Steve Pierce founded HDL
, an Internet and web consultancy, in Albuquerque, New Mexico in 1987. Now living in Ypsilanti, he came to Michigan for true love; his wife moved here for a job.
After relocating to Michigan, Steve was a partner in OpAve, one of the first wireless Internet service providers in Michigan. He relaunched ABQ.com as an email services company and for five years ran MailAndNews.com. With nearly two million subscribers, MailandNews was the second oldest free email service provider behind Hotmail.
Steve launched Wireless Ypsi
in February 2008, a free Wi-Fi Internet service in Ypsilanti. Since then, Steve has assisted the cities of Dearborn, Lincoln Park, Trenton, Wayne County and a number of other communities in deploying similar free wireless networks using the Wireless Ypsi model. Today over 475,000 people have used free Internet networks Steve helped design and install.
He has also been involved in a number of start-ups (including US1 Internet), has been a consultant and teacher for the military, universities, the U.S. Department of Energy, and NATO, and has worked for a number of major corporations as a network designer, support technician, and trainer.
Steve can frequently be seen about town riding on a Segway. In his spare time, he and his wife Maggie have restored three homes in Ypsilanti, including the historically significant Glover House. He shares his home with two German Shepherds named Dax and Ryker. Yes, he is a Trek fan, too, and yes, he knows Ryker isn't spelled that way. (Ask him sometime for the background.) You can reach Steve by email: Steve@HDL.com.
Several years ago I was in a meeting with an executive from a large computer company trying to resolve problems with a soon to be released software product. My job was final testing of the software and training their support staff. The release was not going well. In fact, it stunk.
The software was way behind schedule. Hardware that was supposed to be available wasn't and early adopters were up in arms threatening to withdraw contracts if we didn't get things fixed and fast.
I kept trying to explain to the head of tech support his staff wasn't ready. He kept saying he was and said it was my fault the software was late. That part was true, we were late. I wasn't helping things; every time we ran through our test suite, we kept finding bugs. Yet, I knew we would soon be ready with the software, and I also knew if we had crappy support, no matter how much better this new version, customers would revolt.
I was getting nowhere with the manager and was finally brought into the CEO's office for some talking to about my attitude and lack of team playing. Furious, I sat there as they ripped into me. After a while, realizing I wasn't saying anything, the CEO turned to me and asked me if I had anything to say.
I reached over to the conference phone on the table and called the 800 number for tech support for his company. The call was immediately put on hold. Two minutes passed; five minutes; then eight minutes. The tech support manager sat there glaring at me. At 10 minutes, the CEO said, "Alright enough, I get it."
I would love to say the company saw the light and fixed their problems. They didn't. They were bought out several years later by another large company that spectacularly went into bankruptcy this past year. Was their tech support problem part of the problem? Yeah, it was. They knew their support sucked, but never could figure out how to get it fixed.
What too many companies fail to understand is what happens when a customer calls for help.
Comcast, a company most people love to hate, has one of the worst phone systems in the world. As a business customer with multiple accounts, I dread having to call them because I know it is going to suck.
One of the problems is having to enter your 16-digit account code. Get it wrong, and you start all over. Then you have to enter it time and time again as your call gets routed without ever talking to a human.
One well known secret to calling tech support is to not enter any account numbers or phone number when you call. That way your call isn't ‘directed' by the interactive voice response system, or IVR. Instead you get to a human, describe the problem, and they transfer your call to the right department.
Comcast doesn't work this way. If you don't enter in a number, you are stuck in an endless loop. The system will never hang up, nor will it transfer the call to a human. However, while I work in the 734 area code in Michigan, when I need to call about one of my accounts in another state, the IVR just pukes. The computer looks at my area code and determines the call is coming from Michigan but I am calling about an account in New Mexico; therefore, that account can't be right. The IVR doesn't transfer me to Michigan or New Mexico; it transfers me to their New England call center which can't look up my account info. They then transfer me back to Michigan and the first thing it asks is, please enter your 16 digit account number.
Worse, after 5pm, my business call is not transferred to New England; it is transferred to residential tech support. They can't help either.
What I wouldn't give for just five minutes to show the head of Comcast tech support their IVR sucks.
Yet Comcast residential service has a very cool service, one where the IVR will call you back while still keeping your place in line. The process is simple. Instead of leaving you on hold for minutes or hours, the IVR system will record your name and number and then when your place in line is nearing the front, the IVR will call you back and after a minute or two, you are talking to a live person. Problem is this system isn't used for business customers because too many business customers have an internal phone system with extension numbers and the IVR has no way to direct dial extensions.
We all have these sorts of stories about other companies and bad support. From Sears to Bank of America, we can all tell you horror stories of calls left on hold for hours. My favorite is when the office is about to close and you are still on hold. You know the office has closed when suddenly your call is automatically disconnected.
