Oracle Criticized For Questioning Google's Supporters In Java API Copyright Case (twitter.com) 47
America's Supreme Court will soon decide whether Google infringed on a copyright that Oracle says it holds on the APIs of Java. But this week Oracle's executive vice president also wrote a blog post arguing that Google "sought the support of outside groups to bolster its position" by using friend-of-the-court briefs to "create the impression that this case is of great import and controversy, and a ruling in Oracle's favor will impede innovation."
"Upon closer inspection, what these briefs reveal is a significantly different picture, one where Google is the outlier, with very little meaningful support outside the purview of its financial fingerprints." As we discussed in a previous post, this case is not about innovation, it is about theft. Google copied verbatim more than 11,000 lines of software code, and now attempts post hoc to change the rules in order to excuse its conduct... As those of us that have watched Google over the past few decades know, Google's view boils down to the self-absorbed position that the work it is doing is of such consequence that the rules shouldn't apply to them. The problem for Google is that very few outside of its self-generated atmosphere agree.
Let's be clear, it is not commonplace or foundational in the software industry to steal other developer's software code. Rather, what is commonplace is a confluence of interests where code is licensed to facilitate its widespread deployment, with the owner choosing the terms... Java embraced choice, with three different licensing alternatives, including a freely deployed open source license, and a commercial license designed to maintain interoperability. And it turns out that nobody except Google found it necessary to steal despite Java's enormous popularity. It is not in dispute in this matter that Google destroyed Java interoperability so it is unbelievable that many of its amici take the position that Google needs to prevail in order to protect interoperability...
Out of 26 briefs, we found:
- 7 briefs representing 13 entities that received "substantial contributions" from Google;
- 8 briefs filed by entities or individuals that have financial ties to Google through grants, dues, cy pres settlement proceeds or employment of individual amici;
- 2 briefs filed by companies with a clear commercial interest in Google prevailing;
- 1 brief filed by several former U.S. government employees all of whom worked for a small government agency run by a former Google executive, despite the U.S. government itself filing a brief in favor of Oracle;
- 4 separate briefs representing a total of 7 individuals;
- A few other briefs where Google financial ties are likely;
- 1 brief submitted by a serial copyright infringer repeatedly sanctioned by the Courts;
What masqueraded as a mass show of support for Google, may not be much more than an exercise in transactional interests.
The groups Oracle is criticizing include the American Library Association, EFF, and the Python Software Foundation, as well as a brief by 83 computer scientists which included Doug Lea, a former memeber of the executive committee of the Java Community Process. Oracle's blog post also makes the argument that besides Microsoft and IBM, "not a single brief from the other 98 of the Top 100 tech companies was filed."
There was a response on Twitter from Joshua Bloch, who worked on the Java platform at Sun before leaving in 2004 to become Google's chief Java architect for the next 8 years. He called Oracle's blog post "nonsense." For example, Doug Lea -- who is in no small measure responsible for Java's success -- accepted one small grant from Google fourteen years ago, and promptly doled it out to deserving undergrads who were testing java.util.concurrent. Have you no shame, Oracle?
We are not Google shills. We are scientists and engineers. Some of us laid the theoretical groundwork for the profession, some designed the computers you grew up on, and some wrote the software you use every day.
We depend on the right to reimplement each others' APIs, and we are truly afraid that your irresponsible lawsuit may deprive us of that right, which we've enjoyed throughout our long careers.
"Upon closer inspection, what these briefs reveal is a significantly different picture, one where Google is the outlier, with very little meaningful support outside the purview of its financial fingerprints." As we discussed in a previous post, this case is not about innovation, it is about theft. Google copied verbatim more than 11,000 lines of software code, and now attempts post hoc to change the rules in order to excuse its conduct... As those of us that have watched Google over the past few decades know, Google's view boils down to the self-absorbed position that the work it is doing is of such consequence that the rules shouldn't apply to them. The problem for Google is that very few outside of its self-generated atmosphere agree.
