Slashdot Mirror
Starting a Local Fibre Co-Op?
swordsaintzero wonders: "I have seen several articles mentioning local co-ops implementing fiber. I am moving to a smaller town to take a position as a Solaris Administrator for a large university. After calling around, I found out the fastest bandwidth package available is half the speed of my previous package. For all of you Slashdot readers, who have worked on getting fibre implemented for your town through a co-op, what was involved in presenting it to the city government? What would be the best way to get the ball rolling on getting fiber for home use implemented?"
As we all saw during the "broadband revolution" of a few years ago (even when Taco couldn't get BB), the big cities get the coolest tech first. They also tend to get blown up first during wartimes, but that's a risk we just have to take.
So to answer one of your questions: What would be the best way to get the ball rolling on getting fiber for home use implemented? My answer would be: "Move to a big city".
I'm just a comedian though. Hopefully someone insightful will post a more insightful answer.
Don't think that a small group of dedicated individuals can't change the world. It's the only thing that ever has.
'Cause frankly I don't believe it. A co-op to get DSL implemented on already laid dark cable is one thing. Most communities don't generally have dark fiber run to every residential address. If it's been laid it's being used. To create a co-op tasked with rolling out a community-wide installation of fiber to each doorstep, with or without the local government's help, would be an astronomically complex and time consuming task. Instead of trying to convince a town to fund such a fool's errand why not try convincing the university to subsidize business-class broadband at home in case you need to work from there ("If the system(s) go down at 3am I can get them back up and running faster if I can just log in from home instead of driving across town." etc...).
If the city were funding it, it wouldn't be a co-op, would it?
Another out-of-the-box answer to the question suggests itself, since the poster missed the other 50% of their former bandwidth: Get 2 lines and aggregrate.
Anyhow, if you want to share fiber to defray installation costs, just start talking to your neighbors about it. If you can get contiguous propery owners to collaborate, no right-of-way is needed, and it becomes much more practical. It's also a good way to get to know your neighbors. Probably, wireless point-to-point links will be a lot cheaper, however. Then suddenly you're a wireless ISP. Lots of folks are doing it.
-I like my women like I like my tea: green-
Why the hell would you even want fiber to the doorstep? You realize how much it costs to terminate the stuff? You can get a much better price/performance ratio from SHDSL if you really need quite a bit of bandwidth (it'd still be excessive for home use). Otherwise, just go with regular DSL, it's good for at least 3 MBits.
If you can get contiguous propery owners to collaborate, no right-of-way is needed
Depending on your state laws, no right of way may be needed at all. Lots of states are so pro-development that they'll bend over backwards to help you take the land you need. Most will happily fine anyone who cuts your lines as well. I'm looking at 150' power lines in my back yard that got built despite rejecting the power companies offer, and despite any condemnation procedures. As a property owner, there's technically not a damn thing I can do about it.
Fudging by way of drainage ditches and easements can also go a long ways.
Probably, wireless point-to-point links will be a lot cheaper, however.
For a few points, maybe, but for a neighborhood, I doubt it. Towers and antennas add up quickly. In a neighborhood with reasonable density and reasonable take-up, fiber should be doable, and scales much better.
"I assumed blithely that there were no elves out there in the darkness"
Price is one thing to worry about, but excessive bandwidth today is perhaps not enough in a few years (for some)?
:)
Remember those fast 56k modems back in the not-so-long-ago day?
I remember 2k!
Not Free SF Reader
While I don't have an answer for you, these people may have more information on the potential for this:
8 6.htm
e twork#Australia_.26_Oceania
http://www.bendigotelco.com.au/
http://www.communitytelco.com.au/
http://www.itel.gil.com.au/
News article: http://www.abc.net.au/southwestvic/stories/s12981
If wires are all too much, then wireless can be done too...
http://en.wikipedia.org/wiki/Wireless_community_n
"Go to CNN [for a] spell-checked, fact-checked summary" -- CmdrTaco
I was used to high speed pr0n, and now I'm stuck waiting 5 whole minutes for a movie to download. I'm usually done by then!
I want to delete my account but Slashdot doesn't allow it.
Like this much?
Not to be rude but that is hardly a comprehesive description on how to implement a fiber network in a neighborhood. All that page really does is lay out a few prices, which I might add are linked to ebay auctions.
Digging up peoples yards, laying pipe, installing equipment would all be time consuming to begin with but getting everyone in on it would be a chore in and of itself.
