Oracle sues Google over Android

pm

Elite Member Mobile Devices
Jan 25, 2000
7,419
22
81
This one completely caught me by surprise and I'm still trying to figure out what kind of case Oracle has but Oracle filed suit against Google over the use of Java JIT and other Java-related stuff in Android's SDK.

http://arstechnica.com/tech-policy/...es-google-over-use-of-java-in-android-sdk.ars

Specifically from the complaint:
http://news.cnet.com/8301-30684_3-20013546-265.html

Google’s Android competes with Oracle America’s Java as an operating system software platform for cellular telephones and other mobile devices. The Android operating system software “stack” consists of Java applications running on a Java-based object-oriented application framework, and core libraries running on a “Dalvik” virtual machine (VM) that features just-in-time (JIT) compilation. Google actively distributes Android (including without limitation the Dalvik VM and the Android software development kit) and promotes its use by manufacturers of products and applications. Android (including without limitation the Dalvik VM and the Android software development kit) and devices that operate Android infringe one or more claims of each of United States Patents Nos. 6,125,447; 6,192,476; 5,966,702; 7,426,720; RE38,104; 6,910,205; and 6,061,520.

Since Google says they built Dalvik using a "clean room" (no Oracle/Sun code included) from the ground up, I don't quite see how Oracle plans on pursing this but one would have to guess that Oracle must think that they have some sort of case.

Also, since I'm the moderator of GG&P, can we please try to keep this discussion on the lawsuit and Google and Oracle and not mention iPhones, iOS or Apple unless it's relevant. Thank you.
 
Last edited:

theeedude

Lifer
Feb 5, 2006
35,787
6,197
126
Why does "clean room" matter? The patents are for software approaches/methods, not specific coding.
 

tatteredpotato

Diamond Member
Jul 23, 2006
3,934
0
76
Why does "clean room" matter? The patents are for software approaches/methods, not specific coding.

link

Clean room design (also known as the Chinese wall technique) is the method of copying a design by reverse engineering and then recreating it without infringing any of the copyrights and trade secrets associated with the original design. Clean room design is useful as a defense against copyright and trade secret infringement because it relies on independent invention. However, because independent invention is not a defense against patents, clean room designs typically cannot be used to circumvent patent restrictions.

I assume since Java is an open source standard that Oracles only argument can be that google used Sun code. If there are patent issues Google can still be liable.
 

Cogman

Lifer
Sep 19, 2000
10,284
138
106
link



I assume since Java is an open source standard that Oracles only argument can be that google used Sun code. If there are patent issues Google can still be liable.

This, It has long since been struck down that you can't patent ideas when it comes to software. You can only copyright code. The EU has tried in the past to pass a software patent bill, but I don't think it succeeded. Oracle doesn't have a case.
 

theeedude

Lifer
Feb 5, 2006
35,787
6,197
126
And yet they have the patents mentioned in the case. So the whole "can't patent ideas" argument contradicts the reality of this situation. It's OK to clean room reverse engineer as long as you find workarounds to avoid infringing on patents. What Oracle is claiming is that Google's "clean room" implementation still infringes on Sun's patents. Whether it does or does not, and whether Sun's licensing of Java allows Google to use those patents is going to be up to the courts to decide.
 

Cogman

Lifer
Sep 19, 2000
10,284
138
106
And yet they have the patents mentioned in the case. So the whole "can't patent ideas" argument contradicts the reality of this situation. It's OK to clean room reverse engineer as long as you find workarounds to avoid infringing on patents. What Oracle is claiming is that Google's "clean room" implementation still infringes on Sun's patents. Whether it does or does not, and whether Sun's licensing of Java allows Google to use those patents is going to be up to the courts to decide.

Except, the ideas they are suing over are already in widespread use. Not just by google. Microsoft as a JIT for their .Net languages.

You can hold a patent on just about anything, that doesn't necessarily mean that that what was patented is patentable (Patent offices make mistakes).
 

theeedude

Lifer
Feb 5, 2006
35,787
6,197
126
Except, the ideas they are suing over are already in widespread use. Not just by google. Microsoft as a JIT for their .Net languages.

You can hold a patent on just about anything, that doesn't necessarily mean that that what was patented is patentable (Patent offices make mistakes).

Sure, and this kind of stuff is decided in the courts.
Widespread use does not nullify the patents, as long as it does not predate them (prior art). Again, I am not sure what the claims are, what the terms are for Sun's open source license, what Google's implementation is, etc. And I am not a lawyer. This is all for the courts to decide. Sun was pretty weak in monetizing it's vast intellectual property. Obviously Oracle is trying to extract some value out of the portfolio it purchased. Could be a case of closing the barn doors after the horses have left, depending on the terms of Sun's open source license.
 