Let's be clear, it is not commonplace or foundational in the software industry to steal other developer's software code. Rather, what is commonplace is a confluence of interests where code is licensed to facilitate its widespread deployment, with the owner choosing the terms... Java embraced choice, with three different licensing alternatives, including a freely deployed open source license, and a commercial license designed to maintain interoperability. And it turns out that nobody except Google found it necessary to steal despite Java's enormous popularity. It is not in dispute in this matter that Google destroyed Java interoperability so it is unbelievable that many of its amici take the position that Google needs to prevail in order to protect interoperability...
Out of 26 briefs, we found:
- 7 briefs representing 13 entities that received "substantial contributions" from Google;
- 8 briefs filed by entities or individuals that have financial ties to Google through grants, dues, cy pres settlement proceeds or employment of individual amici;
- 2 briefs filed by companies with a clear commercial interest in Google prevailing;
- 1 brief filed by several former U.S. government employees all of whom worked for a small government agency run by a former Google executive, despite the U.S. government itself filing a brief in favor of Oracle;
- 4 separate briefs representing a total of 7 individuals;
- A few other briefs where Google financial ties are likely;
- 1 brief submitted by a serial copyright infringer repeatedly sanctioned by the Courts;
What masqueraded as a mass show of support for Google, may not be much more than an exercise in transactional interests.
The groups Oracle is criticizing include the American Library Association, EFF, and the Python Software Foundation, as well as a brief by 83 computer scientists which included Doug Lea, a former memeber of the executive committee of the Java Community Process. Oracle's blog post also makes the argument that besides Microsoft and IBM, "not a single brief from the other 98 of the Top 100 tech companies was filed."
There was a response on Twitter from Joshua Bloch, who worked on the Java platform at Sun before leaving in 2004 to become Google's chief Java architect for the next 8 years. He called Oracle's blog post "nonsense." For example, Doug Lea -- who is in no small measure responsible for Java's success -- accepted one small grant from Google fourteen years ago, and promptly doled it out to deserving undergrads who were testing java.util.concurrent. Have you no shame, Oracle?
We are not Google shills. We are scientists and engineers. Some of us laid the theoretical groundwork for the profession, some designed the computers you grew up on, and some wrote the software you use every day.
We depend on the right to reimplement each others' APIs, and we are truly afraid that your irresponsible lawsuit may deprive us of that right, which we've enjoyed throughout our long careers.
Oracle/Larry lost in the lower courts and has spen (Score:5, Informative)
Re: (Score:1)
Re: (Score:2)
Microsoft thought the same thing once they teamed with Apple.
Re: (Score:2)
Android is slightly older than that. Only a small subset of Java is open source, both Sun and Oracle kept significant portions under BCL except for the JDK which a reference implementation is GPL. This is the whole hangup of this case, Google did not develop based on the reference implementation and Google did not maintain Java compatibility - you can't run Android or its applications on computers directly through the JVM.
This is a slightly modified case of Sun v. Microsoft - Sun won, to great cheers becaus
Re: (Score:1)
I'm just responding to the claims "Java is not GNU" and "Java never was an open platform".
Also, the whole of Java is GPL nowadays: there is no longer any portion of Java that isn't GPL, and OpenJDK contains virtually the same code as Oracle's binary distributions. In fact, a lot of indipendent players have stepped in to support old versio
Re: (Score:2)
The mobile part of Java has never been GPL.
And that's the bit that Google wanted without having to comply witht he licence terms Oracle demanded, so they "rewrote" it instead.
Re: (Score:1)
Re: Java is not GNU (Score:5, Insightful)
Re: (Score:1, Redundant)
Is the issue about an API specification, or 11,000 lines of code?
Different conclusions for me there. It seems like both sides are doing some conflation or obfuscation here.
Re: (Score:3)
So you can implement the Java API, in Java, without using the same names (package, class, method, field), package hierarchy, class hierarchy (which methods go in which classes), without having the same relationships between classes and interfaces, the same method signatures, and the same public declarations (other than parameter names)?
Can you give an example?
Re: (Score:2)
So you say those 11,000 lines are just public API interfaces that must necessarily be the same?
Like the API for the printf library even if the guts are all custom-written?
