IBM sponsored and was the primary contributor of resources to an independently developed Java work-alike called Apache Harmony. (about JDK 1.5 level, IIRC) Apache Harmony was, naturally, under the Apache license.
Google wanted something like Java for Android, so Google used Apache Harmony.
Sun was okay with Google having something like Java in Android and had said so. (perhaps reluctantly)
Oracle sees Sun for sale, and envisions a big "sue google" sign on Sun's back. Oracle buys Sun, to get Java and sue Google. Very soon after acquisition this becomes apparent. James Gosling leaves Oracle for reasons that were not quite clear at the time, but soon become clear.
Oracle claimed Google had used Java code. But Android had Apache Harmony.
So Oracle changes its argument to . . . But the APIs!!!
(Groklaw had documented all this back in the day.)
Google explicitly calls out in its brief the fact that Sun, as the original author of the code, very explicitly did not consider Google to be infringing.
Although I do wish Google tried harder to press the point that, based on Oracle's argument that copying the API infringes copyright, Oracle itself is in violation of others' copyright. Ars Technica has an example of that (Amazon's AWS), and the brief does give a vague allusion to this fact, based on what I presume is the trial records:
> Notably, Java SE itself “reimplements” interfaces from earlier programming languages. JA154-157; see also JA211.
This is inaccurate. According to Gosling in the video you link, Sun believed the clean room implementation probably wasn't clean. That has nothing to do with whether or not Sun believed that usage of the API was infringing.
God I miss SUN and SGI- all those pretty computer chassis....Oracle basically is a lawyer who murdered SUN and wears their rotting flesh around pointing at people and yelling THEIF!
Even Gosling isn't advocating that the APIs were copyrighted. What he advocates is that he doesn't think it was actually a clean-room implementation, and that Google should have been 'nicer' about the whole process in general -- that regardless of the legal points, Google was kind of a jerk.
When it gets to the subject of APIs, Gosling is suddenly a lot less accusatory. Here's the actual conversation:
> Host: "I would hate to see a law or ruling that APIs are copyrightable, because that would have a chilling effect across the whole --"
> Gosling: "Yeah, that would be bad. But it kind of feels like the lawyers and laws and legal teams -- they do what they can."
But our laws don't punish companies for being mean, and issues around Google's actual clean-room implementation are largely irrelevant at this point in the lawsuit. It's pretty clear that Oracle bought Sun with the intention of going after Google far more aggressively and on far more legally tenuous grounds than Sun was ever interested in doing.
" issues around Google's actual clean-room implementation are largely irrelevant at this point in the lawsuit"
Are they? I am not an expert on this, but isn't the fact of actual copying (rather than successful clean-room implementation) a very relevant piece of information for copyright?
Actual copying is important, but not in this particular lawsuit.
Oracle is claiming that the organization of constants/methods in packages as well as the method signatures are copyrighted - they are arguing that it doesn't matter if you have a clean room implementation of BufferedInputReader() if you put it in java.io (like Oracle's JRE) and gave it the same signature as Oracle did, then you're infringing on their copyright.
If this SCOTUS broadly rules in Oracles favor, this would kill many projects, including WINE (which implements Microsoft's Win32 API without copying)
"At this point" is key, actual copying was important earlier in this case but is now a decided issue.
This appeal is strictly about the copyrightability of APIs. The supreme court generally doesn't address cases as a whole, but specific issues brought up to it on appeal.
If I remember the decisions on the actual copying part of the case correctly, it was decided that Google copied something like a dozen lines of uninteresting code directly (not api's, just some method implementation), and that this was de minimis copying and didn't constitute infringement. You should check if you really care though, I might be wrong.
>Even Gosling isn't advocating that the APIs were copyrighted....
And neither did the parent implied he was. He is merely stating Gosling wasn't OK with Google using Java without paying a dime, but didn't have much of a choice.
