Ted Fry is the owner of Long Hammock Wirless, a wireless internet service provider connecting the residents of rural Florida.
What's your background in networking, what kind of experience/education/certifications do you have? How did you end up owning a WISP?
My background consists of maintaining and installing small & medium business networking equipment, computers, and cabling. A lot of it is small office networks, with a couple of regional and national multi-site installations. Most of the time, I was either the team leader, department manager, or solo on each project.
I've held HP factory warranty service tech status for 16 different HP products. Got a couple of certifications from Microsoft for neworking and windows 95, studied for my CCNA but never took the test, studied for MCSE Security but never took the tests, ended up getting my CEH. I'm certified by CommTRAIN in tower climbing safety and rescue. A lot of my knowledge I gain from first hand experience. If there is a project or problem that I'm working on, I'll research everything relavent to it, then setup a test environment at my home. I've been working with computers since I was fourteen. That was twenty-five years ago. About the same time that I started dabbling in computers, I got my Amateur Radio License, Technician class. I would play around with radios in my dad's shack, but that really didn't do it for me. I prefer to be setting up the equipment rather than using it. To me, it was more fun to put the bits together. To this day, it's extremely rare for me to use a computer to play games. I just never go into that aspect of computing.
I first started in long range wireless networking when I was a systems administrator for a local company. I was required to have secure remote access in order to manage and monitor the NOC. My problem was that I live in the middle of nowhere and didn't have access to traditional ISPs. I had DirecWAY at the time. It didn't support VPN connections because of latency issues. I researched local businesses and found that I could get wireless internet access, if I had a tower to get over the trees. So I found a local guy that could install a small tower for me. It was used, and not as high as I wanted, but it worked. The problem was that it took him 3 months to get around to installing the tower. I had it sitting in my back yard on the ground while the IT director was breathing down my neck to get it done. I had researched the ins and outs of building towers in that time and was about to start when the tower guy showed up out of the blue and installed the tower. I finally had high speed internet access!
Time passed and I took a job offer at a local cabinet and countertop company as their IT manager. Long story short, I quit about 2 months before the housing market crashed. I took a month off, and started my first company doing IT consulting and computer repair. During all of this, I was gaining experience in IP-PBX phone systems, and constantly fighting my internet connection for reliable signal. The signal was consistent, but with the amount of jitter and lag, it really wasn't stable enough to support VOIP and video streaming. I found another wireless provider and convinced them to bring me on as a customer. The condition was that I had to install the equipment myself. So I bought a belt and rope. I had worked with their network admin to setup a 14 mile line-of-sight shot to my tower. I spent many hours on top of my tower adjusting and trying out different equipment. My signal was better, but because of the distance, the link was susceptible to atmospheric conditions. So some days it would be great while others it was garbage. Thermal inversion was the biggest problem. There just wasn't enough signal strength to punch through at that distance. I was out of options. Then one day about a year later, I was thinking about the different ways of how to get recurring revenue when I looked into my back yard and it hit me. If I had that much of a need for high speed internet access, then surely the rest of the area had a similar need. I already had a tower, all I needed was the business. I already knew I could install the equipment, be it on the ground or in the air, so I started researching what it would take to start a WISP. I discovered that the county I lived in didn't require anything, and I discovered that the state didn't require sales tax on a service. I lived outside of the closest city, so I didn't need a business license. And I didn't have to deal with the federal government for the tower since it wasn't over 200 feet tall. I talked to my sales rep at my ISP to make sure I could resell their service and got the OK from the owner. I started researching equipment and found out that it can get VERY expensive. Calculating equipment costs per installation, ROI would have been months per install. Then I discovered Mikrotik and Ubiquiti. The price point for their equipment allowed me to have a very short ROI per installation, which to me meant a short time to generating revenue. I had what I needed to go forward with the WISP. I sold my Corvette and asked my parents to help out with the initial business expenses, and finally started Long Hammock Wireless one year after I started researching.
