I'm not saying I support this practice, just that there is more than one way to look at it. And I should have realized a better way to do that earlier, which is: Nobody would be pissed if an employee with a 4 year stock plan left after 2 years because he saw the company was going nowhere. Seems to be the same. The employer and they agreed on a 4 year compensation plan that was supposed to keep the employee motivated to stay. When the company is doing poorly, that's not enough. When the company is doing phenomenally well, it's too much. If you're going to complain when the company skyrockets and they don't need to pay $10M to motivate an employee to stay on two more years, then you also have to complain when every employee leaves a dying company while they still have stock to vest.
Again, not saying I agree with or would ever do what Zynga has done, but how can you not look at these two scenarios the same way?
If you want to ensure that your new hire will stay with the company for four years, you put that in the contract. You then negotiate a compensation package that factors in the opportunity cost to the employee of granting you an exclusive right to their services for the next four years. This compensation will consist not of "maybe-someday-we'll-all-be-super-rich-won't-it-be-grand dollars", but instead "we're-a-startup-and-we-don't-have-many-of-these dollars". This will be expensive.
Fortunately for you, dear founder, there's an alternative (at least there was, until these guys ruined it for everyone). Instead of requiring the employee to commit up front to stay for four years, you structure the compensation package in such a way that the employee has an incentive to stay, but isn't required to do so. These packages are often called "golden handcuffs" because they bind an employee to an employer for a number of years, the result you're after, in a way that relies on a large signing bonus of "maybe-someday-we'll-all-be-super-rich-won't-it-be-grand dollars" that is to be disbursed annually over the life of the agreement.
You and the employee both hope those dollars will eventually be worth something, but right now all you know for sure is that you have a lot of them. It's a trade that works for both parties; you keep your employee if those dollars prove valuable enough to compensate for the opportunity cost incurred by the employee in continuing to work for your company, and the employee has the freedom to move on if they don't. It's the "golden" part that keeps the employee around, not the "handcuffs". And remember that it was you, the employer, that chose to do the deal this way. You would have preferred regular handcuffs, but didn't have enough "we're-a-startup-and-we-don't-have-many-of-these dollars" to afford them.
I think the two scenarios are similar, but in the first case, the employee's ability to leave was de-facto -- of course she can leave any time she wants. That's part of the risk the employer takes. But if you want to balance that equation the other direction, that needs to be just as explicit up front. The possibility of a company 'exploding' is precisely the possibility the employee is hoping for. If you want to put a cap on that, fine, just make that cap explicit up front.
Again, not saying I agree with or would ever do what Zynga has done, but how can you not look at these two scenarios the same way?