« links for 2006-03-13 | Home | Google Buys SketchUp; Google To Map The World in 3D? »
Muni WiFi … A Fairy Tale
By Larry Hendrick | March 13, 2006
This story on Muni WiFi continues to fascinate me because of the different discussions in the online community. At Dwight Silverman’s Houston Chronicle blog, TechBlog, a love fest is taking place over the thought of wireless Internet everywhere in Houston and cheap to boot.
In fairness, let me disclose that I work for an Internet Service Provider and we currently sell DSL in LATA 560, Verizon territory. Without disclosing proprietary information, I want to bring a sense of reality into this discussion … from the real cost side, but first, read this article about the city’s plan to determine the wholesale rate.
Chron.com | Project could help Houston bridge digital divide
The idea of a municipal WiFi is not an easy concept for non-techies to grasp.
The wireless builder would create and maintain the network, while an Internet Service Provider charges consumers to use it. The city would regulate the wholesale rate, which the ISP pays the builder.
The real cost of providing the service seems to have escaped the city. They seem to think that the price is whatever they say, not what it really costs to deliver the service. With city services, the price has never been determined the same way a business is required to, but that doesn’t seem to deter the city. Also, the City of Houston expects the winning bidder to provide the city with free access to the network, a cost that currently runs the city about $2 million according to another Chronicle article. The one customer that could make this whole project fly, wants to get it for free.
Who, in reality, will pay for the cost of this service?
- equipment build-out cost
- incoming bandwidth costs
- outgoing bandwidth costs
- maintenance (24/7)
- help desk (24/7)
- equipment replacement amortization
- backend fixed costs (rent, electricity, telephones, alarm system, etc.)
- backend support (invoicing, collections, FBI requisition handling, etc.)
Let’s look at each of these, and see how a real business must perform in order to deliver a service.
The projected cost of the equipment and build-out for downtown only, is $1 million dollars according to the referenced article. This is an estimate by a government agency, so we know how accurate this figure is in the real construction phase. The real cost will probably be in the neighborhood of $3 million dollars with a total build out for all of Houston in the $30 million dollar range. Now keep in mind, this is just for the equipment and implementation expenses. And just to clarify, you can not build a wireless network to cover any part of Houston with $50 wireless access points from CompUSA. It requires enterprise equipment that runs in the hundreds (up to one thousand dollars) at each location and will serve an area in the range of 100 meters. You read that right, 100 meters of coverage for one thousand dollars of equipment, not yet installed and so far, no connectivity. Currently all enterprise WiFi equipment runs 802.11b, so there will be a limit on the speed and number of people that can access at one time. This tends to limit the revenue numbers if they are kept real.
Incoming bandwidth has a cost that is cheaper in downtown Houston, than where I am 25 miles from downtown. There are two costs associated with incoming bandwidth, access and transport. Access is the price per mbs (megabits per second) of bandwidth that is purchased, and the price range is wide. I have seen this as low as $50/mbs in downtown, but our best price is more than eight times than number 25 miles south of Houston. Transport is the price it costs to get the access to your NOC (network operation center), and this is dictated by how far you are from the downtown POP (point of presence), owned by the big telcos (RBOCs). This price is cheap, there again, in downtown Houston, but increases as you move out, until finally, where we are, this price is as much as the access.
Outgoing bandwidth is the cost of getting the connectivity to the wireless access points. In our case with DSL, Verizon will bring nothing smaller than an OC3 (155mbs fiber) to the prim and will turn up nothing smaller than a DS3 (45mbs), This cost is greater than the incoming cost, but is our only way to tie into the Verizon system. This expense is required to get connections to a lot of the access points, because a wireless mesh network (such as the one proposed) has yet to work in an environment like the City of Houston. Many have tried, but as far as I know, all have failed. Buildings are the main culprit that can reduce the 100 meters to measurements much shorter.
Maintenance is a large cost for providing this type of service, because the equipment will be exposed to the elements and will require frequent repairs and replacement. In a five year period, the maintenance (with replacement added in) cost will exceed the original implementation cost, and I see no way around this.
One of the largest expenses we have as an ISP is the help desk. It requires multiple people to be available 24/7 for anyone that has a problem. It doesn’t even matter that 95% of the problems are either in the user’s settings or user inflicted, you have to take the telephone calls and work with them until there is a resolution. This number grows as the number of subscribers increases.
The equipment replacement and amortization cost is already addressed in enough details for this article two paragraphs above.
