Rural residents' DIY 1Gbps fibre project

[zantzinger]

A community in Lancashire, England, has taken it upon themselves to dig, lay and build a complete 1 Gbps FTTH network. Members of the community have been able to buy shares to finance the project, and then they've mucked in to carry out all the leg-work themselves, with farmers digging trenches and other locals installing the technology. I find it very inspiring: especially given I live in London and still can't get FTTH. Business plan, videos and info on their website: http://b4rn.org.uk

EDIT: title changed from "...DIY Google Fiber project" to "...DIY 1Gbps fibre project"

[nemilar]

This may actually be much easier in a rural area, as opposed to urban/suburban.

Rural landowners have large tracts of land, and the land is in fewer hands. You need less individuals to buy-in, since a single buy-in on the project can mean miles of line.

You don't need a jackhammer to dig up sidewalks and streets, and you (presumably) don't need as many (if any) permits; you don't need to take as much care not to disrupt existing utilities, and you don't need to worry so much about noise and other regulations.

[kbob]

The ratio of landowners to households should be very close to 1:1 in the country. (-: In the city, there will be somewhat fewer landowners because of multifamily dwellings (apartments and similar).

I agree with your permits and jackhammer points.

[ricardobeat]

Shouldn't this be a very attractive market for commercial ISPs then?

[InclinedPlane]

Of course not, the expense of the fiber and equipment itself is the same but the number of customers and thus the maximum revenue is far lower.

[kalleboo]

Communities like this were a factor in accelerating the broadband buildout in Sweden as well.

Here, when we broke up the telco monopoly, instead of splitting it up regionally (like AT&T), they were split into roles. So after the splitup, the ownership of all the POTS copper and related infrastructure (buildings, backhaul etc) ended up as a separate entity[0] who had to treat everyone the same and couldn't favor a single telco.

Come DSL. To get DSL, this company obviously has to upgrade the local switches for DSLAM operations, and for this operation they would charge the first telco to request a DSL connection for the upgrade. In the DSL buildout rush at the turn of the century, the big telcos obviously focused on densely populated areas. So people who lived on switches that were too small to be attractive to any big telco would start up a coop like this to save up to pay for the upgrade themselves, and there were several small broadband operators who would take care of the technical, broadband-providing bit. And once your switch was upgraded, suddenly you could get your pick of any of the DSL providers.

There were/are also a ton of small, local wireless providers who use directional antennas to service people who live too far away from a DSLAM.

[0] http://en.wikipedia.org/wiki/Skanova

[Hoff]

B4RN is using burial, which is a great approach for reducing long-term maintenance, but the extra costs of burial may or may not work in other localities.

Burying wires in gravel, or burying below the usual furrow depths in farmer's fields is fairly easy, and trenchers are commonly available. (And if you bury your wires outside of the growing season, you won't disrupt the farming.)

In an area of the US that I'm familiar with, the granite ledge and glacial erratics (big rocks) would make burial problematic. Nearly all electric and telecommunications distribution wiring is accordingly overhead; on poles. With the problems that brings; ice storms, trees, etc.

Rural distribution does have some advantages in terms of not digging up existing infrastructure, and the ability to easily trench connections across the gravel-surface roads, but the cabling distances are usually (much) longer, and the numbers of and densities of houses and potential subscribers are (much) lower than in urban areas.

This means that the survival economics of the installation are fully in play; how many subscribers you'll gather, and how much to join, and how much to maintain the (overhead) wiring. This is why many rural areas of the US will tend to have DSL broadband, at best. DSL is cheap(er), and it uses the existing copper.

As with the rural electrification and rural telephone efforts that preceded this in the US, the stumbling block for rural broadband is the (large) installation costs; they're not a financially viable undertaking for commercial entities.

Incumbent telcos are seeing their business go to cellular, and wireline subscribers are dropping. Some of the telcos have been in and out of bankruptcy. Because of the budgets and the wireline subscriber trends, incumbent telcos also aren't inclined to install wiring ahead of a requirement; dark fiber isn't commonly installed during repairs in this area.

Yes, you might hope to see some of the incumbents build out broadband as a way to stay relevant. But in rural areas, that build-out involves greater distances and lower (potential) customer densities, and with higher on-going maintenance costs. And around we go...

