Hacker News new | past | comments | ask | show | jobs | submit login
Hate Your ISP? Maybe You Need Community Fiber (eff.org)
355 points by jswt001 on Aug 1, 2014 | hide | past | favorite | 51 comments



I once lived in a city with a community run fiber network. It was pretty damn amazing. They still have great rates for what you get: http://www.sfcn.org/sfcn/internet/

I paid about how much the "Plus" plan cost, without TV, just for 3Mps down/1Mps up in the next two places I lived. It really does spoil you when you've had a proper ISP compared with the TWC/Comcast/Cox monopolies that run in other locales.

Edit: and note, this was a tiny (less than 30k residents) at the time I had it. Incredible value for the service. I'm sure there's lots of happy customers there still.


I think it's telling that they list the cost of a static IP on their home page. My local provider doesn't even list it as a service; and if you find a rep who can sell it to you it's a "business-only" service AND costs you around 60 USD per month.


I wonder if there is a way to adopt something analogous to a franchise model here. It strikes me that only the communities that are really, really engaged are going to be able to pull this off (which is true of most public works projects). That being said, it would be great to not have every small town and borough have to figure everything out for themselves. A franchise operator for a fast food restaurant doesn't have to worry about where to buy burgers, the brand of deep fryers, or even the optimal amount of time to fry the tater tots. If there was a full on recipe for doing this that is proven to work, a small number of citizens might be able to make it happen even in places where the local government is ambivalent. A proper setup would include some national branding to help with PR (I.e. Every article in every newspaper about community broadband would use the same name for it). It would include marketing materials for the public as well as briefing packs for local officials, legal briefs for attorneys, etc... On the technical side it would recommend specific hardware, suppliers, and designs.

The advantage to such an approach would be to harness economy of scale at a national level, but allow local communities to drive the process and own the network.

I'm not sure how you would structure it to avoid creating another monopoly in the parent company of the franchises. But perhaps a nonprofit or a corporation 100% owned by the community franchisees would work.


I am involved in a local community fiber project- DadeCountyFiber.org and I have been thinking along similar lines. I think an association or coop model may work best. I am actually considering a proposal to organize a Fraternal Society that would operate with local chapters controlling the local networks, and leveraging the resources of each other. The Fraternal Society would provide education and social activities to members, and encourage the open development and collaboration of the network.

Any thoughts or ideas are always welcome!


The main need for community fiber is because we've proven that private companies cannot be trusted with a broadband monopoly/duopoly and yet to allow more companies to run lines would create chaos. For that reason, the community needs to control the last mile. But the community doesn't need to be the ISP. Without the barrier to entry that is running lines to residential customers, many smaller ISPs will be able to compete with each other and you'll see the market actually work the way that it's supposed to.

So run the fiber to people's houses and charge ISPs a fixed fee per customer that covers the cost to maintain the network and, potentially, recovers a bit of the initial expense of the network. But then let customers choose their ISP based on speed and features. The worst thing you can do is replace the crap fully-private system we have now with a slightly-less crappy government monopoly.


I tend to like the open access model you are proposing. I am very inclined to operate the network on an open access model. However, depending on our funding sources, we may be forced to be an ISP and phone company to get the funding. Right now we are exploring the FCC's Rural Broadband Experiments. In theory, we could have a partner service provider handle the phone service requirement, but it may not be as easy to do in practice. Thanks for commenting. Even if we do operate our own ISP/Phone Company, it is strongly likely the network will be open access.


I've thought the "open access" type model would work for a couple years now and I'm glad to see it being discussed here. Community Fiber may work in pro-active cities. However problems may reveal themselves in other cities when the politicians make campaign budget promises and usage statistics make some people angry they're paying for people who use more internet than they do. You can't expect every mayor of every small town in the US to understand enough about fiber equipment to make the right purchases for their local infrastructure.

The fundamental problem is the cost of the last mile infrastructure is too great to allow much competition. (Though existing companies have worked with lawmakers to further cement their position.) Open access would allow companies to compete with their service, knowledge, and connection to the internet at large, while reducing the cost of gaining you as a customer to plugging you in at the central hub.

