Oracle Blinks

Oracle is giving up on suing Google over software patents. That leaves suing over copyright violation which is also a very weak argument for them. Basically, they claim Google copied Java to make Android which is false. Android runs Dalvik byte-code, not Java. The apps are written in Java which anyone is allowed to do and the apps are translated to Dalvik for execution. So, if there is any copying it is de minimus…

How embarrassing for Oracle/Ellison. 😉

The patents were being whittled down by appeals to USPTO and were de minimus anyway. Oracle could not even calculate the supposed damages even after two tries. The judge was not amused.

About Robert Pogson

I am a retired teacher in Canada. I taught in the subject areas where I have worked for almost forty years: maths, physics, chemistry and computers. I love hunting, fishing, picking berries and mushrooms, too.
This entry was posted in technology. Bookmark the permalink.

17 Responses to Oracle Blinks

  1. Clarence Moon wrote, “find a reason that is better founded than the notion that you have looked at source code and see it as different.”

    Copyright laws protect against copying without permission not creating products. Google nor anyone else do not need Oracle’s permission to write software.

  2. oiaohm says:

    Clarence Moon
    “I will stand by the notion that overall structure is as copyright protected as verbatim code in the software world and you can say that it is not, but find a reason that is better founded than the notion that you have looked at source code and see it as different.”
    There the computer sitting in front of you cannot exist. That plan and simple.

    The bios int codes and exact struts used are a 1 to 1 copy from the IBM bios that was copy right protected by a process called clean room reverse engineering. Where one group documents and one group make a new copy. Every bios in existence today is based off this event even the one IBM uses today. Historic fact does not agree with you at all Clarence Moon.

    Documented structures are not protected by copyright.

    Issue is the method Google used to acquire the structures.

    If google had followed the standard processes of clean room reverse engineering. Even if the struts appeared the same Oracle would not have a legal leg to stand on.

    Yes MS cannot apply copyright law against wine since it has obeyed the rules of black box reverse engineering.

    Yes clean room and black box are the two known valid past legal question methods to duplicate a copyright protected work. Google took another path that is not legally defined.

    If clean room or black box result in the exact same structure you don’t have a legal leg to stand on with copyright that is known.

  3. Clarence Moon says:

    “There’s no way …”

    Ignoring your dictatorial pronouncements as to what is law and what is not, Mr. Pogson, what do you suppose drives the plaintiff in this case? Why are they working so diligently to get their arguments into court sooner rather than later if the outcome is so certain to be against them? Is Larry Ellison a secret masochist who is longing for a good beating?

    I will stand by the notion that overall structure is as copyright protected as verbatim code in the software world and you can say that it is not, but find a reason that is better founded than the notion that you have looked at source code and see it as different.

  4. Clarence Moon wrote, in ignorance of copyright, “Rather it is the copying of the java system structure that is being tested.”

    “system structure” is not copyrightable. Read the law. It is not a work, performance etc. It is an idea about how to interface to the software. In particular Android/Linux uses that interface but not the software of Oracle so there can be no copyright violation. Also, once the interface gets to Dalvik, it’s totally different even to a layman like a juror. There’s no way a juror will see Dalvik is identical to/copied from Java VM or that Dalvik bytecode is identical/copied Java byte-code. Don’t forget, these things will be available for jurors to examine with their own eyes and through the eyes of experts.

  5. oiaohm wrote, “Neither Oracle or Google knows 100 percent for sure if copyright case is valid or not. Its a untested section of law.”

    Clearly there are judges who do not get the technology and there are judges who are not impartial and there are juries that get snowed… but Google knows their case is valid. You cannot copyright protect ideas. That’s all any software is, a representation of ideas, particular steps in some process. You can copyright some particular edition of a cookbook but you cannot copyright grandma’s recipe for cinnamon buns any more than you can copyright the interface to software like Java when it’s not Java to which the software implemented links. You can copyright a painting but not the colours.

    It may take a while but when the dust settles, Google will win this one. Even if there is a temporary win for Oracle, the cost to Google will be insignificant, a few days or revenue, perhaps. The cost to Oracle’s reputation is much larger. No one will want to do business with them forever. Would it not be ironic if the world’s Java programmers switched to Dalvik just to be rid of Oracle? Anything’s possible once the die has been cast like this.

    The way these bullies, Apple, Oracle and M$ have abused the legal system is quite visible to the world and they soon will have no friends and lots of “partners” only too glad to plunge the knife.

  6. oiaohm says:

    Clarence Moon really you need to take a copyright course.

    “If you took a Frank Lloyd Wright design that was built of limestone and sandstone and replicated it in a way using chicken wire and gunite spray, it would still violate the copyright of the design, I think. Same thing here.”

