Oracle's Lead Lawyer Against Google Vents That The Ruling 'Killed' The GPL
from the but-that's-not-true dept
When the Federal Circuit Appeals Court (CAFC) initially made its nutty ruling saying that APIs are copyright-eligible subject matter, many in the copyright and tech world were not only shocked, but were tremendously worried about how the ruling would impact innovation and software development going forward — while supporters on the other side brushed off such concerns.
Now that the second trial has found that, even if APIs are covered by copyright, Google’s use of the Java APIs in Android was fair use, perhaps it’s only fair that people on the losing side are lashing out in the same manner as people on the other side did after the CAFC ruling.
Annette Hurst, the lawyer who led the case on the Oracle side, posted her thoughts to LinkedIn, claiming that the ruling represents the “death of free software,” and, more specifically, saying that the ruling “killed” the GPL (General Public License, even though at the trial one witness insisted it was the Gnu Public License). From reading her post, it appears that she either doesn’t understand that software and APIs are not the same thing, or that she just doesn’t care. The whole argument is strange, and starts off with a bizarre, and simply wrong, assertion that “no copyright expert” would have predicted this result:
The developer community may be celebrating today what it perceives as a victory in Oracle v. Google. Google won a verdict that an unauthorized, commercial, competitive, harmful use of software in billions of products is fair use. No copyright expert would have ever predicted such a use would be considered fair. Before celebrating, developers should take a closer look. Not only will creators everywhere suffer from this decision if it remains intact, but the free software movement itself now faces substantial jeopardy.
Except, of course, tons of copyright experts predicted exactly this result (and many more argued that APIs should not be subject to copyright at all). Famed copyright scholar Pam Samuelson has been writing extensively about the case, focusing both on why APIs should not be covered by copyright (and, why basically every other court has agreed) as well as why, even if it is covered, it’s fair use. Hell, she even wrote a response to the Hurst piece, explaining why Hurst was wrong. It’s weird for Hurst to take a position that actually seems at odds with a huge number of copyright experts, and then state that none would take the position that many did.
From there, she appears to misunderstand the point made by the other side in the very case she led:
While we don’t know what ultimately swayed the jury, Google’s narrative boiled down to this: because the Java APIs have been open, any use of them was justified and all licensing restrictions should be disregarded. In other words, if you offer your software on an open and free basis, any use is fair use.
If that narrative becomes the law of the land, you can kiss GPL goodbye.
Except she’s exaggerating here and misrepresenting the key issues in the case. No one was arguing, as she implies, that any software that is described as “free and open” or that is using the GPL means that any use is fair. Again, she’s conflating APIs with actual software. The ruling doesn’t impact software the way she thinks it does because she doesn’t seem to want to acknowledge that APIs are not software. They’re just a structure — a table of contents effectively.
No business trying to commercialize software with any element of open software can afford to ignore this verdict. Dual licensing models are very common and have long depended upon a delicate balance between free use and commercial use. Royalties from licensed commercial distribution fuel continued development and innovation of an open and free option. The balance depends upon adherence to the license restrictions in the open and free option. This jury’s verdict suggests that such restrictions are now meaningless, since disregarding them is simply a matter of claiming “fair use.”
This is simply not true. The case revolved around the fact that the API and its “declaring code” are fundamentally different from the actual source code within the operating system. It serves an entirely different purpose. Part of the reason why the use of the same API is considered fair use is because of that very nature of it: the API is more functional — it’s like a pointer or a reference, rather than an actual bit of code. It’s only if you don’t understand that the two things are different that this ruling leads to the problems that Hurst describes. A case with the same facts, but where straight up source code was copied would have a much tougher uphill battle on the fair use front.
Developers beware. You may think you got a win yesterday. But it’s time to think about more than your desires to copy freely when you sit down at a keyboard.
Once again, this shows a rather unfortunate ignorance of how coding works. It’s not about a desire to “copy freely.” It’s about building amazing and innovative services, and making use of APIs to increase interoperability, which increases value. Copying an API structure is also just much more about making developers comfortable in using new environments. You know, like how Oracle copied SQL from IBM. Because lots of people understood SELECT-FROM-WHERE and it made little sense to create a relational database that didn’t use that structure. It’s not about copying freely. It’s about interoperability.
And, really, the idea that an Oracle lawyer is “concerned” about the future of the GPL is fairly laughable. Thankfully, many people have weighed in in the comments — including plenty who are quite familiar with the GPL and software development to explain to Hurst why she’s wrong. Somehow, I think she has some fairly strong reasons to ignore those responses.
Filed Under: annette hurst, apis, copyright, fair use, free software, gpl, software, source code
Companies: google, oracle
Comments on “Oracle's Lead Lawyer Against Google Vents That The Ruling 'Killed' The GPL”
Hurst’s piece was also posted on Ars Technica as an op-ed, where it was pretty soundly lambasted by commenters aplenty. Hope she has a thick skin…
A lawyer can afford a thin skin, because they know how to misuse the courts.