At one point long ago our nation decided that broad communication networks which included every household benefited the whole enough that shouldering the upfront cost was far worth the rewards. We wired up every household with a copper line and access to phone service. Now that the internet has surpassed and commenced the phone we have forgotten the lessons of the past and are struggling along with "good enough" copper instead of again shouldering the cost and laying down fiber lines. It would cost a lot to reach rural areas and connect every last household. However fiber is great in that you can upgrade the equipment at the end to continue to increase speed, and in a country where websites can assume all of their potential customers would have very fast internet radical technological advancements could occur.

The technological advancements will occur in the end. The biggest question is whether or not the US will reap the rewards that it has been these last several decades, or if another country will. There is no faster way to irrelevance than doing nothing.


Where it only makes sense to build one of some piece of infrastructure, isn't it really obvious that it should always be owned by the people? I suspect that this applies to quite a few things: cable data networks, wireless data networks (phone masts), electricity transmission grid, railway lines, roads, etc.

Historically, national infrastructure seems to tend towards becoming a monopoly anyway under private ownership. Further, when these are privately owned, we seem to end up with a lumbering private company regulated by a lumbering state bureaucracy which duplicates a large part of the accounts department of the private company to keep them in check.

This means that we have the worst of both worlds; no real choice, patchy service, high prices for the service and a whole load of hidden regulatory costs which come out of our taxes.

Anecdotally, I've consulted for a state regulated privately owned monopoly company, there is no way that private sector efficiency is driving these places, the bureaucracy is intense. Example: A decision from the company on whether it was acceptable to use a different, cheaper but better looking style of suspended ceiling took 6 months and went all the way to the board of directors of this multi billion pound company.


I'm especially enthusiastic about their endorsement of open access.

I'm generally skeptical of such things, but I can see the point in local government (community != government, that's newspeak) taking a lead in running fiber in much the same way they do electricity and plumbing, mostly because it seems that fiber is sufficiently future proof - it seems plausible that transmission technology on fiber can keep up with domestic applications without having to rely on horrible stop-gap tech like ADSL.

But one thing is laying a cable, that's a largely neutral and objective undertaking, another is running a good ISP - that's a completely different beast. I strongly doubt a government run ISP is going to be great, especially if something happens and it turns into a cost centre, or if it's "awarded" to the highest bidder to run, or stuffed full of overpaid incompetents by way of political nepotism. As with anything government run, when first things starts going down that route, it's incredibly difficult to reverse.


This is exactly what we've been doing in Sweden the last 10 years, at least.

Local electrical companies usually offer a city wide fiber grid in small towns. Sometimes housing companies that own a few apartment buildings have their own fiber grid. And ontop of that there are communities in more rural areas that come together and pay for the work to have fiber dug to their houses.

But that's just fiber, they still need an ISP for internet service and there we usually have at least a dozen different companies to choose from.


"Community fiber" is working towards a practical implementation of ambient connectivity that Bob Frankston's been talking up for years through IEEE and elsewhere. Some of his essays lay some good groundwork for governance and expansion. http://frankston.com/Public/


I've been wanting to push for community fiber in my area, but I have no idea where I would even begin. Are there any resources available for those of us who want to get involved?


Just as an aside on one of the points they try to make (not about residential internet access):

"Imagine the Director of Public Works using her smart phone to reschedule all the sprinklers in the city with just a few clicks."

This makes me really uncomfortable. I think infrastructure needs to be physically separated from the public networks. Everything else just asks for somebody to cause mayhem, un- or intentionally.


So, would you propose a dedicated physical wire for every service?

You have to begin weighing the differences between physical vs. logical separation in network design at a point. MSR did two really good studies on network reliability, when they found that larger Aspen trees are significantly more reliable than smaller ones. If you're to build a large tree for every system, you're going to have a prohibitively expensive infrastructure.

In addition to this, the complexity of having multiple bespoke systems for every time someone calls something special, and dedicated results in higher system, and network administration cost. Because the scales of various networks are going to be different, their implementation will probably vary greatly as well, giving your engineers massive headaches. It's been shown that engineers, and software result in the highest number of failures in production systems.

