05.18.2016 by Andrew M. Seybold
No matter how many different financial models have been run on spreadsheets, Muni-Wi-Fi has, to my knowledge, never reached even breakeven anywhere—not from city investment, not from selling advertising, and not from any other model that has come and gone.
It’s Muni-Wi-Fi time again with two Wi-Fi ideas currently being floated. First, New York City wants to provide Gigabit Wi-Fi to all who live and work in the City. Second, the website “Open Wireless Movement” wants everyone in the United States with Internet access points to set up a guest port so anyone in the area can use it.
The City of New York has a long and sad history of trying to be a leader in providing free broadband services. Over the years there were to be pay phones converted to Wi-Fi access points (now back agaian) and many other attempts. Some believe NYC will make it this time because of the demand, but along with that demand come all of the laws of physics, the interference, and the congestion that will make it more difficult and more expensive for the city of reach its goal. As for Open Wireless Movement, it is an interesting concept but I can tell you that when you walk or drive past my home and office you will not find my Wi-Fi open for your use. I pay for it, and I pay for it to be mine all of the time, which means I won’t be sharing my bandwidth with people I don’t know.
Muni-Wi-Fi has failed ever since Metricom, founded in 1985, finally gave up the ghost in 2001 even after Microsoft’s co-founder Paul Allen purchased a controlling interest in the company. Next time you are in Phoenix, look at lampposts along almost any road and you will see what is left of Metricom — white antennas still proudly sticking up into the air. In 2005, EarthLink came on the scene and won Muni-Wi-Fi networks in Anaheim, CA, Philadelphia, and other cities. By 2008, EarthLink was ready to turn off both systems. Philadelphia purchased the network for $2 million and struggled for years to turn it into a network that could carry its own weight in expenses.
In cities in Silicon Valley, in San Francisco, Atlanta, and many other cities, Muni-Wi-Fi was the next big thing. Portland Oregon’s experience cost it a lot of money only to end up disassembling the network. During this time I wrote many articles about the pros and cons (mostly cons) of Muni-Wi-Fi stating the obvious: The lack of inbuilding coverage, the presence of network interference, and the lack of a sound business model.
After a while things went quiet in the Muni-Wi-Fi space but then grew hot again as round two started. In Silicon Valley, Google installed a free system with no support and no way to contact anyone. It was a free network but it only sort of worked. A utility company decided that when installing smart meters it could also install Muni-Wi-Fi and it would be paid for by some magical calculation.
That brings us to date with the exception of one or two places in the United States where Muni-Wi-Fi is alive and well and in some city parks and other confined areas where people gather. Otherwise it has been a dismal failure. No matter how many different financial models have been run on spreadsheets, Muni-Wi-Fi has, to my knowledge, never reached even breakeven anywhere—not from city investment, not from selling advertising, and not from any other model that has come and gone.
Now New York City and a few others are back. This time claims are that it will work because people have to pay network operators too much for broadband and much of that broadband is doled out in 2-GB or 3-GB monthly increments. The disadvantaged certainly deserve access to the Internet and according to the Open Wireless Movement website, open Wi-Fi access using your router and mine will spur innovation, is crucial to wireless privacy (?), benefits emergency services, and conserves a scarce public resource in the form of radio spectrum. The site also delves into the “myths and facts” of open wireless including explaining how opening up your access point to others will not slow your own access, how you will not be legally responsible for illegal actions of others (at least they “don’t think so”) and, of course, the fact that opening up your router, even in guest mode, will not put you or your own systems at risk from hackers.
What the site does not cover, and apparently the City of New York does cover, is that laws of physics come into play with radio or wireless communications. Some of these laws have been bent a little here and there over the years but not broken. They have to do with how many people can make use of the same spectrum in the same basic location at the same time. How much spectrum is available in the Wi-Fi 2.4-GHz and 5.8-GHz bands, how much interference there will be, and how that interference can and will cause problems.
In addition to these issues, I did not see any discussion of the changing environment in wireless either, such as the advent of LTE Unlicensed (LTE-U), which is supposed to co-exist on the same spectrum as Wi-Fi today, or about the new “citizens band” (the FCC could have called it ANYTHING but that!), for users, or the upcoming advent of 5G, which is based on small cell deployment and that includes small cells that today might be referred to as Wi-Fi Access points. Remember, too, that the wireless spectrum Wi-Fi employs is unlicensed for Wi-Fi use but it is licensed to others. The rules state that Wi-Fi cannot interfere with these licensed users and unlicensed Wi-Fi users must accept any and all interference they encounter.
The reality of Wi-Fi is that in New York City if you stand on any street corner anywhere in the City and look for open and closed Wi-Fi access points you will find dozens and dozens. Some are in the stores on either side of you, some on upper floors in offices or apartments, some are being carried down the street (many people don’t realize when their device is active in Wi-Fi mode and can be seen by others), and you will see more and more cars driving by with their own hotspot connected back to the Internet over a wide-area LTE wireless network. Add to these the fact that many cable TV companies ship cable modems and TV cable boxes with Wi-Fi that is active and perhaps hardwired to a specific channel.
