MetroFi network has been turned off
Posted by Russell Senior on 1 July 2008Apparently sometime around 5pm this afternoon, the MetroFi network was turned off, sources say. For the last few weeks, the ads and splash page had been turned off, and today, a day later than they’d indicated to the press, the MetroFi-Free and MetroFi-Premium SSIDs apparently got switched off. We are going to make an brief foray this evening to confirm, but it looks like the MetroFi experiment in Portland is over.
Unwired PDX Watch seeks MetroFi insiders for explanation
Posted by Russell Senior on 25 June 2008For the last year or so, in furtherance of the educational mission of the Personal Telco Project as well as to satisfy our own curiosities, we have been seeking insights and explanations of some of the engineering decisions made by MetroFi. We requested speakers for the Personal Telco Project’s monthly meetings from MetroFi via Denise Graab. Regrettably, despite some promises, we never got any followup, perhaps because Denise left the company last autumn. We remain interested in understanding what went right, what went wrong, and the perspective of the network engineers at MetroFi. We had the good fortune to have an extended conversation with technical people from SkyPilot Networks shortly after our Proof of Concept study, and we would certainly welcome similar interactions with those at (or formerly at) MetroFi. If anyone is willing to share their experience and perspective, please get in touch with us and perhaps we can set something up. Thanks!
Our Methods to Appear at AccessNets
Posted by Caleb Phillips on 21 June 2008Research based on the performance and coverage testing methods developed for our testing of the MetroFi Proof-of-Concept network in Portland has been accepted for publication at AccessNets 2008 which will be held in Las Vegas, Nevada in October. The paper is titled “Robust Coverage and Performance Testing for Large-Area Wireless Networks” and is co-authored with professors Dirk Grunwald and Douglas Sicker at the University of Colorado at Boulder. We’ll post a copy of the final camera-ready manuscript here, as soon as it is ready (presumably sometime in July).
MetroFi schedules Portland shutoff for June 30
Posted by Russell Senior on 20 June 2008The Portland Business Journal, Associated Press and the Oregonian are reporting that, according to officials at the City of Portland, MetroFi plans to turn off their Portland network on June 30th. The equipment is to be removed during the month of July.
Meanwhile, the Personal Telco Project is reminding people that there are still an amazing number of wifi networks in Portland (about 100 times what MetroFi was ever going to deploy), and that ordinary citizens and businesses can help make ubiquitous free wifi a reality by participating in its provision.
MetroFi Network is for Sale
Posted by Caleb Phillips on 16 May 2008As reported on wifinetnews.com, muniwireless.com, and the oregonian blog, MetroFi is throwing in the towel and has suggested that the city of Portland purchase the network as is. There is an interesting discussion about this on the Personal Telco mailing list. Here is the letter sent to Logan Kleier from MetroFi president Chuck Haas:
May 13, 2008
Mr. Logan Kleier
City of Portland
1120 SW Fifth Ave, Suite 450
Portland, Oregon 97204
Logan,
I am writing today to request a call or meeting with you to discuss the City of Portland network. MetroFi's goal was to have the City of Portland network producing enough revenue to cover expenses and even with the roll out of Microsoft SideGuide, the best advertising platform we have for Wi-Fi revenue generation, we are still not covering our costs for network operation and maintenance. As I discussed on the phone, MetroFi has three options to consider. The first is for the City to purchase the network from MetroFi; the second is to sell the network to a 3rd party; the third is to shut the network down.
The City of Portland network is comprised of 598 access points and is used by approximately 16,000 residents and visitors per month. April had 306,000 hours of use. Our operating expenses include pole attachment fees and power, the BAP locations, backhaul transport via microwave the Pittock Building and Internet transit. These expenses are approximately $15,000 per month. To purchase the network, we would propose a price of $1,500 per AP or $894,000. MetroFi would train the City on the network operation and transfer assets, spare equipment and knowledge to the City. MetroFi is also discussing network purchase with other prospective buyers, but there is no assurance that we can complete a sale.
The third option is to remove the equipment, beginning in June 2008. I understand this is a lot to digest, which is why I have requested a meeting or call to allow us to discuss these options in more detail. We are looking to understand the city's position as soon as possible as our intent is to begin network shut-down and equipment removal the by the end of June if we do not have a buyer.
Please let me know a convenient time to meet and discuss these options with you at your first opportunity.