Lastly, physical separation is no longer an easy excuse for security. At some point, that separation, intentionally, or unintentionally is going to get broken. It's best to begin operating from an idea that shared infrastructure will be more resilient, cheaper, and easier-to-operate.


> So, would you propose a dedicated physical wire for every service?

That's actually in a twisted sense what the article is advocating. Instead of relying on commodity communications, they advocate the city putting in their own infrastructure so they can do things like this.

The article is generally good, but that section is strange. You unequivocally do not need to run high speed internet to homes in order to move towards a "smart city" - wiring up those things they mention is orders of magnitude cheaper than running fiber to every home (funny how surveillance cameras is a suddenly a good thing, by the way).


I think what they are going for is: the Director of Public Works can call the sprinkler maintainers directly. Not that they literally have an app, just lower bureaucracy.


I live in a small village (2,000 people) and would love to implement something like this but the costs, as mentioned in articles about communities that have done it, seem prohibitive in the extreme (many millions of dollars). The article talks about underserved rural communities but how in the world do such places afford this (aside from the wealthiest locales, anyway)?


The main thing is economy of scale.

Fibre that has rolled out in Australia by the Government owned NBN Co averaged around $2300 per premises (for the access nework - you'd need to add on a bit for the transit backhaul and interconnection point with providers). We have a ridiculously high cost of labour too so I'm sure it can be done cheaper. So while it would cost millions, it's actually not that much when you spread it out across enough people. 2000 subscribers would probably be enough to get around that ballpark if most people were up for it.


Generally its best if an existing utilities company, like power, can be included as a partner. And then you can use existing power line infrastructure to guide the optical cables.


My understanding is they are heavily subsidized by federal/state grants, local community development loans, and tax breaks.


Community fiber optics is the way to go. In Bucharest, in 2000, when fiber was really expensive a small community, like a block of apartments, would get much better quality service by sharing a business type of connection than buying for each apartment a "home/personal" service. The initial cost is that someone needs to step up, build the infrastructure and start the sharing service. In Western countries this might be a little bit more difficult than it was in Bucharest in 2000 because of the regulation regarding wires, cable tunnels etc. However it is not impossible and the benefits of such a service in the community exceeds the initial pain of setting up the service.

In Bucharest, experience showed that once a community connected to internet via a dedicated fiber service, self maintained, the neighbouring blocks and adjacent residential areas start to press for connecting to the same service. Soon a network serving 30 families, grew to 100 then 500 or even more. A single connection became 4 load balancing the load. You had a problem with your connectivity? Just pick up the phone and call a guy 10 houses away from yours.

I may not be very accurate about the prices in 2000 but the first connection we had in our block was a 350 euro/ month + 150 set up fee for a 100Mbs bandwidth with no traffic limit. In my block 27 families out of 36 decided to join the network and the initial set up of the fiber + local ethernet infrastructure cost was split among residents. No filtering, no connection throttling, no interference. We were paying 15 euro per month to cover some maintenance cost and future failures of cables and routers. These 100 Mbps are not much they were 100 Mbps were steady. The thing is that when we upgraded our link to a 500 euro/month we got 250 Mbps. So bandwidth and cost don't scale linearly. The more people in your network the better the service.

Maintaining a large network can be a pain. However, if you give incentives to the sys admin, and the does a good job by deploying a network with little quarks, he can actually have a passive income without much of a hassle. Networks with more than 2000 families generally had a team of 1-3 sys admins taking care of the network full time.

We also had the traditional cable companies offering a package of TV + phone + internet based on coaxial cable. Their offering was so inferior to the "neighbourhood network". The large companies ended up by stepping up their game and offering similar quality services. They were actually struggling to keep up the pace. The local networks, created around 2004-2005 a backbone network which allowed every "neighbourhood network" to connect to the other networks via high speed 10Gbps fiber cables. This became later the ANISP a national association. Yes, the small networks were in the end bought by large companies but at least the standard was set high enough so that it cannot be rolled back.

http://en.wikipedia.org/wiki/Internet_in_Romania


No seed accelerator? No big name co-founder? No funding round? No buzzword-laden bullshit talks at fashionable conferences (ted)? What about 'professional' ex-lawyer CEO paid >$100K with no clue about networking? Saas? Cloud? can this even scale? Where is the growth?

