Following up on my previous post "Institutional Networks and Cherry Picking," I want to briefly note that the U.S. should reform how it funds Internet connections at schools and libraries. Let me start with an assumption: we do not want to use federal taxes to support these local institutions except where most necessary. It strikes me that wherever possible, communities should take responsibility for their own community institutions. With that in mind, the eRate program concerns me. Basically, eRate is a means for the federal government to aid local schools and libraries in affording broadband. I'm afraid that it indirectly encourages monopolistic service providers (mainly telephone incumbents) to overcharge for T-1 lines while removing any incentive for the school or library to invest in a better connection. If a school or library is only paying 20% of the cost of a slow and overpriced line, it has considerably less motivation to seek a better connection -- especially as the only alternative to an existing connection may be building new fiber paths - as noted in "Libraries dying for bandwidth."

But another problem is simple availability. As the ALA's report (PDF) points out, "moving from a 56Kbps circuit to 1.5Mbps is one thing. Moving from 1.5Mbps to 20Mbps or to 100Mbps or even to a gigabit—depending on the size and need of the library—is another." Even when they can pay for it, many libraries are finding that higher speeds simply aren't available.

This program has been around since 1998 and has paid out $25 billion.