Response to Senate Questions on Internet2

NACSE - Northwest Alliance for Computation Science and Engineering
NERO - Network for Education and Research in Oregon

Response of
Cherri M. Pancake
Professor of Computer Science and Intel Faculty Fellow
Oregon State University
to
Questions of Senator Conrad Burns
Chairman
Subcommittee on Communications of
Committee on Commerce, Science, and Transportation
Hearing on Next Generation Internet
June 3, 1997
Questions for All Witnesses on Both Panels

1. As I understand it, the schools and libraries program, for all practical purposes, forces the telecommunications industry to pay $2.25 billion per year to fund the build-out of the Internet, which ultimately will lead to massive investment in computer hardware and software.

Do you see any danger in the telecommunications industry subsidizing the build-out of a network which could be used by the computer industry to compete directly with the telecommunications industry?
It's not clear that the subsidizing of Internet lines to schools and libraries will result in computer companies competing with the telecommunications industry. Computer hardware and software sales will certainly increase, but Internet access doesn't replace voice communications in any significant way; rather, it opens up a new set of possibilities for learning and access to remote resources. I suspect that what will actually happen is that increased Internet use at school will engender increased demand for Internet access from home - which should actually drive up the consumption of telecommunications services.
This almost seems backwards, since the computer industry has a much greater total market capitalization than the telecommunications industry, wouldn't you agree?
It's hard to compare market capitalization figures for two such different industries. Certainly, it's always tempting to make the beneficiary of new services help pay for those services. This isn't the normal model in communications, however. The manufacturers of telephones (the apparatus itself) certainly benefit when telephone lines are extended into remote areas, but we don't tax them or ask them to pay for the necessary infrastructure. Similarly, the manufacturers of televisions, VCRs, and so forth aren't expected to pay for new or improved broadcasting services. It is the service provider who is expected to pay for extending service infrastructure, just as the equipment provider pays for extending its own production and marketing infrastructure.
Moreover, the computer companies are not exactly getting a "free ride" as Internet access is extended to schools and libraries. Educational institutions will expect much of their new hardware and software to be donated by the manufacturers, or at least very deeply discounted. It simply doesn't make sense to expect the computer industry to subsize the line charges as well as deployment of the equipment.

2. Do you see an difference between sending a voice message over the public switched telecommunications network and an electronic message over the Internet?

No. In fact, we can send electronic messages over the public telephone system, and voice messages over the Internet (with appropriate conversions, of course).

As I understand it, access charges and universal service support are paid if the message is sent over the telephone line, but there are no such charges if the message is sent over the Internet. We seem to be setting up different regulatory schemes for these merging technologies, wouldn't you agree? Do you believe this disparity should continue?

This is just one example of many services that involve transmissions, and there are actually a a number of costing options. For example, it might be possible to receive transmissions free of charge, charged according to the amount of actual use, or charged with a recurring "flat" fee. The following table shows examples of communications systems in common use and the kinds of costing models typically available for each.

	Use based	Flat fee	No fee
Local phone calls	x	x
Long-distance phone calls	x
Cellular phone calls	x	x
Broadcast TV			x
Cable TV		x
Pay-per-view TV	x
Satellite TV		x	x
Internet member (e.g., universities)		x
Internet via access provider	x	x

Note that it is the communications infrastructure, not the transmission of bits or waves per se, that costs money. Charges based on actual use are simply a way of re-packaging these fixed costs to make them more marketable.

Charges are often structured into many re-packaging "tiers," with some providers wholesaling services to others who dispense in smaller and smaller units. When the end-user appears to pay nothing, or pays on a per-use basis, that only hides the fact that some entity is paying flat charges for the availability of communications infrastructure. That re-sale entity is acting on the assumption that while some individuals may impose high use, most individuals won't.

Even a university must pay hard cash for its Internet access. Oregon State University, for example, pays for its portion of telecommunications lines - which it shares with other Oregon universities in the NERO regional network - to connect to the Internet backbone. This is billed as a flat rate per month, amounting to $360,000 per year; an additional $48,000 per year is required to pay so-called "port" charges. Further, the universities much purchase, operate, and maintain a variety of routing and switching equipment. These are the very same types of costs that are rolled into the rate a consumer pays to purchase dial-up services from a network provider.

Regardless of the actually billing scheme, there are real costs associated with anyone's Internet access, just as there are real costs associated with a voice-based phone call. Similarly, there are many options for how those real costs are transmitted to the consumer or his/her organization. The structure isn't uniform for voice transmissions over phone lines, so it should be no surprise that it isn't uniform for electronic transmissions either.

This page is available on the Web at http://www.cs.orst.edu/~pancake/internet2/response.html

Related information:

http://www.cs.orst.edu/~pancake/internet2/internet2.html

Original Testimony of Dr. Pancake before the Subcommittee