Re: (Score:2)
I'm a C guy, and in C I'd disagree. But I get the impression that in modern high level code, 90% of everything is API boilerplate, class wrappers, frameworks, interfaces, and comments about all of that, leaving 10% of the code that actually does stuff.
If Oracle is counting Doxygen comments, then 11,000 lines for APIs seems kinda small to be honest.
Re: (Score:1)
Sure. Any implementation anywhere. If you end up with 11,000 lines of code that are -exactly- the same (if it's as claimed), you cut and pasted it.
Re: (Score:2)
package java.net;
How would you write that so it isn't exactly the same?
How about
public final class URL implements Serializable
and then later on
public URLConnection openConnection() throws IOException
and then later
public boolean sameFile(URL other)
How do you propose writing those lines so they aren't exactly the same? Change the parameter name? Here's Google's version of that last line:
public boolean sameFile(URL otherURL)
You can't use the same names, package/class/method hierarchy, inheritance hierarchy (
Re: (Score:2)
Both. Android Java VM is different enough from actual reference implementations of Java that it violates the Java license (look at Microsoft vs. Sun). Android also copied a ton from the Java SE which has always been under a Binary Code License and did not implement a VM based on the reference implementations.
As much as I hate Oracle, Google was in the wrong all along, if their stuff was based on a reference implementation, we'd be able to copy and run any Android app in any Java VM, that'd be bad for the Go
Re: (Score:2)
But the Java license is for the code and Google didn't use Oracle's (Sun's) code. Android never claimed to be an implementation of the Java run-time. In fact, they went so far as to produce their own (or use somebody else's) open source byte code standard. What Android developers cared about was the Java language, not the ability to portably run Java code written for the sake of that portability. Oracle wants to have their cake and eat it too. They open-sourced Java so that it would have a future, and
Re: (Score:2)
Re: (Score:2)
Which is good that the Senate aren't the ones making the decision.
Re: (Score:2)
To focus on any of this is kind of dumb. Andy Rubin;'s email account would say otherwise. Google was in a panicked raced to match Apple for smartphones.
They knew that they needed developers. They knew they didnt have a platform to win them over. And they knew Java a licensed language was the only alternative if they wanted to "catch up".
google never had any race with apple. google just wanted to have millions of devices all over the world they could provide services to and gather valuable data from. they were dismayed at the stagnant state the sector was in because of lack of innovation and counterproductive competition (see blackberry, windows phone, java me, etc). they wrote android and even started producing their own phones just to kick off the market. plus they started a revolution.
Now, Java has been a very freely licensed technology since its inception. There was really only ONE stipulation that might cost you some money. SUN was providing software for dumb phones that ran Java. So if you wanted to use Java in that way, on a phone, you were REQUIRED to license it.
java on phones has never really worked. again, see ja
Re: (Score:2)
I think you're getting Google confused with Apple. Google just saw Apple do and said "we have to have some of that" and copied it all, slightly differently.
Re: Java is not GNU (Score:2)
Are you smoking crack? Under the hood iOS has absolutely nothing in common with Android. Absolutely nothing. iOS is a locked down, stripped version of OSX with a touch oriented UI. Android is based on a portable cross-platform mobile implementation of Java slapped on top of a Linux kernel. "Slightly different" LOL The only similarities are the fact that you poke the screen to do stuff and they both have home screens, similar gestures for zooming and an app drawer. When you've written any code for both
Re: (Score:2)
Tat's exactly what I meant - Google saw Apples syuff and copied it. Sure they used a different implementation, but that's a detail to the whole "its a smartphoen with a grid of icons and an app store but totally not iOS at all".
Apple was there first, google copied them
Re: (Score:3)
Google doesn't use Java the platform. They use Java the language. They made their own compatible platform around it (actually they made two).
What next, sue React OS because it can run Windows binaries?
Re: (Score:2)
And that is the whole hangup. Java the language's license requires strict compatibility with Java the platform. This would be similar to Wine implementing a slightly different version of .NET so the apps developed for it only run on their platform and not on a commercial version of Windows, provided Microsoft licensed their .NET so you could use their reference implementation as long as it remained compatible.