Call back and the after-hours prompt says the office is closed, please call during regular business hours. I did call during business hours, your stupid #^)#%^)@ hung up on me. Grrrr.
OK, so we know other companies' support center sucks, but wait, have you ever called your own company's service department or help desk? If you haven't, reach over and call right now. I will wait for you. OK I am waiting … more waiting … still waiting …
OK, good, now you are back. While you are on hold, I will finish the story. Comcast's problem could be fixed if someone from their company would simply test their own phone system to see the hell their customers go through to reach a support technician.
The solution is simple: it is called Secret Shopping. You need to secret shop your own company. And not just once; it needs to be a regular part of your business. Executives and managers should regularly call into their customer service center or help desk. If you need an account number to get service, have them set up an account for you and fire anyone that flags the account for special treatment.
I am convinced if the CEO of Verizon had to call into tech support instead of handing his phone to his assistant and saying, "Here, fix this." Verizon tech support wouldn't suck.
Oh, by the way, this goes for internal help or services desks inside your company as well. You need to find out what happens when your own employees call for help. And don't call them from your office phone, they know your extension number. Call them from the front desk, the bullpen, and see how your employees are treated by your own support staff. You will be shocked.
We see some of this in the TV show, Undercover Boss, where the CEO of the company goes undercover in their own company, usually in a disguise. While fun to watch, you would think some junior guy just starting out probably doesn't rate a full-blown TV crew following them around. If you are one of the other employees, you have to sort of wonder if something is up. But the concept is good. If your company is in the burger business, taking off your tie and flipping some burgers is probably a very good idea. Tom Peters used to call this Management by Wandering Around, though some claim Abe Lincoln did this by informally inspecting the troops during the Civil War.
I think aimlessly wondering around is like a blind pig looking for a truffle. The pig might find a truffle, but he could just as easily starve. Wandering around looking for problems is probably not a good use of your time. However, I do agree that simply wandering around is a good way to better see your business in action. But unless you want to see what happens when your customers are wandering around looking for help, this approach isn't the best.
Costco uses a different approach. The managers come in and do a thorough walk around the entire store, with department managers in tow. It is a grueling couple of hours and it is tough on managers and staff that aren't prepared. I have been in enough Costcos over the years that whenever I see one of the hives making the tour, I go up and ask a question. Like, "How come you stopped carrying Doritos?" or "The O.J. used to be in aisle 12, but now I can't find it." And then I wander off looking lost.
So how is that call going? Are you still on hold? You are? OK, I will keep going.
Companies will also hire secret shoppers to routinely check up on the company and provide continual feedback. But you have to watch out. Employees will oftentimes figure out who the secret shoppers are. I have been in places where in the back room, they will post up a list of names or descriptions or patterns used by secret shoppers. I have even seen managers pay employees for spotting the secret shopper.
People are people. If you measure teachers' performance by the scores of their students, don't be too surprised that teachers will teach for the test. Employees will figure out a way to game the system.
Your job is to make sure employees and managers understand that instead of trying to game the system, the goal is to provide good customer service. Reward employees that excel at customer service, punish or retrain those that don't.
The first step to providing better customer service is answering the phone. If the IVR is the barrier between you and the customers, get rid of the IVR or make sure that it is programmed to handle all sorts of input so it doesn't loop the customer through phone help.
Now back to your call. Are you still on hold? If you are or have had to enter your 16 digit account number three times only to be asked by the support representative for your account number, well now you know you have some work to do. At least you know what your customers are going through and when you fix it, your customers will thank you.
I was talking with a manager a while back and she was bemoaning the fact that employees seem to be getting worse and worse. She kept saying, "If I pulled that sort of crud when I was just starting out I would have been fired."
I asked, "When was the last time someone was fired for poor job performance?"
"In the last five years, never," she responded.
Are you kidding me? Never.
I asked, "Do you ever return things you buy?"
She said, "Yes, all the time." She had just returned a door handle that didn't fit to the hardware store.
So you are telling me that you made a mistake and bought the wrong door handle, but in five years you have never made a mistake and hired the wrong person?
Now I am not suggesting you fire someone for fun. And there is a big difference between a door handle and a person. If you start treating employees like door handles, the next person fired will likely be you.
But what this manager was complaining about was the lack of fear, and complacency by her employees. Fear is a powerful motivator, but if there are no consequences for failure, then why try? Worse, bad employees were bringing the rest of the organization down.
Using fear to stifle innovation and risk taking is bad. But a little fear is a good thing. If you aren't pulling your own weight you could find yourself collecting unemployment. Fear is a wonderful attitude adjuster and improves productivity.
