US Supremes urged by pretty much everyone in software dev to probe Oracle's 'disastrous' Java API copyright win The US Supreme Court has been urged to hear Google out in its long-running copyright battle with Oracle over the search giant’s use of Java technology in Android. Some 14 amicus briefs have been filed with the top court in support of Google, with Microsoft, Red Hat and Mozilla, along with the Python Software Foundation, …
I do not claim to be fully versed in the seemingly never-ending saga of Google v Oracle (a pox on both their houses, I say!). But consider the Gof Adapter pattern. If I understand the appeals court ruling correctly (and IANAL, fer sher), one could make the claim that this pattern is inherently in violation of copyright (or, more accurately, is a tool for violating copyrights). If, for example, one created an Adapter that adapts the JAVA API to some other operating environment, have we run afoul of the ruling? What is someone else adapts some other "copyrighted" interface to its environment? One could argue that the recently-released Qt-under-Python interface is an adapter of the Qt API to the Python environment1 And so on.
Seems to me that the Oracle decision was "wrongly decided".
1This wouldn't happen, because the Qt folks were actively engaged in this one. However, Qt itself has undergone several changes in ownership over the last decade, and whose to say whether down the road Oracle might make a play for it?
Please read my post on this...
The short answer is no. That is not what the court ruling is saying.
Your example would be covered under fair use.
This is not what Google was doing.
The courts actually got it right.
If I had to bet. SCOTUS will again refuse to hear this.
Both companies are dicks , weasels, scum of the earth and evil.
But the underlying facts are that Google was caught trying to pull a fast one and Oracle had deep enough pockets to hire really good lawyers and fight this to the end.
Google can easily write the check and walk away.
@Ian Michael Gumby
Either you do not understand how the law works in large parts of the world (including the US) or I somehow missed that this decision would not set a precedent. In case this is the latter, please kindly show the source.
In case there is a precedent, the decision basically means that whoever first grabs copyright for a function name and signature will be able to prevent everyone else from using the same name and signature in their own code. Which is very bad news for developers, because there are only so many sensible names and signatures (as well as code idioms and design patterns) which we can use to solve a common reappearing problem. It also makes it very difficult for developers to change jobs or contribute to open source projects because anything we have done in the past becomes huge legal liability.
"...whoever first grabs copyright for a function name".
Uhhhhhh, from all the articles I've read, it's more like using the interface no matter what is a problem, which is worse. For instance, they can't sue for int main(blah), but they can sue for the very concept of int main(), no matter the syntax. It's crazy!! In the legal filing I read (~8 years ago), Oracle intentionally used API examples and grammar in ambiguous ways.
Java... Even if the devil had the sweetest cake, would you make a deal?
they can sue for the very concept of int main(), no matter the syntax
Since copyright covers expression, not concept, "all the articles [you've] read" seem to be wrong. Or you've misunderstood.
Don't get me wrong - I think the CAFC ruling was wrong, I think Oracle are full of shit, I've yet to see an even slightly persuasive argument from their supporters, and I'm hoping SCOTUS takes the case and reverses. But I don't believe you're correct about the consequences of this ruling.
Yes, you are correct, he got it wrong.
There's more to it than that, but its not what killed this case for Google.
If you bother to read the judgement, the courts found that Google failed to explain why the appellate court should reverse the decision that APIs are copyrightable. (Its in their decision.)
To be clear... Google blew it and now they can't go to SCOTUS and argue it. SCOTUS won't hear that issue.
So you get down to the argument that Oracle made. What Google did wasn't fair use.
And the courts explain why they agreed with Oracle.
Sorry mate its you.
You now have a court ruling in favor of Oracle.
In future court cases, lawyers can cite this case, however its very possible that a judge in a different federal district court would rule differently.
There are cases of this happening.
Unlike a SCOTUS decision which becomes a precedence, this may or may not set precedence.
And again, you also fail to understand the ruling as well as copyright law.
I'm not going to sit down and explain it to you, you should try to learn it on your own.
Read the case judgement and you might learn something.
Or you can try to do just what you suggest and then try to sue someone on a single API call. ;-)
I'm pretty sure I'm going to get down voted by many but F' it. I think its important to really understand what is at stake here.
This isn't about copyrighting code but more of fair use.
If you go to the Wikipedia entry on the case you will find this gem:
The Court found that as a matter of law, Google's use of Java could not have fallen within fair use, even if all factual matters decided by the jury had been in Google's favor. The Appeals Court found that Google's use of API code declarations had not met any of the four current criteria for fair use, but was merely untransformed reuse. It had not been transformative, since it was used for the same purposes without even minimal changes or rewrites. It was not minimal, since it was agreed that only 170 lines of the 11,500 lines copied were needed for Google's purposes. It was not within any example of transformation, nor intended to permit third party interoperability, since Google had made no substantial efforts to use them for the purpose of third party interoperability. (In fact it found that Google had tried to prevent interoperability with other Java and had previously been refused a license by Sun for that reason.[12]) It was not transformative in the sense of a new platform either, since other Java smartphones predated Android.[62] It was plausible that the use had harmed Sun/Oracle – perhaps to a great extent if Oracle were to be believed – since as a result, vendors began to expecting Oracle to compete on price with a freely available derivative of its own language, and to require very steep discounts and undesired contractual terms.[62] Therefore, Google's use of the Java code and APIs failed to meet all four of the currently accepted criteria under which fair use would be possible.[62]
You really need to get a single malt, pint, coffee, or tea and sit down and read the following:
http://www.cafc.uscourts.gov/sites/default/files/opinions-orders/17-1118.Opinion.3-26-2018.1.PDF
The courts are saying that Google's use was not 'fair use'. So for everyone who's up in arms that now this ruling breaks the industry... its false. Fair use still applies. So you can freely use APIs as they were meant to be used.