The backend costs are fixed cost, but for a real business plan, this must be included in the cost of delivering services. This price can be high or low, depending on the resources and location of the offices, but can not be ignored in determining a cost model that is reasonable.
Last, but certainly not least, is the backend support where all the money stuff takes place. The easiest way to handle this is to require a credit card from every subscriber that connects, and pre-bill them each month. However, if you only provide the service with credit card, you defeat the original purpose of the project … to provide service to the under-served. You see, from my experience at an ISP, the majority of the under-served don’t have credit cards, so this can’t be required for this project, which means that your staff of bill collectors will eclipse your sales force.
This last point is probably the most important with this project, because the city has expressly made it known that this project is to “bridge the digital divide” and I don’t even see this happening with this proposal.
In the referenced article, the city has stated that they intend to cap the price at the $15-20/month range and I am here to tell you that the true cost will be much higher. There is no way around it if you look at the true cost to provide this service, and not the fairy tale the city is proposing.
Topics: Business, Technology, WiFi |
April 11th, 2006 at 12:20 pm
City-wide wifi is good - but good for all of us?…
Tonight is another technology geek / non-profit group meetup with the Houston NetSquared group. We’ll be meeting at the Stag’s Head Pub ( 2128 Portsmouth St.) at 7pm for drinks, munchies and discussion about Web 2.0 technologies and their……
April 19th, 2006 at 9:34 pm
Bridging the digital divide - the series…
Earlier this week, the Chronicle’s Alexis Grant provided the latest installment of the newspaper’s series, Bridging The Digital Divide, Downtown WiFi Bubble Style:
Cyberspace may be a freewheeling……
April 19th, 2006 at 9:47 pm
For another insider look at the WiFi project check out Michael Garfield’s blog. The High-Tech Texan on radio and TV. He’s been a proponent of a citywide network for a while IF the right business is secured.
Looks like he has some insight into city hall according to this latest blog.
http://hightechtexan.blogspot.com/
April 20th, 2006 at 5:52 am
Thanks for the tip, Dustin. I will go check his blog out and see what his take is.
April 20th, 2006 at 1:04 pm
I suggest you look at 802.16 (WiMax Broadband for fixed stations at 30 miles and 3 - 10 miles for mobile stations). Makes a big difference in equipment costs.
I agree that the city is just making WAGS on the costs.
The city already collects monthly bills from citizens without credit cards - utility bills.
April 20th, 2006 at 3:53 pm
Charles,
Thanks for the comment and participating in the discussion.
I agree that WiMax might be a better solution, but as I have written as recently as Tuesday, there is not any WiMax equipment available in the USA. Two companies have approval in Europe, but the frequencies are not approved in the US, so it will not translate to here. No other companies have even presented equipment for review to the FCC, so that boat is still a few years out and by then 802.16 will be replaced by several other standards that are in the approval process. In the sales game 802.16 is called vaporware because having a standard and having a product that ships and works is a different beast.
You are correct on the utility bills, but if they don’t pay the water bill, they lose water, if they don’t pay the Internet bill, oh well … I’m just glad it wouldn’t be me having to collect the extra money. I wonder what the payment schedule for the really poor is, what ratio pay on time and what it is for those that are 90 days or more behind. That might be good information in this game.
May 15th, 2006 at 3:58 pm
[...] Business Unusual » My Clarified Stand on Free WiFi: [...] June 12, 2005, I wrote an… Muni WiFi … A Fairy Tale: [...]
August 7th, 2006 at 9:46 am
[...] Larry Hendrick criticizes the City of Houston’s municipal wifi project as a boondoggle. I’ll say more about this later. [...]
December 15th, 2006 at 10:33 pm
There is no way around it if you look at the true cost to provide this service, and not the fairy tale the city is proposing.
December 15th, 2006 at 10:59 pm
There is no way seems around it if you look at the true cost to provide this service, and not the fairy tale the city is proposing.
May 17th, 2007 at 7:28 am
[...] Muni WiFi—A Fairy Tale [...]
August 15th, 2007 at 3:37 pm
[...] an article I wrote in March, 2006, I outlined the costs associated with deploying this type network, and showed that [...]
September 21st, 2007 at 4:05 pm
[...] addressed the topic of money (business model) and Muni WiFi in March of 2006 with an article titled Muni WiFi … A Fairy Tale. This article presented the case for the pricing structure for Internet services, and was a result [...]