In the world of a little is good, a lot is bad, don't do what one Silicon Valley firm does. They grade all their employees on a curve, which means during each cycle, 20% of the employees on a project get a failing grade. That means even on a great team with everything clicking, someone is going to get a bad review. That is dumb.
It leads to employees undermining each other and primping themselves up like peacocks whenever the manager is around. It also leads to higher turnover and good teams that are productive get broken up.
There has to be a balance between never firing people and cutting people simply because the rule says 20% of the worst performers have to go.
Never underestimate the power of a good firing. If you want to shake up the organization, fire the problem person.
Why is it that sports teams often times win a game after the manager is fired? I argue it is fear. It usually doesn't last very long before the team returns to its losing ways, so you can't fire your way into being a winning team and not change the things that are causing you to lose every year. If you want to destroy a team quickly, bring in a crappy employee and watch them bring the entire team down. Even a brilliant employee that is a jerk can be a disaster.
So launch the jerk, too.
At a real estate firm, I once saw one of the high-grossing sales people get launched by the broker. Though he led the office in sales, this guy was a jerk. He was abusive to other employees, stole deals, and there was chaos and dread every time he came into the office. The guy was a selling machine but he was a liability. So the broker said that he thought it would be best if he went out on his own, and within in a week he was gone.
The rest of the office was stunned. Part of it was a little fear; geez, if they whacked a top sales guy, who was next? But it also sent a message that jerks were not going to be tolerated. I checked back several months later and I asked how sales were. Sales volume was up and the office chaos and stress was gone.
When I was at my first real job, I was really raising heck over a deal that had gone south. My manager called me in and sat me down. He said, "Steve, I want you to imagine there is a bucket of water in front of you. Now put your fist into that bucket, splash around, make a mess and thrash about as much as you want. Now pull your arm out of the water and the hole that remains in that bucket of water is how much you will be missed."
OK, I got it.
Now if you are going to fire someone one, you have to document why. Work with your personnel and legal folks to make sure you do it right. Moreover, give people a chance to improve. I was glad to get the water bucket story. I was especially happy my boss gave me a second chance.
But when you are on your third and fourth chance and the employee still doesn't get it, throw them over board. Don't turf them onto a project where they can do little damage, and don't give them a glowing review hoping they will move to another department. If the employee isn't working out, fire them. Then learn from the mistake, figure out what went wrong or what you missed in the interview process, and move forward.
Your customers will appreciate it and the rest of your employees will appreciate it too.
Wireless Ypsi is a free wireless Internet service that covers all of the downtown and Depot Town in Ypsilanti and is branching out to other business districts and neighborhoods in and around Ypsilanti.
In my first two posts I described Wireless Ypsi and how we went about deploying Wireless Ypsi from concept to usable network in two weeks.
By the second month, my business partner Brian Robb and I had covered much of the 10 block downtown district. By month three we covered Depot Town. Over the last five months we have grown the network to include apartment complexes, retirement homes, and a community center. Since I started this blog two days ago, 244 new users signed on for the very first time. Since yesterday, 439 people have used the network.
Hopefully we have shown that Wireless Ypsi and the Meraki mesh radios is a viable free community based wireless Internet Service that has a very low cost of entry, scales well, is reliable and fast, and can be quickly deployed. By the time you read this post, the world famous Bomber Restaurant on Michigan Avenue will be the newest node on Wireless Ypsi.
So how to do you get a Wireless network going in your neck of the woods.
Remember our earlier tips. Don’t throw in the kitchen sink with every possible use. Keep the goals of the project simple. The goal should be a usable wireless service to surf the Internet and check email.
Define the area you want to cover. It should be small to begin with and grow with use. Maybe an apartment building, a one or two block area downtown, a business or neighborhood park or common meeting area.
Build the network where you have a lot of people. It could be your downtown with lots of lunch traffic and offices, the fairgrounds or community center, or a river or lakeside park where people gather for festivals and weekends.
Don’t promote the network, at least not at first. Get it running and seed a couple of the blogs or business groups and encourage people to try it out. Let them know this is a special project and you are just inviting a few special folks to try it. Encourage feedback and make it easy to get a hold of you.
Not every network will follow the Wireless Ypsi model. Be flexible and adapt the network to what works best for you.
Here is how three communities got Wireless Internet going and how we got them started.
The Congressman calls
Ok, the Congressman didn’t really call, but a staffer, Jeff Donofrio from John Dingell’s office did call us in March and said, "The Downriver Conference wants to look at doing wireless Internet, why don’t you come speak to them about Wireless Ypsi."
So we go to a working session in late March that is getting ready for the big conference with 300+ attendees in May. They want to talk about Wireless. I said that Wireless Oakland/Washtenaw was a failed model and blasted other communities including San Francisco and Philadelphia for screwing up their networks.
OK, I should have asked first who else they talked to. Wireless Oakland had visited the same group a couple of months earlier. Oops. Well, I have never been bashful about sharing my opinion.