Sorry, but while both companies are evil, if you're being objective and take the time to read the court's decision... It was the right one.
What I found interesting in the Wikipedia doc:
(In fact it found that Google had tried to prevent interoperability with other Java and had previously been refused a license by Sun for that reason.[12])
So there is more to this case than just Oracle being dicks.
The issue is not whether or not it is fair use, but whether copyright should extend to what is essentially a list of names. It was the case previously that an API was treated like a directory listing or a phone book, neither of which are subject to copyright under either us law or Berne. The court, failing to understand what an API is, extended copyright protection to that list, which is clearly well beyond the scope of copyright. Fair use is a red herring.
Look, I'm not going to play lawyer here... but clearly you didn't bother to read the judgement.
(I even provided a link FFS!)
SCOTUS already decided not to hear Google's first appeal. As to APIs being copyrightable... they are. To a point. There is no SCOTUS decision but a lower court's decision.
From the actual judgement:
Oracle now appeals from the district court’s final judgment and its decisions denying Oracle’s motions for JMOL and motion for a new trial. Google cross-appeals from the final judgment purportedly to “preserv[e] its claim that the declarations/SSO are not protected by copyright law,” but advances no argument for why this court can or should revisit our prior decision on copy-rightability. Cross-Appellant Br. 83.Because we conclude that Google’s use of the Java API packages was not fair as a matter of law, we reverse the district court’s decisions denying Oracle’s motions for JMOL and remand for a trial on damages. We also dis-miss Google’s cross-appeal.
Please read the part in BOLD.
As you can see the courts are ruling that Google infringed on their copyright because it is a fair use issue.
I don't know what Google will do besides try to appeal to SCOTUS.
Will they get heard? I don't know but based on the earlier attempt, I doubt it.
Pretty well the entire software industry is saying that fair use requires a different interpretation to that which the court gave in order for software development to continue. But you're arguing that the court knew more about software development than the industry.
Is that a fair summary?
No, he's arguing that the courts know more about "fair use" and that the interpretation that the software industry "require" doesn't exist in law (that's what the judgement is). So the court doesn't care that it's software industry, or that software industry has always worked this way; rather the law says you can't rip off someone else's product unless you meet these criteria.
The Wikipedia page has a good summary:
https://en.wikipedia.org/wiki/Oracle_America,_Inc._v._Google,_Inc.
"during the second appeal hearing, Google had used this code for commercial reasons to rapidly complete Android and to avoid the "drudgery" of recreating the code."
As an old-timer, it kind of bemused me that you could take, say, WordPerfect, rip off the look and feel of it, even to the point where you could save documents in WordPerfect format, and then walk away with money.
As an old-timer, it kind of bemused me that you could take, say, WordPerfect, rip off the look and feel of it, even to the point where you could save documents in WordPerfect format, and then walk away with money.
So, you would reverse the 20-year-old decision in Lotus vs. Borland https://en.wikipedia.org/wiki/Lotus_Dev._Corp._v._Borland_Int%27l,_Inc. ? And kill LibreOffice as well?
It isn't about ripping-off someone else's product. It is about re-implementing it to do the same thing without copying their code. That has always been allowed.
In the case of an OS/platform/framework the API is the user interface so that precedent should apply here as well.
It isn't about ripping-off someone else's product. It is about re-implementing it to do the same thing
In much the same way as MIcrosoft did with the OS/2 API calls - they essentially duplicated them (same parameters and all) and just changed the names slightly.
(And yes, I know that Microsoft worked with IBM on OS/2 until the toy-cot-ejection but the API stuff happened well afterwards. Our OS/2 devs at the time had an incredibly easy job to convert OS/2 code over to NT code - essentially just use sed/awk to change the API names..)
"So the court doesn't care that it's software industry, or that software industry has always worked this way; rather the law says you can't rip off someone else's product unless you meet these criteria."
Originally the law was built as a means of respecting customs. In fact one of the reasons behind introducing juries in the first place is that they can say what the custom is in a particular community. Nowadays it seems that that particular role has fallen to amicus curiae briefs.
It would help immensely if the jury actually consisted of real peers of any particular community rather than just anybody that couldn't get an excuse from duty. At least in the military I could ask that a jury at my Court Martial consist of fellow enlisted members of the service. [Not a good idea, actually, as they are far less forgiving than officers.]
[There's zero chance I'll ever serve on a jury. My knowledge of the stupidity of the criminal justice system is far too detailed about both sides of the process.]
anybody that couldn't get an excuse from duty
Some of us are willing to do our duty, rather than trying to get out of it.
There's zero chance I'll ever serve on a jury
I wouldn't be so sure. I've been surprised at how poor some lawyers are at voir dire. I served on one jury where defense failed to use one of its peremptory challenges and included a juror who was almost certainly not good for the defense, even if she was as well-intentioned as reasonably possible. (Among other factors, this juror was blind and the better part of the defense hinged on visual exhibits.)
