Google v. Oracle (Supreme Court Copyright Case)

[TwiceHorn 31,440]

Another blog predicts an Oracle win.

This guy is interesting.  He's German, and not a patent lawyer, but follows a lot of standards-essential patent litigation both in Germany (a popular European patent jurisdiction) and the US.  He seems to be smart as fuck and rarely errs in his understanding of patents and patent litigation.

He is kind of anti--patent, anti-IP though.

[TwiceHorn 31,440]

Ars Tecnica with an article arguing that Google, and particularly counsel Tom Goldstein, screwed the pooch.  https://arstechnica.com/tech-policy/2020/10/googles-supreme-court-faceoff-with-oracle-was-a-disaster-for-google/

Goldstein has handled a number of recent patent cases before the Supreme Court.  But patent ain't copyright.

[immamac 42,838]

That's because Google's argument is entirely fucking bogus lol

[Dahobbs 17,527]

2 minutes ago, TwiceHorn said:

Ars Tecnica with an article arguing that Google, and particularly counsel Tom Goldstein, screwed the pooch.  https://arstechnica.com/tech-policy/2020/10/googles-supreme-court-faceoff-with-oracle-was-a-disaster-for-google/

Goldstein has handled a number of recent patent cases before the Supreme Court.  But patent ain't copyright.

Goldstein is usually very good. I need to listen to the argument, because I cant believe he was as unprepared as that article suggest. Further, while I love Ars, its legal coverage is more than a bit suspect. They just don't have anyone on staff that really understands the law, particularly procedure.

[Deej 76,755]

17 minutes ago, immamac said:

That's because Google's argument is entirely fucking bogus lol

When aggy tells you something is bogus, you better well listen. 

[Dahobbs 17,527]

14 hours ago, immamac said:

That's because Google's argument is entirely fucking bogus lol

Why do you say that? 

[Captainant 27,638]

https://www.supremecourt.gov/opinions/20pdf/18-956_d18f.pdf

In a 6-2 ruling, the Supreme Court has ruled that google's use of the java API constituted fair use. The concluding paragraph of the ruling authored by Breyer basically sums up the arguments:

Quote

The fact that computer programs are primarily functional makes it difficult to apply traditional copyright concepts in that technological world. See Lotus Development Corp., 49 F. 3d, at 820 (Boudin, J., concurring). In doing so here, we have not changed the nature of those concepts. We do not overturn or modify our earlier cases involving fair use—cases, for example, that involve “knockoff ” products, journalistic writings, and parodies. Rather, we here recognize that application of a copyright doctrine such as fair use has long proved a cooperative effort of Legislatures and courts, and that Congress, in our view, intended that it so continue. As such, we have looked to the principles set forth in the fair use statute, §107, and set forth in our earlier cases, and applied them to this different kind of copyrighted work.

We reach the conclusion that in this case, where Google reimplemented a user interface, taking only what was needed to allow users to put their accrued talents to work in a new and transformative program, Google’s copying of the Sun Java API was a fair use of that material as a matter of law. The Federal Circuit’s contrary judgment is reversed, and the case is remanded for further proceedings in conformity with this opinion.

Thomas and Alito dissenting and Barrett did not participate in the case.

 

[TwiceHorn 31,440]

18 minutes ago, Captainant said:

https://www.supremecourt.gov/opinions/20pdf/18-956_d18f.pdf

In a 6-2 ruling, the Supreme Court has ruled that google's use of the java API constituted fair use. The concluding paragraph of the ruling authored by Breyer basically sums up the arguments:

Thomas and Alito dissenting and Barrett did not participate in the case.

 

Interdasting.  Doesn't create a bright-line rule, but the fair use evaluation seems a bit off.

Actually, I think I like the outcome as fair use is a pretty fact-intensive inquiry.

What strikes me as a bit odd or off is the calling of another commercial computer program a "transformative use."   Usually, that tends to mean some form of artistic transformation or from a "serious work" (Orbison's Pretty Woman) into a parody (2 Live Crews).