Those pushing for Muni-Wi-Fi and free access anywhere and everywhere will tell you that since each access point only covers a small area (smaller indoors than out), none of what I have just stated above matters. But they are wrong in several ways. First, when we ran tests on the Anaheim Muni-Wi-Fi system we could see some outdoor access points but could not connect to them because of interference from other stronger access points on the same channel. Let me put it another way:
You have an access point in your apartment in New York City. Your neighbors on both sides of you and across the hall from you do, too. Those a story above and below you have them as well, and the folks on the other side of the street have them. If all of them happen to be set to Channel 6 in the 2.4-GHz Wi-Fi band, coverage from each will be limited. You might have already noticed that the access point you installed in your home or office a couple of years ago does not seem to have the same reach or speed it used to. The defense for this will be that Wi-Fi is smart and can sniff out available spectrum and if set up properly (ah, that is a big if!), it will not interfere. Perhaps this is true in some cases.
But NYC is talking about fast Wi-Fi at gigabit speeds not accomplished by using only one portion of the 2.4-GHz band, say channel 6. Instead, this is accomplished by using the entire 2.4-GHz band or most of the 5.8-GHz band. Yes, multiple users can co-exist, but remember those pesky laws of physics? You cannot break these laws, only try to work within them. Putting a lot of people and devices on the same system in the area will, at some point, mean that none of the users will be happy with the coverage and/or data speeds.
The same people that tell you access points don’t cover large areas so more of them can provide more capacity are only partially correct. They are correct to a point where the number of access points, users, or interference issues reach the limits of physics. (How many of you have noticed a drop in your Wi-Fi range or speed after installing LED lights, which introduce interference?) In other words, while today’s technologies can offer up much better coverage and speeds, users are demanding more, using more, and in major cities are in closer proximity to each other. Now add the fact that outdoor Wi-Fi won’t work well indoors but it can interfere with indoor systems and vice versa depending on where they are located in a building, and further add that it will take a great many outdoor access points to cover NYC and it perhaps becomes clearer that those trying to make all of this happen are bucking some issues over which they will have no control.
Two companies are pushing New York and other cities. The first is Intersection, the company that administers LinkNYC, and the second is Sidewalk Labs, a company funded by Alphabet (Google). Their business model is not unique in that it relies on advertising to pay for it, but it is unique in that the advertising consists of signs at kiosks spread out around the city. According to the companies, they are also in talks with other cities but are not ready to make any announcements yet. Will this “new” model work? Will there be more interference from other Wi-Fi access points, and how will they achieve citywide coverage?
The EarthLink system in Anaheim started out with 70 access points per square mile. Now, there is a huge difference between Anaheim and NYC in the number of buildings, heights of the buildings, and other factors, but EarthLink had to increase from 70 access points per square mile to 90 and then 120 and the system was still not reliable or, in many locations, even usable. Remember, too, that depending on the type of network configuration, every third or fourth access point or perhaps every access point needs to be connected back to a network so it can be connected to the Internet. An access point sitting on a pole not connected to the Internet does not provide any type of useful service.
The theory was that with the next, faster Wi-Fi technologies, and the new insatiable demand for more data and video, the economics would be there this time around. The demand would be such that we would be able to take on the wide-area wireless and wired commercial broadband providers by offering the service for free.
So once again I have dissed Muni-Wi-Fi, just as I have in the past, but the only thing that has changed in the world of Wi-Fi is that the new, higher-speed Wi-Fi access points make use of more spectrum in order to reach faster speeds, and since Wi-Fi spectrum is a limited resource (as is all spectrum), the more access points and users in a confined area, the worse will be the broadband experience.
Fiber and microwave are today’s ways of transporting broadband wireless back to a point where it can be integrated with services and the Internet. Fiber and microwave equipment and service are expensive, and fiber costs a lot on a per-foot basis to install if you have to dig up a street, install the fiber, fill the trench back up, and repave. In many places fiber is already installed but some of you may remember that a number of companies offered rooftop wireless broadband. Towerstream, which is still in business, still offers rooftop wireless broadband. One reason Towerstream and others expected to be profitable in major cities was the fact that most buildings do not have any capacity for new cables (such as fiber), or cannot bring it in from the street at a reasonable price.
There is not much magic in providing broadband. It costs money to connect to the Internet and it costs money to put in fiber and/or wireless devices, even small ones such as access points. Bandwidth, regardless of where you are, is not unlimited. Radios interfere with each other, and other devices such as LED lights and even computers can interfere with radio communications. Perhaps one way to better understand how important it is to have a quiet radio environment is to think about how a grain of sand feels like a rock in your eye, but that same grain of sand between your two fingers feels very small and insignificant. A radio receiver (access point) in a quiet radio environment will hear your Wi-Fi signal as though it were that grain of sand in your eye, but if the radio environment deteriorates over time and becomes worse and worse, soon that access point will hear your signal as though it was that grain of sand between your fingers. The term for interference noise is “RF pollution,” and it can turn a really good system into a really bad system over time.
The bottom line is there is nothing for nothing. Every way in which we gain broadband access to the Internet today costs money. Even unlicensed Wi-Fi is not free; someone is paying for it. If it stops working as well as it did because people listened to the folks at Open Wireless Movement, they will turn off their guest access to their access point and make other changes, and return to receiving what they are paying for.
Will Sidewalk Labs and Intersection be successful in New York City and other places? Some will say that with Google behind them success will be assured, but will that hold true for this venture? Only time will tell.
Andrew M. Seybold