I laid out my idea on how other communities can make Wireless Internet work. After an hour, someone asked if I would come back for another meeting. "Uh, no." I was here for free, just talking about Wireless, I didn’t want to come to more meetings. Besides, I said, "No one ever deployed a wireless network in a meeting."
So then they asked me to work up a cost proposal. "Uh, no again." This is simple. The indoor radios are $50 and the outdoor are $100. The price has since then gone up, the indoor is $149 and the outdoor is $199.
If you want to do a typical downtown block with just indoor radios, figure 5 per block using a zig-zag pattern from one business to the next. Outdoor deployments, figure one to two outdoor radios per block. That is it, do the math, there is your cost proposal.
No one was very happy with this. For one, several communities already had $100,000 plus quotes, what I was proposing seemed impossible.
So I made this offer. I will come back for one more meeting but you have to do something first.
Step 1. Raise $1,000 to buy radios. I don’t care how you do that. Get it from your DDA, Chamber of Commerce, City Hall, or write a check, but before we meet in April you have to have a firm commitment for radios and have placed the order for 4 outdoors and 2 indoors or 10 indoor radios. Your choice.
(Note: I changed the numbers to work with today’s pricing from Meraki, in March I said they need to raise $500.)
Step 2. Order or secure a high speed Internet connection you can share. Again you may be able to use the existing connection at your DDA or Chamber office. A local business like a coffee shop or bar may let you share their connection. Make sure you get a connection from an ISP that allows sharing. It didn’t have to be working connection, not yet, but it had to be ordered.
Do that, and I will come back in three weeks for one more meeting and I will help you get the network up and running. Then in May, instead of talking about Wireless Ypsi, you guys will talk about your own Wireless networks at the Downriver Annual Conference
My plan was in May at the big Downriver conference, these communities will talk about Wireless Internet in their own communities. Not a proposal, not a plan. A real, honest to goodness, working wireless network in their downtown, built start to finish in 6 weeks.
Surprising to me, four communities right there in March said they would commit to our two step plan. When we got back together in three weeks, three of the communities had actually followed through with Steps 1 and 2. They were the Cities of Dearborn and Trenton. Lincoln Park was not that far behind.
Remember, one of the other steps that I said in an earlier post was importance of local knowledge and trust. It wasn’t Steve Pierce that was installing this network. It was people closely tied to the community that were doing the work. They could get things done and they knew who to talk to. This was critical to get a system up quickly.
Also there was a ringer. Jeff Donofrio from Congressman Dingell’s office played a critical role. He knew every person from these communities and behind the scenes he helped persuade them to take a chance on the Wireless Ypsi model.
Frankly, the model sounds nuts so without a giant leap of faith the idea of free Wireless system installed in a downtown in 6 weeks, well it is madness. It can’t possibly work.
Here is a guy from Ypsi telling you that you can build a working free wireless Internet system for $1,000 and a $50 a month Internet connection when the experts tell you it will cost $20,000, $50,000, $100,000 or more and you read the headlines in the tech journals of community after community that failed to ever get a working network.
This guy from Ypsilanti must be a nut.
I am a nut, but that aside, they trusted Jeff to give me a chance to make it work. Truth is, they did all the real work, I just had the confidence that it would actually work. Plus, we had our experience of two months with our own network. I can’t write this with a straight face. Two months, and look, we are helping three other communities to start their own wireless network. But we were and people were following our lead.
Each community took a little bit different approach, but each was able to make it work and their networks are continuing to grow every day. Here are two of those stories.
West Dearborn Michigan goes wireless
The project was headed up by Dearborn CIO, Doug Feldkamp. Doug knows networking and he know wireless. Dearborn has a number of wireless connections that they beam around the city for different offices and facilities. So he caught on very quickly how this could work.
Dearborn was able to get commitment from several businesses for their Internet connections to start.
Doug then ordered 10 indoor Meraki radios. He would then visit a local business, order lunch and then ask if he could install a radio in the window so the business could have free WiFi for their customers. After a couple of weeks he had covered a city block.
Doug’s tech geeks in the office wanted to setup the network, but he wouldn’t let them. He didn’t want to distract them from the priority projects they were working on and besides it gave him something to do on his lunch break.
Doug said by doing it on lunch hours he was able to meet business owners and it presented a positive image to the business community that Dearborn was working on a project that would benefit the downtown businesses. Doug then ordered two dedicated connections to feed bandwidth into the system. He also secured $5,000 from the DDA to pay for the bandwidth and buy more radios.
Since April, Open Dearborn has had over 4,500 people use their network and they have about 35 radios covering several blocks of their Western DDA Business District.
Doug says they are working on a similar project for their Eastside business district and has already secured financial commitment from the Eastside Dearborn DDA as well.
