Authored by: hardmath on Monday, April 16 2012 @ 09:55 AM EDT |
You hint at another glaring problem with Oracle's copyright
claims, which is that the APIs are the mechanism by which
Java-the-language evolves, and thus are both integral to and
proof of the non-fixed nature of widely purposed programming
languages.
regards, hm
---
Do the arithmetic or be doomed to talk nonsense. -- John McCarthy (1927-2011)[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Monday, April 16 2012 @ 11:06 AM EDT |
What about the Patent issues?
Or have they died?[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Monday, April 16 2012 @ 11:49 AM EDT |
In computer science (and apparently in formal language
theory) a "language" is defined to be the set of all strings allowed by the
syntax. That is "the Java language" is all syntactically correct Java programs.
So assuming Oracle is not yet ready to claim copyright to all Java programs that
have been or can be written, I wonder what is meant by "can Oracle copyright
[...] a computer language". They probably do have copyright in the language
specification and their implementation of it (i.e. the oracle java compiler).
But those are not what is at issue here.
I guess a similar argument applies
to the API, but the fact that the API spec uses Java language syntax to specify
the method signatures might confuse.
Would copyright law prevent me from
using the character "Harry Potter" in a completely new story? [ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Monday, April 16 2012 @ 12:56 PM EDT |
LOL [ Reply to This | Parent | # ]
|
|
Authored by: greed on Monday, April 16 2012 @ 05:26 PM EDT |
The only way they stand a chance is, through something like:
-
The
language and APIs are described in specifications.
-
The specifications are
protected under copyright.
-
The language is a derived work of the
specification, but it cannot be protected--it cannot be fixed in a tangible
medium.
-
But the interpreter/compiler is a derived work of the language, and
therefore a derived work of the specification.
-
The interpreter/compiler can
be fixed in a tangible medium, and so can be protected.
Ignoring all
of the "functional stuff can't be protected" and all the other problems with
Oracle's claims.
I sure hope copyright cannot adhere via an unprotectable
step. In this case, as has been said, it would be copyrighting ideas.
It
actually reminds me of the hoop-jumping people were going through to try and
break the GPL (and Apple's licenses) via first sale.
Which also reminded me
of a "proof" a math teacher showed us which had a series of steps, all appearing
valid, which resulted in the nonsensical proof that "1 = 0".
[ Reply to This | Parent | # ]
|
|