|
|
|
| Authored by: Ian Al on Thursday, May 17 2012 @ 04:58 AM EDT |
I insist that my view on all this is final and incontrovertible.
I don't know.
Here's some speculation and innuendo. That's a sort of evidence, isn't it?
The original rangeCheck code was copied from Sun's java.util.arrays. It got into
the two timsort files and was used in Android and was donated back to Sun as
part of the OpenJDK and published under a GPL licence. Google did not use the
GPL licenced timsort.
We know, because Sun told us in the registrations and Oracle told us in this
case, that Oracle do not own the copyright on all the APIs that they use in Java
SE.
The Java SE registrations are stipulated by Google to be owned by Oracle.
I have year's of experience reviewing copyright markings in the files that
comprise registered operating systems (how long has Groklaw been going, now?). I
know that, even if the file has a copyright marking for the owner of the OS,
there may be additional copyright markings for blocks of code owned by other
people.
Looking at java.util.arrays, as I understand it that is a collection of files
rather than a single file. Which is the actual file that contains the rangeCheck
code? What are its copyright markings? Does Sun own the copyright of the
rangeCheck block of code as indicated by specific copyright markings that
encompass that block of code?
The third party APIs that Google were considered by Oracle to have 'copied'
include java.util.concurrent, java.util.concurrent.atomic,
java.util.concurrent.locks. They agreed that they had no standing to assert
these APIs.
The asserted 37 include java.util, but do not include java.util.arrays.
Thus we know as a fact that some of the java.util API code is not owned by Sun
and that they have not confirmed, by asserting it in the 37 accused API
packages, that they own all or any of the code in java.util.arrays.
My understanding is that there has to be an explicit registration of the
asserted copyright in order to bring alleged infringement to court. The Java SE
is a collected works or a compilation that includes components not owned by Sun
as confirmed in the registration. The registration cannot be taken as
registering the individual components.
In summary, Oracle have offered no proof that the java.util.array file(s) has
their copyright marking, nor the data of copyright, nor whether the rangeCheck
block of code is included in the copyright ownership of the file in which it is
found. Neither have Oracle asserted which file it was copied from. Oracle have
not registered the unidentified file as a file of which they own the copyright
(under copyright law the Java SE registrations cannot do this, even if they are
not deficient registrations).
IIRC Google said that they would not challenge Sun ownership of the Java SE
registrations, but that Oracle still had the burden of proof of ownership of the
components. That has been restated during the trial, itself. The Judge still has
the option of throwing out the eight decompiled files and rangeCheck as a matter
of law, on the basis of failure to prove ownership of copyright and registration
of copyright.
---
Regards
Ian Al
Software Patents: It's the disclosed functions in the patent, stupid![ Reply to This | Parent | # ]
|
- Original sin - Authored by: Anonymous on Thursday, May 17 2012 @ 05:44 AM EDT
- Original sin - Authored by: Anonymous on Thursday, May 17 2012 @ 06:04 AM EDT
|
|
|
|