Anyone who's eligible and not willing to outright lie during voir dire has a chance of serving. Trial lawyers are an unpredictable breed, and they're gambling during selection anyway.
"Pretty well the entire software industry is saying that fair use requires a different interpretation to that which the court gave in order for software development to continue. But you're arguing that the court knew more about software development than the industry.
Is that a fair summary?"
...which is precisely why this case really does need to go before the Supreme Court to be considered in full because of the significant implications for developers. Kudos to all those other companies for submitting depositions about this matter and special thanks go to Satya's Microsoft for joining the ranks of the good guys (more of the same, please).
As to APIs being copyrightable... they are. To a point.
That point being fair use, which most people here, and all of the amicus filers supporting Google, are arguing is the wrong standard to apply to software.
Your argument seems to be that everyone arguing against you misunderstands the CAFC ruling. I believe you're wrong. We understand it just fine. We think it's a bad ruling.
There is no SCOTUS decision but a lower court's decision.
Yes, everyone understands that. We're hoping SCOTUS takes the case this time and reverses.
Fair Use is an abysmally unhelpful standard to apply to APIs. For emulation, interpolation, and other well-established and commercially-productive uses, the new use cannot be transformative, and often cannot be limited. It can't satisfy any of the other fair-use tests either. That is the essential point.
CAFC has historically been notoriously conservative and narrow in IP cases. It's often bad to let it set precedent. Applying copyright and the Fair Use standard to APIs was a poor interpretation of USC 17.
Google may have copied more than APIs. Fine; remove copyright protection from APIs and remand to the lower court to determine how much of a penalty Google still have to pay for violating copyright on the remaining protected material. I don't care what that figure is, frankly; it can be the entire $8.8B or whatever.
The fact that the courts have been treating it as a fair use issue is the entire problem. Headers and API descriptors should not be subject to copyright. Oracle managed to get copyright extended to cover things that it shouldn't cover, and the amicus briefs urging the supreme court to consider this appeal are making clear that the entire argument - the claim that a list of function names is subject to copyright - is wrong on its face.
The initial ruling in Google's favour made clear that a list of function names in a header file is not subject to copyright, because it is merely an index of what functions exist within an API, like a telephone book or index cards in a library. Copyright does not apply to lists unless there is some creativity in how they are organised (and even then there are tests), and it is safe to say that a list of function names in a header file, organised either at random or alphabetically, does not qualify.
Oracle's appeal managed to overturn that sensible ruling and place a new, senseless one in its place, one that has consequences far beyond android and java, or even the software industry as a whole.
As I said, whether or not it is fair use is a red herring. The argument being made by the amicus briefs is that fair use should not need to be considered because copyright doesn't apply to lists.
I would also note that the appellate court, generically referred to as the patent court, based its decisions on facts not presented in the original case, which as the Supreme Court has noted repeatedly in slapping them down, is wrong.
Appeals are supposed to be decided on issues of the law, with the lower court being (pretty much) the arbiter of issues of fact.
The reason that SCOTUS has been reversing them regularly lately is because the patent court is bat guano insane, and has been for years. (They literally allowed the patenting of a rainy day)
And you and the above posters just don't get it.
I mean what you are saying is correct, however the courts have to rule based on the evidence that is presented to them.
The appellate court will hear facts that may or may not have been presented in the original court case.
This means that if the judge errs in his decision because he misinterpreted the facts then they have grounds for an appeal.
Or if there's case law that wasn't presented or allowed they will hear it.
Or if there's an argument that the judge's ruling ignored... they will hear it.
But you have to make that argument to the appellate court.
Google didn't do that when it came to the decision that APIs were in fact copyrightable.
Got it? So the court didn't err. Google's lawyers did. (oops!)
The appellate court didn't hear any argument from Google that the APIs were not copyrightable.
So its a moot point. So the court has to start with the earlier decision that they are copyrightable.
Did you or any of the others actually read the decision?
Or any of the earlier decisions?
The ruling doesn't say what you think it does.
Again, my money is on SCOTUS not taking the case.
This case will not be a precedent and the world of software development will not melt down over it.
Ok you make some fair points, but the ruling itself shows a lack of understanding of the basis of software API's.
The biggest one is the 11.500 lines of code allegedly copied by Google. That sounds to a layman like a slam dunk case. However when you analyse it as a software developer it is less so.
What are the 11500 lines of code
The Java library code consists of 2.86 million lines of code, therefore Oracles case rests on 0.4% of the code base being copied. Basically those 11500 lines are the Java 2 API headings, virtually everything else is Google's own work . Google's purpose was to to create a language that was familiar to Java developers but fitted in with Android. To most software developers, there is nothing wrong with that. If i want to use C ion a new hardware, I would recreate the C standard library and write the back-end. Or say I wanted to write my own version of SQL, like a small company like Oracle did way back.
I cannot see where the court gets 170 lines from. How would you write a compatible JAVA library using 170 lines?
Its a bit like Collins claiming copyright over oxford dictionaries, because they used the same word headings even though the definitions were totally original.
The rulling is a total mis-representation of software API's and opens the door to any other corporate vulture claiming ownership over software re-use
Google's purpose was to to create a language that was familiar to Java developers but fitted in with Android.
I'm not sure if this is entirely correct. Google could have continued to work with Java but decided for commercial reasons not to, ie. there is some degree of bad faith on Google's part. However, it's still reasonable to argue that this was still fair use.
