License to Copy: Your Software Code Isn’t Safe After Google v. Oracle

By Gene Quinn
April 6, 2021

“What is the point in obtaining a copyright in software? Not much, if Google can copy 11,500 lines of your code and it can still be a fair use. Copyrights just don’t protect computer code from actors like Google.”

copyright, softwareIn characteristic form, the Supreme Court has once again managed to blow it in another intellectual property case. This time, the Justices blessed Google’s copying of Oracle’s code and called it fair use despite the fact that Google copied that portion of the Sun Java API that allowed programmers to use the task-calling system that was most useful to programmers working on applications for mobile devices.

In the infinite wisdom of the Supreme Court, the copying of this code was found transformative because Google only used it to circumvent the need to license Java from Oracle with respect to Android smartphones. Of course, that isn’t exactly how the Supreme Court characterized it, but make no mistake, that is what they decided.

A Tortured Attempt to Excuse Google’s Behavior

The Supreme Court explained there was evidence that Google’s copying was limited to Android, and the deployment of Google’s infringing product limited to smartphones. Java, on the other hand, was developed for use with desktop and laptop computers. Of course, if this were a software patent case, smartphones, laptops and desktops would all be synonymously considered generic purpose machines that are indistinguishable, but this was a copyright case where Google stood to benefit, so no such distinction was made.

“Google, through Android, provided a new collection of tasks operating in a distinct and different computing environment,” Justice Breyer wrote for the majority, as he strained to find a difference without a distinction.

The Supreme Court made much of the fact that Google wrote millions of lines of code originally. And all they did was copy from Oracle what was necessary to forward their own desires. The Court explained: “Because Google wanted millions of programmers, familiar with Java, to be able easily to work with its new Android platform, it also copied roughly 11,500 lines of code from the Java SE program.”

Obviously, since it is in the interest and benefit of all those millions of programmers who are familiar with Java, Google should be allowed to copy Java rather than be expected to negotiate with the copyright owner and negotiate a license.

What is the point in obtaining a copyright in software? Not much, if Google can copy 11,500 lines of your code and it can still be a fair use. Copyrights just don’t protect computer code from actors like Google. In a sad, ironic twist, the infringement at question isn’t one of the free services that Google gives away. Android is extraordinarily commercial. According to the dissent, in 2015 alone, the year before the fair use trial, Google earned $18 billion from Android. And as of 2015, Android sales had totaled $42 billion.


What About Harper & Row?

Excusing the intentional, knowing and overwhelmingly commercial use of Google calls into question the vitality of copyrights and the future applicability of fair use in the digital age. If copying the heart of Java is fair use, this decision implicitly overrules the Court’s decision in Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 U. S. 539 (1985).

In 1977, former President Ford contracted with petitioners to publish his as yet unwritten memoirs. The agreement gave petitioners the exclusive first serial right to license prepublication excerpts. Two years later, as the memoirs were nearing completion, petitioners, as the copyright holders, negotiated a prepublication licensing agreement with Time Magazine under which Time agreed to pay $25,000 ($12,500 in advance and the balance at publication) in exchange for the right to excerpt 7,500 words from Mr. Ford’s account of his pardon of former President Nixon. Shortly before the Time article’s scheduled release, an unauthorized source provided The Nation Magazine with the unpublished Ford manuscript. Working directly from this manuscript, an editor of The Nation produced a 2,250-word article, at least 300 to 400 words of which consisted of verbatim quotes of copyrighted expression taken from the manuscript. It was timed to “scoop” the Time article. As a result of the publication of The Nation‘s article, Time canceled its article and refused to pay the remaining $12,500 to petitioners. Petitioners then brought suit in Federal District Court against respondent publishers of The Nation, alleging, inter alia, violations of the Copyright Act (Act). After considering the four fair use factors listed in 17 USC § 107, the Supreme Court found the use to not be a fair use, in large part because The Nation copied the heart of the story.

License to Copy

As the Federal Circuit explained, and Justices Thomas and Alito agreed with in their dissent, the teaching of Harper & Row is that the single most important fair use factor is effect on the market for the copyrighted work. And, until today, Harper & Row stood for the proposition that a small fraction of the whole 300 to 400 words of an entire Presidential memoir, can be infringing when the value of the copyrighted work is taken. Here, Google’s copying of the Java code and use in the multi-billion-dollar Android platform has unequivocally destroyed Oracle’s ability to license. And why would anyone else license Java from Oracle if Google is allowed to copy it as a fair use?

This decision by the Supreme Court is a travesty of the first order. It will only embolden copyists, harm creators and disintegrate value. If something like this happened with Disney being on the short end of the Supreme Court decision and compromising the value and copyrights held in Mickey Mouse, there would be a bill filed to overrule a renegade Supreme Court decision before the close of business. I’m not going to hold my breath, which means Google appears to have a fully paid-up license to copy whatever code they want, regardless of how they destroy the copyright owner’s market and regardless of how many tens of billions, or hundreds of billions they make.


Image Source: Deposit Photos
Author: PirenX
Image ID: 12182996

The Author

Gene Quinn

Gene Quinn is a Patent Attorney and Editor and President & CEO ofIPWatchdog, Inc.. Gene founded in 1999. Gene is also a principal lecturer in the PLI Patent Bar Review Course and Of Counsel to the law firm of Berenato & White, LLC. Gene’s specialty is in the area of strategic patent consulting, patent application drafting and patent prosecution. He consults with attorneys facing peculiar procedural issues at the Patent Office, advises investors and executives on patent law changes and pending litigation matters, and works with start-up businesses throughout the United States and around the world, primarily dealing with software and computer related innovations. is admitted to practice law in New Hampshire, is a Registered Patent Attorney and is also admitted to practice before the United States Court of Appeals for the Federal Circuit. CLICK HERE to send Gene a message.

Warning & Disclaimer: The pages, articles and comments on do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of Read more.

Discuss this

There are currently 59 Comments comments. Join the discussion.

  1. The sky prince April 6, 2021 10:12 am

    I mean, what were you expecting exactly?

    Suppose I write a function:

    loopntimes(n) {
    for (int i = 0, I <n, i++) {

    There can be something inherently expressive about the name, loopntimes, but it exists so people can call it right? So other programmers can do


    And print it. If Java held copyright that extended to the mere usage of loopntimes that would defeat the whole point of having an API in the first place right?

    Oracle never seriously contested this, their point was that if someone created their own loopntimes:

    loopntimes(n) {
    I = 0
    while I < n:
    print i
    I ++

    That that is not ok. Which is not a totally unreasonable point, but not a great one. Copying loopnfimes simply says to a programmer you will be calling this functionality, it doesn't copy the expressiveness of the code. And Google used its own implementing code.