Sincerely,
Chuck Haas
President & CEO
Reporters sometimes get basic facts wrong
Posted by Russell Senior on 22 March 2008The New York Times is running a story that mentions the Portland wifi network operated by MetroFi in passing. Unless something happened recently without us knowing about it, the story gets things substantially wrong. Here’s what Ian Urbina wrote, in part:
In Tempe, Ariz., and Portland, Ore., for example, hundreds of subscribers have found themselves suddenly without service as providers have cut their losses and either abandoned their networks or stopped expanding capacity.
Urbina, or to be fair perhaps his editors under pressures of space, conflate the Tempe, Arizona and Portland networks into a single sentence and consequently propagate an inaccurate perception of what is happening in Portland. First of all, as we understand it, MetroFi Portland has very few paying subscribers. According to Logan Kleier as of a few months ago, certainly not hundreds. The vast majority of its users utilize the advertising-supported service. Second of all, MetroFi has not (again, unless something has happened very recently) abandoned their network, suddenly cutting off service. There are plenty of places in Portland without service, and without any prospect of service, and even places “with” service might not actually have service. But that is different.
The MetroFi Portland network has substantially stalled. The outward signs provide no obvious prospect it will ever be finished as originally envisioned. However, hundreds of subscribers in Portland have not suddenly lost service they had previously, as far as we can tell.
A quiet path to termination?
Posted by Russell Senior on 6 March 2008The Portland Mercury published a story recently that casts doubt on the longterm viability of the MetroFi network in Portland. The two notable quotes are:
“We’ve been trying to contact Haas, in order to redefine the relationship, but he’s been difficult to get a hold of.” — Logan Kleier
and
City Commissioner Dan Saltzman […] has said that MetroFi’s conduct has been “regrettable,” though he will allow the project “a quiet path to termination”
Interview of Chuck Haas on OPB this morning
Posted by Russell Senior on 21 February 2008We were alerted to an interview (unclipped version) with Chuck Haas (MetroFi CEO) that OPB ran this morning on its radio service. It was interesting to us in that we finally heard a (still incomplete) justification that indicates the basis of their ~30% coverage claim. He said it was based on a population weighting, claiming that 150,000 out of 540,000 people had access to the service. He didn’t get into any details of the assumptions that go into the 150,000 people number. We have previously estimated that on a spatial basis their existing network covers only about 4.2% of the 134 square-mile city’s outdoor areas to the “90% probability of a connection to a 30 mW client device” standard. Our estimate is based on our recent survey of the SkyPilots we did in December, and a 300-foot coverage radius that we found during our testing of the Proof-of-Concept network last spring. Logan Kleier has said that the City and MetroFi are using a 500-foot radius, but even assuming that, you only get about 11-15% of the area covered (depending on how many of the SkyPilots are actually functional). To get to 30% on a population basis, roughly speaking, you’d have to assume that the population density of the covered areas is twice as high as the city as a whole. We wonder if that’s true.
Also, as part of the interview Chuck was asked about SideGuide. He described it as “much less obtrusive”, while the interviewer expressed skepticism of that claim, comparing it to pop-ups. In response to a question about the privacy of surfing behavior, “data on the sites you are visiting are tracked … what is that information used for?”, Chuck ducked and pointed at Microsoft, which in turn was not specifically responsive to the question.
Portland Linux/Unix Group talk tonight
Posted by Russell Senior on 7 February 2008Russell Senior is going to give a talk to the General Monthly Meeting of the Portland Linux/Unix Group this evening about what he has learned in observing the MetroFi network in Portland over the last year, and how he learned it.
MEETING ANNOUNCEMENT
The Portland Linux/Unix Group
will meet
7 PM Thursday Feb 7, 2008
at
Portland State University
in
New Engineering Building
Room 325
The building is on SW 4th across from SW College Street.
See location H-10 on map at http://pdxLinux.org/campus_map.jpg
Here is a Google Street View shot of the building.
MetroFi Portland week-in-review
Posted by Russell Senior on 19 January 2008In the Willamette Week piece last week, Lucie Poulicakos (vice president of operations at MetroFi) said that corrections to the map could be expected in a few days. This week we drove back to the 36 locations that the MetroFi map showed were “In service” but that we had found were not. In our recheck we found that there were indeed some incremental improvements, and found evidence that at least 6 of the SkyPilots had been physically touched. Here’s what we found:
(Read on …)