If the decision does get upheld by the Supreme Court then I wonder if Oracle is prepared for the inevitable countersuits on things like its core business: lots and lots of database code is based upon the principle of reimplementing existing APIs and this would be a treasure trove for patent trolls.
This is probably why so many companies are against the ruling as it could open the floodgates for all kinds of cases, which even decided against in the courts would divert huge amounts of resources.
Yes and no.
First, read the judgement. The court explains why its not fair use.
For this to go to SCOTUS, Google or an amicus brief has to show that the court erred in their interpretation of fair use. (You are correct on the bad faith argument)
Because Fair Use is complex, this is possible however it would have to be a really good argument.
This is why I suspect that SCOTUS will not want to hear this case.
If the decision does get upheld by the Supreme Court then I wonder if Oracle is prepared for the inevitable countersuits on things like its core business: lots and lots of database code is based upon the principle of reimplementing existing APIs and this would be a treasure trove for patent trolls.
There likely won't be. You'd have to show bad faith and that you own the copyright on the APIs along with a bunch of other things.
You would also have to show how a patent troll could sue over a copyright but that probably won't stop someone from trying it. In the US you can sue anyone for just about anything, however it would probably get dismissed outright or before heading in to a court case or discovery.
Sigh.
The courts are using the evidence of what is being presented to them.
Of the code copied by Google, which Google doesn't deny, a very small percentage of the code was different. (Using your number 11,500 only 170 lines were specific to Google)
The courts ruled that this wasn't transformative but mostly copied.
Suppose I write a song. You take my song and change one sentence.
The melody is the same, most of the lyrics is the same. You would lose the lawsuit because your work was not transformative enough. Its theft.
See, I just explained why the courts ruled against Google without this being about code.
The analogy doesn't hold up because Google didn't wholesale copy the 2million LOC with just a few changes, they copied the interfaces. To go with the song analogy it would be like copying the song titles of an album but rewriting each song, still a bit weird but their own work nonetheless.
I think the reason it's still in the courts is that the initial judgement was also well argued. On issues like these there are often (at least) two contradictory but well-founded opinions. Google's behaviour was certainly questionable, but Oracle's assertion of scope is even more egregious and with potentially far greater consequences for software development in the US.
IOW: be careful what you wish for.
Even after many decades of reading relevant case law and stupid software patents I found myself just shaking my head again and again while reading the opinion of the Appeal Court. Hardly surprising considering the utter lack of relevant experience or expertise from the three judges in question. One doddery, one no relevant legal experience and one who seems to have dabbled a bit in patent law. The original trial judgement was completely correct and based on relevant and knowledgeable real world experience. The Appeal Court opinion retreats into legal trivia and arcana . Always the sign of the work of second rate legal minds. Which is documented in this opinion in abundance.
The fact that they talk about SSO copyright for the core language library like J2SE shows the depth of their ignorance. There is not a single piece of functionality as covered by this interpretation of SSO copyright that has not been implemented multiple times by equivalent language libraries in the previous three decades. There is really only one way you can open a file. Or states for a thread. There again, having read a very large number of software patents over the years, I have yet to read one that is either original, novel or non obvious. But given the fact that the USPTO has such low grade examiners, they are only working there as its a fast track route to becoming a patent attorney, its hardly surprising.
The statement that J2ME was the most successful mobile platform pre Android was probably the biggest laugh in the whole opinion. There was a Symbian and even a Brew software ecosystem pre iOS. Not a sign of J2ME. You could be sure of the core libraries being on most devices at the time but not enough to actually write any useful software for.
Larry is what, 74, he should just hurry up and die. Assuming that he has not asset striped Oracle to insolvency before that happy day. Or been sent to prison for doing same. Once dead or departed all these viciously stupid lawsuits will be dropped and maybe Oracle can start doing useful things again and stop being the biggest dickheads in an industry that is stuffed to the gills with utterly insufferable pricks. By definition, anyone who brings a LLD to a business meeting is automatically a dickhead.
You really don't know much about the judicial system.
The judge rules on what is argued in Court.
And in this case... Google didn't do a good enough job.
If you read the judgement. The final appeal which Oracle won... It wasn't about copyright but fair use.
Fair Use is a very complex issue under the copyright law.
While everyone is panicking they need to see what the court actually said.
I'll get down voted, but hey, too many clueless fucks don't want to take the time to read, nor really try to understand the US court system. (Yeah I've had to spend far more time than I care dealing with lawyers these days.)
> You really don't know much about the judicial system....
Actually I do. Wrote my first expert witness / statement of fact exactly thirty years ago. For one of the biggest ISV's at the time. Never went to trial . Other side wrote a very big check. State of California Court. Been chewing up and spitting out second rate lawyers who know nothing about tech on a semi regular basic ever since. Patent lawyers are particularly inane. Last time I talked to lawyers about expert witness work, oddly enough on an android case, no connection, the lawyers decided to pass as I was vastly overqualified for what they needed. Pity, because it seem like an interesting case. But they were right. I was overqualified.
So I do know what I am talking about. As a whole bunch of lawyers over the decades have found to their cost. Because it turns out after learning how to quickly dig through / fix / then ship very large many multi 100k commercial code bases in a very short turn around time frame, wading through the relevant case law / statute law verbiage is a doddle. Legal cases are just games. Nothing more. Law is simple, but very deliberately obfuscated. Pure lawfare. Whereas shipping products that work for millions of customers, now that's difficult. Which is why its my day job. Much more challenging. And rewarding.