    And the court simply made a fair use argument, it didn't even reach that apis were uncopyrightable. It should have, but it didn't.

  2. John White April 6, 2021 10:44 am

    Ah, another victory for the monopolist. Isn’t this great? Copy what you want, destroy the value and incentive structure of the IP system. This is plainly wrong. Period. But, for now, that doesn’t matter.

  3. Anon April 6, 2021 10:48 am

    The sky prince,

    I hear what you are saying – and the legal principle is basically a coalescence of expression into function.

    If I recall correctly, that legal argument was not persuasive to the court (and not revisited in the appeals).

    Once again, a critical theme must be reinforced: protection is NOT binary between the different realms of intellectual property protection.

    An item (such as software) may have multiple aspects, and thus may earn multiple forms of protection. An item may earn protection under patents for its utility aspects, and simultaneously earn a different protection under copyright for the different aspects of expression.

  4. Gene Quinn April 6, 2021 11:13 am

    The sky prince…

    You miss the entire point. Copyright is about copying. What Google did was copyright infringement up until yesterday because they literally copied. If Google did not copy and did it themselves they would not have been liable for copying because there would not have been copying.

    Google took the Java code because programmers were familiar with the code and they didn’t want to have to create their own code and require programmers to become familiar with the way they would do it. That was accepted as true by the Supreme Court. So, Google stood on the shoulders of Oracle for free.

    There are now no limits to the copying Google and others similarly situated can do. Copyrights in software are now useless. They were only ever good against cut and paste copying, and now even that is a fair use.

  5. BP April 6, 2021 11:50 am

    The oral arguments pointed to Breyer notching a win for Google, just like he did in Mayo. Maybe Thomas saw the damage he inflicted with Alice? From Mennell 2018: “Copyright’s low threshold for protection, complex scope, broad array of rights, and long duration created a risk of overbroad protection for computer software products.” And, certainly: “Patent’s low threshold for eligibility . . . created a risk of overbroad protection for computer software products.” Hence the SCt delivers fair use (Google v. Oracle) and efficient infringement (Mayo/Alice). Mennell’s position (as supported by Google) is quite clear and Breyer likes it. Ignoring IP rights is OK if you’re Google.

  6. Judith Perkins April 6, 2021 2:10 pm

    @BP “Maybe Thomas saw the damage he inflicted with Alice?”

    Have you seen the damage inflicted by software patents?

    Copyright is fine for software, although it lasts too long, but patents for software development is a perfect nightmare.

  7. Pro Say April 6, 2021 2:11 pm

    “All your IP are belong to us.”

    — Google

  8. John April 6, 2021 2:16 pm

    This article is on the money and shows what a ridiculously low caliber of brainpower sits on that court. This holding manages to be more moronic than the case that gave the little old ladies house away to the golf country club. No good people want these jobs anymore, so we get a court full of morons. These days, when the court gets something right it is the exception, rather than vice versa of old.

  9. CuriousObserver April 6, 2021 2:29 pm

    What if the copyright was only on the API? That is what Google essentially copied, and that was all that was needed in Android to allow any java program written by third party to work on Android. Had the API been the focus of the copyright, Google’s implementing code in the Java Virtual Machine (JVM) would have been irrelevant. The value of the Java API is allowing third party applications to run on any platform (windows, android, linux, mac-os, etc).

    Hypothetical question, considering this is a copyright and not a patent, what would have been a reasonable expectation for royalty payments had the decision gone the other way?

  10. Pro Se April 6, 2021 2:46 pm

    Oracle can take their Copyright fight outside of the U.S. to Berne Convention Treaty jurisdictions and win. This is every major market outside of the United States, 170+ countries.

    The United States is currently in a war against patents and copyrights, not the rest of the world.

    Oracle can take this same fight up to Canada or down to Mexico and easily win, the U.S. skewed IP protections (like the need for registration before you’re able to go to court) is not followed by all of the other major Berne member countries.

    Oracle has bigger and better global options of enforcement.

  11. Joachim Martillo April 6, 2021 3:24 pm

    There is a confusion in the discourse over Google v. Oracle. An API in the C programming language is an abstraction that C programmers use to describe a set of source and header files (or functions and data structures). In Java an API is program code and an interface is a reference type that is implemented with actual Java code.

    One must contrast Oracle documentation

    with Feist Publications, Inc. v. Rural Telephone Service Co., 499 U. S. 340, 345 (1991), which explains (a) that copyright requires some original “creative spark” and (b) that therefore a phone book is not sufficiently creative.

    A C API might possibly be analogized to a phone book although the argument would be forced because creating the abstraction of C API at least requires the “spark” that SCOTUS seeks.

    I can make the point in another way. Any idiot can compile a list of phone numbers and names. Creating a C interface requires much deeper knowledge and training in software design.

    A Java API is a beast, which differs from a C API, as the link to the Oracle document indicates.

    Is SCOTUS telling us that a prolific writer is much less protected by copyright than a writer, who only writes infrequently?

    Maybe Java programmers should switch to Scala, which is another JVM language that does not have an interface reference type.

    I will have to read through the filings of this case. I have to wonder whether the issues were poorly explained.

  12. Curious April 6, 2021 4:59 pm

    Under 17 USC 107, the law sets forth four factors that SHALL be used “[i]n determining whether the use made of a work in any particular case is a fair use.” The first of those factors listed is:
    (1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes

    Clearly, not every fair use (or not) can be pigeon-holed as “commercial nature” or “nonprofit educational purposes.” However, I do believe that these represent good examples as to the two ends of the continuum. Notably, both refer to business in the making of money. On one end of the continuum is a use that is of “commercial nature.” If you are making money of the copyright, then odds are you are depriving the owner of their ability to make money — and the statutory law intended that this is very likely NOT to be fair use. Conversely, if the use is “nonprofit” (i.e., not making money), then the use is far more likely to be a fair use. Additionally, the added factor of “educational purposes” adds the idea that “education” is an important factor that could offset the rights of a copyright owner.

    What Gene has noted and what no one has disputed is that Google’s use was certainly a commercial endeavor. This is what the Court itself wrote:
    “So even though Google’s use was a commercial endeavor—a fact no party disputed … —that is not dispositive of the first factor, particularly in light of the inherently transformative role that the reimplementation played in the new Android system.”

    The term that I found interesting was “transformative,” which is something discussed both in the majority opinion and in the dissent, which the majority stating that the “transformative role” of Google’s use was enough to overcome its commercial nature.

