Google LLC v. Oracle America, Inc. (Supreme Court 2021)
In a 6-2 decision authored by Justice Breyer, the Supreme Court has held that Google’s copying of the JAVA API naming convention was a fair use as a matter of law. The court did not decide the question of whether the API was copyrightable in the first place.
In his dissent, Justice Thomas (joined by Justice Alito) argues that the majority opinion “disregards half the relevant statutory text and distorts its fair-use analysis. . . . Oracle’s code at issue here is copyrightable, and Google’s use of that copyrighted code was anything but fair.”
- Opinion: 18-956_d18f
The key lines from the case for patent attorneys:
Majority: [U]nlike patents, which protect novel and useful ideas, copyrights protect “expression” but not the “ideas” that lie behind it.
Dissent: Computer code occupies a unique space in intellectual property. Copyright law generally protects works of authorship. Patent law generally protects inventions or discoveries. A library of code straddles these two categories. It is highly functional like an invention; yet as a writing, it is also a work of authorship. Faced with something that could fit in either space, Congress chose copyright, and it included declaring code in that protection.
= = =
The case involves the software interface or Application Program Interface (API) for Oracle’s Java programming language. What we’re really talking about here are the names of the various functions that a programmer might memorize such as “Math.max(a,b)” which returns the greater of two inputs. Java is divided up into “packages” of function calls such as max.
Java is a popular language with many millions of skilled programmers. As Google developed its Android platform, it wanted to simplify the adoption process for app developers. As such, it chose a language that the programmers know – Java. The one problem is that Oracle wanted a substantial royalty and so Google instead recreated the entire language, including the functionality of thousands of function-calls.
What did Google copy?: It copied the naming convention of the functions (Math.max(a,b)) and their organization. This is thousands of function names organized into 37 Packages. Oracle claimed copyright to these names and their organization. The Federal Circuit agreed with Oracle that (1) the API is protectable under copyright and (2) that Google’s use was not a fair use.
In its fair use analysis, the court placed substantial weight on the fact that the value of Java’s API is based upon the fact that “those who do not hold copyrights, namely, computer programmers, invest of their own time and effort to learn the API’s system” rather than in the inherent creativity of the expression. Thus, the “‘nature of the copyrighted work’ points in the direction of fair use.” Although the Supreme Court found Google “precisely” copied the JAVA API, it still found the use transformative — because it was being used on a handheld rather than a laptop. For substantiality, Google only copied 11,000 lines of code. The court found that insubstantial since it was less than 1% of Java as a whole. Finally, regarding market impact, the court found that Sun was unlikely to be able to compete in the Android marketplace and that the copying by Google created a lots more market opportunity for others.
Obviously, there is lots more to this opinion. Read it via the link above.
Surprised they came out the right way on this after the oral arguments. Glad someone (probably one of Breyer’s staff) took the time to get educated on the issue.
Again, Oracle didn’t come up with the “max” computer call, Google didn’t copy the “max” code that Oracle came up with. They just named their call the same thing as the Oracle call. That copyright question dies as soon as you recognize “max” was designed to invoke an image of what the function did. Clearly at least one member of the court felt it was not copyrightable, but they essentially took the functional merger logic and used it to fuel what they view as a less-precedential fair use holding.
I question if the court recognizes that its holding is less precedential than holding it non-copyrightable. At least with the copyright question you could make it expressive by naming the call something unrelated to what it does (i.e. a “widget” call that returned the highest value of numbers is expressive and unrelated to function compared to “max”). With the fair use question it makes any name unprotected so long as someone writes their own implementing code. I suppose they could get out of future issues by calling it insufficiently transformative, but I’m not sure the federal circuit will be so nuanced.
tl;dr – If you write a book about how to hunt bears, and you name your book “how to hunt bears” you shouldn’t be surprised or go running to copyright when someone else tackles the subject of bear-hunting and they also name their book “how to hunt bears.”
This is just a ridiculous analysis. It has nothing to do with the name evoking what the function does. They are many other names that could be used to do the same thing. Your analogy with the bear is brain dead. It would be more like you named your book “How_to_Hunt_Bears_427” and someone copied the name of your book because your book was selling so well.
Just ridiculous. What is the matter with you?
It has nothing to do with the name evoking what the function does.
Of course it does. Copyright protects expression, not rote description.
They are many other names that could be used to do the same thing.
