Way back in September of 1997 , I crossed paths with George Gilder . He was speaking at a conference run by my former employer at a swank resort hotel in Orange County, in front of hordes of corporate execs.

I was not impressed. At the time, I knew of him tangentially, and had already classified him with all the other crackpot futurists that lived off the largess of the corporations they did consultancy performance art for. And I said so:

Futurists have a great line of work. They get paid to make broad generalizations, wild assumptions, and baseless conclusions. And all they have to do is be right about 20% of the time to be taken seriously…futurists outsource reality to someone else.

George had gotten up my ire by talking about IT outsourcing–and I still think I was right and he was wrong on that subject, but that's academic at this point. Apparently, George took himself a little too seriously; in the wake of the collapse of his beloved “telecosmic” industry, George Gilder is now bankrupt.

Gilder's fatal flaw was his undying telecom boosterism. Doing a riff on Moore's Law, he made up his own eponymous theorem that predicted bandwidth would double every six months. John Robb thinks Gilder was right about technology, but wrong about the timeframe:

“The problem he didn't anticipate? That the baby bells and the cable companies wouldn't install fiber in the last mile…The decision of these last mile players not to adopt the right technology was a monsterous blunder based on greed, a lack of vision, and a cozy oligopoly. ”

And John's right. That's part of the story, at least. But wait–isn't greed supposed to be good? Shouldn't greed have made Gilder's dreams of “infinite bandwidth” come true?

Perhaps in a perfect, anarcho-capitalist wet dream of a world it would. But George forgot one little factor: government corruption.The FCC has all but abandoned its role as the representative of the interest of citizens , and has instead become beholden to the corporate interests of a few large companies. It has become, essentially, the digital convergence kingmaker.

So, as a result, the big communications/media companies can ignore the unwired masses because :

1. There's no universal access law governing broadband–and there probably never will be.

2. Poor people with Internet access are not high-margin customers.

3. Only the demographically blessed, falling within well-defined wealthy areas where advertisers and retailers can cherry-pick customers, are worth the time to invest last-mile infrastructure in. (Baltimore City will *never* have fiber-optic broadband, because the cable company will never invest in replacing coax with fiber to the curb. The demographics just aren't profitable enough.)

In the end, everybody else loses. But George, the undying optimist, could never have predicted this nightmare scenario–only a cynic would have thought that the entrenched cable and local phone companies would use government regulation to crush George Gilder's dreams.

And only the most bitter cynic likes to be proven right.