First I should explain that my experience is in fixed wireless networks[0] more than fiber.
First, I would spend some time on Google Earth looking for good neighborhoods to serve. A good neighborhood is typically a dense suburban development with low tree growth. Then I look for potential places to put the access points to serve the area - you really need line of site (no trees) to each home to provide good service. Water tanks, mountains/hills, office buildings and even residential homes make good access point locations. Often you can negotiate a rooftop lease with a place like that for a few hundred dollars a month (or even less). You can also use commercial cell phone towers, but they're more like 1-2k/month. You really want to be able to see a few hundred rooftops within 5-10 miles of your access point location to make it worthwhile.
Next you need to figure out how to get service to your access point location, like you mentioned. You really only need a few hundred mbps circuit to start though, a 1gbps circuit will last you up to a few hundred customers at least. Your first circuit shouldn't cost more than 2k to maybe 3.5k / month. The hard part is finding it in a place where you can then get it delivered to your access point location. Sometimes you get lucky and find an office building near a suburban neighborhood that already has a fiber demarc in the basement - that's a slam dunk. More often you have to lease space on an office building in the city, buy your fiber circuit there, and then use a high capacity wireless backhaul to relay the service to your access points. I would usually just drive around the main roadways looking for the orange stickered fiber whips to see where the fiber lines run, then call the company to get pricing. (can't find a photo at the moment, I'll keep looking.)
Then, you just start building and signing people up. Lot's of people make the mistake of trying to go cheap on the customer install, many WISPs start by trying to install their customers with in-home radios rather than putting them on the roof. That doesn't work. Regardless of the type of equipment you use, you have to put a good radio on the roof with a reasonably high gain directional antenna pointing to your tower. The other common early pitfall is to try to use regular WiFi devices for your network. WiFi is designed to work well in buildings / office parks, not so much outside and long-ish distances. Ubiquiti equipment is a good start and very cheap, although there are better ones.
The hardest part about signing up customers is paying for their equipment up front. You're looking at 250-350$ per customer installation (including amortizing infrastructure build) if you're in an area where the incumbent is really bad you can sometimes charge some of that to the customer. The ROI isn't too bad even if you eat that cost - less than a year - but it can be hard when you're starting out.
The next hard part is jumping over the scaling humps. You'll need a staffed support center and NOC before you're fully able to afford it, for example. Disclaimer - this is what I'm working on know, providing a variety of the backend services that these WISPs usually need / benefit from but can't afford to build on their own.
[0] https://en.wikipedia.org/wiki/Wireless_Internet_service_prov...
And how different is that from a neighborhood underground line?
This would be for Washington DC suburbs - Germantown, MD.
I mostly want to know about the backbone-level work. We already have pretty fast Verizon FIOS, so I'm trying to figure out what it would take to go to the next level - 10gbps, etc..
may be somewhat cheaper in US than Australia due to lower labour costs. e.g. Australia's minimum hourly wage is about double that in the US.
can reduce costs by ~70% if you are able to re-use existing duct / conduit in the ground, but you'd have to negotiate usage of that
I was told $15000, "most" of which was the trenching.
This is a rather built up suburban area.
If you were to bring in unlimited bandwidth to your complex and look at a traffic graph of the usage, those 70 customers would only use more than 100mbps for maybe one hour out of the day, probably more like 20 minutes. If you had a 1gbps link feeding them, you would only be slowing them down at all for at most one hour a day, and even then you wouldn't be slowing them down in any noticeable way. It only takes at most a few dozen mbps to stream a 4k movie, for example. There would almost certainly never be a time when one of your customers could not stream a 4k movie.
If it made you feel better you could sell it as a 900mbps product, or even a 500mbps product. Above 100mbps for residential it's really all marketing anyway - there are no services that a residential customer uses that provide a noticeably different experience at 1gbps vs 100mbps.
Also see sathackr's post: https://news.ycombinator.com/item?id=12520531
EDIT: Consider also that the fiber line you'd be connecting to is probably at most 10gbps, and might only be 1gbps. Residential / muni fiber networks (including Google Fiber) are using at best 10gbps fiber circuits to feed entire neighborhoods with thousands of subscribers.
A big (perceived by management) problem in my particular complex is phone and TV. Would you recommend VoIP for phone? What would you recommend for TV? Sticking with cable? Is there someone who resells cable over IP or something like that?
Netflix is good but a lot of people still want their sports.
And yes I would recommend VoIP for phone, there are several companies that will provide white-labelled VoIP service that works pretty well. You can connect it directly in to the wiring in the apartment so they can plug regular phones in to any jack in the house.
EDIT: Email in my profile, feel free to hit me up there if you'd like to continue this conversation.
Breakdown: Fiber circuit, up to 1gpbs, 2-3k/month. Should support up to 500-700 customers at least. Backhaul - $2k-12k each. Usually need about 1 for every 300-500 customers depending on network topology. Access Points - $200-$600 each. One for every 30-50 customers. Switching / routing hardware - $300-$1200 should cover up to 500 ish customers.
Per customer expenses: $200-$350 - customer CPE equipment and install accessories (cat5 cable, etc.) $3-$10/month - billing/customer management software, 3rd party phone support (if you do that), other misc stuff.
Then you need a ladder and tools and a vehicle that can carry them for doing installs and maintenance.
Obviously the risk depends on the competitive landscape. I've seen some people start by selling an apartment complex or office building or similar on an exclusive / semi-exclusive deal to provide service to the whole complex (which you can do considerably cheaper per customer than the above numbers) and use that to cover the basic expenses and expand from there, that's a good way to do it. In most places in the US people are unhappy enough with Comcast et al to give something else a shot. One issue is that people are usually in year contracts so you have to wait that out. If you have an existing consultancy or some community goodwill that you can leverage in to a customer base that decreases the risk, obviously.
