Commodity markets are large markets

I'm running a bit behind, so I only today read Martin Geddes' contrary opinion, commenting (last week) on a famous paper on Internet economics, the Paradox of the Best Network. 

Thank you Martin!  I too had problems with this paper when I first read it in 2002, but I've never had occasion to write up my thoughts on the matter (until now). I agree with several of Martin's points, but there's a broader issue.

The Paradox paper says the best network is a bit-moving commodity and you can't make money running such a network.

That's a strange assumption.  It arises frequently in high tech but that doesn't make it true.  Actually, commodity markets are usually large markets and those who become efficient producers can make a lot of money.  Oil is a commodity.  There is no shortage of people prepared to invest money in the oil industry. And, high tech examples abound.  Don't tell Michael Dell there's no return on investment in commodity markets.

The biggest difference between oil and high tech commodities is the rate of change.  You have to innovate at a rapid rate if you are to remain the most efficient producer in a high tech commodity market. So there's a lot of money to be made and a lot of risk, but hey, that's true of high tech, commodity or otherwise.

Comments

[David Weinberger]

"The Paradox paper says the best network is a bit-moving commodity and you can't make money running such a network."

Actually, our paper says the first of those independent clauses but not the second. There's good money to be made selling commodities. But it's not the sort of money the incumbent telcos are used to, nor does it suit their business model. Put differently, if you were planning on building a business selling bandwidth as a commodity, you would never ever ever structure it the way the traditional telcos are structured.

I'm pretty sure we're in agreement, Brough. Aren't we?

[Martin Geddes]

I've been meaning to write another article for a long time on how there's no such thing as a commodity. You might equally burn Arabian as North Sea oil, and they might come at the same nominal price, but clearly there's a political risk with one that would influence your buying decision. Or you might want to vary your order by a few million barrels, and only the Saudis can deliver that felxibility. Or you need delivery in specific quantities/dates, and there isn't a tanker coming from Arabia on those dates, so you have to buy Venezuelan oil. Or you need a particular sulphur content which only the Nigerians can supply. Etc.

Same arguments for bit transport. It isn't an undifferentiated commodity -- unless you only have one-dimensional vision.