The following text is copyright 1997 by Network World, permission is hearby given for reproduction, as long as attribution is given and this notice is included.

Virtual reality and the cost of the Internet

The August 1997 issue of the Communications of the ACM has an article on Internet pricing that I don't understand at all. It could be that there has been a time warp and an article submitted for publication 5 or 10 years ago just popped up or it may be that the article is in the spirit of the most of the rest of the issue, which is about virtual reality.

The article, titled "Internet Pricing Vs. Reality" starts out on a tangent to the reality that the title espouses. The second sentence is "Should it [the Internet] be "free", as many computer enthusiasts seem to expect?" I know a lot of "computer enthusiasts" but I don't think I know anyone who thinks that the Internet should be "free" any more than people think that phone service should be free.

The author takes quite a bit of time in the article to try and figure out just what it should cost to send one bit of data over the Internet. He uses a number of pieces of information to make a number of cost estimates. He comes up with a range of cost estimates extending from a low of 0.07 microcents to a high of 1.6 microcents per bit. Even if one could agree that the actual basis for each estimate was reasonable (I don't), the range of estimates would seem to make the results somewhat hard to use.

The author does not seem to know all that much about the Internet itself, claiming at one point that "the Internet is composed of DS 3 transmission lines" as if the Internet were some one "thing". In actuality the Internet is composed of thousands of individual data networks, with backbone links ranging in speed from 9.6Kbps in some parts of the world to 622 Mbps in other places. Each of these networks also has its own costs structures, making any estimate of the actual cost of transporting data on the network quite chancy indeed.

But my biggest problem with the article is not what I see as a poor understanding of the Internet and its technology but with the conclusion the author comes to. In the last part of the article the author says that the government should not be "subsidizing and supporting a public data network". It is very hard to see where this comes from. Years ago the US government provided direct subsidizes in support of the regional and some national data networks but these are long gone. Today the National Science Foundation (NSF) gives some initial funding to help some research universities connect to a very high performance backbone to do pre-competitive research in the applications that will be running on the Internet of tomorrow. (note though that the universities must still buy their normal Internet connections with their own money) NSF and other government agencies also support research into advanced networking technologies. I do not see these activities as subsidies for the Internet any more than support for research in genetics is a subsidy for drug companies. At best the article is a time warp but it is more likely about some other reality.

disclaimer: Even if Harvard is its own reality, as some claim, these words reflect my own opinion on the common reality.