    Of very particular note, the term “transformative” is not found in the statute (does this sound familiar to anybody?). After doing a little bit of research, it appears that this “transformative” factor was added by the Supreme Court back in Campbell V. Acuff-Rose Music. Here, Souter wrote:
    “whether the new work merely “supersede[s] the objects” of the original creation or instead adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message; it asks, in other words, whether and to what extent the new work is “transformative.” … Although such transformative use is not absolutely necessary for a finding of fair use … the goal of copyright, to promote science and the arts, is generally furthered by the creation of transformative works.”

    This is another example where the Court ignores the plain text of the Patent/Copyright clause of the US Constitution, which states: “To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries.”
    The Constitution does not describe promoting the progress of science and the useful arts by permitting copyists to create “transformative” works. Rather, the Constitution states that this promotion is by securing for authors “the exclusive right to their respective writings.”

    This “transformative” factor was created by the Supreme Court out of thin air (and contrary to the Constitution) to bless infringer’s rights. Campbell, which, created this mess could have easily been decided on the third factor (i.e., “(3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole”). However, as SCOTUS is wont to do, they created this factor/test for which they alone know the answer to and with which they can justify whatever result they want. They want copyright infringement, then the work isn’t transformative enough. If they want copyright infringement, then the works is transformative (even though the transformation was solely for commercial purposes).

    I suppose that the rewriting of laws has been going on at the Supreme Court — probably since its existence. However, because justices are so hard to remove (and their writings are so esoteric that the vast majority don’t even understand what they are writing), they know that they can get away with it with little to no repercussions. The Supreme Court really is a sad state of affairs.

    Oh, and for those that think Republican-appointed justices will be more sympathetic to property rights, I need only point at that three of them signed onto this opinion.

  13. LazyCubicleMonkey April 6, 2021 5:15 pm

    If you want to actually support the promotion of public works, then locking in public API’s would do the opposite of that – as it would lock customers into an API and prevent anyone from competing.

    Could you imagine copyrighting a function like Math.add(a, b)?

    The small portion of copying 11,500 lines out of 2.8 million represents the copying of functional/declaritive interface of things like Math.max(a, b), list.size(), etc. Do you think it would somehow be preferable to have to make those Math.maxGoogle(a, b) and list.sizeGoogle()? And that it would promote the arts and sciences by doing that?

  14. John April 6, 2021 6:02 pm

    Judith, your statement is perfectly ignorant. Congrats. You qualify to make it onto the Supreme Court some day.

  15. Anon April 6, 2021 6:08 pm


    smh – so many missteps…

    When you start with “actually support the promotion of public works” in a forum designed to promote innovation with Quid pro quo of private property rights, you should recognize that you are already in the weeds.

  16. Gene Quinn April 6, 2021 7:04 pm

    LazyCubicleMonkey @13…

    Your comment shows the type of lazy understanding of the facts and law that your name prepares us for in your “analysis”, if you can call it that.

    Promotion of public works, where you start, is a total misfire. Copyrighted works still under protection are by definition not public. But nice try.

    Like so many who have tried and failed before you to convolute software and math, protecting a copyright in software (or even a patent for that matter in a software system or product) is not tantamount to granting exclusive protections on a mathematical equation. As Google admitted and the Supreme Court acknowledged, Google could have written original code to do what they wanted, but they felt that would have been a burden on programmers who would have had to learn the Google system. So instead of writing their own independent and original code (which could have been done) they cut and paste the Oracle code to save themselves time and money.

    Only a truly the truly bankrupt could even attempt to rationalize this as fair use, or anything other than stealing.

    Up until yesterday a software copyright protected only cut and paste copying. Now software copyrights don’t even protect intentional cut and paste willful copying. So, ironically, copyrights in software protected very little. That means the overwhelming weight of this ridiculous decision will be born by other creators. The death of fair use when cut and paste is fair use.

  17. BP April 6, 2021 7:58 pm

    @6 Judith Perkins: “Have you seen the damage inflicted by software patents?”

    Of course, billion dollar judgements against infringers. And so? When’s the last time you represented a client that invested millions in product development to have its product ripped off by a larger entity?

    The billion $ judgments sparked heavy lobbying against independent inventors, startups, the USPTO, universities, you name it. They were all labeled “patent trolls”. We ended up with the AIA and an unconstitutional hit panel handpicked by Google’s former head of IP, where ethics mattered little and where panels were stacked to get the “right” outcome, i.e., invalid.

    History is an amazing thing, if people are willing to search for the truth and facts.

    I started programming in the 1970s, machine language, peek, poke, etc., and ended up at a supercomputer institute. But Ms. Perkins says that if I write a code for a new controller (say controlling a complex system, that actually makes a useful product, while maintaining that system in an unsteady state) I can’t/shouldn’t have protection? Because, in the misguided opinion of Ms. Perkins, too big of a risk of “damage inflicted by software patents”.

    Ms. Perkins, stop reading Prof. Mennell, EFF, etc. Talk to people that worked in industry to build real sw driven products that do real things.

  18. A time for every purpose April 6, 2021 11:52 pm

    Everyone is of course aware that copying code, straight up copying, is very common practice in the coding and software industry right? I’ve done it, everyone has done it. People encourage it. Github encourages it. Most software is open source.

    The court did not say you can copy everything. It said you can copy declaring codes. From what was said earlier, I declare Math.add to tell the programmer, hey this adds.

    Am I really going to force programmers to learn 50 different words for adding to call that command? I’m pretty sure you can’t copyright the concept of adding anyhow, and you can’t copyright the label of it as adding.

    Java had every right to charge a fee to use the platform, it didn’t. It made it free. Matlab isn’t threatened by this. Even if someone copies Matlab declaring codes, implements it on their own, they probably cant recreate the matlab environment without violating something, because its completely proprietary. Its also completely incompatible with almost everything.

    Java can’t come back and attack its users from using its free platform. Its interoperable because anyone can use it and attach it to anything. Is using it now a copyright violation? Wouldn’t that threaten innovation through use of the platform?

    And sure, Google business practices were somewhat underhanded here, but thats not the point. The court doesn’t care who is the better business person. It cares who is right.

    The average coder, if Oracle had its way, would have to ask Oracle for permission every time it wanted to use the platform, for fear that using the very thing they are encouraged to use is a violation. That ought to be seen as absurd.

    And at the very least, Oracle would have to justify how a coder calling java.lang.Math by writing java.lang.Math vs. Google creating its own java.lang.Math are different, and how. Because if the former is a copyright violation it is.basically literally impossible to code. The former is like copyrighting the English language and then suing people for speaking. The latter might be reasonable, and the question presented.

    Oracle could have tried better to make that distinction. But it ends up being, perhaps one is fair use the other is not. But the jury said both are fair use. So we are done. What justification does the court have and say, no wait, this isn’t fair use, but this is, and the jury was wrong?