It wasn't about copyright but fair use.
The fair use doctrine is part of copyright law.
Fair Use is a very complex issue under the copyright law.
No, it isn't, particularly. It's actually quite straightforward. Deciding whether fair use applies can sometimes - sometimes - become complex, but the doctrine itself and the tests are straightforward to describe and usually straightforward to apply.
many clueless fucks don't want to take the time to read, nor really try to understand the US court system
If you're setting the standard for "understand[ing] the US court system", then I think you'll find there are plenty of people here who meet it.
Once [Larry is] dead or departed all these viciously stupid lawsuits will be dropped
You think? I doubt Larry cares these days, he's having more fun sailing. My money is on Safra being behind these, and things will not get better when Larry is gone.
Please read the judgement.
It talks about the history of the case and the prior court rulings.
It also talks about why the found in favor of Oracle's argument.
This case isn't the same as any of the other cases and if you actually read the court's decision you would understand why.
First in the latest ruling Google failed! to articulate why the courts should reverse their ruling that APIs are not copyrightable.
Then the courts found that Google's actions were not fair (fair use) and thus they had to rule in favor of Oracle. Simply put... this case is being decided that APIs are copyrightable. And while everyone goes in to Chicken Little mode, you have to read the ruling to understand why the courts ruled in Oracle's favor. Google raised an affirmative defense and failed to defend it. One of the major issues.... the majority of the code is not specific to Google. (You can find it for yourself.)
And the kicker... Google got rejected by Sun, yes SUN MICROSYSTEMS for a license before Oracle bought Sun.
> Then the courts found that Google's actions were not fair (fair use)
And you seem to conveniently forget that it was a jury that found that Google's use of the API's was covered under fair use. That happened at trial.
So what we have here is an appeals court overturning a jury decision.
Normally, right-wingers would scream murder about this. Judicial activism and all that.
“Right-wingers” know full well that juries get things wrong sometimes (e.g., O. J. Simpson). Judicial activism is when a judge creates a law that did not exist prior to his rulings. That clearly is not the situation here. The reason right-wingers object to activist judges is clear; legislatures should make laws, not a handful of non-elected judges. Yet another concept those on the left do not understand (but watch them suddenly understand it if judges rule that a ‘person’ exists at some point before birth).
And wtf does this issue have to do with right/left politics? Are you that obsessed with right-wingers?
> Judicial activism is when a judge creates a law that did not exist prior to his rulings.
You would've been better off had you not typed that gem.
Judges don't create laws - contrary to what right-wingers love to claim and believe. Judges don't have that power.
Judges may interpret laws, and may create precedents, in ways that right-wingers do not like. And that is when said right-wingers get all up in arms and start screaming about judicial activism. Problem is: interpreting laws and creating precedents is precisely what judges - and courts - are supposed to do.
I'm happy to notice that you have no problem with appeals courts overruling a jury decision when it fits your narrative.
Some train of thought consistency would help your credibility, you know.
"Judges don't create laws - contrary to what right-wingers love to claim and believe. Judges don't have that power."
They do. Anyone does if they're creative enough. It's the idea of re-interpreting existing laws to produce something novel. One of the very first SCOTUS cases created a law or the like out of thin air that still exists today: the ability (nowhere explicitly written in the Constitution, I might add) to declare an Act or Order unconstitutional and therefore null and void.
I hold little the Nine Seniles will take this case and if they do by some remote chance overturn the appeals court. You are dealing with many who beyond senile and cannot understand how computer works. Now add in programming and they are so far below water it makes the Challenger Deep look like a wading pool.
First, there's a high probability that they will not hear Google's appeal.
Google has to appeal that their use was in fact Fair. They will have to make the argument as to how and why the Appellate Court erred in their decision. If you read the decision, Google cannot argue that the courts erred on ruling that APIs are Copyrightable. They lost that ability because they failed to argue it in the Appellate Court. (Read the decision)
The Amicus Briefs could probably raise this issue, however it could be moot.
Scotus will look at this based on the argument that what Google did was fair use of copyrighted material.
They don't need to know anything about programming.
To be clear, I've had to experience bad judges and lazy judges. But in this case. The court did its job and SCOTUS may not hear the appeal.
Basically in this case SCOTUS rule on a point of law, not actually the case, they can rule that the AC misinterpreted the fedral copywrite law and an API structure is not a copywriteable.
Google are appealing the lower court's decision that the API is copywriteable, as if the API is not copywritable, then there use, wether fair or otherwise is not relevant.
if SCOTUS set the president that API listings are akin to dial plans, then the industry is safe.
The Supreme Court has been pretty good on patent and copyright cases in the last 5 years or so.
The FC appeals court has been advancing the scope of patents and trademarks for a couple decades. But the Supremes Alice decision and a few others has slapped the FC down, reversing much of their scope-creep.
I think there is a good chance they'll take it. What they do with it I can't say, tho my personal hope is that they overrule the FC's API copyrightability ruling, in which case the fair use question becomes moot. However, if they do uphold copyrightability, then the fair use question becomes important, and I hope they'd overrule the FC here.
