Bubbles aren't defined solely by the extent of their damage post-pop. They're defined by the delta between bubble valuation and real value. In admittedly simplistic terms: the size of that delta, multiplied by the number of people who buy into it, usually dictates the impact of the burst.
Right now, we're arguably in the early stages of a bubble. Almost everyone admits that startup valuations, not to mention a few recent, high-profile IPOs (Groupon in particular) are inflated. Money is easy.
I'd argue that today's operative question isn't "by how much are valuations inflated," but rather, "how many people are buying into them?" And where will the contagion spread from here?
We have a few troubling indicators. Celebrities like Ashton Kutcher are leaping into the angel and VC scene. I have nothing against Ashton Kutcher, but I do question his fundamental grasp of the technology business. And I wonder whether his sloshing easy money around the startup world is inherently constructive or destructive to rational valuations within that world. Furthermore, Ashton's just the tip of the iceberg. CAA, Hollywood's leading Hollywood talent agency, is launching a VC arm focused on tech companies. Precisely what a Hollywood talent agency understands about tech firms is anyone's guess -- but CAA has more money than God, is extremely well connected to investment banks and hedge funds, and could serve as a bridge to the general public's interest in tech equity. (We'll recall that the 1999-era bubble was marked by a similar influx of carpetbagging investors from outside the tech industry).
People often point out that the frothy tech startups of 2012 are fundamentally unlike the frothy tech startups of 1999, in as much as 1999's startups went IPO, and today's startups generally get acquired. This is fair. But in both cases, investment money became (is becoming) very easy and very loose. This leads to a propagation (and propping up) of fundamentally unsound startups. History indicates that, eventually, there can be so many bad apples in the batch that people stop being able to tell the good ones from the bad ones. That's when the trouble begins. We're not there yet, but why wait around and let it happen?