Patranus

Diamond Member
Apr 15, 2007
9,280
0
0
Bye bye Android.
It will suffer the same fate as J#.

I love how everyone is saying how "evil" Oracle is being. Guess what, if you started a search engine that infringed on Google page ranking patent you would certainly be sued.
 

Cogman

Lifer
Sep 19, 2000
10,284
138
106
Bye bye Android.
It will suffer the same fate as J#.

I love how everyone is saying how "evil" Oracle is being. Guess what, if you started a search engine that infringed on Google page ranking patent you would certainly be sued.

J# didn't die because of a sun lawsuit. J# died because nobody saw the point of a language that is more like java than C#. J++ died because of a sun lawsuit in 2001. J# started in 2002.
 

Patranus

Diamond Member
Apr 15, 2007
9,280
0
0
J# didn't die because of a sun lawsuit. J# died because nobody saw the point of a language that is more like java than C#. J++ died because of a sun lawsuit in 2001. J# started in 2002.

Ya ya ya.

The point is google did exactly what Microsoft did with Java. They were even warned by Sun before they were bought out by Oracle. The transaction was recently completed and this lawsuit is the result.

I wonder how long it would take for Google to retool Android if an injunction is enacted by the counts. Even if it is only 6 months that would do irreparable harm to the band and result in the App store count resetting at 0.
 

Cogman

Lifer
Sep 19, 2000
10,284
138
106
Ya ya ya.

The point is google did exactly what Microsoft did with Java. They were even warned by Sun before they were bought out by Oracle. The transaction was recently completed and this lawsuit is the result.

I wonder how long it would take for Google to retool Android if an injunction is enacted by the counts. Even if it is only 6 months that would do irreparable harm to the band and result in the App store count resetting at 0.

Interesting thought, but I don't think it would take as long as you would think.

If the shit does hit the fan, I wonder if google will switch to a .NET language. Microsoft would be more then happy to let google use something like mono on their platform.

That being said, I really do hope that Oracle doesn't have a case here.
 

SunnyD

Belgian Waffler
Jan 2, 2001
32,674
146
106
www.neftastic.com
Despite what everyone things, the end result outcome UNLESS everybody jumps on board the Google bus (EFF + all other major players) and presses the fact that it's OPEN SOURCE SOFTWARE and gets Oracle to back down, Google and Oracle will end up in a cross-licensing agreement, Oracle gets money, Google continues to use Java, and life continues on as if nothing happens.

Long and the short of it, Android isn't going away.

And Patranus, J# didn't die because of Sun. J# died because it was a pointless language.
 

Patranus

Diamond Member
Apr 15, 2007
9,280
0
0
Despite what everyone things, the end result outcome UNLESS everybody jumps on board the Google bus (EFF + all other major players) and presses the fact that it's OPEN SOURCE SOFTWARE and gets Oracle to back down, Google and Oracle will end up in a cross-licensing agreement, Oracle gets money, Google continues to use Java, and life continues on as if nothing happens.

Long and the short of it, Android isn't going away.

And Patranus, J# didn't die because of Sun. J# died because it was a pointless language.

Correct, it was J++. My mistake.

A common misconception is that open source = do whatever you want with it.
Open source is far from that.

One thing you forgot is that if Google moved to a .NET language (which is doubtful as Android competes with Windows Mobile) every single program in the marketplace would have to be rewritten.

Let me ask you this, what does Oracle gain by letting Google continue to infringe on their patents? Oracle has a fiduciary duty to its shareholders to defend its intellectual property. Oracle didn't buy Sun for the fun of it.

People who are here claiming that Oracle is being "evil" only needs to think what would happen if you wrote a search engine that uses googles page rank system. You would be sued in a second by Google (or Stanford depending on their arrangement).
 

pm

Elite Member Mobile Devices
Jan 25, 2000
7,419
22
81
First a caveat, I'm an engineer and not a lawyer. But I have written several patent applications and I have a patent pending, and I have worked patent lawyers as an expert to determine potential infringement in circuitry. So I have some experience with patents.

Patent ligitation is very rarely black and white. First, as an engineer, I've never heard of anyone looking through patent documents in order to figure out how to do something - about the only reason to check the patent database is to make sure something you are doing hasn't been invented or to figure out what sorts of things companies are working on so that you know what they are up to. Patents are confusing legal documents that are particularly hard to for laypeople engineers to read and thus any idea that engineers at one company deliberately copied patented ideas from another company seems unlikely based on my experience. What tends to happen is engineers come to an idea separately and independently as a solution to a problem, someone patents it and then later people come back say that the solution infringes on other designs that were designed in parallel.