Who here remembers Groklaw? PJ was covering Oracle vs. Google with the same level of detail that she used on SCO vs. IBM. Coverage ran up into the first appeal, when she stopped updating the site.
If the NSA -- and other state / state-like actors -- are omnipotent, and can always find you through your email, etc, you either have to stop what you're doing or be willing to deal with being caught or exposed.
If they aren't all-powerful, but people stop using email to communicate and stop carrying out their business because of the belief they are, that also serves the same purpose as catching or exposing the people in question -- it stops their activities.
This reminds me of the idea behind WikiLeaks -- not just to expose the leaked documents, but to make it so costly to carry on secret communications and operations that people stop trying.
no donkey in this race but how sure this is not you and your circumstances rather than the site itself? Living in certain countries my ISP used to inject all sort of nonsense into every unencrypted site they could.
Anyone not using tls these days is intentionally exposing their readers to harm, which you seem to be experiencing.
Personally I won't view such sites on principle and block them with https-everywhere, though guessing many others do allow it, due to being the default setting worldwide.
Reading Groklaw when it was updating was amazing. I remember it as clear and concise writing that always seemed to illuminate the facts and angles to best grasp the topic/issue with. I wrote a thank you email to the author for all of the great articles when it was shutting down.
I still occasional think about it and wish it was around to illuminate contemporary issues.
If you have recommendations for a similar source of writing/insight please let me know.
I bought Caldera Linux in a box at one point in my life. I also wasn't the founder, but the first team member of a Linux distribution, and the thing was very, VERY personal to me.
At a minimum, it will cover the briefs of both sides, and maybe some amici briefs, in an argument preview; then a summary of the oral argument in an argument analysis; and then a final summary of the ruling in an opinion analysis.
This coming from the company that files patents like crazy and gives out t-shirts that say “patent heroes wanted”. Call me cynical but I feel like google is narrowly interested in open innovation when they are at risk for billions in fines, and definitely also supports openness in other ways, but still supports a lot of proprietary and closed source work. There are lots of business reasons to be closed source, but don’t talk up a big talk about the value of openness when you’re really just trying to cover your ass. Lots of google tech fails on openness.
This is coming from a whole lot more than just Google. They are the ones who will lose money in the immediate lawsuit, but we all stand to suffer if Oracle's interpretation of the law is held up. Don't take my word for it, take the word of the entire software community who has weighed in on this to support google in the form of amicus briefs (which I really do encourage you to read).
Here is a partial list of other groups who have already filed motions in support of Google: Microsoft, EFF, Python Software Foundation, Mozilla, RedHat, (see the docket for a complete list).
In addition their was a very interesting brief submitted by a bunch of famous computer scientists, instead of an organized company. The names attatched included Edwin Catmull, Alan Kay, Brian Kernighan, Bjarne Stroustrup, Andrew Tanenbaum, Ken Thompson, Guido van Rossum, Steve Wozniak. The total list of people is 78 long and they are all of similar caliber (though I did pick out the names I had the strongest recognition for), see their brief: https://www.supremecourt.gov/DocketPDF/18/18-956/89487/20190...)
This case isn't about Google making or loosing some money, Google being open or not. None of the amici care about that, the supreme court doesn't care about that. Only Google and Oracle care about that. This case is about the law that governs us all and what effects it has on us all.
And yes, Microsoft is agreeing with the EFF in a lawsuit...
> This case isn't about Google making or loosing some money, Google being open or not. None of the amici care about that, the supreme court doesn't care about that. Only Google and Oracle care about that.
... I don't think there's much support here for Oracle winning on its claims.
But this particular subthread was pointing out the hypocrisy of Google trumpeting "Openness in all cases!"
These are two separate and unrelated topics. We can simultaneously support Google winning the case (for the outcome on case law) while also believing Google could be a better corporate steward of open source (mostly looking at you, Android).
> Android's source code does not contain the device drivers, often proprietary, that are needed for certain hardware components. As a result, most Android devices, including Google's own, ship with a combination of free and open source and proprietary software, with the software required for accessing Google services falling into the latter category.