  19. B April 7, 2021 8:01 am

    @ BP

    Anyone ever notice that Google HQ is about a 30 minute drive from the EFF?

  20. B April 7, 2021 8:26 am

    From the Breyer decision: “Google copied approximately 11,500 lines of declaring code from the API, which amounts to virtually all the declaring code needed to call up hundreds of different tasks. Those 11,500 lines, however, are only 0.4 percent of the entire API at issue, which consists of 2.86 million total lines. In considering “the amount and substantiality of the portion used in this case, the 11,500 lines of code should be viewed as one small part of the considerably greater whole”

    I think that everyone should start stealing Google’s code — 0.4% at a time until the entirety of Google is no longer Google’s.

    “Google’s copying suggest that the better way to look at the numbers is to take into account the several million lines that Google did not copy.”

    I didn’t take all the bank’s money. Why is the bank upset?

    I DEFINITELY would never hire a Supreme Court justice to do my accounting.

  21. BP April 7, 2021 9:27 am

    @B Yes, EFF actually had some good points in its early days, today, it’s a paid-for mouthpiece. The politics, thievery and outright discrimination of Silicon Valley is an amazing thing.

  22. Night Writer April 7, 2021 10:43 am

    Thanks Gene. You truly are the last bulwark before the collapse of the IP system in the USA.

    I wonder if this applies to all the internal APIs as well. I am a former software engineer that designed a couple of substantial systems.

    The way it works is you design the external APIs like in Java and then a bunch of internal APIs to system functions so there may be like three levels of APIs. I think under this decision that someone can just get your documentation and copy your system by sending out work based on your APIs.

    Also, saying that 10,000 lines of code are insignificant because there are 1.5 million lines of code is absurd and offensive. The 10,000 lines of code can be the core of the system that was difficult to design and the other 1 million lines of code can often just be coded by lower level coders.

    And yes I have advanced degrees from top schools in computer science.

  23. B April 7, 2021 10:45 am

    @ A Time “Am I really going to force programmers to learn 50 different words for adding to call that command? I’m pretty sure you can’t copyright the concept of adding anyhow, and you can’t copyright the label of it as adding.”

    What Google might have done: made a set design parameters, use the same labels, pack a bunch of coders into a vault to produce 11,500 lines of code, and document everything.

    As the name/label of a given callable routine isn’t subject to copyright, there will be no infringement, and end-users need not learn anything new.

    This approach isn’t unreasonable.

  24. Night Writer April 7, 2021 11:07 am

    @12 Curious

    Really great comments. Also note as anon has pointed out that “transformative” is what they are saying now to justify the copying but they also are saying that computer programs are not transformative implicitly in Alice. For 101 purposes if something is transformative or transforms then it surely must be eligible.

  25. A time for every purpose April 7, 2021 11:56 am

    Its not about 11500 lines of code, its about what those lines do. Counting lines is ridiculously stupid (and I was somewhat disappointed the majority opinion kept citing it.

    “As the name/label of a given callable routine isn’t subject to copyright, there will be no infringement, and end-users need not learn anything new.”

    Corrent. But its correct because that’s what the majority held! No one is saying that you can’t copyright implementing code, they are saying you can’t copyright declaring code. In fact they didnt even say that, they said you can’t within the bounds of fair use.

    The ruling was somewhat more expansive than that, but not a lot more expansive. An API is a domain specific language. You can’t copyright a programming language. It really isn’t the end of the world everyone is claiming it is.

  26. LSSMIT02 April 7, 2021 1:46 pm

    The part that is galling about the use of the term “transformative” is that the statute grants the copyright owner the right to control derivative works, which are not copies but extensions of the original, i.e., LITERALLY transformative of the underlying work! So the court has read one of the listed exclusive 106 rights to be the justification for denying copyright protection under 107. It’s like how they have read “process” out of section 101 of the Patent Act.

  27. Sean Reque April 7, 2021 2:00 pm

    Google won this case at the local level. It was the corrupt, patent- and copyright- biased federal circuit that forced this case to go the supreme court in the first place. Frankly, the supreme doesn’t do enough to shut down the antics of the federal circuit. No one in the tech industry actually thinks oracle ever should have won this case.

  28. Jason Lamb April 7, 2021 3:06 pm

    Gene Quinn @16…

    Your article and your response to LazyCubicleMonkey @13 are spot on. The sad part is this: what Google co-opted through their copying was the army of programmers that Java had built up over years of painstaking community building, marketing, and refinement of their tools. That is what was locked up in those 11k+ lines of code. That this theft would be brushed off as fair use, and somehow framed as transformative, boggles the mind.

  29. Jeremy Lea April 7, 2021 3:47 pm

    Firstly, the Supreme Court left in place the status quo, so the sky is not falling. Secondly, APIs are blank software forms. They tell you what information to fill in where, and where to submit it. They are not code, they do not perform any function. They are actually purely decorative – if you reverse engineer the system you can fill in the parameters and make the call without ever seeing the API definition, and things will still work.

    The biggest problem with this case from the start has been that the lawyers have consistently failed to define APIs correctly, instead search for all kinds of weird analogies that are not at all like APIs.

    And yes, in object-oriented programming especially a huge amount of thought goes into the API because the classes are integral to the design of the code. However, this is also true of paper forms – just ask the IRS… Yet blank forms are explicitly excluded from copyrightability and have been found to be fair use when copied. In those cases, the argument was exactly the same as Thomas’s discent – that the business model was built into the design of the forms, so they should be protected. The counter-argument is also the same – that the business model is not copyrightable.

  30. Steve April 7, 2021 4:50 pm

    Ignore for the time being the actual declaration lines of code, ok? Look at what those lines define: a set of names, in a hierarchy, for the programmer to use to carry out operations on data. Should those names, or the set of functions that the names invoke, be protected by copyright? If I write a program that works using Oracle’s Java API library, should I have to rewrite it if I want to use a different library (changing not only the names I use in my program to call those functions, but also changing what those functions do and how they work together)? Or should someone be able to re-implement those same routines, including naming them and using them the same way, so that my program can work using either library? Start there before you start thinking about copying actual lines of code.

    Remember that § 102(b) says that copyright protection does not extend to any “idea, procedure, process, system, method of operation, concept, principle, or discovery.” In particular, procedures, processes, systems, and methods of operation are generally taken to refer to functionality , such as a recipe (process), rules of a game (system), or the way you do something (method of operation).

    The expression of a recipe may be copyrightable, but the underlying process (the ingredients used and the steps used to prepare the food) are not. An instruction manual on how to play a game or operate a machine may be copyrightable, but not the underlying game or the interface of the machine.