Not without running up against a functional barrier. The point of a call is to truncate the typing and storage of the code. The reason you call a Max function is because you don’t want to cut and paste the Max function every time you seek the max value of a set. There’s a functional reason people nominate the call with a “maximum” descriptor rather than use flowery prose, and there’s a functional reason Maximum is almost universally shortened to Max.
But let’s assume you’re right and pretend the word “max” is protected expression. Oracle didn’t invent the max function, and weren’t the first to call it max, so why would it be *their* copyright that’s being infringed? Isn’t Oracle just an infringer of some unicoder that wrote a protectable max function in that case?
It would be more like you named your book “How_to_Hunt_Bears_427” and someone copied the name of your book because your book was selling so well.
I hope you realize this is backwards analysis. The motive for someone copying only matters if you have a copyright to infringe for fair use or willful infringement. If you have no copyright and I copy your name, I could have the worst reason in the world and still have no liability. This is black letter, unassailable law.
Since “how to hunt bears” is merely descriptive of the contents of the books and not creative expression, it isn’t protectable. That’s also beyond debate, though you seem to have problems porting this analysis to the computer function realm.
Then the only question is whether “427” in conjunction with the name is creative expression. If I copy “427” but it isn’t expression, and I do so “because your book is selling so well” you have no copyright claim against me. I dispute that the “427” even exists in the actual fact pattern here, but assuming it did, you’d still have to explain why this is a creative expression that renders the title copyrightable if you’re going to prevail (something a jury made clear Oracle didn’t do). You can’t just point to it and say “See, 427, and you also say 427, and my book sells well, I win!”
Just ridiculous. What is the matter with you?
Let’s try an analogy for you to see if you can figure out who is ridiculous. You write a fictional tale like Lord of the Rings. I sell a magazine like entertainment weekly or some such, which I charge money for. In my magazine, I review lord of the rings. I do not copy all of the text in my magazine but do give some samples passages and summarize the plot. I definitely repeat the title of Lord of the Rings over and over.
In this fact pattern the title is unquestionably protectable, which is not true of Oracle’s name. The amount of the actual content besides the title I have taken is more than zero, i.e. more than what Google took. I unquestionably am making subscription money in part because of my use of your book title.
Do you think, given all of these advantages you have over Oracle’s position, that you would prevail in a copyright claim against my magazine?
“Not without running up against a functional barrier.”
See link to patentlyo.com below.
You’re right. I agree that statement was not related to the argued issue, but that’s because Night’s statement (“They are many other names that could be used to do the same thing.”) was not related to the argued issue. Night advances so many illogical arguments at once I mistakenly let the tail wag the dog by stepping into merger for a sentence.
I’m kind of shocked allegedly educated people are having such a problem with this. I could get a folder and name it Tax Documents and put my tax returns in there, but if I went around suing everyone else who also names a folder “tax documents” for copyright infringement I’d be laughed out of court. The fact that I could also name the folder “tax returns” or “income statements” or “government papers” doesn’t make my name expression, because multiple options does not equal selected option being expressive. Nor does the fact that I ontologically group my papers by category or function (i.e. I focus on the “tax” content rather than say the 8.5×11 size feature) make it expression. Moreover, it really doesn’t make any sense that *I* get to sue, because I’m sure I didn’t come up with either the “tax documents” name nor the ontological grouping even if they somehow were found to be expressive.
Now it’s also true that even if you found it expression that the merger doctrine would apply, and that would be an additional reason others are not liable. And then I would assume the vast majority of uses would also be fair use uses, and that would also render people not liable. But that’s besides the point, and you’re right I shouldn’t have gotten into it.
One example that Breyer focuses on is java.lang.math.max(). Max is descriptive (not expressive) is a functional descriptor (merger) and was not created by Oracle (not the holder if there is any protection). For the same reasons, Math is not an expression oracle created. Grouping Max under Math is not an ontological expression that Oracle created. Making math a category within the overall library is not something Oracle created. The only thing that oracle arguably created is the truncation java.lang as the library name. But java lang is just a description that the library is in the java language. It is akin to a folder named Tax Documents or a bear hunting book named How to Hunt Bears. For this clearly protectable and wonderful advancement to human expression, Oracle proposes with a straight face that it is not manifestly clear unjust enrichment to give them millions of dollars. We all know that java lang has no actual expressive worth, it’s just a hook by which Oracle wants to profit from its underlying functions and system that Google did not choose to license and DID NOT COPY. Unsurprisingly, a jury found for Google. But no, it’s *my* position (that commands both the opinion of the lay jury and the supreme court) that is ridiculous according to Night.
