Google Wins a Part of Its Motion for Summary Judgment

Google argued that APIs are not protectable by copyright. Judge Alsup did not accept the argument:
“Google’s argument that APIs are unprotectable methods of operation attacks a straw man.
It is not the APIs but rather the specifications for 37 API packages that are accused. Even if Google can show that APIs are methods of operation not subject to copyright protection, that would not defeat Oracle’s infringement claim concerning the accused specifications.”

While the preliminary discussion in his ruling shows Judge Alsup knows a lot about programming, he seems to miss the fact that programming in source code is text-based and one cannot use the APIs without listing them. The compiler/translater needs to know/have available the API so that the syntax, and data structures align properly. He also misses the fact that “Java is free for anyone to use” includes the use of the APIs without which Java is useless, even in “Hello, world!” applications. Come to think of it, Google’s lawyers did not do a very good job of presenting their arguments. Everyone in the room knew Google was right but Google failed to present a logical and complete argument. It’s a fact that Java is free to use and that includes the APIs. No one disputes that. To use the APIs you have to have copies of them in your code.

“Oracle’s opposition brief seems to use the term API to refer to API packages and API package specifications. If this interpretation is correct, then the parties’ arguments concerning whether “APIs” are methods of operation simply swipe past each other, with each party using the term in a different way. Because the issue is not properly teed up for summary judgment, this order does not decide whether APIs are methods of operation.”

AARRRGGGHHHH! Talking about dancing around the truth. The lawyers really want to make this case expensive. Imagine how much time before a jury will be used to spread these concepts thin enough that a jury can see through them. What a waste of words/time/energy.

At least he does rule that Google is not guilty of violating copyright on the names of the APIs. Why does that concept not penetrate when it comes to names of the variables and structures without which the API is useless? Good luck using those APIs without the names. Sigh… Let’s hope jurors are awake.

see Ruling

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.

21 Responses to Google Wins a Part of Its Motion for Summary Judgment

  1. Contrarian says:

    “You presumed something really wrong Contrarian.”

    Between your extensive duties in day to day administration of large, enterprise level networks and servers and your round the clock hawking of posts on this site, #oiaohm, it is only natural to presume that you have no time to configure and sell customized versions of Android phones and tablets. Of course I do not believe that anyone with your sort of attitude towards things as you express here is doing any of that regardless of your protests.

    If anyone else is convinced of the veracity of these claims, please sound off.

  2. oiaohm says:

    Do you make tablets and phones? answer is yes. Contrarian. Not no. Customised for business usage. Hardware from china software from here. So yes people who will pay me for customising android or meego or what ever else to suit their ends. So yes I must know what is on offer and where.

    You presumed something really wrong Contrarian.

  3. Contrarian says:

    “Contrarian really ask yourself how I know the code is hidden in plain sight. Unless I have been working with it.”

    Perhaps you waste your time doing that, perhaps your claim is as specious as most of your other postings, #oiaohm. “Working with it” is doubtless just another lie, though. Do you make phones? No. Do you make tablets? No. Do you work for anyone who might pay you to make changes to Anderoid? No. Do you just fool yourself? Probably, it must be rather easy to do.

  4. oiaohm says:

    Issue is even if Google are the next evil empire they will never get as strong as Microsoft oldman.

    Reason Microsoft got a lot of exclusive hardware. As time is going on less and less of android devices are being exclusive as more companies are adding the option to open up bootloader. So there are markers that they will not be able to built an empire of absolute rule instead will have to compete to stay the best.

    Already we have sections of the open source world probing for weakness. Notice Inferno-OS is taking advantage of the Android kernel without its userspace.

    There is meego as well. So the battle is no where near over. The battle against Android is lot more fair same hardware vs same hardware. It will become a quality issue.

    Google raise up a OS off the base of Linux kernel. Nothing stops someone else doing the same.

    Google currently is no threat that FOSS cannot handle or battle.

    Patent battle against Microsoft is a battle FOSS has no option bar to win or suffer many times more damage than Google can do.

    Basically google does have a threat level to FOSS possible. But Microsoft threat level is higher. Even with google worse level of threat is nothing FOSS cannot handle and if required defeat.

    So basically bring on the Google evil empire we are ready for it.

  5. oldman says:

    “Hidden version is a very interesting model to say the least. Google is really the first to do it this way”

    Again Mr. oiaohm, all interesting observations. We shall see whether this works out for Google, as we shall see whether they are on the way to becoming the next evil empire.

    Remember Mr. oiaohm, be careful what you ask for, you may get it..