Oh, don’t worry about that, she does.
It is difficult to get people to understand something, when their salary depends upon their not understanding it.
Her salary depended on rubbing off her non-understanding on judge and jury, not on exposing herself as incompetent and clueless in public.
The op-ed is nice to read if you want to understand how the jury could find for Google after a comparatively short time of deliberation.
It’s not even that her reasoning is stretching it. Nothing is being stretched because she never even tries closing the gaps in her reasoning.
It’s the classic Chewbacca defense. Which is a gamble in court and stupid in public.
“Developers beware. You may think you got a win yesterday.”
We will certainly give your warning the consideration it deserves. You know, considering it’s coming from someone who clearly doesn’t know the first thing about what APIs are or what they are used for.
Everybody but Oracle got a win.
Re: Re: Re:
No, even Oracle got a win. If they had won that case, the rest of their products would have been in for a trillion dollar pummelling, considering the number of other vendors’ APIs they “steal”.
Oracle is just trying to spin a loss (not getting any money for Google’s Java usage) into a win (trying to convince companies to move away from GPL-licensed software, such as OpenJDK, to proprietary ones, which Oracle would love).
And it works as good as their case went. Talk about butthurt lawyers.
Sounds more like wishful thinking…
How long before a senator has a new law proposal based on what Annette Hurst believes? You can bet Oracle will be lining pockets.
Perhaps, the concept of an API should be described in
I guess I should copyright:
Noun, auxiliary verb, verb, conjunction, verb, pronoun…?
Why stop there? Copyright the English language. All those words that can construct things are just API’s.
Re: Re: Hmmm...
You can’t copyright that, because it already entered the public domain (being more than 150 years old).
You CAN however copyright any extensions you’ve invented in the past 150 years.
So who is going to claim copyright over ebonics? SMS abbreviations?
I claim copyright over use of emoticons in place of punctuation to communicate tone as well as phrase ending. I was doing that back in the early 90’s 🙂
Re: Re: Re: Hmmm...
So? It’s constantly being rereleased with minor modifications and a number of coauthors are even alive yet.
I mean, try going into a music store and buying something from J. S.
Bach. You’ll not find any score you are allowed to reproduce. And he’s been dead a lot.
Re: Re: Re:2 Hmmm...
So, by that logic, Webster, Randomhouse, Oxford, Fowler, and a few others could end up battle it out in the courts over who owns the copyrights on the English relatively current English language publishing…
Maybe describing an API in language a lawyer understands.
Tell them that an API is like a contract between the developers of a library and the users of the library. It describes what tasks will be performed, but DOESN’T describe how the tasks are performed. For instance, assume you have a sort function described in the API. The contact specifies the name of the sort function and the parameters to the sort function. The library guarantees that the result of the sort function will be a sorted version of the input. However it does not specify exactly what algorithm is used to perform the sort. It may be a bubble sort, a shell sort, merge sort, quick sort, or any other algorithm.
Perhaps that could be described to a lawyer. After all, they ought to be familiar with the concept of a contract.
Re: Maybe describing an API in language a lawyer understands.
“Perhaps that could be described to a lawyer. After all, they ought to be familiar with the concept of a contract.”
I’m sure she’s familiar with the concept. She’s just pissed off she lost and wants to start putting out ideas Oracle can capitalize on. She’ll be in front of politicians very soon.
Re: Re: Maybe describing an API in language a lawyer understands.
No, it’s further than that. She needs to spray arguments she can refer to in the appeal all over the internet, so she can the “everyone* says it’s true” argument.
She’s attempting to win hearts and minds because that’s what will be needed for the appeal to succeed.
*where everyone is myself in multiple forums, with everyone in the comments disagreeing with me.
You think that Oracle gives a whit about interoperability? Think again. Oracle’s idea of a whopping good time is a world where *they* control, and profit from, any product that wants to exchange data with anything from the Oracle world. You know, sort of like Apple.
Somehow, I think she has some fairly strong reasons to ignore those responses
Did you mean $$$trong reasons?
Tipping your hand
Clearly, this entire lawsuit was and
*still is*, all about trying to kill GPL.
The fascist darkside can not deal with
GPL and other FLOSS Licenses. It is all
about monetization and control.
This is not over by any means.
Can someone please help me out here?
Notwithstanding that Google won this round, what would be the implications that an API can be copyrighted?
Would that mean that if Company A had an API with a method called “Find” that accepted an author name, that Company B could NOT also have an API with a method called “Find” that accepted an author name? (unless Company B licensed from Company A)
Am I understanding that part correctly? Cause if that’s so, it seems completely and utterly nucking futs. Beyond insane.
