Speaking from a strange perspective perhaps but here in Sweden (the land of broadband) there was a little rural town in the south that couldn't wait to get fibre so the locals organized all the necessary things to start a town fibre grid themselves.
And now every house has 10Gbps with free choice of operator.
Normally what most communities do is wait for one of the big fibre operators to reach them and start digging. But these locals did all the practical stuff like digging, laying fibre and building the actual grid themselves by hiring their own contractors.
Larger communities, towns and cities, usually have a city wide fibre grid managed perhaps by the local energy company or by the municipality. They in turn might out-source the fibre digging but they still manage the grid and determine what operators are offered to the citizens.
In big cities (we have three of those) you can find multiple city wide fibre grids. One managed by the municipality/city and some y large property owners that take it upon themselves to equip their buildings with fibre.
Regarding the wireless setups in the article I am very skeptical because this was a trend 10 years back in Sweden and a lot of people lost money by investing in it when they could have just waited a few years and had real fibre.
But the US is special because of its size. So wireless might be all they can get for many years.
The US has a population density of 35.0/km², while Sweden has one of 22.0/km². So it shouldn't really be bigger problem in the US because of geography.
Exactly. Exclude the counties with NYC, LA, Chicago, Houston, etc, and it's a very different story. What works in Manhattan could never work in Wyoming or even Wisconsin.
Fixed Wireless has been a marginal player almost everywhere it has been tried.
If you are in a place with favorable geography it can sorta kinda work.
If you are in the Appalachian mountains, on the other hand, you are in a dissected plateau and most people live in valleys that are overlooked by a single ridge. You have to build a road over the ridge, run electricity, maybe run fiber, and with a really tall tower (remember we have trees on the East Coast!) you might look into the valleys on each side, and if you are really lucky, another valley over.
The kicker is that most of the places which are favorable for wireless already have cable. So the addressable market is this tiny sliver which can be served with fixed wireless and can't get cable.
Frankly, what this article tells me is that the ISP conglomerates haven't yet bribed the state officials of Kentucky or Ohio to ban or minefield local governments from being involved in broadband initiatives.
What this article tells me is that these areas are so unprofitable that the telcos don't actually care.
I can tell you that they know exactly which areas can be profitable, and if one of those areas suddenly decides to try to do something on their own, all of a sudden lots of obstacles magically appear.
Actually, Frontier charges $90 a month for 1 Mbps DSL.
The problem is not that they could not afford to offer me a gigabit at $120 a month, but that they can't afford to give up $90 a month that they get without investing in their networks.
In the case of Frontier, it pays a dividend greater than earnings. It even pays a dividend in quarters where earnings are negative! It is not a "profitable" phone company, it is a company that is bleeding a copper network dry.
Well, in actual rural places, as in not dense enough for any particular road's worth of internet connection to matter, there are point to point radio ISPs (built off stuff like what Ubiquiti networks makes), and sometimes just a verbal agreement to beam a network over a distance.
I'd like to understand more. Are you saying that communities and individuals should also manage communication platforms, public transport, law and order themselves?
Those services turn out better when managed by the city compared to when it's managed at the state or federal level. The problem is that rural towns usually don't have the resources to pull it off.
Yes, outsourcing law and order to local communities has never resulted in race biased enforcement of the law has it? Local communities to electing judges and DA's and Sheriffs to be tough on crime by minorities if vastly superior no doubt.
They don't. Federal regulation is vastly superior to state and local regulation of utilities. Remember, it's local governments that created telco and electric monopolies in the first place, a mess which Congress has spent decades trying to clean up.
Your comparison doesn't make sense. Local, state, and federal levels are all part of one country. There is no equivalent institution at the global level.
For global challenges like climate change, regulation at the global level seems like the right responses.
Local municipalities pass their own ordinances, laws, regulations, etc all the time. They just cannot contradict the laws of the state / federal. That is absolutely communities managing their own affairs, writing their own local policy.
That's a good point. I was thinking on a community level, and expanding the group size as necessary in order to provide these things... but it does tend to be a bottom up type of thing (as opposed to top down).