I don't think anyone expects the Android OS to be a collection of entirely open source device drivers.
Those proprietary drivers are specific to the myriad of hardware that the OEMs are running the OS on. You can still run Android on any hardware you have the drivers for, or emulate it in a virtual machine.
Google has systematically moved a lot of apps from AOSP to closed source Play Store equivalents. The AOSP apps are still available, but aren't maintained by Google anymore.
I very strongly support Google’s side of this case. But them claiming to champion open innovation is incredibly condescending and frankly bullshit. They’re pursuing this case exclusively for financial motives (which I don’t think there’s anything wrong with to be clear). But the fact that there’s some public good behind it is just a happy coincidence for their marketing department. They’d just as happily be on Oracle’s side if that made financial sense to them.
A lot of the modern web exists on tech that google open sourced. Something it didn't need to do and would have been a huge competitive advantage to it if they had kept it closed source even after they had moved on to something better.
You can argue about patents, but trying to copyright APIs is on a whole new level. It would be utterly disastrous. Regardless of Google's motivations, their arguments here are completely 100% in the right and we should support them -- if not for Google's sake, then for ours.
It's counterproductive to drag a company over purity while it's currently the primary fighter against an existential threat. It's polishing silverware while the house burns down around you.
Yes, Google could do better (particularly on patents). Yes, some of Google's arguments are hypocritical. But openness is a continuum, not a binary state, and it's not helpful to turn away imperfect allies at this time.
> This coming from the company that files patents like crazy and gives out t-shirts that say “patent heroes wanted”
That's not incompatible with openness if they are largely amassing a defensive patent portfolio. A desire for openness does not mean unilateral disarmament in the face of enemies who are very willing to use every status quo legal tool to.prevent openness.
Agreed. Even though Google patents like crazy, I don't recall hearing very much about them suing other people using patents.
What they're doing is trying to make it so that nobody can sue Google over something Google is doing using patents, because Google already patented it.
>This coming from the company that files patents like crazy and gives out t-shirts that say “patent heroes wanted”.
This is a complete red-herring. Patents are not what this case is about. Openness or cuddliness is not what this case is about.
The reality is that Google is in the right because Oracle is willing to set fire to the norms and standards that were developed over the last 50 years just because they want to squeeze more money out of Java.
You can say it even more strongly. Oracle bought SUN with the sole intention of using it as the basis to sue Google. Full stop. This is a shakedown and Oracle needs to go away.
The argument that Google deserves punishment because they didn't buy another company is not one I'm even remotely sympathetic to.
The entire industry would be worse off today if our legal policy on APIs was, "they're copyrighted, but don't worry, you can just buy out companies you want to interop with." Imagine how much smaller and how much more uninteresting the tech world would be.
Any company that wasn't the size of Google couldn't have bought Sun, and any company, even companies smaller than Google, should be free to interop with the Java APIs the same way that Google did. Discussions about whether or not Google could have bought Sun are meaningless to me, I don't think they have anything of value to add to the current case, and I don't think they have anything to do with Oracle buying Sun with the sole intention of suing someone else.
You're replying to a thread where someone complains about Google excessively filing patents as a defensive/offensive measure, and your argument is that Google should also be excessively buying companies as a defensive measure against lawsuits?
This is also an argument that I'm not very sympathetic to -- the idea that we need to lock down an entire ecosystem and consolidate it into 3-4 giant 'stewards' to protect us from those giant stewards. Arguments about bullying aside, Google was honest regarding Java. They built a clean-room implementation. That was their only obligation. Sun saw it that way too.
It would have a major chilling effect on the industry if every company could decide who was allowed to interop with their products. The fact that a few extremely large, extremely powerful players could get around that chilling effect by literally owning everything makes that situation worse, not better.