    Google’s argument is that the “Sun Java API” is such an abstraction. I would describe it as a “system of processes and methods of operation”, where the way the API is used includes the names and the package and class hierarchy.

    This abstraction is what should be considered an uncopyrightable idea. Yes, a lot of work can go into designing an API, and it can be creative, but copyright isn’t supposed to protect functionality like that That is the creativity of invention, not authorship.

    If you copy the Java API abstraction (e.g. by writing Java code while looking at an English description of the Java API specifications), you will most likely create new code that carries out the actual tasks, since there are usually many ways to perform most functions. Google’s so-called implementing code replicated the functions,without copying the code, and none of that code was found to be infringing. The processes in Google’s library were copied, but the code was different.

    NOW consider the declarations. In Java, there is only one way to write the declaring code that will create that same Sun Java API. The only way to create a class called Math that can be used by the application programmer is to use the line

    public class Math

    The only way to say that the Math class can be found in a package called “java.lang” is for the line before that to be

    package java.lang;

    The only way to specify that a routine called “max” is part of that Math class, and that it takes two integers and returns an integer, is to use the line

    public int max(int a, int b)

    following the class declaration (except that “a” and “b” could be different).

    Once the API is designed, there are no choices in how to right the declaring code (except for some minor aspects, aspects which Google did not copy).

    That is the merger argument which the Supreme Court declined to take up.

  31. Steve April 7, 2021 5:16 pm

    “What Google might have done: made a set design parameters, use the same labels, pack a bunch of coders into a vault to produce 11,500 lines of code, and document everything.

    As the name/label of a given callable routine isn’t subject to copyright, there will be no infringement, and end-users need not learn anything new.

    This approach isn’t unreasonable.”

    Except for two things. What Google actually copied was from the Harmony project, which was in fact developed as a clean-room project, and the ruling by the Federal Circuit was that it is the API itself that is protected by copyright, so the set of names, the hierarchy, and the relationships between classes are all protected expression. It is impossible to write a compatible implementation of the Sun Java API without using almost identical declaring code, whether you literally copy the lines or you create them from the API specifications (which will necessarily be the same, except for the specific order of a few keywords, the order some things are defined, and the names of parameters – and just looking at the code shows that the declaring code in Harmony was not identical to the declaring code in Oracle’s implementation of the Sun Java API).

    What the Federal Circuit said in effect was that any compatible library would be infringing on Oracle’s copyright, no matter how it was created.

  32. Night Writer April 8, 2021 10:11 am

    A big question this raises is open source software.

    Basically for those that don’t know open source software has a license with the code that basically says you can use it as long as any code it is included with is open for public use too. So you can’t take open source code and add it to your commercial product and sell it.

    But maybe now you can just take any open source code you want and incorporate it into your commercial product?

    I think under the logic of this opinion that open source license is unenforceable. But given that the Scotus and the CAFC has no care about consistency or logic and just makes things up as they go, I’d say they would hold for the open source license.

    Still probably means open source is open game to take the APIs and take the core of the code and then recreate the other code or just juggle it around.

  33. Anon April 8, 2021 1:33 pm

    Night Writer,

    Your point here intersects with the point that I have provided in regards to “transformative.”

    As open source code is not likely to be merely taken and used “as is,” most all takings may easily be deemed “transformative” under the reasoning available in the Supreme Court case.

    If one has not entered into a contract with an open source provider, then what are open source providers left with as an avenue of enforcement?

    (And let’s note for argument’s sake that the Supreme Court opened the door to all code with their “all software is functional” statements)

  34. Steve April 8, 2021 1:42 pm

    Night Writer:
    Most open source software can be used in commercial products, as long as they follow the license requirements.

    As many open source projects are based on using APIs from commercial products, this decision helps, not hurts, open source. But if it did go the other way, and proprietary software did use an API created for an open source projects so what? It’s still a two-way street, the open source project can provide that API as an alternative, and if the commercial project changes the API, the open source project is free to use their changes.

    This ruling prevents open source software from being locked out of being interoperable with commercial software.

  35. Anon April 8, 2021 5:50 pm


    The immediate point here is that those license requirements have no enforcement force anymore.

    Google won with a “Screw you, we are just going to take and not license” approach.

    There will be ramifications for this. I am not certain how the Open Source community comes out ahead in the long run.

  36. J April 8, 2021 8:37 pm

    *yawn* these are the kinds of arguments that you get from people who have clearly never coded in their entire life. It’s common sense that APIs aren’t copyrightable. Any developer would tell you that. But hey, ignorance is bliss and there’s no point in explaining this point to anyone who can’t be bothered to read beyond the headlines

  37. Anon April 9, 2021 6:40 am

    explaining this point to anyone who can’t be bothered to read beyond the headlines


    I suggest that you read the actual decision here.

    Your “not copyrightable” won’t be found.

  38. BP April 9, 2021 10:29 am

    25. @A time for every purpose: “Its not about 11500 lines of code, its about what those lines do. Counting lines is ridiculously stupid (and I was somewhat disappointed the majority opinion kept citing it.”

    Sorry, you don’t understand copyright. You’ve probably never represented a client in a copyright case or won at the highest level in the copyright office. You have no clue what is “creative expression”.

    “[s]cantiness may exist because the work is composed of elements in the public domain, and it is only the organization of those elements that is protectable”.

    Copyright exists well beyond “scantiness” in those 11500 lines of code. The copyright is substantial. Anyone arguing to the contrary knows nothing about copyright (or music, or literature, or art).

    Given the substantial “thickness” of the copyright, well beyond the standard articulated in Feist, the SCt handed Google a win in Fair Use – because there was no possible way it could decide that a copyright did not exist.

    Fair use has turned into the new “know it when I see it” judicial tool to deliver a victory to special interests, just like 101 in patents (particularly in early dispositive motions or at the PTAB).

    The US legal system is being degraded to erode IP to the benefit of special interests. Ignorants weigh in with uninformed drivel. We’re supposed to be a country of laws, which, as shown, is being hijacked by Silicon Valley’s advertising/lobbying industry. That same industry is under the lens of regulators throughout the world. Breyer needs to step down.