A year later, I switched ISP's to Brighthouse to gain bandwidth so I could support more customers. I made a deal with one of my customers to use his 40' TV tower as the backhaul to my tower. Six months after that, I started putting up my second tower. A very good friend and neighbor, and customer, had some acerage and tower (go figure, it was sitting on the ground right next door and I didn't know about it!) and was willing to help out building it. One stipulation was that the guy wires be 10 feet off the ground so that the pasture it was in could still be used for livestock. I went about designing the site and searching for suitable materials to make it happen. All the while, talking with my friend about different things we could do. We ended up running conduit to each of the guy anchors so that we could install solar panels some day, and building a cabinet that can house all of the equipment as well as batteries for the eventual solar power. The second tower is 130'. we also build a gin pole and yard arm to assist in installation of the equipment on the towers.
Could you give us an idea of what your network looks like? How many end users? How many access points? What does the backhaul to the ISP look like?
My network to me seems a bit hap-hazard, although it does work reliably. My customers rarely call with issues, and jitter and latency are very low. I have a few users that are heavy into online gaming with game consoles. From the cable POP, it's a pair of Ubiquiti Powerbridge M5's setup in transparent bridge mode to my 2nd tower. From there it's a pair of Nanobridge M5's in transparent bridge mode to my 1st tower. Total wireless distance between the cable POP and NOC is 4 miles, split up between the 3 towers. I use Mikrotik on an old PIII 800MHz PC with a 4 port gigabit card as the gateway for my 1st tower. On the back side of that are a Mikrotik RB411UAHR with a Ubiquiti XR9 and 11Dbi omni antenna for 900 MHz unlicensed operation and a Ubiquiti RocketM5 with a Ubiquiti 120 sector. That 5GHz unit was originally intended to be just testing, but eventually was pressed into service to supply bandwidth to customers. The 2nd tower has a Mikrotik RB493AH acting as the gateway, and transparent bridge between the POP and NOC. Up top are 3 RocketM2 GPS units on 120 deg. Sectors, and 3 Pac Wireless 900MHz 13dbi 120 sectors with LMR400 cable down to a Mikrotik RB600 with 3 Ubiquiti XR9's.
My NOC is a couple of server racks in my family room. It was the only place I could find with enough space and was isolated enough to house all of the noisy equipment. I also have a 2 post relay rack in the Laundry room that holds the PBX, a 100Mbit switch, the gateway server for the local tower, all of the power bricks for the tower equipment, and a couple of Axis 2401 video servers. When it comes to servers, switches and routers, I try to purchase used equipment. The cost savings is huge.
The Primary NOC currently consists of 6 servers. One primary NAS, one secondary NAS, and 4 servers configured to host virtual machines. All production servers with the exceptions of the secondary DNS server and PBX server are virtual. There's also spare capacity to host additional virtual servers. I'm planning right now to provide VPS hosting and IAAS services. I have a mix of Linux systems running everything. Fedora, Debian, CentOS and Ubuntu are the 4 main distros I use. Fedora runs my website, web hosting services, and monitoring services. Debian runs Freeside billing, Ubuntu runs FreeRadius and Xen, and CentOS runs the PBX. I have a virtual machine running XP, and my laptop has windows 7 dual booting with Linux Mint. I have to keep them around for remote support of some customers. I have primarily used MySQL for databases, but do have one PostGRE SQL database. Nowhere do I use Microsoft products for mission critical applications.
All server to server communications are gigabit, and the rest of the copper network is 100 Mbps. I use Cisco 2950's and a small Cisco/Linksys unmanaged gigabit switch. I have a few Cisco routers of different varieties, but they didn't fit into what I had designed for the network. So they are just sitting collecting dust.
Everything is on surge suppression and battery backup. I've been lucky since starting LHW in that I have only had lightning take out one switch in 2 years of operation. I have a a lot of ground rods sunk into the ground.
CPE on 2.4GHz is a mix of Ubituiti gear. LOCO M2, Nanostation M2, Nanobridge M2 and Airgrid M2. Some are installed in the customers home, while others are installed either on a j-mount or on a TV antenna tower.
It seems like there's a big market for WISP's at the moment, especially factoring in developing countries, but what do you see as the biggest threat to your business? Advances in satellite communications? User operated mesh networks? Big ISP's expanding their reach?
