Definition of a Bubble

This question has been floating around the economics blogosphere recently. My consideration of the topic came after reading Arnold Kling's What is a Bubble? I started this as a blog comment, but ultimately realized I had typed too much and decided to cut-and-paste it here:

All popular definitions of a bubble come down to this: "Bubbles occur when most people get the future price of an investment incorrect."

When a minority of investors do this, however, we don't call it a bubble. We call it poor market savvy.

For me, it's difficult to understand asset bubbles in any context other than Austrian theory. Bubbles occur when investments in a given asset exceeds the economy's ability to realistically make use of that asset for its intended purpose.

It should be obvious that the problem with most bubble definitions provided is that they provide no reference to the actual economic use of a good. They speak only to esoteric figures for price, interest, and expectations as though the numbers themselves were ultimate abstract unquestionable concepts like the set of All Real Numbers.

So maybe bubbles occur when a critical mass of investors forget that the price of an asset reflects its price as a utilizable product, and instead come to believe that the price is a reflection of expectations about future prices, i.e., when perhaps the majority of investors plan to sell the asset to someone else later, rather than consume the asset themselves.