Compare the recent 2nd Circuit ruling that Warhol's "transformation" of Prince portraits was not transformative.  https://www.plagiarismtoday.com/2021/03/30/understanding-the-andy-warhol-ruling/

Under this ruling, "normal" use an an API, which doesn't attempt to replace the underlying program, wouldn't be actionable.

Also, Thomas' dissent is surprisingly sane for him.  Somehow, Clarence seems to "get" IP cases to a larger extent than you would think.

Edited April 5, 2021 by TwiceHorn

[bernorange 8,359]

I had to re-read (skimming) this thread to remember all the legal issues at play.  Sorry 2x, but I had to laugh when reading post #246 on page 5.

[TwiceHorn 31,440]

34 minutes ago, bernorange said:

I had to re-read (skimming) this thread to remember all the legal issues at play.  Sorry 2x, but I had to laugh when reading post #246 on page 5.

I don't really think fair use was strenuously argued by either side, although it was at issue in the case at all times.

It's not all that uncommon for a court to dredge up an issue neither party urges strongly and resolve the case on that.

The reason fair use I don't think was strongly argued is that when you have fair use claimed for copying purely commercial work by purely commercial work, it's almost never applicable.  The factors are:

• the purpose and character of your use
• the nature of the copyrighted work
• the amount and substantiality of the portion taken, and
• the effect of the use upon the potential market.

In a lot of ways, this seems to be a good test for evaluating such things, but the fourth factor where the commercial realities get involved, is often dispositive of the issues.  For example here, where Google's virtual machine is a market substitute for Oracle's Java, fair use is out the window, in the typical case.  Indeed, fair use probably isn't argued much in software cases for this reason.

Take as a counter-example the 2 Live Crew case.  Both were musical works that ultimately had a commercial purpose for their authors and associated entities, but the reality was that the nature of 2LC's rap work meant that it wasn't going to have much effect on the market for Orbison's country or country rock work and that market differentiation tied directly into the parodic, transformative nature of 2LC's copying.

Edited April 5, 2021 by TwiceHorn

[Dahobbs 17,527]

6 hours ago, TwiceHorn said:

Interdasting.  Doesn't create a bright-line rule, but the fair use evaluation seems a bit off.

Actually, I think I like the outcome as fair use is a pretty fact-intensive inquiry.

What strikes me as a bit odd or off is the calling of another commercial computer program a "transformative use."   Usually, that tends to mean some form of artistic transformation or from a "serious work" (Orbison's Pretty Woman) into a parody (2 Live Crews).

Compare the recent 2nd Circuit ruling that Warhol's "transformation" of Prince portraits was not transformative.  https://www.plagiarismtoday.com/2021/03/30/understanding-the-andy-warhol-ruling/

Under this ruling, "normal" use an an API, which doesn't attempt to replace the underlying program, wouldn't be actionable.

Also, Thomas' dissent is surprisingly sane for him.  Somehow, Clarence seems to "get" IP cases to a larger extent than you would think.

I think finding yourself on the Alito and Thomas side of an argument is ample reason to reconsider pretty much everything. 

N.B. not meant as an insult toward you at all. Just seeing those two both dissent makes me feel confident that the right outcome was reached.  

Edited April 5, 2021 by Dahobbs

[LABEVO 928]

Wonder what will be all the cascading impacts to API first companies. Eg value prop is connecting a lot of disparate systems while core IP is the APIs that enable it....

[TwiceHorn 31,440]

14 hours ago, Dahobbs said:

I think finding yourself on the Alito and Thomas side of an argument is ample reason to reconsider pretty much everything. 

N.B. not meant as an insult toward you at all. Just seeing those two both dissent makes me feel confident that the right outcome was reached.  

I didn't say they were right.  I just said Clarence's dissent was not insane.  For a guy who is mostly insane, when he writes on IP, he seems to at least "get it" and appear sane.  It does show what butt buddies he and Alito have become, which seems to demonstrate that Alito is out of his depth.