  39. BP April 9, 2021 11:00 am

    If I use this paragraph as a password “He lay flat on the brown, pine-needled floor of the forest, his chin on his folded arms, and high overhead the wind blew in the tops of the pine trees. The mountainside sloped gently where he lay; but below it was steep and he could see the dark of the oiled road winding through the pass. There was a stream alongside the road and far down the pass he saw a mill beside the stream and the falling water of the dam, white in the summer sunlight.” I have copied Hemmingway’s work, which is copyrighted, and I have turned it into something “functional”, i.e., a “computer code” that is recognized by a cryptography system for authentication. While the letters, words, grammar are not new, it is now the only way to authenticate because I said so. I don’t need to change my mind. If a hacker breaks in by “reverse engineering” my password, that hacker infringes Hemmingway’s copyright. However, Breyer would hold that to be a “fair use” of the copyright – because it was the ONLY computer code that could function to grant authentication. Hopefully you can see why Breyer could not hold there was no copyright. Breyer, an enemy of IP, hints at it time and time again, even saying copyright is to promote advancement and not to promote financial gain (but why is that then important in fair use analysis?). Breyer knew he was at the edge of a slippery slope and he could not cross, even though he really wanted to as he proved in Mayo, he hates IP rights and loves illegal monopolies.

  40. Anon April 9, 2021 1:41 pm

    Ignorants weigh in with uninformed drivel.

    I noticed that as well – and chalked it up to a lot of coders that may have been following this case (and the abysmal Kool-Aid that is so constantly served to them in their industry blogs).

    This is being “spun” as a victory for Open Source.

    It appears that no one in that ideology is capable of anything beyond surface thinking.

    The only winner here is the “don’t care what protection you think that you might have, I am just going to take it” view.

  41. Night Writer April 9, 2021 2:44 pm

    BP @38&39

    Great posts!

  42. A time for every purpose April 9, 2021 3:25 pm

    “I have copied Hemmingway’s work, which is copyrighted, and I have turned it into something “functional”, i.e., a “computer code” that is recognized by a cryptography system for authentication. While the letters, words, grammar are not new, it is now the only way to authenticate because I said so. I don’t need to change my mind. If a hacker breaks in by “reverse engineering” my password, that hacker infringes Hemmingway’s copyright.”

    No, the hacker is guilty of violating the law by hacking into your system, not of violating copyright. Do you realize how ridiculous that comment is? Even Oracle did not go that far. Read the briefs. Oracle specifically targeted what Google did, and said numerous times that what you are arguing is specifically not what they are arguing, because they know how ridiculous it is.

    You cannot copyright a password then sue people for entering that password. The password serves a purpose. Merely entering or naming something isnt creative under copyright law. You might be able to trademark a name, but copyright it?

    If you use Hemmingway’s work FOR THAT PURPOSE, it is not a copyright violation. If you use it for a different purpose, say, in a poem or creative work, it is copyright infringement! I dont know why thats a difficult to understand, BP claims to be a lawyer, it should be self-evident that context matters. That is the principle underlying fair use.

    And Oracle conceded all of this. Their point was you can copyright an API so that another entity cannot copy that API, not that programmers using the API are suddenly guilty of copyright infringement. Coding would be impossible if you did that. Thats still unreasonable, but not as unreasonable. You are arguing a point Oracle conceded … and then claiming it is the end of the world that they didnt adopt it.

  43. BP April 9, 2021 4:10 pm

    Oh my, A time for every purpose: “ask not for whom the bell tolls, it tolls for thee”.

    You’re probably a programmer that never stepped foot into an artists studio, a composer’s studio, . . . . You know nothing of what is “creative expression”. Read this case: Satava v. Lowry “a thin copyright that protects against only virtually identical copying.” Yes, “distinctive curls of particular jellyfish tendrils” are copyright registerable. I have no idea what foundation you are standing on, it seems like one that has nothing to do with law.

    Section 107:
    Notwithstanding the provisions of sections 106 and 106A, the fair use of a copyrighted work, including such use by reproduction in copies or phonorecords or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright.

    Please old wise one, identify the “purpose” in the statute that makes the hypothetical use “not an infringement of copyright”. Oh my, my, . . ., looks like it’s not in the statute but old wise one seems to infer that a court would twist the nose of wax to reach his conclusion. Do you get it now?

    Don’t know the last time you spoke face-to-face with Nimmer or Donaldson? Doubt you even know who they are . . . .

  44. Anon April 9, 2021 4:37 pm

    Ah BP, coders have been provided gallons and gallons of Kool-Aid. There is no thirst there, as the lemmings march up their hill. All they see is the warm “promise of sunshine” at the crest of that hill, as they charge full force towards that crest.

    Free copying. NO price to pay of any kind for taking. What could go wrong with that? A dawn of a glorious age beckons, and there is only a few more steps towards that precipice.

    Law? bah, they understand what their overlords have been preaching to them: all intellectual property protections are ‘bad.’
    No patents – copyright will do.
    No copyrights, all software is functional, and Fair Use sets you free (and soooooo “transformative.” Maybe patents will do.

    Gee, maybe Trade Secrets and employment constraints are what the lemmings should be taught to embrace…

  45. Steve April 9, 2021 5:43 pm

    Anon: “The immediate point here is that those license requirements have no enforcement force anymore.”

    That’s nonsense. You can’t just “copy code”. In order for this fair use logic to be applicable, the literal code that could be copied would have to be the minimum that allowed for some measure of compatibility.

    I would argue that the interface itself (the abstraction described by the source code, or the API specifications) is a process, system, or method of operation, and is outside the scope of copyright, the same way the rules of a game, or the ingredients and steps of a recipe are outside the scope of copyright.

    It is not the case that “code is functional, therefore code can’t be copyrighted”. It is the function that can’t be copyrighted, not the code.

    If you look closely at the decision, you’ll see the bones of a merger argument in there, with the API an uncopyrightable idea, and the declarations being necessary to express that idea.

  46. Anon April 9, 2021 6:33 pm

    Sorry Steve but no — transformative (per the Court’s reasoning) is just not so limiting.

    Also – as noted already — what you would want to go to is NOT what the Court went to (by the by, the legal term you are looking for is called the Merger Doctrine). That path provides that item is NOT copyrightable in the first instance, and Fair Use would not even be reached.

    Your attempt to see “bones” only reveals that you do not understand the law at point.

  47. A time for every purpose April 9, 2021 7:04 pm

    Ok, I’ll ignore the fact that apparently Anon and BP have no idea what an API is. I’ll also leave aside the fact that you … lost, and yet are still claiming the law is in your favor.

    Are you telling me that Microsoft, Apple (who is decidedly not a fan of open source. They have made many decisions with the explicit purpose of crushing open source coders!), and all the big tech companies want to give up copyrights for code? You are aware that is exactly backwards right? Why would they want to give up their licensing agreements? Google in particular … so much of their code is public. You can’t use any of it … so why would they endorse a decision to change that?

    Unless, of course, that isn’t actually what they are doing. I can call a map to appear on a webpage by typing “google.maps.Map().” If I copied the data used to actually make that work, that would be copyright infringement under the decision! It is just the words “google.maps.Map()” that are unprotected, because I need to type them in to do my job.

