The most engineer-ish thing to do is a back of the envelope calculation:
1. Estimate how much it'd cost to fight the suit.
2. Estimate the probability of winning.
3. Estimate the damages if you lose.
4. Estimate Oracle's ask for licensing the patents.
If (1 + 2 * 3) < 4, fight, else settle. They probably have pretty good numbers on one, three, and four considering they likely discussed this with Oracle before fisticuffs began. The only perilous part is estimating two.My work has had bad experiences with Oracle license chenanigans. Of course Google probably has better lawyers than my work does!
I mean, obviously, this doesn't mean 'always fight' but it's another value you need to punch in to your equation, one weighing in on the 'fight' side.
In my business, a similar example would be the money I spend on running an aggressive abuse desk. Running a less-aggressive abuse desk would be cheaper, both in the work it takes and in customers I've lost, at least in the short term.
But in the long term, /because/ I have succeeded in making my service a hostile place for spammers, my abuse complaint rate is much lower than it would be otherwise. for a while, another VPS company rented a few servers from me... and in spite of being something like 1/60th of my size, they produced more abuse desk work than the rest of my customers combined. I believe, because I have an aggressive abuse desk.
(I mean, there are other costs... I've lost at least one legitimate customer I know of, because he was compromised and his box was spewing ssh attacks over the network. I shut him down with an email notice, the email went to the shut down box, so he never got it. )
my point is that there are real business reasons to fight that sometimes go beyond the immediate results of that fight.
Giving into Oracle (who owns Sun's patents which were vast and probably only topped by IBM in the industry) is hardly going to give one the reputation for being patent troll bait.
And doing what's best for the rest of the industry obviously isn't something Google cares about.
I am sad that over the last year or so this has gone from a quiet fact of it being a corporation to a in-your-face "Google: do no evil unless you can make money". I guess each generation needs a company to fantasize about being a righteous white knight.
If enough people stop trusting google, they will no longer be able to do this. Consumer trust is integral to their business. and there will always be tension, as to make money, they need to use at least some of that data to serve us ads, and some of us will take offense to that. I mean, really, there's no way they can run gmail profitably without doing things that I feel violates my privacy.
Now, in the short term, they could earn boatloads of money by violating this trust in a more direct manner... but in the long term, it would kill them. I think that ever since they went public there has been a constant war there, short term profits, vs long-term viability.
My own personal expectation was that this trust of google would not have lasted much past their IPO... it has lasted much longer than that.
Personally, I find it odd that now that this trust is eroding, people seem to be complaining more about net neutrality than about google pushing the limits of what they can do with our private data. Of course, I'm a hosting provider, and thus oppose regulation of my market.
Markets must be regulated. I know of no credible market theory that thinks there should be no regulation. As far as I can tell this is just a bizarre invention of American Libertarians.
So here is the cheap way for Google to attack Oracle: Fund startups that research how much it costs to switch from Oracle servers to Postgres, then sell reports to Oracle customers so they can get lower prices.
I can't see that funding some startups makes you vulnerable to suits, but I am not a lawyer.
Another one is that after all the denunciations and sabre-rattling, Google hand over a wad of cash to licence the technology and Oracle score a slice of Android sales.
Merely the threat of future lawsuits... justified or not, could harm Android adoption, so Google may settle (after posturing and sabre-rattling of course) on that basis alone