“ because Night’s statement (“They are many other names that could be used to do the same thing.”) was not related to the argued issue.”
Actually, that IS a solid legal point.
That you think it to be a tangent, well, that’s you being you.
“I’m kind of shocked allegedly educated people are having such a problem with this.”
LOL – no kidding. There’s a phrase that describes this type of false confidence that you have.
And hint: your position (just the end result, but not understanding the proper means) is NOT something that you should be patting yourself on the back for. That’s part of your tendency of serving yourself only those things that align with your desired world view.
You are analogizing fair use in reviewing a book with copying an API and code for the purposes of competing with Oracle’s product?
I hope you are joking.
Come now Night Writer – you know that he is not.
Do you really think that Random has ANY clue about what he is spouting?
^
smh
This (Random’s musing) the type of Ends justify the Means reasoning and lack of touch with the points of Law that stops us from having nice things.
lack of touch with the points of Law
Just another Scoreboard example that I called way back during brief filing at the CAFC level, iirc. How terrible that the crack in the Supreme Court’s scoreboard has now extended to copyright law in addition to a myriad of patent issues, huh? Surely Oracle will abandon Java prompting Congress to spring into action to fix all of the catastrophic actions by the supreme court, right?
“that I called way back during brief filing at the CAFC level, iirc.”
LOL – and what did I call (hint: it was a lot more involved than merely recognizing that an anti-Article I/Section 8/Clause 8 Court would do.
You do not win even a Kewpie doll for that call.
As I also recall, you have no reply to my calling out the “Copyright is enough protection” walkback going on – OR on the duplicity of “well, all software is functional, of course (that is, until the patent people start talking)”
You still don’t.
This case does not stand for the proposition that you can copy code. It does not even stand for the proposition that you can write your own code that achieves the same function (though I think that is independently clear). It only stands for the proposition that when you name your function, you don’t have to provide a translation key that says that My Name X achieves the same result as Oracle Name Y. It is entirely consistent with copyright everywhere else outside of computers – Dennis came up with the name PatentlyO, but you’ll notice I don’t incur liability just by typing the word PatentlyO. Give it up with the sky is falling nonsense.
“well, all software is functional, of course (that is, until the patent people start talking)”
All operative code by definition is functional, and I don’t know of anyone who seriously advances an argument otherwise. Speaking only for myself, I only argue that the fact that it is functional doesn’t mean it can be identified in claim solely by its functionality, which has been the law for hundreds of years and well predating computers. All machines with utility (which has always been a requirement) can be described by the functional result of their use. But that is not a proper way of claiming, because functional result is descriptive but not inherently uniquely identifying. To claim by functional result rather than by describing posited means which achieves the result inherently threatens to sweep in unposited means within the claim scope, and is overbroad of what was invented.
“This case does not stand for the proposition that you can copy code”
Actually, yes it does. Maybe you want to revisit the decision again – the copying was held to be Fair Use copying.
Your “not identify by its functionality’ was viciously torn about by David Stein. Repeatedly, as I recall.
Also, you err (gee surprisingly) on the difference of PURE functional claiming, words sounding in function, and Prof. Crouch’s coined phrase of Vast Middle Ground.
But we have ridden that merry go round enough, eh?
You know a decision is going to be light on actual reasoning when it starts with hyperbole… computers give answers “instantly” (p. 4) and Java includes an “uncountable number of method calls” (p. 9). At least I assume it is hyperbole and the court doesn’t actually think these things are true!
Two questions:
1. When was the last time SCOTUS ruled in favor of IP rights . . . of the right for creators to keep for themselves that which they have created?
2. Will they ever do so again . . . even with the addition of 3 new judges . . . whom one would expect to uphold such rights?
Bonus question: Is Google “too big” to be held responsible for their IP theft?
1. When was the last time SCOTUS ruled in favor of IP rights . . . of the right for creators to keep for themselves that which they have created?
Bonus question: Is Google “too big” to be held responsible for their IP theft?
You know google wrote its own code here right?
You mean, except for the code of the lines in question, right?
You mean, except for the code of the lines in question, right?
Yes, except for those.
So, you are not correct on that basis….
Thanks.
Another observation too is that who would ever develop another system like Java after this? The company must know that a large tech company can just come and take the system if they feel like it and there is nothing you can do about it.
So probably in the short-term this decision helps businesses and the economy but as it the case with most of the recent decision gutting the patent system, this decision will put a halt to the development of future work.