Law creation is probably the biggest outlier though
You are right. The sheriff that covers Ryegate, MT commutes in every day from New York City, and the school bus drivers commute from Seattle, likewise the mailman commutes in from D.C., and the garbage man from San Francisco.
Yes, or hire some entity to manage it. Which is quite different than waiting for it to be managed.
Not to mention the scale of rural communities makes the task proportionately less complicated. Scale matters very much. For example, if you are on the subway, how do you know who owns the magazine in your lap? There's a system of justice at work, it's very small in scale and you didn't need a lawyer to figure it out.
Sadly I've never had a good experience with "wireless" internet. I've looked into building a WISP a few times but the cost of hardware and no guarantee of actual data throughput or quality keeps me away.
Most hardware for WISPs have back haul throughput rates advertised at 100-400 mbs, yet most end points at the customer premise will see data rates of 6 - 20mbs. Including the bad jitter, packet loss, and burstyness, and you've got something that is only good for browsing facebook.
All the hardware I am aware of still uses the old RTS/CTS methods for media access borrowed from the wifi protocol. Which works fine for 100 end points all within 100ft of each other... But stick 100 end points in a 5 miles radius all trying to get a "clear to send" ack and your in a world of hurt.
Yes I know... some tuning will go along way, but at least with the Motorola canopy family, they defaulted to supporting a 15 mile (or was it 20?) radius. And with the WISP's network admin (which is normally always the CEO's young nephew) never changing the defaults because it "just works". This puts a ton of latency on the network since everything is assuming to be waiting on hearing from the node 15 miles away (that may or may not be there). There are some newer systems out now that use TDMA for media access, but from the reviews I see, still getting more than 20mbs to a CPE is unlikely.
You best bet would be a gigabit back haul with some p2p access points covering no more than a mile or 2 each, but that means many more antennas and no place to put them. I've seen many WISPs "cover" 100+ square miles with just two access points. It just doesn't work.
Roughly right, although with enough attention to detail, low subscriber/AP ratios and some luck with spectrum availability it is possible to do better.
Source: I built and ran a small WISP for 15 years.
IF you are willing to spend the money and invest the time to get it properly deployed and configured, it is possible to achieve decent performance even with 802.11-based gear. For example I'm typing this over a connection with two wireless hops that goes 10 miles to the nearest town that achieves 150/150 with sub-5ms RTT. All the links are point-to-point though.
btw rule of thumb is to take whatever advertised throughput a piece of wireless gear has and divide it by four to get a realistic actual performance. 1/2 for the loss between "air" tx-rate and achievable packet payload throughput, then another 1/2 for the difference between full duplex and half-duplex. So that "400Mbps" radio will actually deliver 100, best case.
I'm typing to you from a rural WISP; 5GHz, 12/1 MBps asymmetric service. My link involves about 3 wireless hops in my ISP before I get to wired infrastructure. It's not terrible and is definitely good enough for streaming video and playing games. I do wish I had wired service though.
The biggest problem is the latency. I'm about 40ms from the edge of my ISP before I get to the rest of the Internet (compare 5ms for wired urban Internet.) Also, as you allude to, with a smaller ISP it can often be the CEO's nephew in charge of stuff and professionalism suffers.
My older service was 900 MHz 1/0.25 MBps service. That was pretty awful, just too slow to do anything modern with it. The nice thing about 900 MHz is you don't need a clear line of sight.
My parents have two ISP options and they're on the east coast (like literally, the town has waterfront on the Atlantic and all the BS that goes with it). They are by no means rural. They have two ISP options. One offers 1mb down. The other is Comcast.
6-20mbps with terrible quality would be a godsend for them and everyone else in the area.
They're fine places to vacation. They're terrible places to live.
Everyone is out to screw everyone else in order to make a buck. Shady used car salesman is a level of ethics and honesty that's about median when it comes to any sort of business (and it's more of a bell curve than bi-model IMO). The quaintness that makes the place marketable breeds busybodyism. There's a sense of community in the same way that companies that don't care about their employees loudly preach that they do. The roads are clogged with tourists from memorial day through labor day. From labor day until memorial day hospitals and holding cells are clogged with drug users and dealers. You get to watch the corrupt town government pour tons of money into special interest projects.