[morphle]

We build a fiber optic link of 10 miles in rural areas for less then $400 per house. I think many of the assumptions you make are incorrect, as we find no rural area that is to expensive to service. Just don't ask the incumbents.

[Retric]

The problem with incumbents is there already making money selling the old product which means Fiber instillation needs to be paid back on their 'extra' profits from running Fiber.

[EliRivers]

Here in the U.S., the local monopoly would have had the law changed to make this illegal :)

[profquail]

Like this?

http://www.engadget.com/2009/04/22/time-warner-and-embarq-ca...

See also: https://savencbb.wordpress.com/

[sdoering]

Probably not only in the US.

[sdoering]

What I really like about projects like this is the fact, that local communities organize themselves and get things done, that the bigger entities are not able to tackle.

The more I read about such endeavors and the more I think about it, I believe, that local communities in a connected world have more power than

a; they thought for a long time, due to the fact, that a lot of things are decided far away. b; a lot of bigger players would like them to have.

They might not be as "efficient" or as "moneymaking" but that might be actually a good thing.

For example, building this network underground is much more resilient, than on poles. But I also believe, that the bigger players, will try to enforce their role, as such initiatives get more common.

The same might happen, as power generation becomes much less centralized or as food production (hopefully) returns to a much more local situation.

I really hope to see more such initiatives, not only in Britain, or Europe, but on planet-scale.

[geuis]

Interesting project, thanks for sharing.

The link, http://b4rn.org.uk, should have been directly used when submitting this.

This is not digg, nor slashdot, nor reddit. A description of, or commentary about, what you are submitting is not required, as the content should speak for itself.

[vitno]

How is this "Google Fiber"? Can't we just call it fiber?

edit: title changed, thanks!

[Retric]

I think they meant "Google Fiber" as "Google class Fiber" meaning that 1Gbps standard. But, I agree "1Gbps Fiber" is clearer.

[johnohara]

I applaud the effort but not the method.

The last video on their website shows the trenching, laying and backfill operation. IMO, it's better to lay a 2" or 4" continuous PVC pipe between the access points rather than lay the outdoor fiber directly into the trench.

It's better for protection, diagnostics, additional capacity, and if needed, other types of cable (twisted pair, coax) can be run alongside.

I think what they are doing is great, but seeing the workers walking all over the fiber while backfilling the trench made me shiver. Isolating and repairing a break is going to be difficult downstream.

[zzuser]

Trenching a conduit is multiple times more expensive than laying fiber direct. More than likely using conduit would not be cost effective. 

With a conduit you have to: (i) purchase the conduit, which is more expensive than the fibre cable itself, (ii) install the conduit, (iii) blow the fiber cable into the conduit and (iv) install all the support infrastructure required for a conduit system, such as manholes and cable wells. 

Furthermore access points have to be installed at close enough intervalls to make air blowing or water jetting possible. In rural settings this might well be far more often than otherwise needed for a direct burial network. In all the steps along the way, a conduit based network drives up the cost, resulting in a multiple on the buildout cost.

Protection, diagnostics, additional capacity and other media. None of the aforementioned are necessary any better or easier in a ducted network. A duct does not protect against backhoe fade any better than a proper direct burial cable. Diagnostics are done the same way in a ducted and a direct buried network. However direct burial cables with metallic streng members are easier to find than a non-metallic ducted cable. Adding more cables to a small diameter duct is not always easy, feasible or straightforward, especially if the duct was not properly or professionally installed. As to other media, such as metallic cables, the whole point of installing a fiber network is to only need to maintain one outdoor plant, not multiple legacy networks.

In addition to the extra cost, there is a lot of fun and games to be had if you install ducts in regions with subzero temperatures. Water is going to get into the ducts and then you are going to have giant icicles with cable in the middle and duct on the outside.

Walking on a proper direct burial fiber optic cable is not even going to register. Anything less than a backhoe is unlikely to even leave a mark. On the other hand if you even look funny at a ducted cable, you might have problems.

Repairing a ducted network or a direct burial network is basically the same. You excavate the problem area and you replace the faulty parts. The main difference is that in a ducted network, you are going to spend most of your time fiddling and fixing the ducts, less time with fixing the fiber optic part.

Now, what a ducted network gives you, is flexibility, but at a greater cost. If properly built that is. A ducted network means you might not have to retrench later, but this might not be an issue in a rural setting.

[kbob]

There is some financial trickery going on here. It's buried in Section 6.1.3 of the business plan (pp 23-24).