Big ISP's are secondary competition. They don't want to spend the money to reach the lower population densities. For them, a buildout is potentially millions of dollars spent to reach only a few hundred potential customers. It doesn't make financial sense. For a WISP, the cost to bring service to the same area can be as little as a few thousand dollars. In higher population density areas, having a 3rd option (Cable, DSL and WISP) only helps. There are customers from traditional ISP that just aren't satisfied with either the service or price, or the fact that the big companies aren't a “local” - some people want to know who they are dealing with.
Satellite internet access for residential and most businesses is dead in my book. It can't support the speed necessary for the majority of websites and services, and it has high latency. I'm not saying that it needs to die - there are situations that only satellite can provide. But for residential and small to medium businesses, it's not an option anymore.
The biggest threat that I see is the cellular companies. With faster speeds and tethering coming out, it gives people another, albeit pricey, option to get online. And because of the pervasive advertising that cellular companies are constantly spewing out, it brainwashes the average person into thinking there is no other choice when it comes to wireless. I actually had someone call asking for a particular cell phone! I apologized and explained that I provide internet access only.
And then there are the games that cellular companies play with data rates and caps. It's just plain bullshit. Sorry, but it’s a hot button for me. Data caps had a purpose before the backbone was built, but now they are just a gimmick used to suck more money out of the consumer. It's just plain greed. I provide the bandwidth, and don't care how much data runs through it. I have an analogy that's probably used by quite a few, when describing my services, “I provide the garden hose, and don't care how much water you run through it,” and people get it.
User, or community mesh networks, where no payment for services are required, are run by people like me, but they don't want the extra complexity of the business model, or believe that the Internet should be free. What I believe is that it should at least pay for itself. Either way, I'm good with it. From a WISP standpoint, it only widens the knowledge of wireless networking, and puts more momentum behind the wireless movement. When a person using a mesh network gets tired of the saturated link, they will search out a service and pay for it. Or they may use it for some things, and have a paid service for others. I've looked into OpenMESH, and I like what it offers. I could even use it myself in a paid model. The opportunity just hasn't presented itself yet.
Would the average net admin working in a corporate environment have the technical skills to start a WISP? What would they need to learn?
I'm not quite sure I can say with certainty that the “average” net admin would have the understanding of RF communications, or the intestinal fortitude to hang themselves a couple hundred feet in the air from a rope. My background with Amateur Radio helped a lot in understanding how microwave communications behave over long distances. But for the actual networking aspect of the business, they would have that knowledge. It's basic subnetting, firewalls, and NAT. Yeah, I'm oversimplifying it, but that's really all that's needed. Everything else is to manage those three. IPv6 is happening, and that will play into the knowledge base, but I have to wait for Ubiquiti to get off their asses and build support into their equipment for IPv6. Everything else in my network supports it now. They are the holdup.
What were some of the biggest challenges you faced when starting out that you had not anticipated at all?
I had a good understanding of RF before I started this thing, then I discovered trees. Damn trees all over the place! With the majority of my footprint being in the country, you can imagine how covered some homes and businesses are with trees. Foliage attenuation has killed many a site survey. So in order to get service to some places I have to install additional equipment. Mostly small towers and push up poles. The additional cost can be a deal breaker for some people. Just getting service to some areas is a huge challenge.
The biggest challenge I face is money. I had barely enough to start the business. Two years later, it's paying for itself, but not providing much more. I'm always looking for investors. I have plans for getting on commercial towers that can't go forward until I get the financial backing. Know anyone with some extra cash?
Some of the problem is the economy today. People aren't taking as many chances with their money as before. More of the problem is lack of knowledge. People hear about WISP’s and don't understand that it's a real thing. Much less that it can be a solid investment with a relatively short ROI, given the right amount invested. I've had people call asking if it was legal, or if it was a cellular service. When they see the starting monthly price, they can't believe that it's legitimate. I believe it's because of the advertising and billing from the big ISPs and cellular companies making them believe that every advertised price has extras tacked on after the fact. I have a setup fee, and 5 different plans. The advertised price is what the customer sees on the invoice. Nothing extra, unless they ask for it.
The only other thing I hadn't anticipated was my electric bill going up by over 150%. That really sucked at first.