On Mon, Jul 03, 2006 at 11:29:19PM -0400, Bob Keyes mangled the electrons to say: > > While we're here, let me talk about some muni wifi things that I've > figured out. > > While this post is long, please don't consider it a finished document. > There's some wordsmithing, and documentation of some of the claims I make, > which still needs to be done. This will happen. Eventually. > > Firstly, there is a big difference in service levels between different > technology. Fiber is the most reliable, followed by coax, followed by > 4-wire twisted pair circuits (T1), followed by DSL, followed by dialup, > and WiFi brings up the rear ( I leave out cups and strings and its more > problematic cousin BPL - Broadband over Power Lines). Consumers expect to > pay more money for more reliable services, and LESS for LESS reliable > services. But of course speed is an issue, and also availability. Many > parts of the country are stuck with dial-up right now, which is more > reliable than WiFi but also damned SLOW.
You left out a major, but often silent player in the wireless space - Licensed microwave. Properly engineered links can deliver fiber reliability at T1 prices. Also, properly constructed licensed systems in the 2.5G space can deliver T1 reliability at cable modem prices. David Beery > > The complaints about WiFi relate to municipal/suburban systems, where > there is a competing use of the 2.4 ghz band. Rural WiFi may work better > because of less interference, but they also have greater distances to > cover. > > As long as WiFi is stuck on the part 15 spectrum (2.4 ghz) there will > always be the problem of interference with other part 15 users and the > primary users of the space (Amateur radio, for example). While the U-NII 5 > ghz band is a lot less crowded, it has problems related to its > considerably higher frequency: namely, it is more dependany on > line-of-sight to have good coverage. But I am going to ignore 802.11A for > now. More on that later. > > So, wireless isn't that reliable. This isn't a huge problem for many > people, if the service were CHEAP. But the $21.95 Earthlink is charging is > just too much money, when there are advertised $15 rates for DSL from > Verizon and the like. It sounds to me like anyone who is in a service area > for for Verizon DSL would be more likely to go that route. > > So what if Earthlink dropped their prices? What level would be cheap > enough? Perhaps $10 a month. But this might be too little for them to pay > for the infrastructure they've built. Regardless if it is, or isn't > economically viable to charge $10, there is the problem of customer > expectations. Once a customer spends any money whatsoever, they have > expectations that they will be able to call someone and get any problems > they have fixed. Or at least they'll get put on hold for a while, and the > bureaucracy will eventually creak along and fix the problem. This is the > same, if someone spends $1 or $21.95 or $100. So, there's the expectation > of some level of support, but this becomes particularly difficult when the > system isn't particularly reliable. And, as mentioned before, WiFi is NOT > particularly reliable. So, there are more disgrunteled customers sitting > on hold, requiring more customer service respresentatives to tell them > that the problem is being examined. Or to tell them to get close to a > window and move around till the signal gets better. > > In addition, there is the increased percentage of the fee going to > transaction costs, with a decrease in price. What I mean is that certain > transaction costs are fixed. For instance, credit card processors > typically charge a transaction fee of something like fifty cents, plus a > 2% commission. When you're a merchant involved in large transactions > (diamonds, a full tank of gas for a SUV) you are concerned more with the > percentage, when you are a penny retailer (linux CDs on ebay), the fixed > transaction fee is more important. WISPs start becoming more > concerned with the transaction cost as their monthly rate is forced down. > Besides credit card rates, there are issues like postage and paper costs > when sending bills, labor costs for processing payments, etc. > > What if you were able to eliminate expectations and overhead and just make > the service FREE and get money in another manner? This is the attraction > of advertising supported Internet access. It's pretty tempting, but the > returns from said advertising do not currently justify the cost of wifi > infrastructure. This MAY be because many of the WiFi equipment vendors are > still trying to market towards traditional WISPs that are still trying to > compete with wired ISPs in regards to reliability. > > Another potential funding source is public coffers. Many people object to > this usage of public funds, but we're already funding public radio & TV, > newspapers, bulletin boards, etc. In addition, by providing a public > Internet service, other problems are addressed. One of these is the > infamous issue of the Digital Divide, i.e. the tendancy of the lower > income people to not have broadband Internet access, therefore much less > education in computers & technology, and therefore less job opportunities. > a Public Internet would obviously give these people much more access to > the Internet. Another cost center that could be addressed, is the cost of > public information dispersal: If a Public Internet became available, it > could replace other, less efficient modes of official communication (for > instance, the 'official notices' section of the newspaper), saving time > and money for a municipality. > > So perhaps what is needed is a FREE, PERVASIVE Internet that differs from > the wired providers in its level of reliability and support. It would be > the equivalent to Public Transit, in that it's available but if you can > afford it or have a special need that makes it worthwhile, you might want > to just pay for a commercial service (buy a car, or take a taxi, or pay > that Comcast fee). > > This is only an outline of my thoughts. Yes, I am writing a book ;) And > yes, we are building something like this. But not in New York. It's in > Cambridge, Mass. It's called Cambridge Public Internet. > -- > NYCwireless - http://www.nycwireless.net/ > Un/Subscribe: http://lists.nycwireless.net/mailman/listinfo/nycwireless/ > Archives: http://lists.nycwireless.net/pipermail/nycwireless/ -- NYCwireless - http://www.nycwireless.net/ Un/Subscribe: http://lists.nycwireless.net/mailman/listinfo/nycwireless/ Archives: http://lists.nycwireless.net/pipermail/nycwireless/
