How Oracle’s fanciful history of the smartphone failed at trial

by Joe Mullin

Aurich Lawson

Despite a final verdict, the recent Oracle v. Google trial leaves plenty of questions about the future of APIs, fair use, copyright, development, and more. While their views do not necessarily represent those of Ars Technica as a whole, our staffers wanted to take a look at the outcome and potential ramifications from both sides. Below, Joe Mullin says Google's win sends a powerful message against a familiar legal tactic. Elsewhere, Peter Bright argues that software is about to suffer. You can also find guest op-eds from professor Pamela Samuelson (pro-Google) and attorney Annette Hurst (pro-Oracle).

We may never know with certainty why the jury in Oracle v. Google decided in Google's favor, but I can make a pretty good guess.

Like the jury, I'm no expert. I've been reporting on technology law for years, but becoming an experienced journalist is really just mastering the fine art of non-expertise. I have a pretty good conceptual idea of what an API is, derived entirely from listening to more knowledgeable people talk about this case. But if you showed me a block of code, I couldn’t pick out the APIs or "declaring code" at issue.

However, you didn’t need to be a computer expert to see through Oracle’s case. Google and Oracle offered competing narratives about the early history of smartphones. Oracle tried to win by re-writing that history, and it just didn’t add up. The brave new world it presented was contradicted most powerfully by the former CEO of Sun Microsystems, Jonathan Schwartz. In the end, Oracle could not be saved, even by their crack team of JD’s and PhD’s. By the end of trial, their case looked, to me, like an intellectually bankrupt loser’s lament.

It was a hell of a show, though. Oracle was well-served by the new team of lawyers brought in from Orrick, Herrington & Sutcliffe. The firm's New History of the Smartphone was a story well told. From the first minutes of opening statements, their case was simple and sharp, folksy and forceful.

"I always have to think when I write this out, because I'm not used to writing billions," said Oracle lawyer Peter Bicks as he wrote out "3,000,000,000" on a white poster for the jury. Google had "made a deliberate business decision" to "copy Oracle's software illegally" and had reaped "huge profits," he said.

It was a simple argument—and a dangerous one. Accusing someone of wrongful copying has the visceral draw of a school-yard taunt, yet it's a crime that's well-understood, and seriously punished, in the adult world. Bicks knew Google would point out the APIs in question were just one-tenth of one percent of the massive Android codebase; he deftly belittled this argument before it was even made.

"'We left a lot behind,' is what you'll hear," said Bicks, channelling his opponent. "'We took your property, but we didn't take all of it.'" This would be Google's "fair use excuse," he told the jury. The 11,000 lines of code that Google would compare to a restaurant menu were powerful and creative works, Bicks assured them. This was a story about hard work and just rewards. Just don't copy stuff. Is that too much to ask? Do your own work.

Google's lawyers would have to explain to a non-expert jury that, in the software industry, some kinds of copying are truly allowed, even vital. It's a more complicated argument that requires some understanding of industry practices, and having to make it put Google at a disadvantage.

The trial also had a fundamental structure that favored Oracle. The jury was regularly reminded, by a judge in black robes, that Oracle’s code was copyrighted. The mere existence of a high-profile trial—the judge and his assistants, the room full of dark-suited lawyers, the coterie of reporters in back taking notes, the cryptic mention of "billions" at stake—it all drove home another point, a kind of subtext that favored Oracle. The Java APIs at issue here were Very Important Things.

All in all, Google had a high mountain to climb. If I'd been forced to make a bet on which side would win the minute after opening arguments, I would have put my money on Oracle.

Finding "fair"

The 300-lb linebacker in Google's defense play was Schwartz, the ex-CEO of Sun Microsystems who was Google's second witness. To my mind, it's impossible to imagine Google winning without Schwartz's support.

If you don't understand code, it's still possible to understand Sun's economics, as explained by Schwartz. Creating Java, a free and open software language, was a boon to Sun's hardware business. The language was free for starving college students and free for millionaires—it was good for the world and good for Sun's bottom line. On the stand, Schwartz made it crystal clear: the "free and open" Java language included use of the 37 APIs that Oracle had, literally, made a federal case out of.

From that moment on, Oracle was put on the defensive. In Schwartz's cross-examination, Oracle gave the first glimpses of the alternative history it would try to piece together for the jury. There was no dispute that Schwartz had publicly celebrated Android, welcoming the new software in his 2007 blog post. But hadn't Schwartz written that Android was "lame" in an e-mail? Didn't he write privately that Google was a company that played "fast and loose" with licensing rules, that it had little regard for copyright law?

Yes, he'd done all that. He admitted it. But it was far from the "gotcha" moment that Oracle wanted. Schwartz didn't come across as an altruist or an angel; he was a competitive guy, having a difficult time at a company that was flailing on his watch. He badly wanted a deal with Google to work together on Android, but it didn't happen. The few e-mailed potshots he took at Google didn't make Schwartz look like a hypocrite; they made him look real and frustrated and honest.

The unpredictability of asking a jury about the amorphous rules of fair use in copyright law also made for an interesting trial. On one level, the case was about the specifics of Google's behavior. On another level, it was about what it meant to do business "the right way" in America. Who were these men—and they were largely men—who had made so much money, so quickly? Android chief Andy Rubin talked about "wanting to win." On cross-examination, he was forced to acknowledge his own sizable share of the geyser of wealth produced by Google—$60 million in bonuses if he could deliver Android and deliver it fast.

The effort to paint Rubin as someone in a greedy rush didn't hold much sway in the end. Despite his private frustrations, Schwartz had given Google an "A" for fair play without reservation. That meant Oracle had little choice but to engage in ritual denunciations of the former Sun CEO, which continued throughout the trial. In closing arguments, Oracle lawyers showed the jury a slide depicting the "two faces of Jonathan Schwartz."

The continued attempts to tear him down, like entering into evidence an Internet article naming him as one of the "ten worst CEOs," ended up looking like cheap put-downs. Schwartz's answer about the articles could be understood by anyone who'd worked hard in tough times: the economy was in a recession, Sun was failing, and people were hurting. "I was upset, too," Schwartz said.

Schwartz testified about his successes and failures, his company's competitions and compromises. He didn't describe a perfect world; he described something that sounded like reality. Some projects don't work out; some deals can't be done; sometimes the other guy makes a whole lot more money than you, whether he deserves it or not. Sometimes, life isn't fair.