  6. oiaohm says:

    “This having been said, the hiding of the full source in plain sight like this is nonetheless interesting. Google is either stupid like sun was with java and staroffice or far sharper than microsoft ever was.”

    oldman it very smart in fact. Since the source code is out there. Someone cannot take the internal code and sell it under a different license and claim independent development since the code was public-ally released. Corporation thieves have nothing to gain attempting to steal android code basically. Also anyone release a block of real code can be made a laugh stock of that it was just a snapshot of the development repo. They never got it from inside google(even if they did).

    Yes oldman the perfect defence is to the release the code google is a really effective in what they do. Yes they truly make Microsoft look dumb. Yes that google has some brains does cause some fear in some people. But those brains also means you can depend on Google to avoid doing moron things to you like not fixing secuirty flaws properly.

    Sun attempted to keep key sections of java and staroffice secret and iron fist control over the source code. Google has just made it hard. Making it hard it simple to go to google than fork.

    Hidden version is a very interesting model to say the least. Google is really the first to do it this way.

    Contrarian really ask yourself how I know the code is hidden in plain sight. Unless I have been working with it.

    One thing that is being pushed for means unlocked boot loaders on devices to allow devices to be updated at end of life.

    Google android is changing the game. From where no one could install third party firmwares most of the time where that possibility is coming more and more that you will be able to install third party firmware so extending live of devices.

    Google android partners will remain its when you solve want android is replacing is why they will remain.

    Android is replacing what was the mobile phone entry OS –Symbian–. Remember Nokia was the owner of Symbian. Motorola is smaller than Nokia. So Google + Motorola is less of a worry than what Nokia providing Symbian was. Motorola was making there own Linux based phones for a long time.

    Yes Nokia would also release Symbian new features on their hardware first. So really come into the real world Contrarian. Device makers are no worse off with Android than with Symbian even allowing for people dreamed up evil Google might do since nokia was already doing all those evils with Symbian and more.

    Yes the key code to confirm you device is android x version compatible is the kernel and that has been released with version numbers each time so yes you can code your drivers up and so on. Under Symbian this was not the case. Nokia would release there products first then follow with the source.

    Even at the worst hardware makers are better off with Android than Symbian.

    Contrarian please remember device makers were making devices before Android. So Android has to be replacing something. Looking at what being replaced tells you what Google can get away with. Basically Google can get away with lots and hardware makers will not be upset. Minor-ally annoyed at worst but not enough to walk away. Might be enough for them to join up with meego or run there own Linux based OS’s as well so they are on a more stable footing to talk to google.

    Nothing about the Motorola acquire will push them in the Microsoft Direction. In fact most likely the other way. How are you going to win against Motorola if you have to be paying Microsoft.

  7. Contrarian says:

    “Let’s see how this turns out.”

    All this fretting over whether or not Android is open source is rather beside the more important real point, I think. No one is ever going to use the Android source code outside of OEMs who either make a phone or a tablet and they only care to the extent that they might need to create drivers for their specific hardware. #pogson and #oiaohm are not going to do anything with it, certainly, and neither will you or I. If there is something that Android lacks, we will all have to wait first for Google to fix it and then for the device maker to publish a firmware update just like everyone does now for their MP3 players, phones, DVRs, and most other electronic devices that we use. Some very small group of hobbyists might fuss around with the code but they are few and very far between and can be safely ignored by Google and any partners who may remain after the Motorola acquisition is completed.

  8. oldman says:

    Mr oiaohm

    An interesting factoid, but as you have already effectively noted the availability of the source code is rendered useless by lack of an official release.

    This having been said, the hiding of the full source in plain sight like this is nonetheless interesting. Google is either stupid like sun was with java and staroffice or far sharper than microsoft ever was.

    Let’s see how this turns out.

  9. oiaohm says:

    Define not release. Oldman. They are there just not tagged.

    Google is just making it hard to replicate. So you can work out the user-space inside a few development tree revision numbers. Just you don’t know the exact one. Of course someone willing to spend enough time could solve to the exact version number to development tree revision number.

    Basically the source is not secret as such. The secret information is what reversion number is android 3.0, 3.1 and 3.2. Of course that information being secret does not make those closed source. Since you can look at the source code you could be looking at exactly the right revision to each of those version numbers. Just you don’t if you are or if you are not easily.

    So one way the source code to all of android 3.0 3.1 and 3.2 is released. Just it is cloaked.

    So yes all the information is out in the open if someone wanted to take a risk of making there own fork of android 3 series.

  10. oldman says:

    “Android 3.0 Android 3.1 and Android 3.2 kernel code is tagged and released. All the gpl and lgpl parts.”

    So what, the portions that make android android are not released, and until they all bets are off.

  11. oiaohm says:

    Android 3.0 Android 3.1 and Android 3.2 kernel code is tagged and released. All the gpl and lgpl parts.

    The all the stuff under apache license the userspace is untagged. But you can mostly work out from the sdk what revessions were the release.

