Free, Slow, Censored Internet – A Bad Idea

The FCC is looking for an organization to provide free, slow, and censored Internet access. The censorship apparently would include email as well as websites.  According to an article in today's Wall Street Journal:  "Outgoing Federal Communications Commission Chairman Kevin Martin is pushing for action in December on a plan to offer free, pornography-free wireless Internet service to all Americans, despite objections from the wireless industry and some consumer groups [nb. and from me]… The winning bidder would be required to set aside a quarter of the airwaves for a free Internet service[ nb. the WSJ hasn't got that part quite right. More below]."

Here's what the FCC's NPRM (Notice of Proposed Rule Making) says:

"(a)  The licensee of the 2155-2188 MH band (AWS-3 licensee) must provide as part of its free broadband service a network-based mechanism:

"(1) That filters or blocks images and text that constitute obscenity or pornography and, in context, as measured by contemporary community standards and existing law,  any images or text that otherwise would be harmful to teens and adolescents.  For purposes of this rule, teens and adolescents are children 5 through 17 years of age;

"(2) That must be active at all times on any type [emphasis added] of free broadband service offered to customers or consumers through an AWS-3 network."

The NPRM is silent on which community's standards should be used to determine what is harmful to "teens and adolescents"   (I'm glad my children weren't adolescents at five). But note that all services must be filtered for whatever it is that "otherwise[?] would be harmful". The FCC will be the ultimate judge, however, since they enforce the terms of a license and can determine whether or not to renew. They may think my blog is harmful; I think junk science is really dangerous; you may have still some other pet peeve.

This NPRM is a reworking of a proposal by venture-backed and politically-connected M2Z networks, which was setup to provide this free (actually ad-supported) service in return for the right to also use the spectrum for subscription service and may well be the only bidder if the auction goes forward on these terms.

Let's forget for a minute what an absolutely terrible idea it is for the government or its agent to determine what the content of our communication is allowed to be although this whole proposal SHOULD be junked just because of this. The principle DOES matter a lot but, in practice, no one is going to use the free Internet service.

The requirement is that the free service have a download speed of at least 768kilobits/second; this is the speed of the slowest DSL. This is not a mobile service where we might accept a slightly slow speed in return for mobility; it is meant to be primary residential access to bridge the digital divide. But, by the time this service actually exists, the web will not be usable at such a slow speed. Websites get designed for the capabilities of the top 50% of users; that's why dialup is now useless for surfing even though it used to work fine.

By contrast, the Vermont legislature specified that a service won't count as broadband at the end of 2010 unless it is at least 1.5Megabits/second AND realizing that this requirement must escalate, charged the Public Service Board with appropriate and timely upper revision. Nevertheless, this requirement stays the same in the NPRM for the full ten years of the license. 768Kbs will be as obsolete as your old 300 baud modem ten years from now; but it will be the only free service the licensee is required to offer.

The licensee has no incentive to offer better free service because it will also sell a higher-powered service. The NPRM does NOT specify that a quarter of the network capacity be used for the free service; it says "up to" an "as needed". If no one uses the free service, then no network capacity need be devoted to it.

It's nice to think that people who have no service will get it from M2Z; don't hold your breath. The licensee is required to build out to 50% of the population within four years and 95% within ten. The licensee will, of course, build first where the people are and where there are plenty of other service options. The areas without adequate Internet access now might get it in ten years – and might not. Even the poor in the urban service areas won't be helped much. They can't use the service without a computer and a modem which M2Z estimates will initially cost $200. In urban areas DSL at this pathetically low speed is usually available for $12.95/month or less and without a requirement to purchase equipment.

There's another interesting gotcha in the proposed regs: the paid service must be open to all devices and applications which is good. The free service, however, has no requirement that it be open to all applications. In theory the provider could charge Google, for example, to be accessible to the free users – assuming there are some which I don't think'll be the case.

It's good that the FCC recognizes that previous spectrum auctions haven't gotten the US to the point where it has the connectivity it needs. There's nothing wrong with the concept of a "free" Internet service paid for by advertisers; in fact, I'm confident that such a service will be one of the offers that surface in the newly deregulated white spaces. The  spectrum at issue is unused and that's a waste. This proposal, however, should be rejected once on civil liberties grounds and again because the promise of free Internet access is meaningless and unhelpful.