City of Trenton weathers a storm
The City of Trenton took a slightly different approach. Trenton ordered 11 outdoor radios and two additional indoor radios to use as test radios and to learn how to use the system. They also ordered a high speed connection from a local ISP that is dedicated to the free network.
Trenton DPW installed the radios on City owned light poles in the downtown. At first, Trenton had a number of problems with the network that didn’t make sense. One was that it never seemed to work during the day, but then in the evening, when I had time to look at the network, it worked great.
Thankfully, someone far smarter than me in the Trenton DPW department said, "Hey you know, those light poles have photo cells, is that going to be a problem?"
That would explain why the network only worked when the lights came on.
A quick revisit by DPW to rewire the outlets solved the problem. Trenton also had a channel interference problem. It took a couple of days, but we found a wireless channel that seemed to work and we were off to the races.
During the big Downriver conference in May, Trenton City manager Robert Cady said this about their wireless network: "Trenton has not had a lot of positive things to talk about recently. With GM closings and the economy things, are tough. But here comes this guy from Ypsilanti and he says we can get wireless Internet in our downtown for less than $2,000. For that price, I am willing to try anything. Well, all I can say is, it works."
If you have ever met Bob Cady, you know this to be true; he doesn’t get too excited about anything. Bob simply saying "it works" was high praise.
Several weeks after the conference, Trenton was hit be a devastating summer storm. Power was out in much of the city for three and more days. But downtown Trenton had power and their Wireless Network was working.
We read on several blogs how Trenton residents had discovered the free downtown network and took their laptops downtown and were able to get on-line to check email and stay in touch with family and co-workers.
Shortly after the conference, the City of Lincoln Park came on-line as the third wireless project downriver, though we haven’t seen that network in action. We have also learned that community volunteers in Taylor are deploying a Meraki network using the same Wireless Ypsi model.
Build your own Community Wireless Internet Service
I touched on ways to get the funding in your own community and showed you a couple of different approaches. Wireless Ypsi got businesses and residents to contribute $50 or $100 to buy a radio and extend the network. Dearborn seeded the money form city hall and then got the money from the DDA and Chamber to build out the network. Trenton paid for the system out of the general fund.
There is a forth model of either using an ad-supported network or monthly subscription revenue. The third largest Meraki network in the world is in Hawaii and is entirely subscription based. While beyond the scope of this series, Meraki does support both models.
Wireless Ypsi is looking at ad-supported revenue to help expand the network and we are hoping we can get some support from the two DDA’s in Ypsilanti and City Hall to put up radios in our parks and community centers. Surprisingly, City Hall still won’t let us put up a radio at City Hall. But we are working on it. (Hint, hint)
During the Downriver conference representatives from Wayne County were at the conference and were especially interested in the Wireless Project. The next week they called me and asked all sorts of questions. They then said they wanted to hire us to help consult with the IT staff on a wireless project. Hey cool, a paid gig.
So I told them what to order and laid out the strategy to deploy the network at one of their lakefront parks. Using Microsoft Live maps, they have the best aerial views; we flew over the park identifying high spots, buildings, and light poles to install the radios.
Three weeks later I get the call from Wayne County. Oh boy, I thought, time to do some billable work.
On the other end of the line was a quizzical IT geek who said, "Yesterday, Facilities installed the radios on the poles. I then configured the Internet connection and connected the radios and it worked. What am I missing?"
I asked, were you able to surf the Internet? Yes, she replied.
Damn, so much for a consulting gig.
Can anyone deploy one of these networks?
No, it takes some skill and knowledge about wireless networking and especially troubleshooting and effective antenna placement. But as I said before, a competent network geek with good problems solving skills can get the first couple of nodes going in about an hour.
Ed Velmetti commented in an earlier post, "I think people underestimate how much the Meraki technology is a game changer for this…" I agree with Ed to a point. Just as important as the technology are the people installing the system and critical to the success is local knowledge and connections to get things moving.
You have to have local knowledge of the business owners, city and county governments, DDA, Chamber, and much more. That doesn’t mean that Brian and I couldn’t go in cold to a new community and get a network working. We believe we can.
But to be truly successful and to deploy in weeks rather than months or years, you have to have the contacts inside the community to pull it off.
While it is fun to talk about the technology, it was really the forward thinking business owners in Downtown Ypsilanti that saw the benefit of a free wireless network as a way attract people to live, work and play in our community.
While Brian and I may be the face to Wireless Ypsi, it is the 125 business owners and residents that each bought a radio and in a sense became investors, partners in Wireless Ypsilanti that all contributed to making Wireless Ypsi successful.