They seem unable to perform any kind of maintenance without massively degrading network service over a /large/ area.
There are also rumors that technicians will drive to various areas and disconnect (paying) customers in attempts to see if anyone complains about missing service (supposedly as an attempt to boot off freeloaders).
Finally, they're Comcast. If there isn't competition in an area to drive service (it sounds like there isn't here) then they probably haven't pushed 100mbit out.
I've heard about comcast's support. But I am curious if opening up a business account would be a different experience. At least with Time Warner and AT&T, you just make up a fake business name (unless you have one) and set up a business account. That costs more, but it atleast gets you out of the stupid "1 year" package deals where you have to call support once every year and beg for mercy to lower your bill and get back into another package ....
It'll take you 10 seconds to check instead of speculating. The DOCSIS 3.0 upgrade was basically footprint-wide, so there is a good chance you've got 100 megabits or higher.
Realistically just browsing facebook and not having it take forever is going to satisfy these people a lot more than what they have now, which probably takes ages just to load FB.
Of course, 4k netflix would be nice too, but I suspect these people just want some basics to start with.
I think 2 Mbps is minimal for using the web now because every horrible site has an autoloading video on it. The FCC definition of broadband is 25 Mbps.
It really depends on how the infrastructure is built. For instance at our WISP we limit tower radius distance to 1 Mile. This allows us to push higher speeds and have lower latency. I do agree; many traditional WISPs try to maximize tower distance to CPE's in an effort to maximize their ROI, however it comes at the cost of a decreased customer experience (though in some areas this is unavoidable).
We use Cambium ePMP hardware (the Motorola team split into UBNT and Cambium) in 1 mi radius' and see ~40-90 Mbps down and 10-40 Mbps up.
This seems like a pessimistic assessment to me. Both my brother and my parents are on rural wireless networks, and they seem to get good service, though I can't say with certainty what technology is used or what bandwidth they get. Skype calls work fine. They don't stream movies due to data caps, however.
It is not "pessimistic". It is my actual experience both as an end consumer and a (potential) WISP admin. And please don't mention "works fine" and "has data caps" in the same sentence...
What can I say? My actual experience differs from yours. I have no doubt that there are many fly-by-night operations out there that don't know how to configure their networks properly, but I know of at least two examples that seem to do just fine. There are no significant problems with latency or jitter or lost packets. You seem to take strong issue with data caps. They are unfortunate, but I don't really see any way to avoid that in a wireless environment. Heck, even in a wired environment, I suspect you'll find some limits hidden in the fine print. Even then,the data limit appears to be more of an issue of willingness to pay than any fundamental limitations of the technology: i.e. if customers pay for packages with higher caps, the ISP will have the income to maintain the network to support the higher caps.
San Fransisco is not a "rural" area. Even on their home page they mention they have access points on over 1000 buildings. Turn off about 998 of those access points and that is what you will be dealing with in the typical rural WISP.
I am not old enough to know the history of electrification, but it was completed thoughouly into rural America. Other utilities did not. Here in rural Kansas many households were still using wells or hauling water from cities, until they banded together and laid their own water pipes.
Back in 2002 when I was in high school in Coshocton I worked for the local internet service provider, CloverNet. It was mainly mundane tech support and computer repair, but every now and then I got to help with an AirClover installation. AirClover was the medium-distance (3-5mi?) broadband solution that we set up with on-the-roof directional antennas and Cisco 802.11b cards. Really excited to hear that Coshocton is still working on solutions like this.
Per usual on ISP topics popping into hacker news, I implore anyone considering starting an ISP to jump into the ISPSchool Slack channel. PM me or send an email to staff@tsi.io and we will add you.
TL;DNR Montana probably has faster internet than you. http://tsi.io
I was just this evening going through the "what if I started a wisp? How many pops? How much bandwidth? What would it take?" I've done this off and on since 1990 when breezecom was around.
Getting access to rooftops is infinitely easier than trenching fiber.