One sys admin? 500 homes? bootstrapping? low prices? What are you? some hippie commie geek man slave? You want to actually work for a living of something?


I remember this, happened across country actually and we had one in our smaller town but the infrastructure between apartment blocks was built using coax instead of fiber.

A few years later the guy that started it initially(96-97??) wrapped everything in the first CATV local internet and tv network which runs even today.


How would these Internet "coop's" deal with the tech support? I would worry that regular users would constantly have all sorts of problems with spyware, viruses, cheap APs/routers and broken network configurations.


There are indeed a lot of problems with individual users. Small networks as such were generally maintained by sysadmins like teenagers/students/grads which sometime fix these config problems for a small amount of money. Plus, good tech guys were always requested by neighbours to fix problems as such, change a HDD, change a cable or a faulty plug.

It is indeed a local community type of network. Plus, in every family there is a teenager willing to step into network configurations. Teenagers, students and young professionals were those that made these networks so popular because they were demanding high quality connection and they were tech savvy. People like them are generally willing to do the hard work of learning something about setting a network IP to a router or AP. Setting a DHCP server in your network was something that released a lot of pressure in my network. Mac addresses were also used quite often to block unauthorised devices from accessing the network. You had to call your sysadmin and send / spell over the phone your mac address.

When everything was at very beginning, during storms a lot of equipments connecting two blocks were getting burned, UTP cat 5 cables broken and required fixing. People who started this kind of networks were to become country's first big generation of top notch dev ops who faced networks growing from 3 users connected by a BNC cable to thousands of users distributed all over a large neighbourhood.


I am sure their would be an enterprising local computer tech the community Internet "coop" could contract with and recommend for their users.


I ran a similar but smaller network back in high school. I was in Bulgaria but the setup was exactly the same. Me and several other kids wanted to play Starcraft and dial-up was bad. We bought a lot of cable and connected 3 apartment buildings.

Soon our neighbors learned about the network and started asking if they can join. At the peak we had wired about 20 families, this was all supported by me and a friend. The incentive for me was free internet and some side income.

In retrospect I find it amazing that such DIY approach brought us in the top of the internet speed rankings surrounded by the most developed countries in the world.


> "People love to complain about the speed of their Internet access and with good reason."

I'm sorry, but no, in my experience, they don't. I can't think off hand of any non-tech person in my area complain about their Internet speed. People pay $42/mo here for 9/2Meg and because it will play Netflix and get them on Facebook, it's good enough. I'm not arguing against a fiber upgrade, I'd love to see it, but in the end, non-techy people don't seem to see much advantage. You're not going to build a community network unless the community is engaged. How other non American countries have done it I don't know, how you get that level of engagement in the most individualized culture in the world is beyond me. Sadly, I feel the only way this would really happen in most of urban America would be to beat the incumbents on price. Which sounds possible, but not by much, especially with the fact that if you don't offer bundled cable-vision, people then have to break their other provider's bundle and pay more for cable elsewhere so the difference is even less. I hate to be pessimistic about it (I think gig everywhere could have huge economic advantages), but this is a major hurdle. People often hate their providers, but it's often due to price or customer service, not speed.


It has been my experience that "non-tech people" do complain about their internet speed because the major ISPs actually don't provide good enough service for them to play Netflix and get on Facebook as consistently as they want.

More importantly, what "non-tech people" don't realize is what they will want in the future. In 2008, they didn't know they would want to stream Netflix over the internet. I don't think there's a big chicken and egg problem with faster internet; people will want faster internet when they start using services that the current infrastructure doesn't support well. It has happened before and it will keep happening.


I was really hoping to see someone have a different experience than me, so thank you ;). Not sure why the difference though, perhaps we have the speed around here but just not a decent price. I've heard more non-tech people complain about the data caps actually, especially families with children who watch kids shows on Netflix.


My community might get Google Fiber soon. I still am pretty excited about that, but the more I read about community fiber, the more torn I feel.


I may be missing something but is the suggestion that local municipalities dig their last mile and run their own fiber?