Tiny players who get taken advantage of don't need the world you propose -- they need a way to safely, legally interoperate with the software around them. When Microsoft says that its Windows APIs are copyrighted, should the WINE team buy Microsoft so they have permission to get Windows software running on Linux? Should Oracle buy Amazon so it can compete in the cloud?[0]
> Oracle can't simultaneously argue that API-using code does not embody copyrighted material from an API, and yet API-using code embodies all copyrights in the API necessary to give Oracle the right to reimplement S3.
So, in your opinion, WINE is an illegal software product, and Microsoft should sue them into oblivion. And IBM should sue everybody else for copying the BIOS interface. And Amazon should sue Oracle for implementing the AWS APIs.
If you believe that Oracle deserves to win this case, those are requisite outcomes of the case, because Oracle's legal contention is that merely providing an independent implementation of the API is copyright infringement.
> Their first attack came in the form of lawyers using 9,000 IBM copyrights in the hope that Compaq had broken just one. Luckily for IBM, Compaq did break one copyright, which forced them into a corner. Compaq negotiated with IBM and ended up purchasing the copyright for $130 million. This, however, did not stop Compaq overall.
Google also did a clean room reverse engineering. Or at least clean enough that the jury voted to acquit Google of copying the Java source code, and Oracle has not appealed that claim--meaning Oracle concedes that at this point.
The only thing that Oracle claims Google has done at this point is copied the API, which anyone who wishes to do a clean room reverse engineering would also have to do. Oracle is trying to get a judgement that makes such clean room reverse engineering impossible, which is why everyone else is freaking out about this case and asking SCOTUS to reverse CAFC and Oracle here.
>With Google at the steering wheel I bet Java developers would still be enjoying Java 6.
Maybe. Maybe not. But just to be clear, if all this case was about was Oracle trying to extract some money from Google, it would be nothing but a curious sideshow.
What Oracle is doing goes way beyond Java. They are trying to set an insane precedent and that has never existed in our industry, and had it existed it would have drastically reshaped it to the extent that it would be unrecognizable today. The implications of Oracle winning would be far reaching and unknown.
I don't think you're being unreasonably cynical. We have numerous demonstrations of companies pretending to be virtuous when it benefits them, while also being evil when it benefits them. It's fortunate that their posturing lines up with values that some of us hold, but I consider it nothing more than a profitable coincidence.
On the one hand I agree that Google saying that we need Open interfaces is correct, but then I also know that most code within Google itself is closed source. Does anyone know if there is an Open Interface to Google Search?
Actually it's possible that they copied that interface from someone else, basically every search engine uses the same one, it's how your browser url bar works.
Good point. I actually meant that google had an internal API for search which they use for customizing the search engine which I guess that they have not opened
the lawsuit will now be known as Google v. Oracle, since Google asked the Supreme Court to hear the case. After nine years, we’ll have to get used to calling the case by this new name
Interesting.
Poetic.
Google can’t win this fight (if they win) as a defendant, only as an aggressor.
> Google can’t win this fight (if they win) as a defendant
Sure they can (and it's the only way they can), if they win at the Supreme Court, it will be as a Defendant-Appellant. And if, as is often the case, the final formalization of the victory comes on remand back to the trial court from the Supreme Court after the Supreme Court declares the controlling law, that victory will be as a simple Defendant.
Question: My understanding is that the Supreme Court is here to interpret the constitution and other laws. When it now decides that Oracle is right and APIs are copyrightable, can't the other parts of the government introduce a new law making APIs not copyrightable again?
If the supreme court decides for Oracle, congress can absolutely pass a new law that reduces the scope of copyright to exclude APIs. The constitution limits what congress can say is copywritable, but doesn't require that they say anything is.
There's even a good chance that they will in my opinion, given the amount of lobbying dollars on the side of sanity in this case.
Still, I would rather not rely on congress doing anything sensible in a timely manner these days, especially when existing copyright law already sides with Google (IMO). It would also not be retroactive, so everyone under the sun can sue everyone under the sun over previous violations.