Can you scale wireless in a modest city like KC? At what CIR? You would have to make use of extensive caching. Even the best hardware like ubiquity tops out for backhaul at 2Gb. That's always been the problem is wireless can't scale with speed. At least not in any spectrum us mortals can use.
To be fair, distances in rural Britain are much shorter than distances in rural America. That said, I applaud them for doing it, and I have occasionally thought about pursuing a project like that myself.
It would be really fun digging up places to lay fiber. I've wired up 2 offices for new networking equipment when the company moved. It's pretty satisfying seeing the stuff you rigged up light up and actually function. It was pretty fun but I can imagine doing digging with equipment would be even more fun.
Another anecdote, I paid for DSL to be brought to my area. A neighbor chipped in and bought an extra mile to be upgraded. It was costly but worth it. I pretty much considered it a cost of having my house where it is.
About $36,000 and I am about seven miles from where they installed some new equipment in a large box on a pad. I don't actually know what is in the box. I think it might be called a DSLAM?
I believe I only paid for the equipment and Fairpoint paid the installation expenses. This was about nine years ago and I didn't pay as much attention to it as I should. I was pretty occupied.
There are only six residencies in my unincorporated township. All but one have DSL, now. The one that doesn't have it, doesn't want it. They live beyond where the power and phone lines stop, anyhow.
Anyhow, they are all pretty happy and grateful. As I was moving into the area, this was a good ice breaker. I'd say that I definitely got good value for the money. I get ~12 Mbps down. More importantly, a neighbor brought me some Piccadilly, just today. Worth every penny.
The most interesting part of the article for me was that it mentioned 5G wireless signals. I was under the impression that the standard(is that the right word?) was still being developed.
> We view it as the next economic revolution for coal towns," said Harry Collins, the chairman of the Letcher County Broadband Board, which formed late last year. "The majority of our railroad tracks are ripped up now—that revolution has played out. We feel that this [digital] revolution is just as game changing and life changing as those railroad tracks were in the 20s and 30s."
It's sad, but this is wishful thinking. If anything, tech-related jobs are more concentrated in urban areas than other service industries. These places have no economic viability outside the specific circumstances (coal mining, farming) that caused people to move there in the first place.
> An online freelancer with a low cost of living can live decently.
Eastern European and freelancer here. I can do all that freelance work for much cheaper than any American from a dying coal town. Furthermore, we Eastern European freelancers generally tend to be university educated, are used to working in a global cross-cultural marketplace, and we have a large pool of local talent we can subcontract work out to and thus grow our business. Anyone in rural America who now starts trying to make it in freelancing is likely to lose out to us.
On the other hand, one thing we cannot do – due to our foreign passports – is easily move to a large American city. So, I think moving to the cities is a better option for someone in a dying coal town than staying there and hoping for success in the digital economy.
An average middle class salary in my country would be around 1000€/month. That provides for a pretty good life in my country. In the USA, on the other hand, that annual income would be below the poverty level. The cost of fiber internet, electricity and gas is much cheaper in Eastern Europe than the horrible prices I have seen quoted for North America.
For what it’s worth, I am now a digital nomad. I don’t stay in my country on that salary, I travel among countries that are even cheaper. Someone who intends on staying in their dying coal town in Appalachia does not have that option.
If it costs $5,000 - 10,000 more to hire an American, so be it. The company gets an individual of native culture, native English speaker and a fellow countryman. Plus they are at least within 3 hours based on time-zone - minus Alaska and Hawaii.
So be it to who? The simple phenomenon of outsourcing shows that getting a "fellow countryman" isn't worth such a drastically higher cost to many clients. Yes, hiring people from the Indian subcontinent has burned many clients, but after getting burned they are often more likely to look towards Eastern Europe, where we are "close enough", than to hire North Americans at that much higher cost.
Time zone isn't so important, inasmuch as many freelancers in Europe are happy to adapt to North American hours.
In almost all of eg Appalachia, a 35-40 hour per week, $15 per hour online job (doing anything), is a good job for the majority of people versus the alternatives available.
There's no question it can make a meaningful difference near-term. The obvious long-term solution is further urbanization or near-urbanization, convincing people to leave those dying regions for greater opportunity.
