There is much ongoing discussion of municipally-owned broadband projects, usually portrayed in this manner, as a battle between public-minded, well-intentioned politicians and greedy private firms who want to keep the forces of light from fulfilling the city's broadband dreams. Nevermind that good intentions are rarely sufficient basis for public expenditure. Despite utopian promises of economic development premised on building a a broadband network, this does not account for why private firms aren't doing it if this is the case. (See Laissez le Fiber Roulez)
"This is just like the government building sidewalks or roads," is one supporting analogy that is often used by municipal broadband proponents. The USA Today editorial approvingly quotes the City Parish Manager in Lafayette: "Installing fiber-optic cable, he credibly argues, is no different from laying down sidewalks or sewer lines." Unwittingly or not, the Manager is making what is called a "public goods" argument -- that a city or municipality needs to build this because it is a public goods.
The arguments get more than a little involved, but the chief characteristics of a public good are non-rivalrous consumption and non-excludability. Thus a sidewalk is a good example of a public good -- one person can use it (for the most part) without interfering with others' use and you can't exclude others from its use. Because it's a public good, private markets will not supply it because there is an externality and a free-rider problem; that is, I will be able to use the sidewalk without paying for it. [But see, The Lighthouse in Economics, (discussed here) Ronald Coase's debunking of a classic public goods example that shows private markets come up with ingenious ways to solve public goods problems.] [Another interesting sidelight is that as consumption of roads becomes more rivalrous (in terms of congestion), that is becomes more acceptable to talk about private toll roads as a means to alleviate congestion. This also indicates, as Buchanan discusses, that nothing is a completely private or public good.]
What is striking about municipal broadband networks is that they have few, if any, characteristics of a public good. Broadband connections involve both rivalrous consumption and excludability (indeed, excludability is a rather important consumer demand in the guise of privacy concerns). Therefore, they are only public goods under the common parlance that some people might think they'd be good for the public. What these networks are, from an economic perspective, is a private good. And with private goods, we rely for the most part on private markets, absent market failure.
In other words, this is a meandering way of saying, the proponents will need to come up with a better analogy because the sidewalk and roads one doesn't work.
Finally, there is a curious confluence between the rhetoric of municipal broadband proponents (faster, cheaper, now!) and the defenders of the universal service status quo: these services are necessary, must be provided universally to all, and at a low rate. Watch then for the deal to be cut: In exchange for no more municipal broadband systems, roll universal service subsidies to broadband networks, impose service territory buildout requirements, underprice the service to achieve greater penetration, and, voila!, full blown public utility regulation for broadband. Maybe I am a pessimist...