A lot of the pundits seem to evaluate this as creating sort of a new/different fair use for computer programs.  That's consistent with my initial thought that this isn't a "normal" fair use analysis.

One aspect of it that I like is that a) it doesn't create a bright-line rule and b) it "operates" on the infringement end of copyright analysis rather than the Section 102 copyrightablility/originality end of it.  It's not a perfect analogy, but a decision that had focused on copyrightability would be somewhat like the Section 101 patent cases, which have become bad law of monstrous proportion.

I think it is going to fuck up fair use analysis in more traditional ("arty") copyright cases.

Edited April 6, 2021 by TwiceHorn

[TwiceHorn 31,440]

3 hours ago, LABEVO said:

Wonder what will be all the cascading impacts to API first companies. Eg value prop is connecting a lot of disparate systems while core IP is the APIs that enable it....

Didn't realize there was such a company. They are slightly less fucked than they would be had it come out as a bright line test for copyrightability.

[Captainant 27,638]

3 hours ago, LABEVO said:

Wonder what will be all the cascading impacts to API first companies. Eg value prop is connecting a lot of disparate systems while core IP is the APIs that enable it....

I don't see there being that big of a difference, IMO. The ruling makes clear the copyrightable part is the implementation, not the interface (API) to that implementation.

For example: AWS's S3 storage service has an API interface to store, retrieve, modify, delete, etc. There's loads of open source projects to replicate that API pattern, but implemented against on-prem hardware. That doesn't steal the value of the s3 service - it just let's people access their non-AWS stuff using it's interface and existing scripting/tools

In the case of businesses that integrate and orchestrate many different systems via their APIs, I dont see this ruling changing their business model whatsoever. Because again, the implementation, not the declaration is the creative and copyrightable part

[TwiceHorn 31,440]

5 minutes ago, Captainant said:

I don't see there being that big of a difference, IMO. The ruling makes clear the copyrightable part is the implementation, not the interface (API) to that implementation.

For example: AWS's S3 storage service has an API interface to store, retrieve, modify, delete, etc. There's loads of open source projects to replicate that API pattern, but implemented against on-prem hardware. That doesn't steal the value of the s3 service - it just let's people access their non-AWS stuff using it's interface and existing scripting/tools

In the case of businesses that integrate and orchestrate many different systems via their APIs, I dont see this ruling changing their business model whatsoever. Because again, the implementation, not the declaration is the creative and copyrightable part

If I am reading correctly, LABEVO references companies that are in the business of creating APIs and don't have implementations.  

And the bolded is not correct.  The ruling holds that most copying of APIs is fair use (actually, it only holds that Google's copying of Oracles JAVA API is fair use, but that's probably a fair extension of it).  That presumes that the copying is infringement and the subject matter is copyrightable.  Not that the ultimate outcome is that different.

[Captainant 27,638]

2 minutes ago, TwiceHorn said:

And the bolded is not correct.  The ruling holds that most copying of APIs is fair use (actually, it only holds that Google's copying of Oracles JAVA API is fair use, but that's probably a fair extension of it).  That presumes that the copying is infringement and the subject matter is copyrightable.  Not that the ultimate outcome is that different.

The ruling repeats and emphasizes over and over that Googles unique re-implementation of the functionality is what excuses the copying of the API definitions, which was the central issue being examined in the ruling. By definition, you cannot change an API definition without breaking it's usability. That inability to reexpress the interface without changing it's functionality is what makes it non-copyrightable.

The entire case from oracle hinged on 11,300 lines copied out of 3,000,000+ lines written by Google. The copied lines were only and exclusively the API definitions, which are expressible in one and only way. Because there is no creativity to writing something that can only be written one way, it is not copyrightable

[TwiceHorn 31,440]

4 minutes ago, Captainant said:

The ruling repeats and emphasizes over and over that Googles unique re-implementation of the functionality is what excuses the copying of the API definitions, which was the central issue being examined in the ruling. By definition, you cannot change an API definition without breaking it's usability. That inability to reexpress the interface without changing it's functionality is what makes it non-copyrightable.