That's ... Not likely to succeed surely? I assume the actual dig out is a specialised job, with no private hire market, the back haul ... To where and to whom? Do most towns know where their nearest telehouse/interconnect is? The support tasks are difficult - and scale pretty well, but not if you have no scale.

Ultimately this sounds like "hey let's build a public ally owned nationwide fiber network!" Why not start at that point?


A common approach is to incorporate a municipally owned subsidiary that hires the appropriate expertise. That's how the electric and plumbing networks were built in many towns, for example. A number have since been sold off (turned into privatized utility companies), but some are still city-owned. Other cities have gone for a pooled model, where the each city owns their utilities, but rather than administer them directly, pays into a nonprofit organization that provides the service/maintenance, thereby gaining some economy of scale (e.g. http://www.ncpublicpower.com).


That's what some of us in Seattle are pressuring the city to do. Seattle City Light is a power company owned by the city and is already doing extensive work for the smart meter rehab. The idea is, while SCL is out digging around, lay last mile fiber.


Ok - not a common UK model (at least I think not). But I guess it makes more sense.


I recall reading about this being done, honestly. The community actually running the lines.

I wish I remembered where, I'd link it.


There's a big ol thread [0] on this in NANOG with lots of juicy implementation details. I'd hesitate to use the word 'consensus' in relation to a mailing list in any way but my take home is that if the municipality sticks to providing L1 (i.e. a physical fiber as opposed to an 'internet connection') it requires a manageable level of expertise and still vastly lowers the cost of entry for ISPs to provide the rest of the stack.

[0] http://mailman.nanog.org/pipermail/nanog/2014-July/068527.ht...


There's a small town in the UK that ran their own lines, literally. http://www.tomshardware.com/news/Broadband-UK-Fiber-DIY-Fibe... :)


Do we have something like this in SF Bay Area?


Actually the article includes a link to a study on Community Fiber which included San Francisco: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2439429

I haven't read it, but maybe this means there is an SF neighborhood that's rolled this out or at least was studied?


I just skim through it, pretty good reading to understand where we are now. Main key-points:

- fiber has been rolling out for library, public housing, schools and public Wi-FI. But not on community / businesses.

- SF has the infrastructure, it just needs to build "the last mile" to connect with the houses / buildings

- there has been effort for community-fiber since 2011, but as of 2014, the city is still not in favor of this kind of development.


I would 100% support (and pay premium for) a community fiber in SF instead of paying Comcast or allies.


The sad thing is it is taking local community governments to compete with broadband 'providers'. You know you need some competition when local governments are the tip of innovation and your biggest competition. I hope many more communities and Google Fiber wake up the slumbering cable companies and telcos.


I am currently involved in a community project to build fiber in our rural Missouri county- DadeCountyFiber.org

We are hoping to participate in the FCC's Rural Broadband Experiments for the Connect America Fund.

Any ideas or feedback are most welcome!


Definitely hate TWC, but no other option.


No shit?


Wheeler is still a dingo.


I'd love to have fiber dropped to my home but this article glosses over the privacy implications of having a government entity have direct access to everything you do online.


RTFA. Direct quote from the linked page:

"There Are Good Reasons Not to Want the Government to be Your ISP

"Just because a local municipality might own the fiber infrastructure does not necessarily mean it is also best-suited to act as an ISP to residents. Residents might rightly wonder what sort of information sharing practices would become policy, particularly information sharing with law enforcement.

"This challenge can be addressed as well. Cities can help resolve privacy concerns by adopting the open access model described above. On this model the local municipality merely leases the fiber and never has to have access to the data on the fiber. Local ISPs that lease the fiber can be held accountable by users that encourage the ISPs to adopt privacy-protecting policies and terms of service."


Indeed. Taxpayer-funded broadband has its allure, but the risks of something going very wrong with private data are much higher, in terms of what is at stake if not in terms of odds of a breach.


The solution here is SSL everywhere, at all times. We should be just as concerned about Time Warner and Comcast having access to our private browsing.

It'd be nice to have an intermediate step between HTTP and CA-validated HTTPS. Yes, my ISP could theoretically MITM me, but it'd at least allow websites to prevent casual snooping.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: