As the old saying goes, history never repeats itself, but it often rhymes.
Asset bubbles occur when people are more interested in acquiring assets rather than effort to acquire the money used to buy the assets. Basically, people start throwing money around just to get that asset.
It usually occurs when money is easily acquired through loans, etc, and as the asset prices increase, the general idea is "I can borrow the money, make some money on the asset, and then pay back the loan".
Usually a lot of delusional math occurs that convince people that the price they are paying is actually "cheap". For example, people who made $50k/yr and bought a $1M house would delude themselves into buying the house by saying "Well, I can take the teaser rate for 2 years, and by then the price should go up by $100k, and then I can sell, and use that money for a downpayment on a house I can afford." Or, similarly during the dotcom bubble, analysts would say "Well, the P/E ratio is currently 1000, but based on the projected earnings 5 years from now, it's actually only 20." To be honest, in the midst of the bubble, this type of thinking works well. We all know what happens once the bubble bursts, though.
So when popular products like Instagram with zero revenues are being sold for $1 billion, there's a lot of the bubble math going on. "There are 30M users and we're paying $1 billion, so we're only paying $30/user." Anyone remember the acquisition of broadcast.com by Yahoo as a strategic acquisition? The same goes for the rumored attempted acquisition of Path for $100M a couple of years ago. Or Color.com paying $300k for their domain name. I think these massive Internet companies like Google and Facebook are taking a product, and applying dotcom math to it, and coming up with sky-high valuations given the size of their audience. Maybe they're right, maybe not. But they are the ones contributing to the idea that revenue-less products are worth hundreds of millions or billions of dollars.
So, when VCs know there are giant vacuum cleaners that are willing to pay ridiculous prices for products, of course they are going to drop all their money and invest in as many startups as possible. They would love to invest $250k in another Instagram and make $78M. And this helps sell investment in their funds as well, which causes more investors to pile in, trying to get a piece of a bunch of Silicon Valley startups.
And this is the situation where companies like Facebook or Google, or VCs are more interested in acquiring assets (ie investments in startups), rather than caring about the actual money itself.
My guess is that the bubble will burst soon after the Facebook IPO. Why? Because at that point, paying $1B for a company with zero revenues will likely be the cause of shareholder lawsuits, and there will be a lot more scrutiny involved in these acquisitions. Since selling to Google or Facebook is the exit strategy for most of these SV startups, if that door closes, then funding will get pulled quickly and violently, and it will be the start of the next dotcom bust 2.0.
People will also be closely watching Facebook's financial performance and seeing if they can justify their valuation of $100B. If they miss estimates, or if their ability to monetize consumers flattens, then it's most likely a catastrophic, extinction-level event for most of the startups in the Valley.
I certainly hope this doesn't occur, but I can't see how it's not likely. EDIT: To be clear, I don't mean that I believe Facebook will miss estimates, I do however believe that we are in a bubble that will burst.