The entire case from oracle hinged on 11,300 lines copied out of 3,000,000+ lines written by Google. The copied lines were only and exclusively the API definitions, which are expressible in one and only way. Because there is no creativity to writing something that can only be written one way, it is not copyrightable

You're mostly not wrong.  But I am also right.

The statement that the case holds APIs to be uncopyrightable is incorrect.  As is the repeated assertion that it is "not copyrightable."

[DonkeyCigars 144]

4 hours ago, LABEVO said:

Wonder what will be all the cascading impacts to API first companies. Eg value prop is connecting a lot of disparate systems while core IP is the APIs that enable it....

What are so examples of these types of companies, in your mind?

[LABEVO 928]

Not all of these as many have other functionality...too lazy to dig into a list and give quality examples  

https://techcrunch.com/2019/09/06/apis-are-the-next-big-saas-wave/

[TwiceHorn 31,440]

16 minutes ago, LABEVO said:

Not all of these as many have other functionality...too lazy to dig into a list and give quality examples  

https://techcrunch.com/2019/09/06/apis-are-the-next-big-saas-wave/

This kind of brings up a definitional problem with APIs.  In the usual case, you have software proprietor A, with package A, and proprietor B with package B.  One or both of A and B may generate an API to facilitate and increase use of their respective packages in various environments.  They deliberately expose that code to the public and it's generally free for the taking.

Unsurprisingly, I suppose, there are companies out there that specialize in interoperablity of software packages and their products thus are "APIs."  A company C that generates an API to interface packages A and B, drawing out the analogy.

In any event, it would seem that such an API, being largely dictated by the structures, data, and syntax of packages A and B, is going to be highly functional and of limited "originality" or copyrightability.  Even more so than your usual computer program.

This decision acknowledges the "thin" copyrightability of "APIs," but does not create a bright line rule that an API is not copyrightable.  So those C companies are a bit less fucked than they might have been with such a bright-line rule.  In fact, the fair use analysis might come out in their favor as any infringing API would be a 100% market substitute and replacement for company C's product, which usually nukes the application of fair use.  That is, every infringement of company C's API would be a lost sale to Company C.

But, I might surmise that company C's revenue model is not based on repeat sales of APIs, but rather the service of interfacing software, billed by the project.  The projects are probably sufficiently unique that there's not a lot of market for repeat sales of the same software.

Edited April 6, 2021 by TwiceHorn

[Captainant 27,638]

8 minutes ago, LABEVO said:

Not all of these as many have other functionality...too lazy to dig into a list and give quality examples  

https://techcrunch.com/2019/09/06/apis-are-the-next-big-saas-wave/

But still, the API is just the interface to the functionality written by the company. In API definition is just a description of what you're giving and what you expect to get back. It's not the creative part or portion of building your service. 

[Captainant 27,638]

@TwiceHorn I feel like you're only thinking about this in the way that a lawyer does and you don't really understand technically what an API definition is. In your example, company C is still doing some unique orchestration to integrate APIs A and B that is more than simply defining the API schema (or method header, in the case of this decision). The definition/schema/header is not the hard part or the original thought.

You should read the ruling, it makes the same points I'm making here that you disagree with. 

 

[TwiceHorn 31,440]

8 minutes ago, Captainant said:

@TwiceHorn I feel like you're only thinking about this in the way that a lawyer does and you don't really understand technically what an API definition is. In your example, company C is still doing some unique orchestration to integrate APIs A and B that is more than simply defining the API schema (or method header, in the case of this decision). The definition/schema/header is not the hard part or the original thought.

You should read the ruling, it makes the same points I'm making here that you disagree with. 

 

I don't disagree with your general points about APIs.

