I sort of understand stock pricing for companies that do pay dividends. You try to predict the rate of return, look at the going price for that rate of return, apply a discount for the risk factor... I'm not claiming that I can do those numbers easily myself, but I can follow an explanation and feel that it makes sense.
The other model of valuing a stock is what A Random Walk Down Wall Street called the "greater fool theory"--X is a good price, because someday someone will be willing to pay Y > X. I don't deny that this sometimes happens, but I'm wary of betting on that. It feels like a financial game of Hot Potato. (Except that the financial term for "Hot Potato" is "bubble economy".)
So consider Apple. Apple has never payed dividends (as far as I know). There's no particular reason to believe that Apple ever will pay dividends. So what basis is there for choosing a price for Apple other than the greater fool theory?