  12. oiaohm says:

    Phenom then why did they level the repos open. With the core code exposed.

    This is where you argument is flawed.

  13. Why the hangup about one release of Android/Linux? Lots of people write code and don’t release it because it’s not good enough. M$ releases very little of their code. They are probably ashamed of it as well.

    Android will be FLOSS but not solely GPL. I have said repeatedly that’s unfortunate and the major flaw in Android.

  14. Phenom says:

    Android 3.0 is closed source, and it is not open.

    Ohio, I find Google’s decision to close Android 3.0 absolutely logical, and good. It takes Google one step towards Apple and MS, and one step away from open source, FOSS, Linux, and whatever.

  15. oiaohm says:

    Phenom really Google will most likely release 3.0 after 4.0 is released why. Then there will be no one stupidly porting it in strange directions.

    Like you find android 2.x on tablets even that it was never designed for that.

    Its called fork control. Historically it has worked.

    Development branches containing code from android 3.0 and working to android 4.0 can be viewed from outside google just there is no tag to show you want is android 3.0. Comes down to define. Yes the android 3.0 code is out there but google has made it hard to confirm. So it depends on your define what is open source.

    Could someone make a fork compatible with android 3.0 by what is released yes. Could someone claim that theirs is better than 3.0 nop. Since you cannot directly compared. Smart move really.

  16. Phenom says:

    Wow, Pogson, you really bested yourself! So, Google can close the source of an OS, and still that OS is open source.

    You quote it, too: Until then, we’ve decided not to release Honeycomb to open source
    Emphasis mine.

    Android 3.0 is not open source. Period. Then, why are you so happy about open software?

  17. So, having too many versions of Android/Linux is fragmentation but unifying the branches in Android 4.0/Ice-cream Sandwich is closing the source? 4.0 will be open sourced and will work on smart phones and tablets. It’s not a problem for me.

    see Builds

    “Android 3.0, Honeycomb, was designed from the ground up for devices with larger screen sizes and improves on Android favorites such as widgets, multi-tasking, browsing, notifications and customization. While we’re excited to offer these new features to Android tablets, we have more work to do before we can deliver them to other device types including phones. Until then, we’ve decided not to release Honeycomb to open source. We’re committed to providing Android as an open platform across many device types and will publish the source as soon as it’s ready.”

    see Google keeping Honeycomb source code on ice, says it’s not ready for other devices

    It’s their code. They can do with it what they will. ETA for Ice-cream Sandwich is October/November 2011.

  18. Binkly says:

    Speaking about “dancing around the truth”:

    “In a disappointing move, Google has restricted access to the tablet-oriented version of Android, also known as Honeycomb. Version 3.0 of Android, which many have called a fork of the mobile OS (and now it looks like they were right), is now closed source, with access only going to OEMs and specific developers.”

    Android 3.0 is now closed source.

    Open source – and this is what the GPL states:

    “You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:

    Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange”

    Since Android 3.0 is not available to all, it is no longer open source.

    But I know you Robert, you’ll find some “trance mode” way (using cognitive dissonance) to redirect this point with some silly reference to 2006 and setting up computers for students or a reference to ARM, essentially you’ll pretend you didn’t read this and continue spewing ideas from your delusional mind about how Google is the world of technology savior and how ARM is the messiah.

    It’s just nice to throw these factual pieces of evidence at you and watch you squirm in front of your audience of what, 4 people?

  19. oiaohm says:

    MS-JVM Joey McIntyre microsoft lost that on abuse of trademark.

    Basically ms made .net after it. Issue was Microsoft calling what they did JAVA when it was not passing the TCK or other requirements to use that Trademark so it was infringement on trademark.

    Google did not make this mistake its Android and Dalvik.

    RedHat’s IcedTea passes TCK so can call itself JAVA without breaching trademark.

    Dalvik is a different beast to what JME uses completely.

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

    That is google big problem is Dalvik has a different performance shape and bytecode requirement so could never pass JME certification.

    Simple fact the VM’s are different. The bytecode are incompatible due to the two different ways of doing a bytecode engine.

    Basically google is not being deceptive when they say that Dalvik is not java. JME VM and Dalvik are completely different.

    Google has not used any of the JME API’s in Android.

    Closest java to Android is java se with awt and swing removed and a new graphical library added instead.

    Problem is there is a very big question if the VM patents can apply due to Dalvik and Java VM being from completely different design methods.

    Oracle and Google are not shorted sited at all. All these rulings have effects outside there battle.

    API cannot be protected thank you open assault option for wine.

    “In the end, if all else fails, Oracle will (justly) win on their VM patents, there’s precedent after all, Microsoft had to pay out for .NET and the CLR.”
    CLR is a stack based VM. So Oracles patents did apply. Google Dalvik is a completely different beast. Closer to python bytecode method. There is more than one way of doing a VM.