But the ruling is confined to fair use.

That means the court did NOT specifically decide that APIs are not copyrightable.

Indeed, the implicit holding is that they are or may be copyrightable, and in this instance the copyright was infringed, but that infringement is excused by fair use.

Quote

We reach the conclusion that in this case, where Google reimplemented a user interface, taking only what was needed to allow users to put their accrued talents to work in a new and transformative program, Google’s copying of the Sun Java API was a fair use of that material as a matter of law.

I am also going to wager that the italicized part is going to open a whole can of worms about software copyright infringement and fair use in the future.  Probably not going to amount to much though as software copyright cases are fairly rare.

As a technician (I get it, I'm trained that way too), you have a tendency to want to make broad general rules that fit your conception of the general case.  Rules that would work 99% of the time, maybe more.

But as a lawdog, you learn that such general rules, that don't take into account little factual variations that inhere to "court cases," can be dangerous as hell and that 1% of wrong decisions that result can swallow up the whole field.

Edited April 6, 2021 by TwiceHorn

[TwiceHorn 31,440]

Oh shit.  I just noticed something that law dogs may appreciate. Fair use is a question of fact, or at least a mixed question of law and fact.

When a court concludes that something that is at least partially a fact question, like fair use, exists as a matter of law, the court probably fucked up.

See for example, obviousness in patent law.

[Captainant 27,638]

27 minutes ago, TwiceHorn said:

That means the court did NOT specifically decide that APIs are not copyrightable

You're not hearing me. I'm not saying that APIs in their totality aren't copyrightable - the actual implementation of functionality is certainly copyrightable. The definition - and lines of code in question in the decision - are not creative works and are not copyrightable in this context. 

Edit: or maybe better put, if those definitions are part of a larger copyrighted codebase, any copying of the definitions for a unique implementation is fair use

Edited April 6, 2021 by Captainant

[bernorange 8,359]

Say an author publishes some poems and then a rock group takes one of his published poems and incorporates it into a song (without permission).  I assume this would be a clear infringement of the author's original copyright.

Say an author publishes a declaration header for an API in a book or trade journal and then a software company incorporates that header into a software package.  Is this analogous to the poem appropriation example?

[TwiceHorn 31,440]

15 minutes ago, Captainant said:

You're not hearing me. I'm not saying that APIs in their totality aren't copyrightable - the actual implementation of functionality is certainly copyrightable. The definition - and lines of code in question in the decision - are not creative works and are not copyrightable in this context. 

Edit: or maybe better put, if those definitions are part of a larger copyrighted codebase, any copying of the definitions for a unique implementation is fair use

And i have said repeatedly that I don't disagree with that as a general proposition.  Except I would probably state it as APIs are of minimal and suspect copyrightability/originality because of the factors you have mentioned.

You're not hearing me.  I disagree with your assertion that this case held APIs uncopyrightable, which you said at least twice.  And repeated again above.  That's not what the court held.

This is a fair use decision.  It does not address copyrightability except in the most implicit way.

Edited April 6, 2021 by TwiceHorn

[TwiceHorn 31,440]

7 minutes ago, bernorange said:

Say an author publishes some poems and then a rock group takes one of his published poems and incorporates it into a song (without permission).  I assume this would be a clear infringement of the author's original copyright.

Say an author publishes a declaration header for an API in a book or trade journal and then a software company incorporates that header into a software package.  Is this analogous to the poem appropriation example?

In the first case, there is an argument that that is fair use.  As in the 2 Live Crew decision, if something less than the entire poem is used and there's some kind of commentary on it in the lyrics, that's maybe a good fair use case.  If it's just the poem set to music, it's probably not fair use.

Fair use is very much a case-by-case thing.  I posted the factors for consideration above.  So the answer is "it depends."

Also, it's becoming clear that the Court is trying to create a new "species" of fair use specifically for computer programs.  As a practical matter, that might be the right result as at least it takes into account all of the "facts" surrounding the copying, its purpose, etc.  Doctrinally, it's a holy mess.