I would love to blog about the many other cool things we are doing in Ypsilanti, especially in our two downtown districts. Like video casting the 2009 Elvis Fest live from Riverside Park using Wireless Ypsi. Or YpsiNews.com reporting on our local government with live broadcasting of town hall meetings and debates using Wireless Ypsi.
We'll have to see what time allows. Thanks for stopping by and visiting.
If you are ever in Ypsilanti, make sure you call or email me and we can get a cup of coffee from one of our great local coffee shops or a brew from one of our local taverns and you can check out Wireless Ypsi for yourself. First round is on me.
Feel free to email me directly with questions or comments at firstname.lastname@example.org
Wireless Ypsi has been an interesting project to work on for the past eight months. When Brian Robb and I started it in January 2008, we thought we might have 5 or 10 business hooked up and the 'regulars' that visit downtown Ypsilanti every week could surf the net. We sort of blew over that goal in the first week.
Last night, Wireless Ypsi had its 8,400th new user and 3.2 terabytes of data has been transmitted since we started. In the past 24 hours, 454 people signed on to use the system. According to Meraki, Wireless Ypsi is one the top 10 networks in the world. Stunning.
Beyond the users, beyond the technology, there is an important reason why this has been successful and one we are very mindful of when consulting and advising other communities that want to use our model. It is knowing the people in your community
We could have done the traditional wireless model and pulled in our own Internet connections. We could have negotiated tower and light pole agreements with City Hall. We could have paid engineers thousands of dollars to do electromagnetic site surveys. But that would have taken months, even years, and it would have cost tens if not hundreds of thousands of dollars.
Between Brian and me, we had $1,200 we were willing to throw at this project. We came up with $1,200 as we figured each of us was willing to kick in $50 a month if we could get Internet access in our favorite downtown restaurants and pubs.
Inventory your community assets
Typically a wireless system from one of the big companies like Motorola or Tranzeo might cost between $25,000 and $100,000 per square mile. We had $1,200 and two weeks to make something work.
So why not take advantage of the assets that your community or downtown already has. In our case, a number of bar and restaurants already offered free wireless. Moreover, most every business downtown already has a high speed Internet connection. If we could just tap into those connections, we could get going quickly and not spend a fortune.
That is where personal connections come in. I had been active in the community since I first came to town in 1999. I was a former DDA chair, ran for mayor and lost, and helped start the Ann Arbor/Ypsilanti LDFA which is a major funder for Spark.
If there was a project to work on or plants to be watered I was there. So people knew me. Same can be said about Brian Robb. Brian is a city councilmember, he won, and a tireless promoter for Ypsilanti and especially our Downtown and Depot Town.
So when we walked into a business, we already knew the owner and they knew us. It was easy to get a meeting. The hard part was to get them to let us hook into their Internet connections. But it wasn’t that hard.
Some owners like Dave Curtis at Pub 13, Brian Brickley at the Tap Room, and Derrick Block at TC’s Speakeasy all immediately said, "Yes!". Thirty minutes later we were ferreting around basements running Ethernet cable and sticking the radio to a front window with suction cups.
But some other owners were a bit skeptical. They worried that we would drain their bandwidth. They wanted to know if this was legal and still others asked if it was secure. Well let’s talk about these issues.
It doesn’t drain bandwidth because we monitor the system and will bounce anyone being unreasonable. We don’t restrict what you can see, but if you are connected for 8 hours and download 10 full length movies from iTunes, you will likely get banned at least for a day or so until we chat about how this is a co-operative free network. We explain that no one is making money here and all we ask is everyone to be reasonable. Most folks get it, apologize and we never have a problem again.
Moreover, because we have a high concentration of radios and multiple connections to the Internet, at any one point in time, there are likely just one or two people on any one connection. So the bandwidth use is minimal. And we can prove it by showing the customer the actual stats from Meraki.
Is this legal? Sadly, we had some folks with ties to the Wireless Washtenaw project and the City and County that were telling people what Wireless Ypsi was doing was illegal. It is not illegal. We are not stealing bandwidth from the air or hijacking an unsecured network connection. It took a while for us to first find out this was happening behind the scenes and then dispel the misconception.
For anyone that did not get the memo, "Wireless Ypsi is not illegal."
Good, I hope we can put that to bed finally.
However, some Internet Service Providers like Comcast Home and AT&T Residential service do not allow sharing of Internet connections. But downtown, there are businesses with business Internet service that cost as much as two or three times more per month and they don’t have the same restrictions. So yes, it is legal.
We have spoken to several representatives from AT&T of Michigan asking them to reconsider their terms of service limitation for residential customers. Brian and I are convinced that hundreds of people in Ypsilanti would switch from Comcast if AT&T Residential would allow sharing so they could put up a Wireless Ypsi radio. While nothing has happened yet, we are hopeful AT&T will consider this change.