    No, I will not find it in the statute. The majority opinion did not adopt the statutory argument. Google immediately backtracked on it upon questioning during oral argument. Rather, the court has always balanced congresses right to copyright things with freedom of speech and precedent arguments. Fair use preceded the statute, the court simply made it up, then it was codified. Congress never expressly said you cannot copyright a recipe, yet, you can’t. Same idea. You might not like that. But for copyright that is how it is, and that isn’t remotely controversial.

    And if the entire computer science industry, from big companies to small individual programmers, think something is a bad idea, well … shouldn’t that get deference? Instead of some lawyer? This isn’t like Alice, where there are a lot of programmers who are upset with inabilities to patent … almost everyone thinks this is a bad idea. Except Oracle. Maybe also SAP and Matlab, but they deliberately designed their software to not be interoperable so they should not have as much of an issue.

    Even Sun, the developer of Java, thought Google was in the right.

    It isn’t big companies against individual software engineers, its practically everyone. Believe it or not programmers do understand how code works, which most of SCOTUS and lawyers generally do not. You are saying that knowledge is not relevant at all to deciding a major question that could upend the very industry in which programmers work?

  48. Anon April 9, 2021 8:29 pm

    Ok, I’ll ignore the fact that apparently Anon and BP have no idea what an API is

    Clearly a false statement. Maybe instead look at the actual arguments I have provided.

    I’ll also leave aside the fact that you … lost, and yet are still claiming the law is in your favor.

    Another false statement.

    You clearly lack an ability to understand a legal discussion. There is no “Anon lost” in my comments. There is no “claiming the law is in my favor.”

    If I copied the data used to actually make that work, that would be copyright infringement under the decision

    Now I know that you are not following what I have been saying. If you copy — and transform then you are in the clear for the Court’s (not my) provided reasoning of Fair Use.

    Congress never expressly said you cannot copyright a recipe,

    Not pertinent here. Please read again what the Court (and again, not me) actually stated.

    And if the entire computer science industry, from big companies to small individual programmers, think something is a bad idea, well … shouldn’t that get deference? Instead of some lawyer?

    That is not how law works.

    Believe it or not programmers do understand how code works, which most of SCOTUS and lawyers generally do not. You are saying that knowledge is not relevant at all to deciding a major question that could upend the very industry in which programmers work?

    Both a false presumption and a false equivalency. Most patent attorneys in this space ALSO know how code works — and they have the benefit of knowing how law works.

    You presume a lack of knowledge (when that lack is not evident) because of your own lack of understanding of the law.

    And by the way, THIS Court decision is what is going to upend your world. That’s not my doing.

  49. A time for every purpose April 10, 2021 1:08 am

    “Now I know that you are not following what I have been saying. If you copy — and transform then you are in the clear for the Court’s (not my) provided reasoning of Fair Use.”

    The transformative standard replacing the commercial standard was not invented in Google vs. Oracle. It has been around in some form, and not just for computer code, for 25 years at this point. So, to be frank, idk what you are talking about. .

    The court found it transformative. That is an existing standard. And it is transformative, you are transforming the declaring codes, which themselves do nothing, by changing the implementing code. And yes, by changing the platform, that is because the whole point of Java is that it is platform flexible! So the platform changing matters!

  50. Akash Garg April 10, 2021 1:13 am

    “That is not how law works.”

    I understand that, but you should adjust your views about law based on what the people who are supposed to be following it actually think and do. Breyer’s a pragmatist, that is what he believes, and he wrote the opinion.

    You may feel differently that everything must be super formal, but there are pragmatists on courts and in law and pragmatism isn’t invalid.

    Finally, isnt the purpose of copyright law to increase innovation? How does being able to copyright declaration code … how does that encourage innovation?

  51. BP April 10, 2021 6:38 pm

    A time for every purpose: too funny, you just don’t get it, I’m really getting a good laugh. Diapers for you. If you read carefully, I noted that I started programming in the 1970s and landed at a supercomputer institute (ever hear of Cray?). Most of my programming was for control systems, you know software that has to work. A controller built on some Windows OS without use of APIs? I was calling APIs while you were in diapers. You’re probably just another SV diaper baby. Part of the launch and patch crowd, like those that infected Boeing and Tesla (yes, faulty sw that caused deaths, oh well “we’re working on the patch”). The ones that never learned to program without an error checker. The ones that never had to get it right on the first run, like back in the days when homework in computer sciences classes ran on mainframes as you carried around your shoe box of cards. How many runs do think they gave you? Real programmers do it right the first time. History is an amazing thing but diaper boy believes he has no need for history, heck, he’s living in the “tech” age.

    Great comments by others here, people that actually know something.

  52. A time for every purpose April 11, 2021 10:44 am

    @BP no reason for insults.

    I do find it funny that you are both an intellectual property lawyer who doesn’t understand what an API is (you have never actually articulated why a language can be copyrighted when precedent clearly says otherwise or why an API isn’t sufficiently described as a language) and now a hard-core programmer who works close to the metal and lectures newer developers about “real” programming and “back in the day.”

    “Real programmers do it right the first time.”

    Walk into any software company today and say that and you will be shoved out the door immediately. That is an excellent way to ship buggy and unworkable code.

    And whatever kind of person you are, both are insufferable, and I have zero business carrying this conversation further. This decision would allow me and other developers to further my career, can’t say the same for you parasites. Good day gentlemen.

  53. Mr 259 April 12, 2021 11:41 am

    Well, Ben Franklin has left the room, everything for the public good, as long as I have mine, then off to France. When Thomas Jefferson (Anon) argues, if you build it you own the rights to it, Jefferson stayed in the US.

  54. Jack April 12, 2021 1:08 pm

    Gene I’m afraid this demonstrates both your lack of understanding and your utter commitment to rulings which serve your own self-interest.

    Google did not COPY any code whatsoever. It duplicated declarative headers which are necessary for software developers to use the API in the first place. Every product shipped “copies” these headers because they’re absolutely required for source code to use the API – however by themselves they have ZERO functionality and do not actually IMPLEMENT an idea.

    The ENTIRE purpose of an API is to provide services which other software can use to build more complex functionality. If you don’t want anyone using your API (which is why you need the headers), then you don’t publish the API (IE: The headers).

    Oracle’s claim is utterly meritless and ludicrous. They’re claiming that by attempting to use the API, Google infringed copyright on the headers which are necessary to use the API.

    This is like claiming your patent on the printing press allows you to dictate the uses of paper. The Java API is meant to be used, otherwise it has zero value. The entire value proposition of Java is based on the idea that it’s used to build software. You can’t build software without using the headers which you naively claim Oracle has a copyright on.

