Write up your experiment to determine how much money your app can earn. Then get the story picked up by as many high profile sites as possible. Looking at the story determines the outcome of the experiment.
That's not to say that there won't be a malinvestment of time; the supply should increase to the point where there aren't as many customers per app, so eventually people won't make it as easily. But the whole thing isn't based on a flawed assumption, the customers really are there.
1) A bubble burst weeds out apps that don't have enough intrinsic value to the customer in the first place (E.g. T-pain, fart apps, and so on)
2) Devs, entrepreneurs, and investors will be forced to evaluate and form actual business plans/models as opposed to "let's throw an app out there and make some money"
3) An app won't always be and shouldn't be the end all and be all of a startup business plan - it's another platform that allows you to reach out to more customers
You might not like the T-Pain app, but the code behind it is likely a hell of a lot better than that of a fart app. Other then that, agreed.
What do you do when users seem to actually want software with little intrinsic value?
Keep iterating and maybe you will become Zynga!
Unlike on the web, or Facebook, where you could simply improve the product, he's screwed if it doesn't take off quickly.