Charles 9. You are wrong. If your initial appeal did not include an argument, you can not use that argument in an appeal of that Court’s ruling. You can not appeal based on something that was not ruled on. You must argue that the previous Court was wrong on its ruling of your appeal, not that there are additional issues to consider.
"You are wrong. If your initial appeal did not include an argument, you can not use that argument in an appeal of that Court’s ruling. You can not appeal based on something that was not ruled on."
You can if it's germane. Otherwise, new evidence not known (or possible) during the original case could not be introduced (meaning all those DNA exonerations would be invalid).
Google is Google, can they not just pay for it.
Or have android rewritten to not use Java (it's not as if it's an important unique language that IT can not do without).
Or did they just decide they had the better lawyers....?
If you gamble, you got to be prepared to lose......
.....or gamble again and sue your broker and anyone else you think might be able to convince a judge are at fault.
The problem is that this jeopardizes most cases of reverse engineering, etc. Seattle Computing re-implementing the CP/M interface for 86-DOS (look where *that* went...), Compaq/AMI/Phoenix/... re-implementing the original BIOS for All Those Clones, hell Torvalds re-implementing from the AT&T docs to get Linux (good God, nobody tell SCOX or they'll wake up again).
The previous poster talking about taking an album and writing new music to the song titles and releasing it as new content is probably best; Oracle's claim that API design is Hard Work and should be subject to copyright is so much rubbish. The potential impact to the software industry in general if reverse engineering is effectively banned would be tremendous.
This case has been going on for many years.
I really suggest people actually read the actual judgement.
It explains why the courts did what they did.
If you read it, you'll understand that Google's actions are not fair use which was part of their affirmative defense.
The court case is judged by the evidence that was put in front of a jury and a judge.
The court has actually gotten this one right.
I'll wager that SCOTUS will not hear it and those who are pushing amicus briefs have their own agendas. Fair Use will hold.
This won't be a precedent until SCOTUS gets engaged. (Its in a single Federal District Court and it means other Courts can decide cases differently.)
You really, really, really need to read the judgement.
"I really suggest people actually read the actual judgement.
It explains why the courts did what they did."
The reason the justice system has multiple layers of appeal is that sometimes courts get it wrong.
In this case you have almost the entire software industry, having had innumerable well-paid lawyers read the actual judgement, saying that, in respect of APIs, the courts got this one wrong.
@Doctor
"in respect of APIs, the courts got this one wrong."
Or got it right, but the law is wrong or deficient?
History is full of courts radically revising what is acceptable, usually when an actor takes the existing gray area and drives a truck through it.
History is also full of highly paid lawyers making a cock of technical matters.
"Or got it right, but the law is wrong or deficient?
History is full of courts radically revising what is acceptable, usually when an actor takes the existing gray area and drives a truck through it."
One of the aspects of law is that when it's laid down in legislation it's done by legislators who aren't omniscient (definitely not!!) and can't anticipate the ways it can be applied. We then have judges who can take the law, the particular circumstances of a corner case and say how it should be interpreted in those particular circumstances to provide the best fit. And the higher up the chain it goes it's likely that the circumstances, as here, need to take into account the public good. That way the interpretation of the law can be adapted to suit requirements the legislators didn't and maybe couldn't have anticipated.
It looks like here we have one of those circumstances where an actor has driven a truck through one of the grey areas. It's now up to the courts to come up with a decision as to what the best interpretation of fair use in regard to APIs. And with almost an entire industry lined up in the form of amicus curiae briefs saying more or less the same thing it's difficult to escape the conclusion that (a) the public good is involved and (b) the interpretation given in the previous court does not match what the public (in the form of those working in the area) expect it to be or previously understood it to be. That seems to me to be a circumstance where the court needs to look at that previous decision and provide a new ruling that reflects the industry view of what's workable.
What the majority of you do not understand is that judges rule according to how the laws are written (or are suppose to); not justice, fairness, the American way, or what’s best for an industry (although the can consider the welfare of the public).
Most of the arguments I have read indicate that the fair use definition and patent laws may need to be revised, but that is NOT the purview of Courts. Those disdaining this decision need to be petitioning Congress.
More than a few times I have heard a judge say, “I don’t like making this ruling, but it is the law.” Once I heard a judge say, “I can’t legally rule this way, but I will, because I know it won’t be appealed.” I considered that judge to be unfit for duty (and would never want to have a case in front of him).
"...damage innovation"? Bovine excrement. It will encourage innovation by causing developers to develops other methods and routines to perform the task Java does. The other companies don't want to spend the money to do the development and research but would rather grab stuff that's "free'. Kind of like the businesses insisting on putting everything out on that oh so secure internet.
Java pushed off the mobile environment ? Well yes because it was diabolically crap.
Oracle had the driving seat for this and completely blew it whilst Google came up the rail and shot into the distance. Oracle were too busy working out how much money they could squeeze out of it to see the bigger picture.
I'm seeing more and more projects and products move to OpenJDK just because they they don't want to have their balls gripped by Oracle and are worried that it's going to cost them a licence to use it. Docker is a case in point - I can pull an OpenJDK image but if I want Java 8 from Oracle I have to "checkout" and agree to all my details being supplied.