[Captainant 27,638]

3 minutes ago, TwiceHorn said:

This is a fair use decision.  It does not address copyrightability except in the most implicit way.

Ah I think I see what you're saying now - it IS copyrightable, but because google's copying was fair use oracle can't stop them. Copyright in software (and the digital space, generally) is weird since it has real legal teeth but there's so many circumventions to it

[TwiceHorn 31,440]

Just now, Captainant said:

Ah I think I see what you're saying now - it IS copyrightable, but because google's copying was fair use oracle can't stop them. Copyright in software (and the digital space, generally) is weird since it has real legal teeth but there's so many circumventions to it

Actually the question of copyrightablity remains open.  Most accurate to say it MAY be copyrightable, but regardless it is fair use. https://www.law.cornell.edu/uscode/text/17/107

Fair use assumes that the copying is infringement, which implicitly means the copied material is copyrightable, Fair use excuses the infringement, for lack of a better term.

[bernorange 8,359]

* bump *

Posting from my phone so I'm not quoting it...

https://www.eff.org/deeplinks/2023/04/sas-v-wpl-federal-circuit-finally-gets-something-right-computer-copyright

[TwiceHorn 31,440]

30 minutes ago, bernorange said:

* bump *

Posting from my phone so I'm not quoting it...

https://www.eff.org/deeplinks/2023/04/sas-v-wpl-federal-circuit-finally-gets-something-right-computer-copyright

That's kind of a shitheaded article, as is most stuff from EFF.  They're not always wrong, but they're always smarmy, smug fuckheads about it.

The Federal Circuit is not without its flaws, but it's a good court and head and shoulders above every other court in the nation in IP facility, including and especially the Supreme Court.  Sometimes, their understanding of technology and technological issues gets them in trouble, meaning they have a resistance to drawing bright lines, as in "APIs are never copyrightable," or "copying of APIs is always fair use," while also ignoring the age-old fair use analysis.  Applied to APIs as conventionally understood, those are probably workable rules right in 90 something percent of cases.  The Fed. Cir. concerns itself with the other 10% as well, because its business is mostly the other 10%.

Interesting piece of trivia.  The abstraction-filtration-comparison test was basically invented by Steve Susman in Computer Associates v. Altai.  I've heard him speak on it, and he actually barely understands it.  His associates and expert witnesses invented it, Susman argued it to the courts.

Shit, I missed that Susman had croaked.  RIP Steve.

[TwiceHorn 31,440]

On 4/6/2021 at 9:01 AM, TwiceHorn said:

In the first case, there is an argument that that is fair use.  As in the 2 Live Crew decision, if something less than the entire poem is used and there's some kind of commentary on it in the lyrics, that's maybe a good fair use case.  If it's just the poem set to music, it's probably not fair use.

Fair use is very much a case-by-case thing.  I posted the factors for consideration above.  So the answer is "it depends."

Also, it's becoming clear that the Court is trying to create a new "species" of fair use specifically for computer programs.  As a practical matter, that might be the right result as at least it takes into account all of the "facts" surrounding the copying, its purpose, etc.  Doctrinally, it's a holy mess.

Re-reading this, I feel I should elaborate at this late date. 

If the rock band takes the poem and republishes it verbatim set to music, that is classic copyright infringement.

If the band reworks the poem a bit, say omits parts, or even uses all of it, but interspersed with other lyrics, there is an argument that their use is "transforming" or maybe even "parodic."  It is a commentary on or criticism of the original work.  That potentially places is into fair use, subject to evaluation of the other factors.

That is the issue in Warhol v. Goldsmith, pending at the Supreme Court.  Goldsmith took a photo of Prince (left), Warhol cropped it and watercolored it (right).  It is a wholesale reproduction/copying of Goldsmith's photo, and by that standard, copyright infringement.

