Ask Slashdot: What Is the Best Way To Become a Rural ISP?

hawkeyeMI writes "I live in a small, rural town nestled in some low hills. Our town has access to only one DSL provider, and it's pretty terrible. However, a regional fiber project is just being completed, and some of the fiber is in fact running directly past my house. Currently, there are no last-mile providers in my area, and the regional project only considers itself a middle-mile provider, and will only provide service to last-mile providers. Assuming this will not be my day job, that the local populace is rather poor, and that because of the hills, line-of-sight service will be difficult, how could I set myself up as an ISP? I have considered WiFi mesh networking, and even running wires on the power/telephone polls, but the required licensing and other issues are foreign to me. What would you do?"

Comments

Re:don't

[mwvdlee]

Please explain; how does someone become such a "pro"? Is it perhaps by learning and doing? Or is it by giving up before you even start?

Find someone to help

[ccguy]

Unless you really want to be a one person ISP, which seems like a recipe for disaster before you even begin, find someone else to help. To me it seems like there's 3 primary roles: a) Someone who bankrolls it, b) Someone how deals with bureaucracy (licenses of all kinds), and c) Someone who at least has some technical knowledge to figure everything out.

I guess you'll hear stuff like "leave it to the ones to know to do it", etc. Fuck that. If there's an opportunity, willingness to learn, etc, go for it. Worst case scenario you will fail but probably will be the "one who knows how to do it" the next time.

Good luck :-)

Re:Find someone to help

[Anonymous Coward]

While you make a good point about how to learn about regulations and how to handle the process, you seem to have misunderstood the GP's point about the seperation of roles. He never said, or even implied, the rules were mutually exclusive. He merely provided a reasonable way of turning the problem into a couple smaller problems.

The submitter seems aware of the limitations on his current skillset in both the technological and paperwork side of the business. Splitting the main problem into smaller ones introduces the option of bringing in a partner which would split the amount of knowledge that has to be gathered/learned. As you said yourself, learning about regulations is a lengthy process and if the submitter has a friend who would be interested in that, the submitter can fully focus on the technical choices.

The submitter could easily both bankroll and do the technical side of the job, while the friend does the "paperwork". Or they could divide up the investment cost in any way that suits them, anything is possible.

Re:Find someone to help

[mknewman]

I was a one man ISP for 10 years. It was an outgrowth of my BBS that I ran for 10 years prior to that. It's fun but thankless, and competition is fierce. Packet traffic became commodity, which means minimal profit margins and large companies (read AT&T and before them Southwestern Bell) willing to come in and throw large amounts of money at stealing the market. If you want to sell DSL get an ATM T3 router and get a big PVC to a provider with some ATM bandwidth. Then allocate bandwidth to your customers on SVC connections. You will need some infrastructure servers, billing, DNS, Web hosting, DHCP, etc. I'd say you could get into it in a minimal way for $100k and a few employees. If you are wanting to do a one man operation you are basicly nuts, you will want to sleep, take a vacation or a night out. I made that mistake and would not recommend it to anyone. Contact me if you want more insight. I have a reasonable hourly contract rate :)

learn about 501(c)(12) organizations

[emes]

I would suggest you learn about what are known as 501(c)(12) telecommunications cooperatives. One specific example would be www.rric.net
It would also be good for you to consult the IRS information on this kind of nonprofit organization.

Re:don't

[rtfa-troll]

There are many routes in. You get working for a big ISP then work your way towards a top technical job. You take a degree then get in more directly. Alternately, you stat doing small scale semi-amateur stuff for some years connecting up e.g. local charities and stuff. What you don't do is start without a good idea of the business and and technical side unless you can safely sustain yourself with no return for at least five years. Firstly there are huge barriers to entry. All the good sites for transmission likely are already taken, for example. Secondly the customers are pretty demanding; fail to fixa customer's internet in two hours and you've lost then. Even in the middle of the night. Thirdly the competition is brutal. A Place can sit without broadband for teen years then get the best internet in the country within weeks of a small ISP having completed their new installation.

My Suggestions

[DiamondGeezer]

Definitely it's a project worth doing but you've got to put in some work, both legwork and office work to make it work.

You need to go to the regional fiber provider and talk to them about becoming a last mile ISP and what their requirements are to terminate their fiber in your town and likely licensing issues, service contracts and support.

You need to speak to your town hall about permits and applicable laws.

Depending on where the fiber actually is, you need to pick a business unit where the fiber can be terminated and where your fiber can be run from.

In that business unit you're going to need reliable power and UPS backup to create a small datacenter (2 or 3 racks should be plenty) on raised floors for cable runs. (There are companies out there that ship all of this stuff in a single container, meaning that all you have to do is site it and run fiber and power to it)

You'll need to find out how much it will cost to run fiber from your datacenter businesses (who will be the main consumers) and home users. Get maps and start planning. Your regional fiber network provider should be able to put you in touch with the people who put fiber cables down in streets.