    Basically lot of test casing going on here.

  20. Android/Linux uses JDK…


    “JDK 5 or JDK 6 (JRE alone is not sufficient)”

    “Oracle grants you a non-exclusive, non-transferable, limited license without license fees to reproduce and use internally the Software complete and unmodified for the sole purpose of running Programs.”

    The developer uses the JDK to create a Java language programme which is translated to Dalvik and that’s it. No JME involved.

  21. Joey McIntyre says:

    He also misses the fact that “Java is free for anyone to use”

    And you miss the point that none of this is about the JDK or the JRE which are free to everone, but JME which isn’t and never has been.

    It’s quite sad that the open source fanatics were all on board when Sun Microsystems took Microsoft to court for doing with MS-JVM more or less the same thing as Google is trying to do with Dalvik, but now that Google is involved it’s tantamount to killing and eating babies.

    Google is just playing dirty for the sake of trying to bypass license fees for JME, which is the core platform for Andoid Apps (but it’s nice to see we’ve all gotten past regurgitating Google’s bollocks about how Dalvik is not Java even though it is). It really sad that despite all this chearleading, if someone were to use the same line of reasoning to bypass GPL compliance, you’d be the first in line to tear out their jugular (figuratively speaking, of course) and you know it.

    Oracle isn’t the bad guy here, even aside from everything else, at the end you can’t take JME, call it something else and pretend like that’s okay. even RedHat’s IcedTea and Apache’s Harmony, while not borrowing any code from Sun/Oracle Java don’t pretend to be anything but Java, and even IBM and Apple who’ve long licensed their own JDK/JRE/JVM can’t call it anything but Java (and nor can they, by license break compatibility).

    It’s a fact that Java is free to use and that includes the APIs. No one disputes that.

    It’s not a fact, it’s an ambigeous statement. Java the language is free to use. Java the basic runtime environment (JRE) is free to use, Java the Development toolkit (JDK) is free to use, Java the Enterprise stack (JEE/J2EE) is free to use, but the Java environment being used in this instance, the Mobile Edition (JME) is NOT free to use

    The claim that Java is free to use cannot be disputed purely because it vague, and that’s why Google continues to stress it. They won’t and don’t mention JME directly because they know it isn’t and they know they’re wrong.

    The compiler/translater needs to know/have available the API so that the syntax,

    and there’s another key point, the interprester is the VM and so is the compiler, if they’re arguing about what a compiler and interpreter needs to do their doing themselves a diservice, they licensed neither and tried to rip off their own. google’s cheer leaders love to stress on Java The Language, which isn’t the point of contention, rather than Java the VM and Framework, Mobile edition which is.

    In the end, if all else fails, Oracle will (justly) win on their VM patents, there’s precedent after all, Microsoft had to pay out for .NET and the CLR.

    AARRRGGGHHHH! Talking about dancing around the truth.

    “theTruth(tm)” you mean. Both interpretations are valid, it’s an issue of determining which one is most applicable to the situation, rather than which one aid the prefered party of Mr. Pogson.

    Imagine how much time before a jury will be used to spread these concepts thin enough that a jury can see through them.

    Says the guy trying to argue that you can’t copyright an API because you link against a class in an application, you’re still bound to the copyright on those classes, and you still can’t distribute them if you never had permission to use and thusly distribute them to begin with (again Dalvik is the JME, not the JDK, and not the JRE). Had Google gone with the JRE or JDK instead of insisting on trying the screw Oracle on the JME a lot of these things would not have been issues (insisting that it is not Java when it is would still be an issue, and the question of weather Dalvik apps can run on non Dalbik JVM would as well).

    The fact of the matter is that Google went with the one edition of Java which is not free to use, and is in essence trying to argue that they should be able to do so because the other editions are. To make it worse, they’re being intentionally deceitful by purposely obfuscating the distinctions betweem not only the javrious Java editions, but between the framework and the language.

    Even sadder is that the cheerleaders don;t even care, to them it’s purely a matter of Google being right because it’s good for Android, and therefore good for Linux. Fortunately this sort of trickery won’t work to get around the VM patents which are the ace in the hole, precedent is a beautiful thing.

    FWI. This has nothing to do with Android. Oracle doesn’t even compete with Google on that front, it has everything to do with google trying to screw Oracle on JME/Dalvik. It’s a case of one megacorporation trying to screw another.

    At least he does rule that Google is not guilty of violating copyright on the names of the APIs.

    This will be important when it comes down to the issue of Dalvik being a rip of the JME, it’ll be important to show that Google in fact lied when they insisted that Dalvik is not Java, and will be of use when it comes time to argue weather Google ripped the JME rather than the JDK or JRE. Oracle isn’t as shortsighted as you may be led to believe.

Leave a Reply