Oracle is no saint, but Google isn't a saint in this either. Google is one of the strictest companies I've talked with when it comes to people working on personal projects, contributing to open source, etc. even when in California and on your own time/equipment.
There is a lot comments focusing on which side will win, but I would like to hear peoples thought about how the judges can reach the desired conclusion and what the potential risk or prospects to areas outside of programming API. As long there is not explicit paragraph in copyright about programming API I would think that there will be an effect outside the scope of this lawsuit.
One area of copyrighted works I have seen people also question the legibility of being copyrightable is standard documents and law texts. They seems to me as very similar to an program API, if somewhat less strict when implemented. It would definitive lower the cost of interoperability if such works would be lifted outside of copyright.
On the risk side however, I am wondering if the Idea–expression distinction get pushed towards the idea side, then will companies just use that finding as an argument in favor of patenting API? 20 years is better than the practically infinity that is copyright, but I am not sure it will make the situation much better in the general case.
How come you assume they can be? My understanding is that they have generally been protected by patent but I'm open to being proven wrong (preferably by being shown a court case, but being shown big licensing agreements for copyright would also be interesting).
I’ve just assumed it so because I’ve read it a few times in various contexts, but now that I google it more thoroughly it looks very much like you’re right.
Perhaps this confusion stems from the fact that chip companies tend to license the ISA bundled with documentation and software tools, which are of course covered by copyright (see e.g. the MIPS Open license).
Time for Google to support open innovation for real.
Some Key principles for open innovation
Creating value is not an end but a means for making our world better.
You embrace openness when creating value for your customers.
You use an OSI approved license when developing software.
You use an Creative Commons License to share documentation that has value.
You embrace openness as basic value for knowledge sharing and growth. Think of open science, open data, open access, open research.
Knowledge and improvements should be beneficial and reachable for everyone.
Your organization is inclusive.
This is the new definition of open. A documented API, probably with no guarantees and many requirements. Is the Nest thermostat really open when you can only write values to it via the cloud? Why not a simple local mqtt solution? I would not call this "open".
Last year I went to refresh my memory on the status of the case and all pop press referred to “11,500 lines of code” copier. I couldn’t find any clarification of those are all API calls. Anyone know or better yet have a reference with analysis?
Google's arguments were already refuted in the earliest hearings. There's a reason this is going to the supreme court, the rulings didn't really match up with accepted evidence and testimony of experts in the court.
Google will lose this. You don't explicitly copy code, take other people's engineers after exiting licensing talks with the company you took from, and have it not be about stealing someone else's intellectual efforts.
The courts have already explicitly denied Google's claim they did this for compatibility or interoperability. The grey area is how much damage Google has done to Oracle, and it's hard to put an explicit price on that. But given the popularity of Android and how Google's has massively benefited from the platform, in no small thanks to the development community around it, and given the absolutely morbid failure of its other community dev efforts it's not really hard to see that Android wouldn't be what it is today without Java; having a familiar platform for developers to code against is priceless (but not legally).
Google wanted all the benefits of using Java without actually paying for it.
Sun put a lot, and I mean a lot of money into Java. Java had well defined licensing terms for how to use their code. Oracle bought Java. Oracle has the rights to license and price their code however they want. Google does not. Google was in talks with Oracle to license Java but backed out when they didn't want to pay to use it. Instead Google took engineers from the company they copied code from, and re-licensed said code.
It's black and white but people's blind hate for Oracle leads them down an argument or view point the courts have already denied and are now ignoring.
Don't sacrifice your principles for some cheap hit against Oracle. If Google can just throw their man power around to ignore your license, your open source license will not matter. You will not get credit for the work you do. Google and others can just feel like they don't want to abide by your license, take your code, and re-license it however they want. It's already hard to enforce any type of open source licence but if Google comes out on top it really won't matter moving forward. That is what is at stake here.