Any resident who invests £1500 or more can nominate a property for free connection and one year's free service. The tax authority gives a tax credit of 30% of that £1500, and the investment can be sold back in year 4 (2016?) at full value.

So, pay £1500 (or more) in 2012, get £450 in 2012 tax credit, then sell for £1500 in 2016. That's a 9.2% ROI plus free Gigabit Internet service. Of course, there's risk involved, just like any startup investment.

[morphle]

We run a couple of startups like B4RN, one in the US, one in Europe en one in South America. We build our own 10 GBps routers.

Come join us, we have room for extra founders. info at buurtnet dot org

[wcunning]

I see nothing on that site about the US. Where would I find such information?

[morphle]

I'm not displaying any info on the US on our sites, as we are still in stealth mode and have fierce (but toothless) competitors. If you want info, email me at info at buurtnet dot org

[jiggy2011]

As someone who grew up in a rural part of the UK, I can empathise with how annoying it is to be limited to a 56k connection with daily timed limits (usually around 3 hours or so) to be connected when all of your city dwelling friends have 1mbps broadband.

I wonder if this is as much to do with property values as anything? It must be very difficult to sell a house with no option of fast internet and the modern internet must be basically unusable on 56k.

[andyking]

I live in rural Yorkshire, and there's a village about five miles from me where broadband is not available - at all. There's no FTTC/H, no ADSL, no 3G coverage from any network, nothing. Residents still use either dial-up, or expensive, high-latency satellite connections from an outfit called Tooway.

It's a beautiful little village, not bad for commuting into the nearby city of Sheffield, but last time I looked, there were several houses to rent, that had been up for rent for over three months each, in a rental market that's generally red-hot. The lack of broadband is surely an issue. I'd move there if broadband was available, and I'm sure a lot of other people are in the same boat.

The issue here is that the rest of South Yorkshire has now been wired into a publicly-funded FTTC operation called Digital Region (http://www.digitalregion.co.uk/) which I'm using right now. The connection quality is good, I get a full 40Mbps with the option of raising it to around 110Mbps for more money - but the network is failing financially, because they've wired up places that are already connected to BT "Infinity" FTTC, instead of those areas which don't get broadband/fast broadband. My small town is one of the few places which has Digital Region and not Infinity, and I don't know anyone else with a DR connection.

[morphle]

If you tell us the name of the beautiful little village, we'll be happy to come build our 1-40 Gbps fiber in this village. Finding a local contact to help us get started would speed up that process.

[ianlevesque]

It's really unfortunate that there are still places where people casually call 1 mbps "broadband." It's the same in rural areas of the USA though.

[jiggy2011]

This was ~2002

[walrus0]

If you go active ethernet, not GPON/EPON...

This is now particularly economical for two reasons, aside from the cost of the fiber to the home itself (either aerial or buried, which is going to cost more for labor/bucket trucks/trenching than for the fiber itself). Why?

1) Inexpensive good quality 24-port 1U switches from Taiwan (Planet.com.tw) with 24 1000Mbps SFP ports, 10gigE uplink.

2) Inexpensive $65 1000BaseLX SFPs

3) CPE routers like the Mikrotik RB2011Ls which have a SFP port.

[alt_]

I helped build a couple in Finland back in 2003. The lack of bureaucracy and city infrastructure in rural areas likely helped, but the driving force was my uncle - a CS professor/farmer who retired from teaching and had to find something else to occupy his time ;)

[webmonkeyuk]

This is a great initiative. I wonder if there are any plans to bridge sections with WiMax or similar and/or break out from 1G fibre to Copper GigE or wifi in population points to reduce the client-side install and equipment cost.

[morphle]

We use some of these tricks to reduce cost for people who don't need 1 GBps. But serious users all want 10 GBps fiber. We deliver that over 4 redundant connections of 10 Gbps each.

[cheez]

Haha, I just got a picture of "There Will Be Blood" 1Gbps version :)

[JoshTriplett]

One notable thing I don't see mentioned anywhere on the site: where do they plan to get their upstream bandwidth from, and for how much?

[morphle]

We buy our transit and our peering at neutral not-for-profit internet exchanges.

[Rastafarian]

It's in the FAQs.

[ommunist]

Thnk you for sharing. The business plan is a very eloquent document, that helps.

[cyberdoyle]

Power to the People!