Java license since Sun has always had weird hangups like that. Another one was that you had to lice
Re: (Score:2)
I don't know if using the language has anything at all to do with Java the platform. The intent of requiring Java platform compatibility was so that no single implementer of the 'standard' platform could build a version that was non-compatible. I.e., this was geared toward Microsoft - the only entity with enough market dominance to get away with building a non-portable version of a platform whose primary selling point was its portability. At the time, Microsoft was still in full-on Embrace/Extend/Extingu
Java isn't a platform, it's a compiled language (Score:2)
Re: (Score:2)
jvm is a stack based, bytecode, abstract machine specification (i'm sure there's a stricter definition somewhere)
much like i can take a sequence of x86 instructions and recompile it for 68k,
i can take a sequence of jvm instructions and recompile them for x86
the only differences between the two are the practical value and the existence of a cpu that can "directly" (doublequotes because fuck me if even intel knows what their cpus actually do to the instruction stream internally) execute the instructions (jaz
Re: (Score:2)
Oh boy, you're wrong in so many ways.
The Java programming language is only a small part of the platform. The Java platform also includes a compiled bytecode specification, a virtual machine (Java Virtual Machine or JVM), and a rather large class library. Well, more than one class library with different licensing - there's Java SE (Standard Edition, the ), Java MIDP (Mobile Internet Device Platform, used on "feature phones" in the pre-iPhone era), and EE (Enterprise E
Oracle are harming Java (Score:5, Insightful)
Re: (Score:1)
Re: (Score:2)
And your comment made me 200% confused.
Re: (Score:2)
Which is the way corporations want it.
Worried that they can sue you 100% of time.
And that you can't sue them 100% of the time.
Re: (Score:2)
Just because President Skroob failed doesn't mean it's impossible.
Re:And...? (Score:5, Insightful)
However, Google appear not to be stealing Java. They are creating an API compatible clean room version of it.
That the APIs are the same is because they're published APIs and the entire fucking point of them is to enable interoperability. They're intended to be re-used.
Oracle need to lose this case.
Confluence of interest doesn't invalidate argument (Score:1)
I find it rather striking that so much importance is put on why people say things and less on what they say.
Re: (Score:2)
But without that we wouldn't have influencers! :O
Re: (Score:2)
In science, address the argument, not the arguerer. In politics, the arguments are often cover story memes for useful idiot true believers, and the real reasons are hidden. Or maybe always.
Well of course. (Score:2)
It stands to reason that the groups filing the briefs would get support from Google. Google after all used to run on the basis of do no evil, and advertising pushes and screwing up Chrome aside they mostly still run and donate altruistically.
Oracle on the other hand is run by one rich arsehole how is more concerned about donating to governments so he gets to have his own island. The companies listed have financial ties and receive donations from every non-arsehole company.
Re: (Score:2)
Ellison "I want my own island"
Brin and Page: "why stop there, we want to own the entire world".
Making their argument for them. (Score:5, Informative)
It sure is nice when an opponent makes your argument for you.
1. Google did seek the opinions of outside organizations to bolster it's argument. Just like every other supreme court case ever.
2. They received many friend-of-the-court briefings supporting their case, because copyrighting an API is ridiculous.
3. It's not really shocking that companies and individuals that use open software might want to have a say in your attempt to cause a ridiculous precedent that will forever change the way software interacts, and since Google contributes to many open software projects, there will be a coincidence between these companies and people.
4. This case is of great importance, and indeed will stifle innovation should Oracle win. The primary evidence of that is that you've spent untold resources suing Google over their implementation of the Java API in order to make a product that has unquestionably innovated (Android), and any company besides Google would have folded under the financial burden of this aggressive and punitive litigation by now. And, that product would not exist without stifling license fees paid to Oracle should the Java API become copyrighted, if you even make it available for license. QED.
Re: (Score:2)
Oracle amicus brief (Score:2)
Oracle has their own set of supporters with their own amicus brief, lets see who's on there;
RIAA, Songwriters guild of America, former US senators.
hmm, yeah, that is a rather weak list regarding the subject matter, compared to the people backup Google.
https://adtmag.com/articles/20... [adtmag.com]
Where to send my profanity/whiskey soaked brief? (Score:2)