"What we’ve actually been asked to do (by Larry [Page] and Sergey [Brin]) is to investigate what technical alternatives exist to Java for Android and Chrome. We’ve been over a bunch of these, and think they all suck. We conclude that we need to negotiate a license for Java under the terms we need." - Email from Tim Lindholm, a Google Engineer
From the most recent hearing:
"Ultimately, we find that, even assuming the jury was unpersuaded that Google acted in bad faith, the highly commercial and non-transformative nature of the use strongly support the conclusion that the first factor weighs against a finding of fair use." - https://www.leagle.com/decision/infco20180327178
Google didn't copy code. Even if you believe that APIs are copyrightable, Google's clean-room implementation hasn't been part of the case for years. It's just about the API now.
If you want to make an argument that APIs are copyrightable, fine, make that argument. Make the argument that Google is infringing on Oracles API. But if you're trying to accuse Google of copying code, you're wrong. You're conflating two ideas that are not related.
> It's black and white but people's blind hate for Oracle leads them down an argument or view point the courts have already denied and are now ignoring.
If this was actually as black and white as you say, the Supreme court wouldn't have agreed to hear it, they would have just allowed the lower court decision to stand. Unless you believe that Ruth Bader Ginsburg blindly hates Oracle for some reason?
No, they didn't. If you're referring to the infamous 9 lines of code:
private static void rangeCheck(int arrayLen, int fromIndex, int toIndex {
if (fromIndex > toIndex)
throw new IllegalArgumentException("fromIndex(" + fromIndex +
") > toIndex(" + toIndex+")");
if (fromIndex < 0)
throw new ArrayIndexOutOfBoundsException(fromIndex);
if (toIndex > arrayLen)
throw new ArrayIndexOutOfBoundsException(toIndex);
}
that was removed in Android 4.0 and is no longer being litigated as part of the case. Google is not being sued for stealing code -- Oracle agreed way back in 2012 not to pursue statuary damages for those lines.
The appeal courts reversal of Alsups decision in 2014 was on the basis of copying the API. They wrote "that the overall structure of Oracle's API packages is creative, original, and resembles a taxonomy"[0].
You're conflating two separate issues, copying implementation code vs copying an API -- copying the implementation was resolved in 2012. Copying the API is still being litigated. If you have a docket or ruling or source that says otherwise, then of course feel free to list it.
Or if you want to make the argument that declarative headers for an API are technically code, then... OK, whatever, fine. But it's still wildly deceptive to conflate declarative code and implementation code without mentioning that the law has treated them separately for decades. You're making it sound to a normal reader like Google wholesale lifted implementation logic from Java, which is just not what happened.
> Google's arguments were already refuted in the earliest hearings. There's a reason this is going to the supreme court, the rulings didn't really match up with accepted evidence and testimony of experts in the court.
This is not how appeals works. The Supreme Court is obliged to cover just two questions presented in the petition for cert: (1) Whether copyright protection extends to a software interface; and (2) whether, as the jury found, the petitioner’s use of a software interface in the context of creating a new computer program constitutes fair use.
The reason why it's hearing this case is because precedent on every circuit appeals court other than the federal circuit has answered "no" to the first question, and the precedent of fair use interpretation is such that CAFC overturning the jury verdict to rule not-fair-use is incredibly surprising.
> The courts have already explicitly denied Google's claim they did this for compatibility or interoperability.
That is factually incorrect. Whether or not Google did it for compatibility or interoperability is a factual determination. And facts are determined by jury, not by courts (and definitely not appeal courts). CAFC overturned the jury verdict, which found that Google's use was fair use. And Oracle does not contend that the jury was given bad instructions, which means that the appeals court can only consider if the jury's verdict was unreasonable.
> Google wanted all the benefits of using Java without actually paying for it.
I do not dispute this, and I don't think anyone disputes this. But the question is if it is legal for Google to do what it did... and the contention of quite literally everybody but Oracle is that it is.