Re: Copyright API
You’ve understood it perfectly. And this is why there was so much jaw dropping when the CAFC decision in 2012 said that in this instance, the APIs could be considered copyrighted.
I have to agree with Oracle. In the end, this decision, if it’s allowed to stand, will see content created by other developers used freely without compensation or license. Copyright/trademark/patent is simply a way for someone to protect content that they create.
If developers are celebrating the decision for Google then they clearly haven’t looked long term at the problem this will create.
You sound like the lawyer who doesn’t understand. I’ll also bet that you aren’t a programmer either. Like I said earlier, an API is like a contract between the application developer and the library developer. It specifies how the library interacts with the application, but it does not specify how the library actually does its job. For instance, in the C language standard, there is a description for the function sqrt() which returns the square root of its parameter. But there is no mention as to how the square root is obtained. The same principle applies to the API that Oracle fought the lawsuit over.
I actually have to agree with your first paragraph. Someone creates the APIs. These get to be used freely by other developers without compensation or license.
However, APIs don’t need “protecting” from anything — in order for them to be interfaces for programmers, programmers need to be able to interface with them. If you “protect” your APIs, there’s no reason to HAVE APIs. Just implement it all in-line, and if you want anyone else’s code to interoperate with your own, get together, sign a contract, and show them the parts of the code they’ll need to refer to.
It’s kind of like opening up a new marina, but requiring anyone who ties up to your dock to purchase and use your own ropes, which they return to you when they leave. Of course, they need to get those ropes from somewhere. This means that unless someone was all along intending to tie up at your new dock and already has a relationship with you, they’ll go elsewhere when they arrive without your special ropes.
Of course, what Oracle has done is bought an established marina where people could use whatever ropes they wanted, and then retroactively stated that some other marina that allows boaters to use ropes purchased at THEIR marina before they bought it are somehow stealing their ropes, and their rope interface (the dock cleats) because the cleats look exactly the same as the ones they use.
Agree all you want, but you’re simply wrong.
“if it’s allowed to stand, will see content”
Did you read enough to understand that APIs and content are not the same thing? Apparently not.
“Copyright/trademark/patent is simply a way for someone to protect content that they create.”
That might be the intention, but it sure as hell isn’t how it’s used and abused by some. Which is why this is the correct decision.
By the way, you haven’t bothered to state why you think APIs should be copyrightable. Would you mind discussing this point, since it’s literally what this argument is about? “Content” and source code are completely different subjects.
“I have to agree with Oracle. In the end, this decision, if it’s allowed to stand, will see content created by other developers used freely without compensation or license. Copyright/trademark/patent is simply a way for someone to protect content that they create.”
Why are you even committing if you are totally ignorant in how programs are coded.
I have to agree with Oracle. …blah blah blah…
Let’s not pile it too high and deep.
Here’s a list of this lady’s cases:
It’s a really mixed bag and I’m sure she knows better.
But Mrs. Annette Hurst is, for now, literally Oracle’s paid shill.
This status precludes any personal opinion on this mess.
As long as she’s paid, it’s literally (and legally) her duty to muddy the other side as much as possible. And yes, lying is accepted.
Honestly the same could be said about a lot IP lawyers for plaintiffs. Do they think that the grandmother without a computer downloaded that file? Probably not, but fuck that shit, there’s hundreds of thousands of dollars in settlement money at stake here!
I don’t know if it’s irony or hypocrisy (or maybe even idiocy) that a company that threatened software being freely adapted is now claiming to be protecting freely adaptable software.
Reuse / reimplementation of things like an API is the norm. It’s how interoperability works in technology. Heck, it’s how Java itself works!
I read the op-ed on Ars Technica and I am frankly struggling to understand it. What does SaaS or Google’s advertising have to do with this case? What does Google releasing free software have to do with anything? Software is never “owned” because software is never “sold” – only licensed. Speaking of… Licensing is not copyright either. Breach of a license is not necessarily a breach of copyright. It looks like a childish attempt to say “But look at the bad stuff Google will do!”
Of course, none of this matters when there is potential money at stake.
No, Annette. I am a developer. I know what APIs are. I know how they’re used. I know their purpose. I saw this case for what it was: a vindictative attempt at extracting money.
No, Annette. I am a developer and I was never on Oracle’s side.
And yet another shill claims APIs are copyrightable cause beautiful math, or something.
I got into a Twitter spat with a man claiming that “consume” is a normal, ordinary, everyday term used by developers to describe the usage of APIs, where “consume” = “use.” I told him I had a problem with the loaded language in the post linked above because the author emphasised the word “consume” with regard to the use of APIs but he’s mad keen on the usage thereof. He’s the editor at Ars, it seems. If that’s true it explains the nutty op-eds from the pro-Oracle camp.
Re: Re: Re:
In the comments to one of those op-eds they justified putting them up… rather unconvivingly so I actually forgot their reasoning.