You'll need to talk to your local Chamber of Commerce or Better Business Bureau about likely customers as well as schools and colleges (and the town's own infrastructure like the townhall itself) who will be big consumers of fiber bandwidth and likely to be the baseload of your cashflow. Also likely partners in your state who might like to put their systems in your datacenter to provide services to your town such as VOIP providers, cloud services and storage providers etc. (Speak to them under NDA)

You'll need a business plan, a financial planning showing likely costings and cashflow and a project plan to maximize return by hitting major sources of revenue first.

I would suggest that you go for a low cost base based on opensource software and hardware as much as you can (I hear cheers from Slashdotters!)

Once you've got this done, then find out about likely sources of finance, microloans, angel investors who will need to see the proposed balance sheet and cashflow projections. (You might find that the reason there is only a crappy DSL service in your area is that that is all the demand that there is - economics trumps everything else and the whole idea has to make economic sense)

You will need help. Other people have done this on very limited budgets so use Google and network like crazy. Make contacts with technical people willing to pitch in. You will need to look at project plans created by others and business plans created by others and sources of finance used by others.

This isn't to put you off, but to give you an overview on the size of the mountain you're looking at climbing. Others have started where you are now and made great local companies. But the business must be based on sound economics and a steely concentration on a plan of action.

Options

[twisteddk]

Well, the projects I have been running have all been based on either wifi or xDSL. So I can pretty much only provide my expertise in these areas, BUT......
I see a couple of possibilities:

1) Contact the preexisting DSL provider in your area, and tell them that fiber is now available in the area. Ask them if they would be willing to provide a new DSLAM in the area connected to the fiber, which would boost the speed of the internet considerably (if the DSLAM is within a mile or two you should easily be able to get a stable 20 Mbit connection, which I assume is better than what you have now). Its always easier to lobby someone else to do the job they're supposed to, than it is to start competing with them......

2) Contact the people providing the fiber and ask them what servicepartners they have that are last-mile providers. Contact some of them and ask if they would be intrested in setting up shop in your town. Get the local populace to sign a letter of intent, that they will switch providers, if they can get better or faster internet at the same or lower cost..... Again with the lobbying, but it's an easy way out

3) Consider setting your own lastmile service up. But use xDSL connections or wifi, because FTTH would require that you start digging fiber to each house. I doubt you could make a profit on that if you're a one-man operation. In a hilly area, get a permit to set up repeater antennas on the highest areas. I'm sure you have cell service in the area too, so ask the local cell providers if you could use their towers. Usually, they have the permits in place, and you'd just have to pay rent, or simply swap services with them (your internet for their towers), if you can find someone who'll go for a straight swap..... This option requires a lot of footwork, and negotiation, but it's possible even in an industrialized and regulated society, it's just a lot harder than in Africa ;)

4) Get a group of friends together and work out a division of labour, make plans and set them into motion... More people = less burden on the individual.

I'd say it's possible, but if the market was big enough that you could live of it, then I'm sure you'd have more than one provider covering your area at the moment. So dont expect to get rich in anything except experiences :D

WISP is the only real option.

[wvnet]

I'd go fixed-wireless. It's the only option that you can start on a shoestring and end up with a decent business. Tapping the fiber can get quite expensive. It probably goes through the local telephone Central Office, so your best bet is to find cheap office rental as close to the CO as possible, and then contact the middle-mile provider for a quote to run you a drop. Bonus if you can rent a space in a muti-story building and arrange roof rights for a few antennas.

But this is doable, if you are serious about it.
Ubiquiti wireless gear is the way to go right now, and there's lots of technical help on their forum and others. Their 900Mhz gear will handle SOME tree coverage, as will the 2.4Ghz. Their gear is so cheap that you can afford to make little house-to-house relays to get into hard to reach spots. Their wiki has a decent write-up of how to build a WISP with their gear.
http://wiki.ubnt.com/Building_a_wisp

There are lots of other gotchas in the biz, arranging tower sites (private landowners are good, but you'll need a solid contract), getting customers to actually pay you (at all, not just on time), each install is going to have to be paid for up front ($150-200) and you won't make any money off that customer for about 6-8 months, service truck & tools, insurance (wispinsurance.com) and lots more.

Go lurk on the Ubiquiti and Mikrotik forums for a few months, and you'll start getting a clear picture of what running a small ISP day-to-day is like.

Re:Ubiquity

[Anonymous Coward]

I did the design and operations work for one of the largest WISP builds on the planet.

There are lots of WISPs out there, most of them mom and pop shops. You can make a few bucks but it is very hard work. Do not underestimate this.

Ubquity is great value for money and is perfect for your situation. They've got good forums and you should have no problem getting something going.

Please be safe when erecting towers or other antenna supporting structures. If you make a mistake, you can die. Have someone who knows what they are doing show you proper techniques for working on poles, poletop rescue, etc. Do not skimp on this.

Good luck!

Pretty simple

[interkin3tic]

You've already completed step one, which is to move to someplace rural. Step two is to become an ISP. So basically, you're halfway there.

Re:don't

[ILongForDarkness]

Of course their fake I saw Leonardo DiCarprio in movies that came out years later. His death was a hoax.