Accidently implementing someone's patent is not a defense, but it does allow people to generally be somewhat sympathetic to the accused.

The second reason that patents lawsuits are hard is that patents are written to be as general as possible, patent examiners spend effort narrowing the scope of the patent, and the final result tends to be something in which you can change one or two things and no longer be infringing. So just copying an idea - deliberately or not - is allowed as long as you don't implement it exactly as it is in the patent. Again, patents are deliberately vague specifically to deal with this, but as someone who has analyzed designs to determine potential infringment, I can say that it doesn't take big changes to make something infringe or to avoid it.

Next, here are the patents in question. I looked them up in Google's patent search (yes, the irony is intentional) and then wrote a one sentence summary based on my very quick read of the abstract of the patent:

6125447 - security using protection domains which estabilishes permissions assigned to code entering a device.
http://www.google.com/patents?id=dyQGAAAAEBAJ

6192476 - controlling access by a thread to a resource based on permissions from routines in a stack
http://www.google.com/patents?id=G1YGAAAAEBAJ

5966702 - packaging class files together to remove duplicate information by determining memory allocation requirements in advance to set of classes rather than individual classes.
http://www.google.com/patents?id=TzsPAAAAEBAJ

7426720 - a system for dynamic preloading of classes through memory space cloning of a master runtime system process so that child runtime processes inherit the memory state of the parent.

6910205 – compiling virtual machine (VM) instructions for a function into native machine instructions to improve the execution speed of VM.
http://www.google.com/patents?id=U-4UAAAAEBAJ

6061520 – improves initialization of static arrays by preloading all class files, determining the initialization performed by them and then consolidating them into a single expression.
http://www.google.com/patents?id=mEwEAAAAEBAJ

Again, I'm not quite sure what to make of these. I'm a hardware engineer, and software is not my expertise, and I'm not very knowledgeable in Java or VM's.

If Google sticks to the usual patent lawsuit script, they will file a countersuit allegeding Oracle's infringement of their patents, there will be a long drawn out deposition process involving literal mountains of paperwork and lots of emails. Then generally patent infringement cases are settled - often by some form of cross-licensing agreement and a bit of cash to sweeten the weaker portfolio's side. The cases tend to be long, and they are incredibly risky to the accused since a "jury of peers" generally are not experts in the specifics of, in this case, virtual machines and Java JIT compilers and a guilty verdict generally involves a crazy amount of money and, usually, an injunction which stops the sale of all infringing designs (in this case, it would halt all Android sales, and possibly the App Market (since the apps were written with the SDK), as well as the removal of the SDK).
 
Last edited:

Cogman

Lifer
Sep 19, 2000
10,284
138
106
First a caveat, I'm an engineer and not a lawyer. But I have written several patent applications and I have a patent pending, and I have worked patent lawyers as an expert to determine potential infringement in circuitry. So I have some experience with patents.

Patent ligitation is very rarely black and white. First, as an engineer, I've never heard of anyone looking through patent documents in order to figure out how to do something - about the only reason to check the patent database is to make sure something you are doing hasn't been invented or to figure out what sorts of things companies are working on so that you know what they are up to. Patents are confusing legal documents that are particularly hard to for laypeople engineers to read and thus any idea that engineers at one company deliberately copied patented ideas from another company seems unlikely based on my experience. What tends to happen is engineers come to an idea separately and independently as a solution to a problem, someone patents it and then later people come back say that the solution infringes on other designs that were designed in parallel.

Accidently implementing someone's patent is not a defense, but it does allow people to generally be somewhat sympathetic to the accused.

The second reason that patents lawsuits are hard is that patents are written to be as general as possible, patent examiners spend effort narrowing the scope of the patent, and the final result tends to be something in which you can change one or two things and no longer be infringing. So just copying an idea - deliberately or not - is allowed as long as you don't implement it exactly as it is in the patent. Again, patents are deliberately vague specifically to deal with this, but as someone who has analyzed designs to determine potential infringment, I can say that it doesn't take big changes to make something infringe or to avoid it.