Who would spend time working on something knowing that it can just be taken by a big company if it appears it will be worth something?
This is really what it is all about:
Thomas on the dissent:
Oracle spent years developing a programming library that successfully attracted software developers, thus enhancing the value of Oracle’s products.1 Google sought a license to use the library in Android, the operating system
it was developing for mobile phones. But when the companies could not agree on terms, Google simply copied verbatim 11,500 lines of code from the library. As a result, it erased 97.5% of the value of Oracle’s partnership with Amazon, made tens of billions of dollars, and established its position as the owner of the largest mobile operating system in the world. Despite this, the majority holds that this copying was fair use
I’d point out too as a former software engineer who designed systems that often that “11,500” lines is the core of the design and you can hire lower level people to code the rest of the system.
B-b-b-but it was “transformative.”
Hmm, does this type of “transformative” cross the stream and ‘count’ in the patent sense of “transforming’ a thing to a different state (for satisfying eligibility)…?
“transformative” it was not. And any software engineer would get a belly laugh out of that.
(I know you feel the same way.)
Correct.
But my point is that IF the a court is going to do their handwaving “transformative” (and force that same handwaving to infect and overcome factors to arrive at opposite conclusions of reasonable analysis), then that same notion of “transformative” — in all of its ‘reach’ — should equally apply in the patent sense when evaluating when a patent claim “transforms” something.
As the Court also indicates, software IS functional, and thus DOES something, and that “doing” is the essence of “transforming.”
As I pointed out in multiple discussions in the lead up to this case, the “game playing” to deny IP protection (fostered by the errant ‘binary’ either/or view) necessarily has ramifications that the Court does not either address or even seemingly recognize.
I had put this dichotomy on the table for “academics” to clarify.
Not surprisingly, those that are anti-patent would not touch the necessary ramifications that software has utility aspects — and such are the immediate focus of protection under the patent system.
It is a denial of inte11ectual honesty to not see what underlies these points.
This is a good point anon.
You see lots of comments below as if there would only be one way to write an API that performs the functions of Java, which is a ridiculous contention as the way in which the functionality is divided and provided could be quite different.
That notion is not a ‘bad’ notion in and of itself.
Loss of protection IS a reality when the Merger Doctrine is shown to hold.
That being said, I do not recall precisely, but I seem to remember that the Merger Doctrine argument was tried, lost, and was not preserved through appeal.
Breyer is clueless about how cut throat – and winner take all – the network effect companies are. He thinks he is making a pro-competition decision here!! LOL, all it’s doing is letting google expand it’s advertising monopoly into the handheld market.
Perhaps not as much as you might be imagining.
Stranger bedfellows have been known to happen, and the Big (established) Tech has been more than willing of late to align the the Liberal Left that Breyer ascribes to.
When cert was taken in this Google v. Oracle case I had noted as a warning that the Sup. Ct. had split 4-4 in Lotus v. Borland years ago on comparable fact issues. [But not being a decision it was not going to get cited.]
Paul this was a pretty obvious remark. But one that should be made.
One academic commentator who has followed this situation for 9 years did just note that: “The court’s fair use analysis was influenced by Judge Boudin’s concurrence in Lotus v. Borland, though that concurrence didn’t actually call it fair use, but instead ‘privileged use.’ ” [In the 1st Cir. Lotus v. Borland decision, not the Sup. Ct. split decision that sustained it.]
When you order hash browns at waffle house the ordering is like an API; hash browns (smothered covered diced). this means hash browns withe cheese, onions and tomatoes. The program is the actual cooking, potatoes uses, cheese,onion, and tomato type used. Google copied the API of smothered, covered, and diced but the cooking was original. So could Waffle house copyright smothered,covered,and diced and all other hashbrown topping combinations? Would it make a difference if they tried to copyright cheese, onions, and tomatoes without the added flair?
More like, google copied the entire menu – verbatim, including the organization grand slam, breakfast, lunch, desserts, etc. And yes, why can’t you copyright a menu?
Two quick thoughts:
1) The lack of opinion as to the first question presented of whether copyright protection extends to an API suggests a potential 4-4 split that would theoretically get decided by Barrett if it came up again to the Court any time soon (seems unlikely).