Other providers like Wow, Speakeasy, TDS Metrocom, Cavalier, and Ypsilanti’s very own Provide.net all allow their Internet connections to be shared.
Wireless Ypsi is not an ISP, we are not competing with other ISP’s. Provide.net was initially worried that we were going to take business away from them. The opposite is true, we send customers to them. Once they learned that, they thought it was a great idea.
In eight months we can point to 12 new Internet connections that were specifically ordered by local businesses and residents in support of Wireless Ypsi. There are four more connections that will come on-line just in October. Far from taking business away from the ISP’s, Wireless Ypsi is driving new business to them.
In fact, in several cases, a business ordered a second Internet connection just to connect to Wireless Ypsi which brings us to the next topic, Security
Is it Secure?
It is wireless Internet, it is not secure. I repeat, it is wireless Internet, it is not secure.
If you are looking at patient records, reviewing a legal case for a client, processing payroll, or anything else were personal or confidential information is involved, don’t use a free wireless service, it isn’t secure.
Can you check your Yahoo or Gmail account securely, you bet. Just make sure you use "HTTPS://" and then your provider’s domain name. For example to check your Gmail go to https://www.gmail.com. And for heavens sake, your email password better be different from your banking password. ‘Nuff said.
But Steve, what about network security? You are plugging into the businesses Internet connection. They have computers and QuickBooks, credit card processing, and banking. Aren’t you just opening up their network to problems? The answer is "Yes", but it isn’t as bad as you might first think and most of the time the Wireless Ypsi Radio is more secure than the low bucks Linksys wireless router they are already using.
Remember many of those businesses we first visited already had a free WiFi service and the security was grim. All they did was plug in a Linksys or Netgear wireless router and turn on the network. They had no idea who was connecting or for how long, or how much bandwidth they were using.
Worse, their own PC’s and printers were connected to the same Wireless network so users could potentially snoop the network and learn passwords and other personal information. Unfortunately, that is the state of affairs in most businesses that are providing free WiFi.
The Meraki system is more secure. The radio cannot see any of the local traffic. So when a customer connects to the Wireless Ypsi network, all they can see is the Internet, they can’t see the local computers and they can’t see other wireless users. Already this is more secure than 80% of most WiFi hotspots. So it is reasonably safe to plug a Meraki radio directly into your existing network and you will not be exposing the rest of the network to the community.
However for better security, we recommend firewalls and additional routers to isolate traffic between the business and the free WiFi.
For two businesses in Ypsilanti, they took the best approach. They have an entirely separate Internet connection from a different provider for Wireless Ypsi. We then helped them secure their internal network and locked down the configuration and in one case even turned off the internal WiFi for additional security.
For those businesses that were already offering free WiFi, the Meraki system is actually better and more secure than the system they already had in place.
Closing the Deal
So after we spent 20 or 30 minutes explaining this to folks, all but one business said, sure plug in. We haven’t given quite given on up on that one last business, we keep telling the owner his current WiFi is wide open and anyone can see his QuickBooks PC and even print to his printer. Maybe we can make him a convert in the next month.
To grow the network, it is all about the people in the community and it is about relationship building. I know, this sounds like a sales seminar, but it is true. You are selling a service and you need the trust and support from the local community to make it a success.
I would have a harder time going into Dearborn or Trenton and asking those same businesses would you let me connect to your network. Yeah, right. I would be tossed out on my ear.
To successfully deploy a free wireless network in a new community, you are going to need local knowledge. You need an advocate that has local ties to the community and can make the connections and help sell the service.
A second important aspect of building support is for people to invest in your network. I am not talking about Angel investing or stockholders, I am talking about getting the local business owner to feel they have a stake in your success.
In Ypsilanti every person and every business is a part owner in the network. We do that by having them pay for the radio that goes into the window. Today the radios cost $150 for the indoor units. We bought a bunch of them early on when the price was lower so we sell them for just $50.
And we reward them by providing a link and graphic on our website and for those that also provide bandwidth, they get banner advertising on the Wireless Ypsi service.
But wait, I am getting ahead of myself.
Tomorrow, how you can setup your own free Wireless Internet service in your community and how to pay for the thing.
Questions or ideas, email me at Steve@Ypsi.com
Why is Wireless Ypsi working and so many other community wireless efforts failing? I am asked this question on an almost daily basis.
Is it because we are smarter than everyone else? No.
Did we learn from past failures? Yes, partly.
Was there a fundamental technology change in the business? Yes, definitely. More on that later.
Is it because the business' and residents have a sense of ownership? We believe so.
However, the real reason why Wireless Ypsi is successful is we didn't care about making a profit. If you don't care about making any money, it is much easier to create a successful wireless Internet Service.
Wireless Ypsi is a demonstration project to prove the technology and to show other communities use how they can deploy a wireless network that actually works and can be deployed today.