> Google was in talks with Oracle to license Java but backed out when they didn't want to pay to use it.
Google was in talks with Sun, not Oracle. Sun didn't pursue Google because they couldn't find evidence of copyright violation. (As has been pointed out, many Sun engineers do suspect that Google didn't follow clean room implementation procedures strictly, but the amount of copied code that even Oracle could dredge up is 8 test files and a 10-line method called rangeCheck--not worth pursuing).
If you're making an emotional case based on the investment put into Java, you should be honest that Oracle bought Sun to acquire Java, in large part to sue Google for the use of Java in Android.
You're description of the history is 100% correct. Legally, who knows, but you've got what actually happened 100% correct.
There are no good guys in this situation. The cynicism on both sides is astounding. One company outright stole work and negotiated in bad faith, and another tried to take advantage of that in the courts, which threatens to set a horrible precedent about APIs if they win.
Google could lead by example here - perhaps by letting people download a reliable, usable copy of the data they're making 70 billion dollars a year from. But they won't even do that.
Let's be real; it's not like your data is also purged from their predictive models when it's no longer available in Takeout. They'll still be using it for profit. The least they can do is let you download a copy of it.
Yes your data is purged when you delete it from your activity (https://myactivity.google.com/) (not instantly because there are many systems involved, but within a strict deadline). This is taken seriously, and there are teams at Google making sure this happens. (Not sure how much is public, but some details in this thread: https://news.ycombinator.com/item?id=19809259) (Note that anonymized data, such as aggregate data that cannot possibly be used to identify individual users, cannot be deleted this way: e.g. (say) the number of YouTube views for "Despacito" (if you had watched it) will not decrease by 1 when you delete your activity, because that number was never associated with you in the first place.)
If something used to be in your Takeout data but is no longer there, that's most likely because Google no longer has that data about you.
I'd be surprised if behavioural data over a two years old is useful for any predictive models, and I expect they would remove it from their models to avoid spurious correlations.
First Google should provide the Java community an platform that actually supports Java language alongside its standard library instead of Google's version of J++, just like the NDK supports C and C++ standard library.
None of the other Java vendors has ever had any problem with either Sun or Oracle.
Or just like with Microsoft with J++, they can create their own .NET aka Android with either pure Kotlin/Native or Dart and see how successful it will turn out to be without piggybacking into the Java eco-system (Fuchsia).
What great PR speak. The core topic of the article is an update in the ongoing copyright legal dispute between Oracle and Google. But the title and opening paragraph invoke an ethos of justice, which inherently frames Google as a valiant champion defending freedoms for its users.
Sure, but that's not really my point. I don't really care about Google or Oracle here. I'm saying that Google is spinning it really well, regardless of whether the spin is justified, so that the framing of their position is more interesting than just "Google v Oracle update."
That's a really effective framing for any company in a highly publicized legal dispute. If it happens to be true for Google in this particular dispute, that's nice but not necessary for the PR speak to be advantageous.
... and this is why we're opening our web index so anyone can build applications on top of it?
Google built one of the most valuable companies in the world on intellectual property. They want open innovation when it's a competitor being open, or when being open has strategic benefits. They don't give a damn otherwise.
IBM sponsored and was the primary contributor of resources to an independently developed Java work-alike called Apache Harmony. (about JDK 1.5 level, IIRC) Apache Harmony was, naturally, under the Apache license.
Google wanted something like Java for Android, so Google used Apache Harmony.
Sun was okay with Google having something like Java in Android and had said so. (perhaps reluctantly)
Oracle sees Sun for sale, and envisions a big "sue google" sign on Sun's back. Oracle buys Sun, to get Java and sue Google. Very soon after acquisition this becomes apparent. James Gosling leaves Oracle for reasons that were not quite clear at the time, but soon become clear.
Oracle claimed Google had used Java code. But Android had Apache Harmony.
So Oracle changes its argument to . . . But the APIs!!!
(Groklaw had documented all this back in the day.)