    Copyright is not Intellectual properties rights.

    Copyright does not forbid an item looking the same. IBM bios vs the PC clone bios for example.

    If I had photos of a Frank Lloyd Wright design taken by someone else. They documented on sizes and so on. The processed to make a clone of the object out of limestone and sandstone. I have not breached copyright. Yes I have made a possible counterfeit object. As long as the two objects have different defects due to different production its not copyright infringement even if they are made from exactly the same material.

    Anti counterfeit objects fall under trademarks and patents not copyright. So if Frank Lloyd Wright had patented or trademarked that design the copy in chicken wire or limestone and sandstone would be in trouble.

    “Verbatim copying of source” is 100 percent sure a breach of copyright.

    Making a mold of a Frank Lloyd Wright object is a breach of copyright because its a from of Verbatim copying. And any object from that mold is a breach of copyright.

    Using scraping software(as google did) on binaries to make templates to code in might be equal to make a mold off a Frank Lloyd Wright object. If so Google is in breach. If not Google is free and clear.

    The question is if what Google did is Verbatim copying or not.

    The Google copyright case is quite a important ruling. Clarence Moon Its defining where the edge of copyright is when it comes to computer binaries.

    Clean room coping has been done for years. Were one group takes the program a part documentations it and then a another group that has never seen the object makes a copy.

    Other countries have permitted the same person taking apart and making a copy. As long as it not a exact copy. The copy has to have improvements and alterations.

    The fall out from the copyright case if Google loses is basically nothing. The fall out if Google wins the copyright case is massive.

    Neither Oracle or Google knows 100 percent for sure if copyright case is valid or not. Its a untested section of law.

    How often to important test cases like this come up. Not very often.

    The reason Oracle brought the patents in was that the copyright case was not what you call 100 percent solid and patents were thought as a way to end the case quickly. That has not worked.

    Odds of Google or Oracle winning the copyright case is 50/50. Both have valid reasons to claim that they are in the right. I really do fell sorry for the judge there is no black or white answer here.

  7. Clarence Moon says:

    You have to broaden your understanding of copyright, Mr. Pogson. The issue is not any verbatim copying of source. Rather it is the copying of the java system structure that is being tested. If you took a Frank Lloyd Wright design that was built of limestone and sandstone and replicated it in a way using chicken wire and gunite spray, it would still violate the copyright of the design, I think. Same thing here.

  8. oiaohm says:

    Dr Loser Really the copyright infringement case is the more interesting one.

    Google used software to scrape over stuff to generate basic structures. Now if this is legal its a boost to projects like Wine. Since the full ABI/API of Windows could be stubbed in a few days.

    Dr Loser you have the license wrong. The files in question and from the OpenJDK from Oracle that are LGPL. Apache Open Source License was put on the Google files. But even if they ship the android platform requirements is that the effected files are not modified. Device makers most likely don’t have any major infringement to answer.

    Only section of LGPL the device makers have not obeyed is releasing the source. Due to the version of LGPL used in OpenJDK they have 60 days to conform once notified. So most likely not punishable.

    Copyright case is Google vs Oracle and mostly everyone else not effected at all. LGPL allows making of closed source applications on top even linking the LGPL into the final binary. As long as you give back what you changed in the LGPL code base.

    LGPL is very moderate license. Media coverage of the problem is most likely a large disruption than what the license change will be.

    The ruling is going to be important to anyone reversing program code.

    It what pissed Oracle off the most about it was that in there eyes there was no requirement to change the license on those files at all. Google at the time did to get around device maker GPL and LGPL phobia.

    Today that the GPL and LGPL phobia is mostly dead in device makers the change back will make no difference at all.

  9. Dr Loser says:

    @Robert:

    I’d be interested to hear which parts of the Java source code you looked at, given the fact that you are fairly contemptuous of anything that runs on a virtual machine and have only ever claimed expertise in Pascal, but here’s an interesting link to somebody who basically agrees with you. It’s not shipped; it’s just testing and stuff.

    Unfortunately, as Nitin points out, this doesn’t actually make a difference in court. (Whether it should or not is a different matter. I would side with you in this case and claim that it shouldn’t.)

    An even more interesting point is that Google got around this by … wait for it … replacing the GPL with the [incompatible] Apache Open Source License.

    I’m sure you can imagine how riotously amusing I find this.

  10. I’ve looked at the source code, Clarence. There’s no Java in what Google and others put on smartphones. Oracle and others have no claim on the Java apps that people write and translate to run in Dalvik virtual machines on Android/Linux phones. The only thing I have seen were a few headers from the TCK that were part of the SDK, not what’s on phones. Even if dinged for what’s in the SDK, Google has not $billions of liabilities associated with software distributed to smart thingies.