2) More tangentially, I think the biggest thing that struck me reading this opinion is that while function calls and APIs are extremely simple ubiquitous concepts in software introduced near the beginning of any intro level software class… the explanation of APIs still took up five pages and was treated with 5+ analogies to non-computing contexts. It will be interesting to see what happens when machine learning, gene editing, and other more complicated technologies start to become huge issues, not just in copyright and patent law, but in other areas of law as well. I cannot help but be reminded of Scalia’s honest words in Myriad: “I join the judgment of the Court, and all of its opinion except Part I-A and some portions of the rest of the opinion going into fine details of molecular biology[;] I am unable to affirm those details on my own knowledge or even my own belief.” Ass’n for Molecular Pathology v. Myriad Genetics, Inc., 569 U.S. 576, 596 (2013) (Scalia, J., concurring).
On point (2), you cannot fault the Court for providing an explanation of APIs and how they work to better understand the opinion. But I did find it odd that the Court referred to the Java API as a “user interface,” which it clearly is not.
Agree on both points. On the first point, I should have been more clear, as I in fact think it is unquestionably a positive for the Court to take the time to include such technological background. I was merely intending to note that, in reading the opinion, the care taken to provide that background brought to mind the challenge of taking the same care and providing the same treatment for cases involving much more complex technologies.
So google wants to use Java, so they copy the java.api to the letter. Meh, sounds like the CAFC had it right and SCOTUS wrong. But the finding that Sun was never going to compete with android? Somebody pulled that out of thin air.
Is this insight into J. Thomas’ unexpressed, undergirding for his Alice opinion: “Copyright law generally protects works of authorship. Patent law generally protects inventions or discoveries. A library of code straddles these two categories. It is highly functional like an invention; yet as a writing, it is also a work of authorship. Faced with something that could fit in either space, Congress chose copyright.” Any future SCt argument needs to address this misunderstanding of software protection by J. Thomas, because software includes BOTH protectable expression (e.g. the source code text itself), functional when executed to utilize and control processor and other computer resources as with many patentable processes PLUS creatively functional, such as UI elements, for design patents. There is no technical or legal basis for J. Thomas saying that copyrighting or patenting software was/is a binary decision by Congress.
Yes, but fortunately this is only a small minority opinion.
The point though is that this is a fundamental lack of understanding — there should not be a Single Supreme Court Justice that so misunderstands the underlying nature of intellectual property.
So, maybe it is time to face the facts: Software needs its own sui generis form of protection. Whether one tries to protect it by patent or copyright, Scotus says no.
Remember that AAA JJ says this is OK. That you don’t need a science background to understand engineering and technology.
We are in a “woke” society. If J. Thomas believes this, then it must be true and we must all adopt his beliefs.
Please obey.
I blame Academia – and specifically the woke Academia of “Liberal Arts,” where meritocracy and teaching actual critical thinking skills was long ago replaced with the ‘upward path’ of how well you hew to the orthodoxy of above.
AAA JJ’s desire for this extends well beyond this latest snafu that you have with him, and is evidenced each and every time a Liberal Left dogma point raises its head (and all reasoning vanishes from AAA JJ’s posts).
As usual, Thomas is an iconoclast and Alito just hates giving the peoplez what they want, no matter the dimensions of the legal question.
Also as usual, the only lens that coherently makes sense in the Information Age is the question of the entity using the information. The baseline that copyright is for expression and patents for function breaks down when function and expression are merged, but it need not if a simple corollary is accepted: machines are incapable of producing or consuming expression and human beings are incapable of producing or infringing patentable utility.
API calls are consumed by machines, even if written by human beings. Therefore, no copyrightable expression should exist for an API. Software output, on the other hand, when consumed by persons (like a video game or word-processor) is expressive and should be covered by copyright. The same goes for software in a patent sense. API’s, if new and useful and non-obvious, should be protectable by patent, while information output of software consumed by persons should not be.
I assume that in the loooong run, this lens will eventually be the one that intellectual property protection for information assets will use. There is no other rational frame that I know of. The roots of various rules in statute or common law or the hearts of Congress are meaningless when a rational frame cannot be applied to facts on the ground. Laws that cannot be enforced will not be enforced.
Some sense to what you wrote, but output that is consumed by a human should, of course, be patentable.
Fixed that for you:
“output that is consumed by a human should,
of course, not be patentable.And yet, traffic lights – whose output is ONLY consumed by humans, are eminently patentable.
You continue to push your vain but futile misapprehension of patent law utility….
At least some autonomous vehicles are programmed to obey stop lights as well as stop signs.
Thanks Other — but certainly not at the time of the invention, eh?
