The result? While Oracle is still entitled to a copyright of Java as a whole, they would not be able to assert protection for individual APIs (a claim that was already contentious and is largely without precedent). That would mean the majority of Oracle's claims may be dismissed, and Oracle would be in the much more difficult position of trying to claim that Google stole Java as a whole. Google has filed a rule 50 motion, requesting that Oracle’s claim to the 37 APIs at stake be dismissed “as a matter of law”, since there would be no legal basis to assert them. Judge Alsup may rule on the motion as early as this afternoon.
As if to prove that when it rains it really does pour, The Verge reported this morning Google introduced evidence from the U.S. Copyright Office indicating they have no record of what Oracle submitted to them on Java. How can that be? Apparently in one case a disk was submitted but turned out to be blank, and in a second case the U.S. Copyright Office says they have no record of a disk being submitted at all.
We all know that physical media can go bad, but it doesn’t generally disappear into thin air. If this holds up it could be devastating to Oracle's attempt to win a large paycheck from Google. Oracle wouldn't lose its copyright in this situation (simply publishing a work grants you a copyright) but if you don’t register your work properly with the Copyright Office you are generally not entitled to trumped-up punitive damages.
So the TL;DR version is this: Oracle may be forced to drop their individual API copyright claims (which is most of the copyright case) and even if not they may not be able to ask for anything more than “actual damages” given their apparent failure to properly register their works with the U.S. Copyright Office.
I’d really hate to be the person(s) whose job it was to burn those disks and ship them off, wouldn’t you?
sources: Groklaw, The Verge