I thought I read something about this on Slashdot a couple years ago, where a guy and his dad got a whole neighborhood fibre'd up in Sweden. I can't however, find the link anywhere... Anyone?
I'm not even proposing internet access on that page. Without crossing a street, the most homes that could be hooked up in my neighborhood is 16, and that's assuming 100% participation. Assuming you did get that many, you might be able to bring in 20 or so ADSL loops, at 10mps a piece, you could have a decent upstream. That's assuming you could find a DSL company willing to do such business. But cheap multimode does scale pretty far, all the hardware's cheap...
If a few close neighbors did want to do such a thing, we could probably implement it for $100 per, one time fee. Gotta be something you can do with gigabit, even if its not an internet connection.
I get my power from a co-op. Back in the 1930s the farmers in my area realized that they would not be served by any utility because there was no profit in them. So they got together and started a co-op and wired the county. We are now one of the fastest growing utilities in the nation (Minneapolis is growing right into our territory).
There are likely to be some old laws on the book in your state that you can fudge to help you. Have a lawyer look them up. You will need a lawyer before you get very far, so best is if you can find one to be a partner now, and hire him.
- Ashland Fiber Network http://www.ashlandfiber.net/ (Ashland,OR) is shared by all utilities in Ashland (population @ 21k) -- it is community owned
- The Portland,OR metro still doesn't have fiber access
- Verizon FIOS is not planned to be installed in the Portland metro for quite some time (no ETA)
- World Wide Packets business model is to provide the infrastructure for ISPs to provide fiber to the end user, but they don't work directly with end-users
- 6 years ago [Beaverton,OR], I had Telocity SDSL for $20/mo @ 802/802
- 3 years ago [Beaverton,OR], when I tried to sign up for 1.5/768 Verizon DSL, we found out (after they set us up and started charging for said speed) that the best we can get is 384/384 for $60/mo ($40 Line[Verizon] + $20 ISP [Aracnet]) because of a 26-gauge wire between me and the C.O. They also stated there is no intention to ever upgrade that line unless it is damaged. Of course, they wouldn't tell me where it was.
- A few months after talking with Speakeasy about the last bullet point, they sent me a letter saying that they could get me 6MB/768 over their fiber network without using Verizon's phone line. When I called them, they said that they couldn't because it still has to go over Verizon's line. Of course, they still wanted me to try to set it up, without installing a replacement for that 26-gauge wire.
All in all, I think Ashland has it made. Community-owned fiber network that the cable companies and phone companies have to pay to use -- and you don't HAVE to go through any one of them to get service from a 3rd party. I wish Beaverton would do that.http://www.google.com/profiles/malachid
After calling around, I found out the fastest bandwidth package available is half the speed of my previous package.
Maybe I'm missing something, but the optics aren't going to get you any additional speed to the internet if the fastest connection you can find to uplink it is still slow.
Reminds me of my old school that touted FIBER TO THE DESKTOP!!! - doesn't make much sense when all you have is a T1 and 10k nodes on a single subnet.
I have never felt that a complex task should not be attempted merely because it is complex. I am sure I have no need to belabour the point with examples. While time consuming I did not specify a time frame. My initial thoughts on the matter pointed to a 6 year schedule from the start of planning to actual deployment. The point of the article was not to ask for simple common sense solutions (which I have already thought of) to the tenous issues that got me thinking about doing this, the point I suppose is that I want fiber and the only way to have it is to do it myself! I apreciate the reply none the less it was coherent even if it didn't really add to the discussion.
Panel F, Relay #70
That being said, one could easily provision linear fiber (non-ring) for a lot less, and if you and your neighbors care about speed, etc.. (sharing that mp3 server in your basement, or those trees keep getting in the way of the wireless) a short fiber run put in PVC or similar conduit outside buried not that far down (depending on how you want to do it) would provide high reliability over that short distance. A 3com 100BaseFX card costs roughly US$90, and cabling is not *that* expensive even for multiple pairs or outdoor/aerial cabling. Most of the cost is seen in adding connectors which range in cost as not anyone can go to home depot and get a cable crimper, connectors, or polisher. You may find the cost comparable to some more expensive wireless bridges or cheapos with nice cabling/antennas, but for reliability it's the way to go, and you're not too likely to need more expensive cabling unless you're going to do some fancy DWDM or other stuff over the short distance.