Leisure Larry, being a greedy bastard, could not see the forest for the trees as the adage goes. He failed to grasp the more people are using Java and JVM ultimately the more they can make with commercial support. Also, more people are writing Java code because of Android. By being the greedy bastard, he made OpenJDK much more important and made looking languages like Kotlin or Scala more reasonable as they run well on the OpenJDK. More effort is being made to make the OpenJDK the go to JVM. This can only hurt selling commercial support and harassing customers about their Java licenses.
Yes he might get some cash from Chocolate Factory but he probably has thrown away a lot more money over the long term. Android is not going anywhere but if it is OpenJDK and Kotlin is getting a lot of love from Chocolate Factory he will not get any more money. Of course whether he gets cash depends on the Nine Seniles and whether they rise above their collective IQ just south of a turnip's. In all likelihood he gets his money, I have little faith in the Nine Seniles.
"By being the greedy bastard, he made OpenJDK much more important and made looking languages like Kotlin or Scala more reasonable as they run well on the OpenJDK. More effort is being made to make the OpenJDK the go to JVM. This can only hurt selling commercial support and harassing customers about their Java licenses."
You know that OpenJDK actually maintained, among others, by Oracle, don't you?
I'm seeing more and more projects and products move to OpenJDK just because they they don't want to have their balls gripped by Oracle
Just shows that "the software industry" is a rather thick learner. Once "enough" projects and products are on OpenJDK, the Oracle legal team will come up with a licensing model designed to Squeeze all of that delicious juice out of those too.
"I'm seeing more and more projects and products move to OpenJDK just because they they don't want to have their balls gripped by Oracle and are worried that it's going to cost them a licence to use it. Docker is a case in point - I can pull an OpenJDK image but if I want Java 8 from Oracle I have to "checkout" and agree to all my details being supplied"
Oracle's very poor and egregious conduct in this matter is no doubt doing wonders for the take up of OpenJDK and Kotlin so Oracle's actions are ultimately self defeating in the long term. That said, I still really want to see them lose this case.
In microsoft's response to the issue, there's a section about comparing the java api definitions as a qwerty keyboard and scrambling the keyboard key placement would not promote the progress. This argument is fundamentally flawed, because it focuses on the names and labels of the software modules in question. The names and labels of the software modules are NOT the copyrighted content that the oracle is complaining about. Instead, analysis should be done based on "feature list" of the platforms. When google copies the api definitions, they need to invoke cloning of the feature lists of the functionalities that java apis are implementing. Thus their new platform will be exact clone of the existing software system, and it has no innovation whatsoever. Thus our position is that oracle should be the winning party in this question and microsoft's PDF's position on qwerty keyboard is misplaced.
Our position is that "feature lists" are the main innovative content in the software platorms, and thus the API cloning business is dubious activity as it tries to reach feature parity with an existing system by cloning both the api definitions and the feature descriptions that the software system is implementing.
The names and labels of the software modules are NOT the copyrighted content that the oracle is complaining about. Instead, analysis should be done based on "feature list" of the platforms.
In that case Oracle clearly doesn't have a leg to stand on because the feature list of Java is the feature list of previous platforms. On that basis there must be a score of other businesses that could sue Oracle into the ground, not only over that feature list but also over the feature list of SQL.
> In that case Oracle clearly doesn't have a leg to stand on because the feature list of Java is the feature list of previous platforms.
We can always focus to the area of the code which was copied by google from java platform. We don't need to take all the features into account, just the ones that google copy-pasted.
> On that basis there must be a score of other businesses that could sue Oracle into the ground, not only over that feature list but also over the feature list of SQL.
Well, if oracle didn't get software licenses to it, then it might be possible. But usually large companies have legal department who carefully examine the required licenses and actually sends money to the companies that sell these SQL software. Once they accept the money and give permission, they have no legal remedy to claim copyright infringement any longer.
Whole google vs oracle spatter happened because google wouldnt want to accept the onerous conditions that oracle placed for the licensees. And only accepting the license conditions gives google permission to use the software in question.
> Read file. Modify Write file.
Featurelists that are short are not copyrightable. Only once the list reaches certain length, it becomes unique enough that other companies do not accidentally copy it. But come-on, the java api definitions have like 200 features in it, and the whole list is so unique that without good plan, cloning the features is virtually impossible. Only if you're evil mastermind, and you spend time researching on competitor's feature setup, you will gain enough information to exactly replicate the feature list. API cloning is evil because it requires cloning the feature list too.
Note that there's strict rules for large companies to not replicate or clone features that their competitors are implementing. However, while companies are not allowed to look at competitor products to discover features, they _are_ allowed to listen their customers. The customers will shout all the cool features that are relevant in the marketplace, and if companies listen carefully the customer demands, they are allowed to implement same features as their competitors are doing. But then the featurelists as a whole are not the same, since they're listening different areas of the market.
Another situation where competitors will implement the same features happens when their hardware components available in their supply chains are coming from the same component vendor and the bought components determine what features are possible to implement.
IBM v Compaq seems to rely on cleanroom implementations. That's different can of worms, but google vs oracle is not using cleanroom implementation methods. You need to control information flow between development teams to do cleanroom implementaiton. In current internet based development practises, the cleanroom is bad position to take, since your programmers have access to the information coming from your competitors via internet. These old cases could still rely on programmers not having access to any information coming from the comperitors. Basically it relies on your development to be done in isolated island with no communication to the competitors.
" In current internet based development practises, the cleanroom is bad position to take, since your programmers have access to the information coming from your competitors via internet."
