Report from IPI not surprisingly gets its facts all wrong: It's been a long, long time since I had to apply the "sock puppet" tag to a post, but it's been a while since telecom-funded thinktanks that don't disclose their funding have written reports that allege to explain why municipal ownership of telecom services are always mismanaged. I came to be a fan of public/private operations, in which cities and towns figured out their needs, and worked with private contractors who would own and operate networks that would serve many purposes (public, academic, digital divide, public safety, and municipal). Networks built along these lines have been completed and get good marks; those that were built entirely by municipalities or for entirely public access purposes seem to have faded away, except for a few restricted to relatively small towns.
The latest report, by Barry Aarons, is called "We Told You So! Continue to Say 'No' to Municipal Broadband Networks." I suppose his next report will be title, "Cars Are So Much Better Than Horses," and "DC Power: Work of Satan."
In any case, Aarons, formerly associated with major telecoms and who works as industry consultant, appears to be trying to forestall putting stimulus broadband dollars into municipal hands. I tend to agree: I'd rather see non-profits and local telecom groups use existing expertise and knowledge of underserved audiences to built out infrastructure. Cities, towns, and counties likely have a role in establishing and leasing rights of way and facilitating access for others putting services in.
However, I have to take issue with the facts. There are essentially no municipal Wi-Fi networks of the type that Aarons wants to use as a strawman. Over and over, this report cites private efforts, and misstates facts.
Page 1: "For example, in Tempe, Arizona there were three times as many antennas required at a cost of over $1 million or twice the original cost." True, but that network was built by what became Kite Networks, and which wasn't well designed. The city didn't buy any services from the network, but was supposed to receive free roaming accounts in exchange for the quasi-franchise.
Page 2: "And now that analog television broadcasting has been eliminated it is likely that portions of that spectrum may become available for expanded wireless competition." That spectrum was already auctioned off for $20 billion to AT&T, Verizon, and others.
"...some communities’ municipal wireless projects are, in fact, alive and well. And there still appears to be an appetite for such programs as evidenced by the estimated $900 million invested to this point." Invested, almost entirely, by private entities.
On Philadelphia: "...city officials thought they could get existing companies to let them use refurbished gear and could build the entire project with 'non-city' financial resources." In the very early stages of the plan, there was no mention of refurbished gear; "non-city" would mean that it wouldn't be paid for by the city?
Subsequent paragraphs cite the failure of the Philadelphia network, but that was built entirely with private dollars. The city put in some funds for initial studies and such, but the money spent on the network was from EarthLink. Further, Aarons writes that EarthLink "closed down this project on June 12, 2008." EarthLink exited the effort, but the network is still running, with a private firm having assumed the assets and operation.
Page 3: "Philadelphia’s experience was considered the flagship of government projects covering huge amounts of area with a system that was considered in 2005 to be the cutting edge." Well...no. It was the flagship of privately funded metro-scale networks that cities requested would be built with no public money, ownership, nor control. And because the network started being built in 2006, it's not that odd that it used 2005 technology.
On Portland: "And then there is Portland, Oregon, a system that crashed and burned from the start. The city hired a start up company to construct and install its municipal Wi-Fi system." Bzzzt! Sorry! MetroFi paid all its own costs. The city spent a tiny, tiny sum in services, never contracting for its offerings, even. (Had it, Portland would have been using MetroFi as a service provider, much like a telecom.)
"So MetroFi is in default and millions of dollars are yet to be spent to finish a system that is at best 20 to 30 percent completed. The probability is that the project will not be completed at all." Did Aarons write this report in mid-2008? The footnotes aren't from any later than May 2008. The network will never be finished and no one is interested in that.
"So what in Portland went wrong? MetroFi found that municipal government ultimately was unwilling to provide the subsidy that would be necessary to support the system." Uh...wait...so...doesn't that indicate that a privately funded network that attempted and failed to seek public service contracts can't succeed?
On Ashland, Ore.: Ashland isn't Wi-Fi. And it's only mentioned here because it's an ongoing strawman for thinktanks. Note that the figures here date back to 2001: "Begun in 2000....after only one year." Aarons may be using a 2001 report, that I found largely inaccurate, to pull this example in.
On Lompoc, Calif.: This small-town network had a lot of problems, and, in fact, was the poster child of bad network planning and spending. However, it's a) small time and b) nearly sui generis. Only Saint Paul Park, Minn., a similarly small town, had an equally bad blowout after spending city money for a municipally owned network. I don't know of a similar third example in which public funds were used. (Hey, see this is how you write something that accepts contrary examples to one's thesis! What a concept.)
Page 4: On Orlando, Flor.: I don't know anybody who ever cited Orlando, which Aarons note had a short-lived and very inexpensive hotzone in 2004 to 2005.
I can't make heads or tails of this report follow-up. If you read this accurately, fixing Aarons's errors, he's saying that private companies attempting to build large-scale Wi-Fi networks all fail. In the intro, he writes, "We noted in particular that the expense to the cities and counties would likely make these government owned projects expensive failures." Yet the only example that involved significant city dollars is Lompoc; the rest were private efforts, with private risks (and losses).
Of course, Aarons doesn't point to San Francisco, where a private company, Meraki, is building a ground-up Wi-Fi network; to Minneapolis, where a private firm contracted for city services and built a network at its own expense that appears to have enough customers to make a go of it; at Cablevision's huge system, funded by private dollars for its own customers; and so forth.
I don't quite know how or why this "report" came into being, but it's specious from beginning to end. I would have enjoyed seeing a report that attempted to explain from the regulatory and competitive angle why EarthLink, MetroFi, and Kite failed in their efforts (among those of others), as it might teach something to future businesses launching efforts.
In a bit of not quite irony, Aarons praises WiMax in passing. But Clearwire's WiMax service (and all the independent WiMax companies that might come into being for niche markets) has the same constraints and properties as the private firms that failed to make Wi-Fi work on a large scale. WiMax has superior technical characteristics, but it's still a metro-scale network of a type that hasn't been built before.
With WiMax, however, cities didn't issue RFPs. Perhaps, in the ideological world in which these reports are written, that makes all the difference.
Aarons is like a cold warrior, long after the Soviet Union fell. (Perhaps a bad example given the current leadership in Russia.) Long ago, cities and private firms seemed to have mostly decided that cities won't build wireless networks. The battle was lost with private funds.
Sundance honors `Push,’ `We Live in Public’
(AP)Clearwire Opens Portland NetworkWee-Fi: Intel’s Wi-Fi PAN Released; Eye-Fi iPhone AppFlo Rida Kicks Some Eminem Butt
(E! Online)
Atom Feed (xml)