There are now just too many other countries with young people speaking excellent, near-native English thanks to an entire childhood spent with American movies and television. Eastern Europeans or Filipinos can do call center work cheaper than American rural areas.
Cheaper per hour is easy to measure but its hard to measure customer satisfaction.
When I call a support line and get "Walter" who is going to help me, I know they are in a different hemisphere and am mostly unsatisfied with talking to them because they are trained with scripts and they talk like robots.
If I called and recognized a Kentucky accent, chances are that person is going to be more relatable and personable than "Walter" from 8 timezones away.
Mr./Mrs. Kentucky might crack a joke and get me to lighten up a little about why XYZ isn't working. If both calls have the same result, I walk away from talking to Mr./Mrs. a little more satisfied than with "Walter".
The same could be said if it was reversed, if someone was calling from the far east and ended up talking with Mr./Mrs. Kentucky it is going to be hard for them to be relatable and personable to someone from the far east.
Call centers insist so much on following scripts that I am not sure there would be opportunities to make jokes even if the person was in the same country as you.
Also, do you think that Americans want to hear a Kentucky accent? I would rather speak to a foreigner than hear someone with a village accent from my country. Those places have a reputation for low education and low culture, and so it is hard for customers to trust any employee with that accent who claims he can help you. I know that call centers have moved to my country and to the Philippines because English speakers in these countries have a more "neutral" accent.
Everything from support, to sales, to web development, admin, security, etc.
I'd list all the telecommuting job sites, but there seems to be a thousand of them now.
The US median income is about $37,000 per year ($17-$18 per hour). To get beyond ~$13 to ~$17 per hour doing lower skill telecommuting work, you have to have a resume with something meaningful on it.
If you're from Appalachia, it's not difficult to teach yourself any number of skills (from database admin to javascript), go to a local small college to pad your resume with that, get a certification or two if applicable, and then get a job working remote at an entry level tech job that pays considerably better than anything local (few of those jobs exist locally there). A person can go work at a local AT&T or similar store doing sales in an Appalachian town, get some sales experience, and then go apply for a remote sales job making more than they'll get locally. Particularly right now is an ideal time, the US labor market is getting very tight.
I grew up in Appalachia and entirely self-taught my way out of it in the mid 1990s. It was a helluva lot more difficult to do back then. And that process is trivial compared to working in a coal mine, and it's a lot less soul destroying than toiling away for a decade earning $10-$15 at a local Walmart.
No doubt some people can do that. But the original quote is comparing broadband to the railroads. The railroads (and mining) supported a broad economy with various roles for lots of different kinds of people. I have a hard time believing that broadband is going to have the same effect through remote sales and database admin jobs.
Instead of the kind of policy they had with telephones, with the Universal Service Fund, and the like, to subsidize the massively higher deployment cost. It's not just in the middle of nowhere, the run length are vastly longer. Of course it's going to cost more money. Electrical lines, water and sewer do too.
I don' think you read that link you posted properly. The proposed expansion of Lifeline to broadband was tied up in court because it was entirely Federally decreed, whereas the actual law passed by congress said that it had to be handled by the states. Pai pulled it back from the courts, and determined that following the law as written was a better idea.
Go ahead and kick the FCC and Pai around for screwing up net neutrality, but the Obama-era ULS expansion thing was never going to happen the way the FCC had written those regulations.
And now every house has 10Gbps with free choice of operator.
Normally what most communities do is wait for one of the big fibre operators to reach them and start digging. But these locals did all the practical stuff like digging, laying fibre and building the actual grid themselves by hiring their own contractors.
Larger communities, towns and cities, usually have a city wide fibre grid managed perhaps by the local energy company or by the municipality. They in turn might out-source the fibre digging but they still manage the grid and determine what operators are offered to the citizens.
In big cities (we have three of those) you can find multiple city wide fibre grids. One managed by the municipality/city and some y large property owners that take it upon themselves to equip their buildings with fibre.
Regarding the wireless setups in the article I am very skeptical because this was a trend 10 years back in Sweden and a lot of people lost money by investing in it when they could have just waited a few years and had real fibre.
But the US is special because of its size. So wireless might be all they can get for many years.