I don't think the dotcom bubble comparison is accurate. They're similar only in that (anything) new generates hype in the beginning, before it slows down and generate a more steady incline (whether up or down).

One big differentiating factor between the dotcom bubble and this proposed AppBubble is that the dotcom boom was fueled by investors with no direct knowledge of the market, whereas the app developers have a direct link and feedback from the market; so that they get quicker read of the market and can (if they want to) make the necessary corrections.

The underlying variables to boost this spincycle up:
- Increasing # of potential customers (if you stick to lowest common denominator; in requirements)
- Increasing skill and experience of developers (lower cost, higher quality)