Do you really think that Marty’s point holds?
(That’s an easy “Yes/No” binary question)
A new, non-obvious and fully described traffic light should be completely eligible. Why not?
A method whereby people stop when they see a red light should be completely ineligible.
Making the traffic light infringes. Stopping one’s car does not. For the hundredth time.
“For the hundredth time” — you still do not get the patent sense of utility.
Making a thing — without recognizing the necessary utility of its use — does NOT get you past 101.
Just curious…
How many places do you paste this stuff around the internet?
Ben you talking to me?
The usual spots, plus major media on the unusual occasion of IP news that breaks thru, like this case.
I saw it over on WaPo and was just curious.
IDK, I think one can see why this is probably wrong because if you could copy all the APIs, then you could send out the work to program the functions of the API and thus copy the program without doing any design work.
But, of course, the Pharaoh had city plans, which are very similar to APIs and the city plans could be copied without penalty.
Wow, that really bytes!
But seriously, I am surprised by that ruling as I was thinking the same thing with the book analogy (i.e. You can publish a book but not by copying someone else’s, but by entering all the words yourself). Am I wrong? (Note I’m not a lawyer or patent agent so just curious).
And re: the comment that copyright doesn’t protect software. If that doesn’t, what does protect software? It sounds like patents are out..
yes, you are wrong. entering the words in yourself would clearly be copyright infringement, and would almost certainly not qualify for fair use.
this is more like if a set of really well known books had a very specific chapter naming sequence, then you could potentially use that very specific chapter naming sequence for your own book, as long as all the actual content in your book was new and not derivative. at theoretical baseline there may be some difference between when chapter naming ends and actual content begins, but from afar it is pretty obvious they are different.
meant at theoretical baseline there may be some difficulty distinguishing between exactly where chapter naming ends and actual content begins (e.g…. between where declaratory code ends and implementing code begins)
Ah I think I understand a bit more now. Thanks!
I had a similar thought about chapters, but I think that might not even work considering that same-chapters-different-book could be a pretty direct market substitute, and would then have have a different “effects” analysis.
The key aspect that the “table of contents” analogy misses is that the specific names used in an API must be copied in order to develop an underlying implementation that is interoperable with code written for the original API. If you change a function name, then code expecting the original function name won’t compile or run.
A better (though not perfect) analogy would be a dictionary. The vocabulary words in the dictionary are references to underlying ideas, and those ideas are given form through the definitions. The definitions need not be written in any particular way, as long as they implement the ideas referred to by the words being defined.
What Google did is somewhat like copying the vocabulary words, but then writing their own definitions for those words. If they don’t use those words, then the resulting dictionary is useless for people speaking whatever language it is.
(Again, this isn’t a perfect analogy – in a naturally occurring language, the vocabulary isn’t copyrightable. And one could easily argue that the vocabulary from a constructed language has enough differences under the fair use factors that the test would turn out differently than it did for Google and Oracle here.)
Interesting analogy. Thanks!
I don’t think the chapter naming sequence analogy is helpful. Oracle used that analogy in their first successful attempt at the CAFC – the Java API was compared to chapter names in Harry Potter book which Google copied and then filled in the rest of the novel.
My guess is that the Supreme Court would have decided that this very thing does violate copyright law. They specifically said in their opinion that fair use has to be decided differently for books, movies – and software; moreover within software API declaring code is different from implementing code.
What is very specific about APIs is that they enable developers to write programs. There is no such thing for books: other authors don’t care about the chapter headings of Harry Potter books to write “compatible” books.
The court essentially said that the muscle memory of developers who invested in learning the Java API cannot be monopolized by Oracle.
bigmushroom,
A substantial portion of your comment is focused on a utility aspect of the particular software.
The proper point to keep in mind is that having a utility aspect does NOT prevent an item from having an expressive aspect.
To the extent that those on the Court (and elsewhere) postulate that somehow protection is necessarily binary, these are simply not understanding what the Law is.
the chapter naming sequence works fine if you assume that the books are so well known that readers have trained themselves to recognize the naming scheme. like, readers know when they open the book, they will turn right to chapter XX.Y.ZZZ because thats where some important information is…
Honestly, this reminds me more of the phone book case, where each phone number calls a specific person via a phone network.
However, rather than reproducing the phonebook alone, google built its own interchange and network able to reach the same phones using the same numbers.
I think I’d really want to see what the arguments are for how unique and strong the creativity of the commands are.