The cleanroom implementations of BIOS were, AIUI, done by reimplementation of a a list of procedure calls without access to the actual source code of what was being implemented. Did Google's developers actually have access to any implementations other than macros defined in include files?
> Did Google's developers actually have access to any implementations other than macros defined in include files?
Internet is making cleanroom implementation very difficult. It requires only that one of your employees to read a competitor product review in some scandalous web sites to throw your cleanroom implementation to illegal area. Controlling the information flow is necessary aspect of cleanroom, which means they need to cut the internet access, maybe use programmers that never used a computer and go to the homes of your employees to add firewalls and filtering technology to avoid your people to accidentally ruin your cleanroom implementation. Without all these dragonian measures, the cleanroom is just too risky. Not doing cleanroom is even more risky, but then you don't need to prove that your employees are outside of the scope of digital technology.
Whether google had access to which tech when copy-pasting the java api, we wouldn't really know. The implementation code has not been focus of the lawsuit, instead it focuses on api definitions and verbatim copying of the interface.
There was argument in the paperwork that rules out implementation code: "infringer cannot justify infringement by attaching original code to the infringement -- i.e. the fact that google wrote their own implementation isn't material to the copyright status of their platform" -- it might help them in the damage award section, but we haven't reached that area yet.
"t requires only that one of your employees to read a competitor product review in some scandalous web sites to throw your cleanroom implementation to illegal area"
A product review is not a listing of the software.
I note. BTW, you seem to be arguing on behalf or Oracle here ("Our position is that" a few posts above). Some of us have the greater good of the entire software culture at heart. Not Google's interest. Not Microsoft's. Not the US tech industry's although that has most to lose. Just the interests of those who enjoy or at least live by software development and enjoy the products of that development. The interests of a culture that, for some of us has extended beyond a working life time. We think Oracle has taken a line here that runs entirely counter to what we understand to be the norms of that culture and industry because if they succeed the result will be toxic, at least for software development in the US. Given the fact that Oracle is part of that industry it will suffer the long term consequences along with the rest.
>> companies are not allowed to look at competitor products to discover features
> On the contrary, reverse engineering is allowed.
There's three issues with this position:
1) it's not allowed among competitors
2) and it shouldn't be used to discover features to implement for your product
3) market leaders have stricter rules to follow than your ordinary startup
AFAIK easily sidestepped. They pay another company to reverse engineer, then ask "what features does it have", they get a list.
I saw at least one Defcon video, where I assume this was being done, because they were reverse engineering, and not able to say who the customer was. They were talking about the cool stuff, but not "hacking", very specifically, looked like competitor scouting out hardware features to add into their product.
> I saw at least one Defcon video, where I assume this was being done,
Too bad the practise is illegal.
But what exactly makes it illegal?
When companies invest millions of dollars for their customer experience and they are setting up shops to sell the products, do some market research on what exactly are important features this year for the customers. While that information comes as side effect of setting up the shopping experience, it only works on that local area of the world. When you need global network of shopping centres and shops to do the market research properly, it actually costs huge amount of money to do it.
So it's a significant money-saving technique to do the research on customer expectations only on small area of the world, and reverse engineer your competitors setup to detect what features competitors are implementing. But this is exactly what makes it illegal -- it is too efficient technique to save costs. Your competitors spent millions to do the research and by examining their products, you get all that research for free. When this happens, someone pays millions for the activity and you reap the benefits for free, then there's very high chance of the activity being illegal. What makes it illegal is copyrights and patents. When you use someone elses research for free, you end up violating patent and copyright laws.
This is what happens here too. The activity is illegal simply because the illegal activity saves money by skipping work that would be beneficial for the society. It is important that companies are doing their own market research properly, or else certain areas of the world are not getting their needs fullfilled properly since noone is listening the problems and issues in the area.
It isn't a trivial task to define and document APIs, which includes the interface and all the data objects that pass through it.
..Java has 4000 classes:
https://docs.oracle.com/javase/8/docs/api/
the ruling says that it's not OK to "borrow" all of them.
I agree that Sun, when promoting Java, acted all hippy, open-source, but there is still a copyright statement at the top of each source file.
Day 1 :- Entire US software industry implodes because they can't use APIs without being sued for copyright infringement.
Day 2 :- ROTW carries on regardless.
Day 3:- We put the second "c" back into "software licence".
Day 4:- We put the "u" back into "color" in all existing API strings.
Day 5:- US APIs now incompatible with rest of world.
...been arguing the case wrong for years. From the start it should have been something like this:
https://youtu.be/byFLZAwqY3Y?t=88
Parody is fair use not infringement. The lawyers could have argued that "It isn't our client's fault that none of their customers were smart enough to get the joke."
You know, a lot of this pain could have been avoided if the term of copyright had continued to be the original 14 years specified in 1789, with the removal of the right to renew. Instead, we now have a situation where corporate owned copyrights are enforceable for 130 years. Not exactly a term consistent with promoting "... 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."
Massive civil disobedience would be one route. Done carefully. Not any of that violent nonsense, which is unsustainable, ineffective and, franky, immoral. Just consumers sitting on their hands and refusing to play. Literally refuse to buy any more corporate copyrighted material. Of course it would have to be organized and so in-their-face that the message couldn't be ignored. In other words, generational change. Start now and by the time the future is here maybe we (or those who survive us) will enjoy the benefits.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