That being said, I may be soon using a power washer to dig a trench, or just doing it by hand with a shovel between me and my neighbor to deal with crappy wireless gear reliability.. This will let us communicate over a wired link without fear of the growing trees causing a problem. And the call to miss-dig is free. It also means that we can do fancier stuff in the future at faster than 100M and not have to worry about the limits of copper cabling as much.
Reminder to our readers, if you're going to go more than 2KM, you need single-mode fiber, you can't do it with multimode. You may also need to watch the type of optics you use (SR vs LR, etc..)
go with WI-MAX!!!
Verizon is starting to offe the fiber to home called FIOS.
New developments are already receiving FTTP as it can provide phone, data, and HDTV and whatever else might come along.
I would call the local phone company and see if you can work out a deal with them. Fibre is now cheap to implement and I see it as every home will have like every home has copper today...
There have been at least one good article on AskSlashdot before on this subject.
Akvo.org - the open source for water and sanitation
Here is a link to a co-op that laid their own fibre. Quite informative.
Akvo.org - the open source for water and sanitation
Here are some things one neighborhood did. I was the consulting network engineer in step -1, and I helped them get a 2400 meter monomode fibre from their division to a municipal fibre loop splicepoint. They pay about Eur450/month for leasing the right-of-way from the city, a single fibre pair in the loop, and a port on a switch with a 10Mbit commit. They have a /24 PI netblock hosted by their ISP. Startup costs were about Eur11,000 for 8 buildings, 4 apartments each. Every building has a multimode fibre terminating in a SX to copper el-cheapo switch. Every apartment has a copper GigEthernet connection to do with as they please, no caps, no limits, static IPs, anything but IRC servers.
-1) Before starting, hire (or at least talk to) a competent network engineer, with a working knowledge of fibre, used equipment, local ISPs, data centres and existing fibre runs. This may give you many ideas of how to get bandwidth to your neighborhood at a relatively cheap cost. Google can't help you for local fibre RoWs, that knowledge is arcane, closely guarded secrets.
0) Talk with a lawyer who works directly in "rights-of-way" and municipal law. Get accurate price quotes before proceeding, and understand the project will take a long time (6 months to 2 years) and cost a fair amount of money ($1000 to $2000 per household + monthly bandwidth costs $20/Mbit).
1) Once you and your neighbors understand the implications, costs, time line, and benefits, get each household to put up money. No money == no more participation in the co-op. Promises don't count. You should lock up around 75% of the initial startup costs in a special account, and provide a proper accounting to each investor at regular intervals. You'll need an external accountant for that function, don't let one of your neighbors say "I took an accounting course once, I'll do it for free", for down that path lies embezzlement and jail terms. Count on a second round of houses to join in after the network is up and running, and charge them more for the priviledge of not taking the initial risk.
2) After securing rights-of-way and approval from your local council, start digging. You will probably be forced to hire certified diggers who won't take a backhoe to the gas lines. If you are forced to take the aerial route, then you will need a contractor approved to work on poles around power/cable/telephone lines.
3) You need an external connection to an ISP somewhere. This will be the most expensive part of the project if there isn't one nearby. Negotiate an unlimited port on a router or a switch, and a block of addresses which you can route. Negotiate a reasonable commit price for what you expect to pull, and a burst rate which isn't too much more expensive. If you can't possibly get a fibre to an ISP, then you are stuck with a copper signal, which implies the local telco. Hopefully regulators in your area will force them to provide a T3/E3 at a reasonable price, but then you need a real router with an expensive interface card (or maybe two, if your ISP only deals with 100baseT).
4) You will need some kind of structure to act as your POP. This could be a concrete equipment vault buried at the edge of your division, or a shed on the side of someone's house. It is there you will have to terminate all the fibres from each house on a multi-fibre switch, and place a router to your upstream ISP. You may need an air conditioner if your summers are hot. Put in an electric meter so the co-op knows how much the electric bills are. Put a webcam so everyone can keep an eye on the kit.
5) At the houses, there will need to be a demarc point, where the co-op responsibility ends, and the homeowner takes over. Trust me on this, or your co-op will drown in a sea of pointing fingers and blame. At the demarc point, put a small SOHO 10/100/1000 switch with at least one SX fiber port. These are pretty common now, and you should be able to get them for around $150-$200. After that, its up to the homeowne
Hemos is like...sci-fi fans;he thinks technology is cool, but he hasn't bothered to understand the science it's based on