Right there's where your logic is failing you.
You bought a machine that said it was b/g compatible, not n. That machine was whatever price it was. De facto, then, the sold price was fair for a computer with b/g capabilities.
Were it advertised with 'n' capability you would have paid $X extra. That's $X more Apple could have charged, but did not. Ergo, like it or not, Apple is "out" $X.
Alternatively, if Apple had provided "b/g" hardware instead of 'n' hardware, it could have saved $Y. Ergo, like it or not, Apple is "out" $Y.
No matter which way you look at it, unless $Y is negative, there is a cost that Apple ate.
Completely incorrect (last sentence, not the rest). Reread what I posted. Because the b/g price is higher than some comparable 'n' device, Apple sold fewer Macs. For those that bought b/g Macs, they saw the value of the b/g equipped Mac as greater than the value of the hypothetical 'n' equipped PC. There's obviously more to the buying decision than the particular wireless connection, else an eMachine with an 'n' PCI card would own 100% of the PC market.
Listen: you can, with a lot of work, draw a nice, pretty curve of price versus demand. This is high school economics territory, and one of the few basic principles taught there that really holds up under heavy scrutiny. If X people will buy a b/g Mac for $Y, then X+n people would buy a b/g/n Mac for the same price (assuming additional features increase value, which is a safe assumption here). If X people will buy a b/g Mac for $Y, then X+n people will buy a b/g Mac for $Y-m.
Apple holds a micro-monopoly, aka lock-in. I think we can all agree about that.
At a larger level, however, they are not in a monopoly or even industry control position. Even within the micro-monopoly, people will buy fewer of a product that they don't necessarily need when the value is less. While some people may truly need a new Mac right now, the vast majority will wait for a few more months or years before buying if the value is not seen as significant. Again, the fact that Macs were advertised as supporting only b/g and other makers were selling 'n' hardware, meant that Apple sold fewer computers, either as a result of a buying delay or as the result of a wholesale switch. If I buy a new computer on average once every three years, and this time I wait for three and a half years to buy a computer, Apple has lost 1/6th of a computer sale from me. That may not seem like much, but it's significantly more than the cost of 'n' hardware.
Again, I'm not sure why we're arguing about basic high school economics here. Modifying the value of a computer will affect either its sales (if you leave price constant) or its price (if you aim to keep sales constant). Unless you are a monopoly on a utility-style required service, that is pretty much always true (and even in utility monopolies there is some price/demand response).
As I've shown several times, Apple had major financial incentives to advertise their pre-n hardware. The reasons they did not are another topic, but IMHO center around a focused market message and the risk of shipping networking hardware that for a long time looked like it might be made obsolete when the final standard came out.
