal Circuit is essentially immune to Supreme Court decisions of this type. Aside from occasionally judge Lourie, who wrote the plurality opinion for the CAFC, there are no apparent judges on the CAFC who care what the Supremes have to say. They don't have to; the Supreme Court can reverse on average one or two patent cases a year while the CAFC takes a thousand.

The CAFC will eventually overturn this case just as they have been making progress overturning KSR v. Teleflex, a much more important precedent. [0] Obviousness would have taken care of many more abusive troll cases than this one if it were taken seriously, but the Supreme Court cannot force the CAFC to do anything.

The CAFC judges are mostly chosen from organizations that depend maximizing the power and scope of patents, largely appointed to please that patent bar, esteemed and supported by the social and intellectual environment of the patent bar, invited to speak and surrounded by patent bar lobbyists, and promised their chances to make millions after retirement -- if they wish -- back in the patent bar.

The idea of a single subject court was a novel experiment in 1982 but it's a disaster and a failure from a judicial and technical perspective. It's even worse for innovative American businesses. Probably things won't change unless the CAFC is changed.

[0] final four paragraphs of http://www.ipwatchdog.com/2012/03/20/supreme-court-mayo-v-pr...

> The CAFC will eventually overturn this case

Why would the CAFC, even assuming it could, want to overturn a case in which the Supreme Court affirmed both the judgment and the dominant reasoning of the CAFC? I mean, that would go beyond not caring what the Supreme Court does into active spite.

> The idea of a single subject court was a novel experiment in 1982

The CAFC isn't a single subject court, and both of its two immediate predecessors (the United States Court of Claims and the United States Court of Customs and Patent Appeals, 1929-1982) were more like single-subject courts than the CAFC is, so, insofar as the CAFC could even remotely be considered by such a court, it wasn't a "novel experiment" in that direction (and Court of Customs and Patent Appeals prior incarnation as the Court of Customs Appeals, 1909-1929, was a single subject court, and until 1914 was a single-subject court with no further avenue of appeal.)

The CAFC was part of an evolution away from narrow-subject courts, not a "novel expirement" to try a "single subject court".

insofar as the CAFC could even remotely be considered by such a court

Before the CAFC, patent appeals went to the regional circuits. Now they all -- every one in the nation -- run to the corrupt CAFC. Even non-patent cases like the case where CAFC ruled in favor of copyrights on APIs can be packaged off the the CAFC instead of regional circuits by including unrelated patent issues.

The regional circuits created a mess with patent law. Wasn't it the 7th circuit that never found a patent to be valid. 3M was justifiably upset about that. If forum shopping is bad now, think how it would be with regional circuit patent law.

The 7th Circuit is also not very generous to debtors in bankruptcy practice. Yet the world goes on. This variation isn't a bad thing, IMHO.

Bankruptcy has bankruptcy district courts.

Yes, but they're Article I courts where everything is reviewable by a generalist District Judge.

> Wasn't it the 7th circuit that never found a patent to be valid. 3M was justifiably upset about that. 3M was justifiably upset about that. If forum shopping is bad now, think how it would be with regional circuit patent law.

I started out at the very beginning of the Federal Circuit era. I think you might be thinking of the 8th Circuit (and the District of Minnesota) as implacably hostile to patents. And yes, forum-shopping could be really bad in those days.

> and promised their chances to make millions after retirement -- if they wish -- back in the patent bar.

This is a bullshit angle you're working. It is exceedingly rare for a federal court of appeals judge to return to private practice. The ones who do are usually appointed very young. 75% of former CAFC judges have died on the bench. The rest retired outright, or taught. Judge Rader isn't going to be making millions teaching IP law in China.

Reply on news.ycombinator.com