 

But Warhol's estate argued that the resulting artwork transformed the otherwise carbon copy into another work that constitutes some form of commentary, criticism, or parody of the original work.  That then collapses the first three elements of the fair use test of s. 107 of the Copyright Act:

(1)

the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;

(2)

the nature of the copyrighted work;

(3)

the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and

(4)

the effect of the use upon the potential market for or value of the copyrighted work.

All of which would otherwise favor the copyright owner pretty decisively (as to 1) it's art but it's commercial, as to 2) the original work is pretty highly creative (photography usually is) and as to 3) it's damn near 100% of the photo and certainly the substantial part of it, his face), into something that, so long as the infringing/transformative work doesn't cause the original to lose sales, means the copyright owner loses their case.

That was the holding in the 2LiveCrew/Roy Orbison case, that by using snippets, although verbatim, of Pretty Woman in a raunch ass rap song, they created a parody of the original.  But, big big but.  The holding at the Supreme Court was that this was a plausible legal theory for fair use; that it MIGHT qualify.  It did not hold that it WAS fair use.  It sent it back down for re-evaluation, where it settled, leaving the viability of "transformation" and "parody" unresolved.

This may finally resolve the question.  If it rules against the notion of transformation, which was central to the JAVA API case, then what does that mean?  Hmmmm.

Edited April 11, 2023 by TwiceHorn

[TwiceHorn 31,440]

Something still bugs me a little bit about Google v. Oracle.

Developers put out APIs to make use of the developer's application easier for others, thus making the developer's code more saleable and usable.  A win-win for the developer and for those that wish to interface other software with the developer's application.  That's the usual case for an API.

When a developer's API is used by a third party to make its application a SUBSTITUTE for the developer's application, to actively reduce sales and use, that's a little bit dirty.  Maybe a lot dirty.  It certainly seems to undermine the purpose of fair use altogether, so to call it fair use is rather bizarre.

Is it a dirty that should be provided a remedy by the copyright law?  I don't know.

[Captainant 27,638]

7 hours ago, TwiceHorn said:

Something still bugs me a little bit about Google v. Oracle.

Developers put out APIs to make use of the developer's application easier for others, thus making the developer's code more saleable and usable.  A win-win for the developer and for those that wish to interface other software with the developer's application.  That's the usual case for an API.

When a developer's API is used by a third party to make its application a SUBSTITUTE for the developer's application, to actively reduce sales and use, that's a little bit dirty.  Maybe a lot dirty.  It certainly seems to undermine the purpose of fair use altogether, so to call it fair use is rather bizarre.

Is it a dirty that should be provided a remedy by the copyright law?  I don't know.

You're missing the crux of the argument: Oracle was asserting that they owned the intellectual property of a mathematical max() or min() or avg() function - AND - oracle wanted to copyright not only the mathematical logic, but also the interface contract. That is to say, NOBODY else but them could pass two numbers to a function named min() and get the lower number back. 

 

The court absolutely got it right on this, and I'm not sure what you find so offensive about the EFF article - it's not smarmy or shitty, just lays out the technical facts that the court considered and why they decided how they did. 

An entity should not be able to own the idea of a particular interface format, especially for something as abstract and ephemeral as an API request/response handshake. If such a thing we're possible, we wouldn't have the internet and you'd be metered per TCP/IP handshake because some greedy idiot fuck with an MBA thought of a new way to increase revenue 

[bernorange 8,359]

Re-reading the article this morning and it struck me as amusing - at a high/abstract level - that "intellectual property" (whether patents or copyrights) seemingly does not encompass ideas or processes.   So what exactly is protectible?

[Fudge Nuggets 19,189]

On 4/11/2023 at 8:03 AM, bernorange said:

So what exactly is protectible?

Everything or nothing.

Depends on which side the party handing out the biggest donations prefers.

[bernorange 8,359]

EFF posted another comment on patent issues:

https://www.eff.org/deeplinks/2023/05/why-us-solicitor-general-trying-change-law-benefit-patent-trolls