7 ms·

So Oracle is saying that google violates copyright by implementing certain java interfaces for android? I’m not sure how java and android work together. Does go

So Oracle is saying that google violates copyright by implementing certain java interfaces for android? I’m not sure how java and android work together. Does google provide a java wrapper and then go under the hood and implement those APIs for the android device?

lallysingh 8y ago
They use a different bytecode and implemented the standard Java apis themselves.
snarfy 8y ago
They use the same signatures in the header files, e.g. int add(int, int); //adds two numbers Oracle is claiming copyright on a similar bit of text. If Oracle wins, any developer writing a method with that signature is violating Oracle's copyright. It's insanity.
- treis 8y ago
  That's not true. If you independently come up with that then there is no copyright violation. Especially if it's a single method out of a large API.
  - blihp 8y ago
    But there's the rub: they are claiming the API is copyrightable, not just the code that implements it. Independently coming up with the code is irrelevant if the API being (re)implemented is copyrightable.
    - treis 8y ago
      Yes, I understand that. My point is that simply defining add(int, int) isn't infringement. You have to actually copy something for it to be infringement. Independently coming up with a small portion of an API is not copying and not infringement.
      - dralley 8y ago
        There's not much of a difference, practically speaking. If you can copyright the structure and organization of an API, then interoperability becomes impossible. If every API has to be unique enough that it doesn't violate the copyright of anyone else in the same problem space, you can no longer create compatible reimplementations.
        jdironman 8y ago
        Its almost like trying to copyright a spoken language.
        treis 8y ago
        >There's not much of a difference, practically speaking. It means you can come up with your own API design without infringing on someone else's copyright. >If you can copyright the structure and organization of an API, then interoperability becomes impossible. It doesn't become impossible. It simply means that people need to follow the license that the API has. If the Java API was valuable to Google then they can pay the creators a license fee to use it. If the license fee is too expensive or comes with unacceptable terms then don't use it. Write your own or use some other API with a more permissive license.
  - taftster 8y ago
    Contrast that with musicians who get sued over copyright for using one or two bars of music from another composition. Unfortunately I'm concerned that this isn't even about a single method being OK and a large API not.
  - monocasa 8y ago
    Well, good luck proving to a jury against Oracle counsel that you independently came up with it.
- AnimalMuppet 8y ago
  It's a bit different from that. Any developer who writes a big chunk of the Java API violates their copyright. It's the API as a whole, not one method in it. It's still insane, in my view...
  - monocasa 8y ago
    That's still super grey, the higher courts in this case haven't really been looking at de minimis interpretations in a positive light (overturning a lower court de mininis ruling over the range check code). It just hasn't been a real sticking point yet on the API questions.

lallysingh 8y ago

They use a different bytecode and implemented the standard Java apis themselves.

snarfy 8y ago

They use the same signatures in the header files, e.g. int add(int, int); //adds two numbers Oracle is claiming copyright on a similar bit of text. If Oracle wins, any developer writing a method with that signature is violating Oracle's copyright. It's insanity.

treis 8y ago
That's not true. If you independently come up with that then there is no copyright violation. Especially if it's a single method out of a large API.
- blihp 8y ago
  But there's the rub: they are claiming the API is copyrightable, not just the code that implements it. Independently coming up with the code is irrelevant if the API being (re)implemented is copyrightable.
  - treis 8y ago
    Yes, I understand that. My point is that simply defining add(int, int) isn't infringement. You have to actually copy something for it to be infringement. Independently coming up with a small portion of an API is not copying and not infringement.
    - dralley 8y ago
      There's not much of a difference, practically speaking. If you can copyright the structure and organization of an API, then interoperability becomes impossible. If every API has to be unique enough that it doesn't violate the copyright of anyone else in the same problem space, you can no longer create compatible reimplementations.
      - jdironman 8y ago
        Its almost like trying to copyright a spoken language.
      - treis 8y ago
        >There's not much of a difference, practically speaking. It means you can come up with your own API design without infringing on someone else's copyright. >If you can copyright the structure and organization of an API, then interoperability becomes impossible. It doesn't become impossible. It simply means that people need to follow the license that the API has. If the Java API was valuable to Google then they can pay the creators a license fee to use it. If the license fee is too expensive or comes with unacceptable terms then don't use it. Write your own or use some other API with a more permissive license.
- taftster 8y ago
  Contrast that with musicians who get sued over copyright for using one or two bars of music from another composition. Unfortunately I'm concerned that this isn't even about a single method being OK and a large API not.
- monocasa 8y ago
  Well, good luck proving to a jury against Oracle counsel that you independently came up with it.
AnimalMuppet 8y ago
It's a bit different from that. Any developer who writes a big chunk of the Java API violates their copyright. It's the API as a whole, not one method in it. It's still insane, in my view...
- monocasa 8y ago
  That's still super grey, the higher courts in this case haven't really been looking at de minimis interpretations in a positive light (overturning a lower court de mininis ruling over the range check code). It just hasn't been a real sticking point yet on the API questions.