How is the connection shared between customers? Can one kid downloading blu ray rips over NNTP eat half of that 1gbps?
What's the price range for the end customer?
4:1 on a 1Gb connection gets you 4Gb/s. Divided into 500 connections is 8Mb/s per connection. I've fought this battle before on HN, but everyone does not need a 150Mb/s connection. Yes, some people do, but most do not.
I haven't checked recently but last time I looked an HD Netflix stream used about 5mb/s of bandwidth. I think your assumption that half of your subscribers would be streaming video is a bit on the high side -- in some communities it may be accurate but not most by my observations. For planning, I would say 10%, maybe 20% tops.
If you're selling a 100Mb/s to residential customers, you could probably even oversell at 8:1 or higher, since their usage compared to total will be much lower.
Your oversell rate will also heavily dependent on how much your base is commercial/business customers. Residential use is peanuts during business hours, business use is peanuts during residential hours. So you can almost oversell by a factor of 2x over your normal rate for business vs residential, and it's also been my experience that as a whole, businesses use far less bandwidth than residential.
There are some outliers that will really skew your numbers. If you sell a 100Mb/s connection to a college that is providing it to their dorms with 1000 students, you can almost guarantee they will be nearly continuously saturating their connection.
Edit: adding answer for 2nd question. For the most part, without any throttling or queuing users downloading will get an equal share of the bandwidth available. 10 simultaneous users would get 100Mb/s, 100 simutaneously downloading users would get 10Mb/s each.
Upload is what will kill your bandwidth. One user saturating your connection with outbound traffic will slow everyone down.
But most ISPs queue/rate limit customer traffic. It's integrated into most commercial wireless equipment and routers. On top of that, depending on what technology you use, CPE(customer premise equipment) generally doesn't have the ability to pull more than 100mb/s. Some are lower, a few can go up. I'm not a Mimosa fan yet, but there is a guy woth a usually well informed blog[1] that sings their praises in providing multi-hundred megabit service to residential customers. My experiences with Mimosa gear have been such that I can't recommend them right now, but they have potential.
EDIT: I have a few other ideas/suggestions depending on your situation. Email in my profile if you'd like to continue the conversation.
i consulted them on biz and sales side so i don't remember their gear, but i remember both had quite a basic setup that worked pretty well. mind you, it was in urban areas (a 600k city in bolivia and a 2m city in serbia). serbian guys gone as far as building a file sharing network and setting up game servers
Also, how would you go about making an ISP for a dense urban area with no real higher-than-everything-else structures? Like, say you wanted to build out fiber in Berkeley: there is dark fiber going down the middle of the city, you'd need to get a physical wire from the cable to some room where you split it out in to other cable bunches? Is most of the difficulty in doing so from the difficulty in physically putting down wires?
Not necessarily, depends on what equipment is used and what the topology of the network is. It's not impossible to get 40-50ms latency on WISP networks out to Google, which is similar to what I get on Comcast.
To your second point - my specialty is wireless, but the reality is that wireless doesn't work everywhere. If there really is no single location that can see at least a few hundred homes with full line of sight then wireless probably isn't the answer. So then you're looking at fiber, and I think you're exactly right - the difficulty is in getting that wire down the last mile from the center of town (or whatever) to the customer. Sorry I don't know Berkley very well.
Properly done, no. It depends on several factors.
I have wireless links pushing hundreds of megabits per second across several miles with sub-millisecond latency.
Are the margins really razor thin? I always thought I've been getting price raped for years. $100/mo to provide a service that's just a pass through doesn't seem like thin margins.
I pay €1/day for gigabit service, so yeah pretty thin.
edit:typos
The convincing part is the hard part at least where I live. There's 13 different ISP that offer fiber service, a cable ISP and a dozen or so DSL providers. You can get gbit service starting at €1/day.
With about two dozen ISP's available at most locations and several different plans per ISP it has become so difficult for consumers to choose that several websites have popped up allowing you to compare all the different options available at your address.
We're still trying to solve the 'last mile' problem.
The irony for Phoenix suburbs & county islands is these areas haven't been "rural" since the urban sprawl encompessed them in the 90's. Yet, the baby bell won't invest in infrastructure where people have no choice but to pay top dollar for pre-existing shit.
Wut ? There are 13 fiber ISP's that offer service at my address alone, then there is a cable ISP and at least a dozen or so DSL providers. You can get gigabit fiber service for €1/day.
> The big cable co's have de facto monopolies and can charge whatever they want, because it's so expensive to run lines through the ground.
Why would you need to run lines, the fiber is already there and completely open for anyone to use. The fees for access are the same for everyone no matter how big/small you are so you have a level playing field.
And that's just from my casual knowledge.
(This reminds me...Vivint aquired SpaceMonkey and there was talk of including SM's distributed storage devices in the routers.)
The ONLY internet access available was provided by the apartment complex itself. I think they had a few T1s split between a couple hundred units. Even 10 years ago, that was still embarrassingly bad. It was unusable in the evenings or weekends. You could do a traditional ARP spoof and read everyone's traffic. Really bad stuff.
I had no option but to deal. I set up a VPN to protect and compress my traffic and spent weekends at the homes of friends and family.
These days, I hear wireless ISPs are far more versatile and reliable. They're great options for people who get stuck like I was (as long as your apartment points the right direction!).