With the exception of the wisest man in Christendom, Homer Simpson, who once exclaimed: "The internet?

Is that thing still around?", most of us have come round to the idea that the digital revolution is the most significant change of our lifetimes.

It affects almost everything already, and can and does make money (Ryanair will charge you £60 this weekend to board a flight you've booked unless you've checked in online, even though the airline has shut its website for maintenance).

In the future which, as the prophet of cyberspace William Gibson has reminded us, is already here, just not evenly distributed, digital technology will affect even more. So much so that I confidently expect that at least some of the jobs my children will end up working at – if there are any jobs for young people to work at – do not yet exist.

The dotcom crash a decade ago was not a refutation of these extravagant claims. Tech stocks fell then because too much was being promised too soon. What is usually described as Web 2.0 was no more than the realisation that the machinery, software and media interactions which were promised in the late 1990s had now arrived.

The shining example, to which even your parents and grandparents have signed up, is Facebook, a company founded eight years ago in a Harvard dorm which went public yesterday on the NASDAQ (the American exchange for technical stocks) with a valuation of around $104 billion.

This makes its founder Mark Zuckerberg, who turned 28 on Monday, worth around $20bn. It gives Peter Thiel, the co-founder of PayPal and the first real investor in the company, a return of $2.2bn on his investment of $500,000. Bono, the lead singer of U2, whose investment company stuck $120 million of secondary funding into the site less than two years ago, now has a stake worth not far short of $2bn, instantly and easily outstripping all the money he has ever made from music, and making him much richer than any other musician ever. More than 1000 people who work for Facebook are now millionaires.

Unless you happened to win the £50m EuroMillions lottery last night, you might be forgiven, in the face of such preposterous figures, for staring at the NASDAQ tickertape and willing the share price to fall. On the face of it, the valuation is madness. Facebook brought in $3.75bn in revenue last year, of which (estimates vary) between $500m and $1bn was profit.

You wouldn't keep your money in a bank deposit account that paid 1% interest. Any sensible investor wants a return of at least 5-10% on stocks and shares – perhaps more in Facebook's shares, since Zuckerberg and his associates control 57% of the company: what is being traded is not usable voting stock.

But then Google, now priced at $200bn, and making much more credible returns (being valued at six times sales to Facebook's 25 times sales) went public in 2004 at a valuation of $23bn, which was then seen as highly fanciful – and Facebook is a much more established brand than Google was then. It's the second most visited site (after Google) amongst UK users, and almost one in seven of the planet's population has an account.

Many of those who are sceptical of the digital economy make two fundamental errors. The first is to imagine that the whole thing is like the Darien scheme, in which all the money invested will be lost. A much better analogy is railway stock in the mid-19th century, in which lots of people will lose money by investing in the wrong companies, but those who bet right will make a great deal.

The second is to imagine that the internet increases profits. It doesn't in general terms (though, as Amazon demonstrates, it can, by increasing supply efficiencies and maximising buying power).

It is much more nakedly capitalist than that. The internet destroys existing profit models, and simultaneously offers the most open and competitive playing field there has ever been for traders to enter markets.

As canny a media operator as Rupert Murdoch learnt this to his cost when he bought MySpace, then the most successful site on the internet, for $580m in 2005.

Five years later, it was losing $180m a year, and he sold it for $35m. Its users simply moved elsewhere – many of them to Facebook.

Yet that is why, though I may well be proved wrong, I tend to think that this particular IPO is overvalued. Facebook is, like MySpace, a platform for content generated by its users. That is not the same as (say) a newspaper or an online news platform, which presents what the digital world calls "well-curated" content. And entertaining user-generated content has, for the most part, already migrated to Twitter.

Nor is it (say) a search engine such as Google, which most people regard as their essential gateway to the rest of the digital world and that can – to use the horrible technology venture capital phrase – "monetise" such activity.

Facebook's problem is that the more it attempts to make money from those who visit it, the more they are likely to resent the experience. This applies in spades to the really parasitic companies such as Zynga, whose games, such as Farmville and Mafia Wars, are one of the primary drivers of Facebook's traffic.

No-one is going to visit Facebook in order to click on adverts; the figures suggest that few of those visiting bother to even when tempted, which may be why General Motors has just decided it's not worth advertising there any more.

This leaves Facebook with two assets; one for it, one for the users. For the company, it is a mine of personal data to be sold to marketers and advertisers. But the more it uses this facility, the more its users will resent it and, potentially, desert it.

For the users, it's a helpful address book and storage for pictures and links. The problem is that that information could migrate almost anywhere else immediately. And if the future's the Cloud, Facebook's valuation may be Cloud-Cuckoo Land.