I'm going to be very interested to read this series by Marc Andreessen, but I rather strenuously disagree with this:
Overt sexism aside, from an incentive standpoint the result of shifting from stock options to restricted stock should be obvious: current employees will be incented to preserve value instead of creating value. And new hires will by definition be people who are conservative and change-averse, as the people who want to swing for the fences and get rewarded for creating something new will go somewhere else, where they will receive stock options -- in typically greater volume than anyone will ever grant restricted stock -- and have greater upside.
And sure enough, in the wake of shifting towards restricted stock and away from stock options, Microsoft's stock has been flat as a pancake. The incentive works.
Now, against that, it is true that stock options, particularly for public companies, have an often-destructive random component: they tend to increase in value in rising stock market environments and decrease in value (potentially to zero) in falling stock market environments, regardless of whether value is being created inside your particular company.
For that reason, in the long run it probably makes sense for some new approach to stock-based compensation to be developed that both preserves the motivation to create as opposed to preserve value, but factors out the environmental swings of rising and falling stock markets. Some form of indexing against market averages would probably do the trick. This has been tried from time to time, and I expect it to be tried more in the future, at least for public companies.
This is not the major problem with stock options. The major problem with stock options is twofold: out of the money options encourage managers to take excess risk, because they get nothing if they preserve value, so even a remote chance of boosting the stock price that carries a hefty risk of failure is a good idea for the managers--but a terrible idea for the shareholders.
The other problem is that they encourage short-term misinformation about a company's prospects in order to boost the price long enough to excercise your options and sell the stock. The time of option excercise is one of the few times when an executive can sell his own company's stock without triggering a market reaction.
You will notice that while Microsoft's stock price may not have done much, it also didn't go the way of Pets.com, or even Sun. No compensation system is perfect, but stock options have big problems.