    Headers are not a copyrightable work. They are a mandatory component of interoperability and they were made freely available.

    Consequently, they cannot be patented, cannot be copyrighted and Oracle attempting to do so is a misuse of IP law. This is a win not only for Google but for every small developer who would be crushed by Oracle’s extreme misuse of the IP system.

  55. Anon April 12, 2021 1:48 pm

    Mr. Garg.

    Respectfully, your statement of “but you should adjust your views about law based on what the people who are supposed to be following it actually think and do is the antithesis to the Rule of Law.

    It is the very opposite of what one “should do.”

    I really do not have any issues with an appropriate change in law – by the appropriate means.

    I DO have issues with inappropriate changes — as should everyone in a civilized society.

    Do NOT confuse “pragmatism” with proper application of the law. The Rule of Law is critical, and is so in part to safeguard against such things as “tyranny of the majority.” In a hyper-example, it would be extremely expedient and “pragmatic” to dispense with the criminal notion of “innocent until proven guilty.” While intellectual property law does not carry the same immediate impact of loss of personal freedom (up to and including loss of life), our country was founded on a triumvirate of critical freedoms that include property rights. Intellectual property law falls under this penumbra.

    This is not merely an item of formality (let alone super formality). Rather, the larger grasp so absent from so many of hte ‘tech’ ilk more (poorly) reflects what it takes to intelligently operate in a world of both law and tech.

    There are simply far too many people who have no clue of law, and merely think that their level of technical understanding compensates and allows them to speak on something that they just do not understand.

    That is not to say that all comments – with huge numbers coming from non-regular posters – are worthless and such and can be summarily dismissed. Some are indeed respectful and acknowledge that they are only speaking on a particular aspect.

    Sadly, many are not – and show more emotion than reason. Many are of the mob-mentality type that “this is what we do,” regardless of whether or not that doing is appropriate under the law.

    Some even show a complete disregard for the essence of a legal view. I see many that declare “wrong,” with it evident that the speaker does not even grasp the legal points being discussed. Gene’s article here (and the Absurdity of the Result) are at heart legal pieces.

    Many seem unable to even recognize that. Some what to immediately state the Court’s reasoning only apply to the immediate facts of the particular case – and that is just not so.

    Fact’s of a case are applied for the immediate case. But the case law is necessarily than used and applied to other cases that do not have the same fact patterns. It is far more than “formality” to be rigorously trained (at law) to understand just how that Rule of Law is taken from the writings of the Court.

    For a quick example, many are OVER-focusing on the functional difference between functionality of “declaring” and “implementing,” but this misses the larger point given by the Court in its statement (so I am only supplying the message) that ALL software is functional. It also misses the larger point that the copying was considered Fair Use (and NOT that copyright was somehow precluded in the first instance). Many are attempting to use a Merger Doctrine position without understanding that the Court simply did not Hold so. The larger point under the legal reasoning is that by stating that all software is functional, and that the reason why Fait Use was obtained was a reliance on “transformative” (to the rather large scale detriment of the actual Four Factors of the analysis), is that the SAME TYPE of ‘analysis’ for the extremely wide notion of ‘transformative’ is what will control going forward — across ALL types of software functionality (and NOT limited to the facts of THIS case and ANY ‘distinction’ between a functionality of ‘declare’ or ‘implement.’

    It is NOT ‘the function’ that is a driver, as much as it is a transformation of ANY function.

  56. Gene Quinn April 12, 2021 2:01 pm


    It i asinine to say that Google did not copy any code. And it is even more asinine to say that an API is not code. Only a truly ignorant person would every say such ridiculous things. Please inform yourself in the future before you make such idiot commentary.

    Your comment speaks volumes about how little programmers know about what they are actually doing.

    And for all those who are convinced I’ve never written code, wrong again!

  57. Steve April 12, 2021 6:26 pm

    Google did copy code, but they copied it from the Harmony project. The Harmony project developed their code based on the specifications, not by copying any code.

    An API is not the same as the code that implements it, any more than an algorithm is “code”. The “declarations” at issue are code (which has no legal significance), and are part of a copyrighted computer program. The API is what is expressed by the computer program. Just as a “recipe” is the expression of an uncopyrightable process, or a rule book is the expression of an uncopyrightable system, a computer program can expresse many uncopyrightable processes, systems, and methods of operation. The whole “code is functional” is not accurate. Code is the expression of function, just as a cook book expresses function

    Whether Google’s declarations are infringing is a question of merger. Whether the API described by the specifications (including the names and hierarchy) is a system or method of operation is strictly a 102(b) question.

    § 102
    (b)In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work.

    If the API is a system or method of operation, then you need to show that there are other declarations you can write in Java that express that same uncopyrightable API. I assert that there is no such alternate language.

    That’s the question that the Supreme Court didn’t address, but it is still very much an open question, and if this comes up again in the Ninth Circuit, they are very likely to find that an API is such an uncopyrightable idea. See Bikram v Evolation, Sega v Accolade, Sony v Connectix.

    As for the fair use ruling, it seems almost inevitable that something that shouldn’t be copyrighted in the first place will meet the standards for fair use, particularly as to nature of the work. With merger, amount copied also favors fair use (minimum amount needed). Purpose of the use is also favored, even if it is not “transformative”.

    Y’all are acting as if “transformative” is the only thing you need to show for fair use. All they did, in effect, is say that when something is functional, you can’t say it isn’t transformational simply because it has the same (uncopyrightable) function.

    Function in a computer program is behavior. If two components produce the same behavior, they have the same function.

  58. Anon April 12, 2021 8:39 pm


    You are incorrect on pretty much every single statement related to law and this case.

    Whether Google’s declarations are infringing is a question of merger.

    Try again – this is quite not correct at all. Perhaps you are reading an entirely different case.

    As for the fair use ruling, it seems almost inevitable that something that shouldn’t be copyrighted in the first place will meet the standards for fair use

    Not even wrong. To BE Fair Use, it first needs be copyrighted.

    Y’all are acting as if “transformative” is the only thing you need to show for fair use. All they did, in effect, is say that when something is functional, you can’t say it isn’t transformational simply because it has the same (uncopyrightable) function.

    This is not correct – in multiple dimensions.

  59. BP April 12, 2021 9:38 pm

    OMG, the arrogance of some programmers is incredible! They are hacks and I don’t mean hackers, because many hackers are intelligent, though misguided. Picasso? Hey, is that a website? No appreciation for creativity, why SV is such a cultural wasteland of fraud and discrimination in the name of $$$. Craping on the shoulders of giants. Disrupt copyright based on ignorance and greed.

Post a Comment

Respectfully add to the discussion.

Name *
Email *