Next, here are the patents in question. I looked them up in Google's patent search (yes, the irony is intentional) and then wrote a one sentence summary based on my very quick read of the abstract of the patent:

6125447 - security using protection domains which estabilishes permissions assigned to code entering a device.
http://www.google.com/patents?id=dyQGAAAAEBAJ

6192476 - controlling access by a thread to a resource based on permissions from routines in a stack
http://www.google.com/patents?id=G1YGAAAAEBAJ

5966702 - packaging class files together to remove duplicate information by determining memory allocation requirements in advance to set of classes rather than individual classes.
http://www.google.com/patents?id=TzsPAAAAEBAJ

7426720 - a system for dynamic preloading of classes through memory space cloning of a master runtime system process so that child runtime processes inherit the memory state of the parent.

6910205 – compiling virtual machine (VM) instructions for a function into native machine instructions to improve the execution speed of VM.
http://www.google.com/patents?id=U-4UAAAAEBAJ

6061520 – improves initialization of static arrays by preloading all class files, determining the initialization performed by them and then consolidating them into a single expression.
http://www.google.com/patents?id=mEwEAAAAEBAJ

Again, I'm not quite sure what to make of these. I'm a hardware engineer, and software is not my expertise, and I'm not very knowledgeable in Java or VM's.

If Google sticks to the usual patent lawsuit script, they will file a countersuit allegeding Oracle's infringement of their patents, there will be a long drawn out deposition process involving literal mountains of paperwork and lots of emails. Then generally patent infringement cases are settled - often by some form of cross-licensing agreement and a bit of cash to sweeten the weaker portfolio's side. The cases tend to be long, and they are incredibly risky to the accused since a "jury of peers" generally are not experts in the specifics of, in this case, virtual machines and Java JIT compilers and a guilty verdict generally involves a crazy amount of money and, usually, an injunction which stops the sale of all infringing designs (in this case, it would halt all Android sales, and possibly the App Market (since the apps were written with the SDK), as well as the removal of the SDK).

Reading through those, it sounds like a lot of "prior art" is going on here. Especially "6192476 - controlling access by a thread to a resource based on permissions from routines in a stack
http://www.google.com/patents?id=G1YGAAAAEBAJ" Can we say, "Mutex".
 

theeedude

Lifer
Feb 5, 2006
35,787
6,197
126
Looks like the main issue is that Google is not implementing the precise Java spec, in which case it would be covered by Sun's license. Implementing either a subset or superset of the spec is not allowed by the license, to preserve compatibility. This was at issue when MS tried to do its own JRE but extended it to not be backwards compatible. If Google wants to be covered by the license, they'll have to match exactly the Java spec functionality. That would mean that Android apps would run on all Java enabled platforms, not just Android.
 

tatteredpotato

Diamond Member
Jul 23, 2006
3,934
0
76
One thing you forgot is that if Google moved to a .NET language (which is doubtful as Android competes with Windows Mobile) every single program in the marketplace would have to be rewritten.

Let me ask you this, what does Oracle gain by letting Google continue to infringe on their patents? Oracle has a fiduciary duty to its shareholders to defend its intellectual property. Oracle didn't buy Sun for the fun of it.

Google could just make a Java .Net compiler so the source would stay compatible, however that most assuredly won't happen.

I'm not surprised Oracle is doing this, but I agree this will be little more than a waste of time that might end in some cross licensing agreement. Anyone who thinks this is the "death of Android" is just being naive.
 

Patranus

Diamond Member
Apr 15, 2007
9,280
0
0
Feb. 6 (Bloomberg) -- Google Inc., fighting a patent lawsuit filed by Oracle Corp., lost its appeals court bid to keep secret an engineer’s e-mail that says the Web search company should negotiate a license for Java technology.
http://www.businessweek.com/news/20...-bid-to-exclude-e-mail-from-oracle-trial.html

Thats not good. I am pretty sure most juries would rule in favor of Oracle if they hear these emails. They certainly don't make Google look good when it comes to this litigation.
 
Aug 23, 2000
15,509
1
81
http://www.businessweek.com/news/20...-bid-to-exclude-e-mail-from-oracle-trial.html

Thats not good. I am pretty sure most juries would rule in favor of Oracle if they hear these emails. They certainly don't make Google look good when it comes to this litigation.

“The e-mail’s discussion is directed at a negotiation strategy as opposed” to a legal strategy, Circuit Judge Alan Lourie wrote for the three-judge panel. “The e-mail does not evidence any sort of infringement or invalidity analysis.”

So the Judge says the email doesn't prove that there was any copying going on.
It's possible the guy just wanted Google to license the Java tech so they didn't have to re-invent the wheel.
 

bearxor

Diamond Member
Jul 8, 2001
6,605
3
81
Are those the emails where the engineering team tried to create an alternative from scratch and said 'eff that, let's license java'?