  11. Clarence Moon says:

    “Without prejudice” means just that, Mr. Pogson. The case is not pre-judged in any way and can be refiled at any time and the previous filings are not evidence of anything and cannot be referenced in the new action.

    “If they do not push the violation of the patents, how is it possible to push the violation of the copyright? It’s software after all. The patents are for the software.”

    I can only give my understanding of the matter, Mr. Pogson. There is obviously more than one side to the argument, since it is in the courts and expensive and erudite lawyers argue about it. It is not a simple matter of copying code, binary or source. The issue turns on how the very structure of java can be a unique expression of an idea. You may want to pooh-pooh that, saying it is “only headers”, since you want Google to win, but it is not so easy to dismiss and Oracle, with their legal advice, seems eager to put that very notion to the judge and jury.

    Maybe you know something that they do not know, but you have to at least understand that someone with a law degree is willing to argue the other side.

  12. Clarence Moon wrote, “What you suggest is evidence of “Oracle giving up” is more of an instance of a change in tactics and timing.”

    How can Oracle, on the one hand, tell the court it is being damaged, and, on the other hand tell the court that fixing the damage can wait? I expect this move at least will prevent a “preliminary injunction” which is the key thing that Oracle seeks and $billions is now out of the question in damages. The judge even suggested $0.1 billion would be a more reasonable starting point. That Oracle cannot even calculate its damages strongly suggests they do not exist. The concept of “without prejudice” just means putting off a decision possibly for years and is an admission Oracle knows its case is very weak. It’s like a bar brawler asking the other side to “take it outside” and then skipping out the back door.

    The issue of the e-mail being coloured as evidence that Google did wrong and knew it completely misses the point that the e-mail was sent after Oracle threatened legal action, not anywhere near the time of the “crime”. It was triggered by enquiries within Google by the lawyers. Either it will be thrown out on appeal (imminent) or the jury will get the full context and they will discount the interpretation that Oracle makes and the judge entertains. That and a few header files is all the copyright case is about, incredibly weak. What’s the math on 0.1% of the code being copied, or not? How does the fact that Oracle allows any developer to use those headers with no paid licence affect the maths? If this development does not explode Oracle’s balloon it at least is a flitting-around-the-room rip in its fabric. If they do not push the violation of the patents, how is it possible to push the violation of the copyright? It’s software after all. The patents are for the software.

  13. Kozmcrae says:

    “The core issue here is not patents,…”

    The core issue is no longer patents. And no longer patents under protest by Oracle.

    Clarence the spin doctor running out of spin.

  14. oiaohm says:

    Clarence Moon the dismissal without prejudice is mostly because Oracle could not get past the courts the normal patent damages maths that has been used to work out most patent deals. Also lot of Oracles patents were found to be void or not able to be applied. So really Oracle needs time to back off and recheck if it has a leg to stand on here.

    Dalvik is clearly different to a JVM.

    “Unlike Java VMs, which are stack machines, the Dalvik VM is a register-based architecture”

    Yes Dalvik design is more like a general cpu than a JVM. Even Oracle is not attacking Dalvik with copyright there is nothing there Clarence Moon.

    Its sections of the class libraries Oracle is accusing google of nicking. Not the bytecode engine.

    You are mixing the two cases up with each other. Patents might have applied to Dalvik the copyright case does not.

    Dalvik and the android runtime are in fact two different things Clarence Moon.

    The name of the android runtime that the case is about is android libcore.

    Yes google android libcore kinda does not ring off the tongue very well. So leading to the common mistake.

  15. Clarence Moon says:

    A dismissal without prejudice or the stay requested by Oracle in the article does not end the case. Rather it just delays adjudication to some day in the future. Meanwhile the whole thing hangs over Android.

    I doubt that Google is in favor of this move.

  16. Clarence Moon says:

    I wold caution against taking one’s legal advice from Mr. Oiaohm, Mr. Pogson. What you suggest is evidence of “Oracle giving up” is more of an instance of a change in tactics and timing.

    The core issue here is not patents, which would be licensed anyway as they are with Microsoft, but the notion of copyright, which has the spectre of injunction buried within.

    The copyright issue does not turn on the bits and bytes, I am told, but upon the overall structure of the java framework, which is copied by Dalvik out of necessity. An analogy is along the lines of you cannot publish a novel that is different from some other popular work merely in that you have changed all the character names and re-phrased a number of sentences. If the plot is identical and the storyline the same, it is a copy violation. Ditto for Dalvik.

  17. oiaohm says:

    Really Oracle case proves patent damages can be far less.

    Now thing is remember this Google legal team is tied up in this battle with Oracle. When it ends Google legal team will be free to go after other problems like Apple or Microsoft.

Leave a Reply