Today Wireless Ypsi has had over 8,300 different users since startup. We don't double count users. If you use the network today and then come back next week, we only count you once. We average from 400 to 500 users per day. We have 125 nodes with 33 separate connections to the Internet.
Having been in the Wireless business since 1999 and managed a successful wireless ISP in Chelsea for nearly four years I know how difficult it is to set up and more importantly keep a wireless mesh network going.
A mesh network is where each radio talks to every other radio it can see. For it to be a true mesh, the network must be able to have one or more radios removed from the system and the rest of the radios automatically reconfigure themselves to re-route traffics. You improve reliability by increasing the number of radios.
A mesh network is not the same as hooking up three wireless routers you got from Wally World and connecting them via WDS or bridge mode. In this case, if one node fails, the network stops functioning. While there is a mesh standard coming from the IEEE, it isn't here and even after adoption, it will be an additional year or two before manufacturers begin making mesh radios. Today most mesh solutions are proprietary.
Wireless Ypsi started in Ypsilanti in January 2008 when Brian Robb and I bemoaned the fact that one of our favorite restaurants didn't have Internet service. I said that I had 10 Meraki radios that I never setup. Maybe he could get them to work. I had received the radios as an early Beta tester in March 2007 but never did anything with them.
I have done hundreds of wireless networks; I figured I wouldn't see Brian for weeks. Brian called the next day and said they work! Sure enough, he had a working mesh network.
Learning from Past Failures
Southeast Michigan had one of the first wireless high speed networks. Called Ricochet, it covered from Detroit to Royal Oak and all along I-94 from Detroit to Ypsilanti. Sorry Ann Arbor, Ricochet never worked there. It cost $90 a month and you got 128KB service. Sadly they folded in August 2001. Ricochet spent $500 million to cover 17 metropolitan areas in the U.S.
In Detroit, they had 384 customers. That is not a typo. At $90 a month it would take 89 years to payback the investment here in Michigan. The problem with Ricochet was that they were competing with ISDN, DSL and cable --all faster technologies at a much lower cost.
Ricochet could have been successful if they had deployed their networks in communities that had no other service but dial-up. But competing against wired services, they could never be cost competitive.
They should have focused on tourist destinations like Aspen, Lake Tahoe, Traverse City, or Mackinac Island. Business would gladly sign up for the service and tourists and vacation home owners would also gladly pay $75 to $90 a month.
Then in 2003, Washtenaw and Oakland Counties as well as many cities across the county announced grandiose plans for Community Wireless Internet. And with few exceptions, they all failed.
They failed because they tried to be all things to all people. It many ways it was like a Swiss Army knife. These communities threw in every possible wireless use hoping to get these diverse constituents to kick in funds.
Police and Fire want Wireless for Secure Mobile Access. Add to the RFQ. Water departments wanted wireless for meter reading. Add it to the RFQ. Building Departments wanted it for plan review. Housing Commissions wanted access for low income housing. Add it to the RFQ. City Hall wants Voice over IP to replace cell phones. Add it to the RFQ.
With so many requirements, the costs grew exponentially and so did the complexity. So when the first networks were deployed, they didn't work.
Wireless Ypsi had one goal. Provide reliable wireless Internet service. No meter reading, no private secured network. No sharing computers or printers. Most importantly it wasn't intended to be your primary Internet service. If it worked, great, if not, oh well. We constantly remind people, the network is free, and it is worth what you paid for it.
If you need high speed Internet service for your business or a full time connection for school, Wireless Ypsi is not for you. But if you want to check your email, chat with a friend, or search the Web, Wireless Ypsi is perfect.
At Wireless Ypsi, we under promised and over delivered.
New technology changes the game
The Meraki mesh radios work without a lot of hassles inherent in previous mesh technologies. It isn't plug and play, but it is very close. Any competent networking geek with decent problem solving skills can get a Meraki mesh network up and running in an hour.
The system is fully managed via web control panel and you can throttle the bandwidth delivered to each user. This keeps bandwidth hogs from abusing the system and then, if you still have a bandwidth hog, you can block them from accessing the network.
Meraki has a fundamentally changed the way mesh networks are designed and deployed. For one the nodes are cheap. The indoor radio is $150 and the outdoor is $200. By comparison, just one year ago we were purchasing outdoor radios using similar technology that cost $1,500 each.
Wireless Ypsi is possible because the cost of deployment has dramatically dropped and Meraki has developed a system that actually works.
In San Franciso, Meraki has deployed a free wireless service that today has over 150,000 users. Meraki uses the San Francisco network to demonstrate how a large scale community WiFi network works and as a laboratory to test new products and features.
Tomorrow, I will talk about the need to build a sense of community ownership and how to finance a wireless community network.
Feel free to email me directly with questions or comments at email@example.com