[toomuchtodo]

The same could be said about Frontier [1] and Windstream [2]. I like that Vermont keeps trying, my internet should be like my water, sewer, and similar local services: a utility. The unemployed can be trained to run fiber, and the Fed is going to keep soaking up muni bonds [3], not much to lose by trying again. Capital is cheap, treat failure as just as cheap.

It can be done [4] [5]. But you have to care. You have to build public systems and goods that are loved, because to love them is to want to provide ongoing care for them, and to invest yourself in them (financially and otherwise).

[1] https://www.eff.org/deeplinks/2020/04/frontiers-bankruptcy-r... (EFF: Frontier’s Bankruptcy Reveals Why Big ISPs Choose to Deny Fiber to So Much of America)

[2] https://arstechnica.com/information-technology/2019/02/winds... (Windstream, ISP with 1 million customers, files for bankruptcy)

[3] https://www.cnbc.com/2020/04/27/the-fed-says-it-is-expanding... (CNBC: The Fed says it is expanding its municipal bond buying program)

[4] https://muninetworks.org/communitymap (US Muni Fiber Map)

[5] https://b4rn.org.uk/ (England: Broadband for the Rural North)

[zajio1am]

> I like that Vermont keeps trying, my internet should be like my water, sewer, and similar local services: a utility.

It is practically impossible to run multiple competing water and sewer pipes or other utilies. But there is no such problem with fiber. Or there could be public passive fiber operated by multiple competing ISPs.

Proper role of government is to actively encourage competition, not destroy market by subsidised public service. But it seems that US government instead actively discourage competition, which leads to bad commercial services, which leads to public preference for monopolistic government service, with all its problems well known from Europe past.

[toomuchtodo]

You're entitled to your opinion of course, but it has repeatedly been shown that ISPs will not make the necessary investments, and instead will over leverage with debt while milking the customer base for the most dollars they can while providing the lowest quality of service tolerable.

Private enterprise had their chance, decades even. Fool us once.

[bhupy]

I'm not sure how you can come to that conclusion...

Telecommunications investment per capita is higher in the US than all other OECD countries, and by a lot[1]. The gap in telecom investment between the US and the EU is similar[2].

The robust and enduring gap in American telecommunication investment really paid off as COVID began to take over the world. American internet was the most well positioned to withstand the lockdown-induced surge in internet load. Meanwhile, internet speeds in Italy and many other European nations were slowed down to less than half what’s standard in the US, partially due to older infrastructure[3]. In Europe, Netflix actually reduced its traffic there by 25 percent and YouTube promised to limit quality in order to to free up bandwidth for other services. None of this was necessary in the US.

The fixed broadband speed in the US is higher than that of Portugal, Poland, Germany, Belgium, South Korea (!!), Austria, and Italy [4].

[1] https://pbs.twimg.com/media/EVHQddtUcAElcjq?format=jpg&name=...

[2] https://www.orange.com/fr/content/download/47277/1372866/ver...

[3] https://www.vox.com/recode/2020/3/25/21188391/internet-surge...

[4] https://www.speedtest.net/global-index#fixed

[charwalker]

That's lovely. Now please explain to so many of my friends and family why their only broadband option with remotely decent pricing (if they have an alternative) is Xfinity, and why many of them had to pay hundreds to get a line run to their house to use that service.

The market will not solve access issues, regional monopolies by ISPs, or similar matters without intense regulation. Even then they will use tax subsidies to fight those regulations tooth and nail.

I'm running 100/10 on a dsl type line right now because it's my only other option than Xfinity and it is literally as fast as I can get outside of Xfinity as they own the infrastructure here for anything better. At least I don't have a data cap and can skate by at these speeds, though my upload hurts quite often.

[bhupy]

I'm sorry to hear about your friends' experiences.

While anecdotes are important, the data just doesn't corroborate the notion that American broadband internet is worse, on average, than that of its peer countries. You're right that the experience of some individuals does not match up to the aggregate statistics, but you can't draw broad conclusions from those data points.

The aggregate data captures the experiences of your friends, just like the aggregate data for other nations captures adverse experiences of their citizens. And what the data does say is 1) investment in broadband infrastructure is higher in the US (per capita) than most other peer nations, and 2) average broadband speed in the US is higher than most other peer nations. That's not really a matter of opinion, it's strictly an observation.

[heraclius]

I looked at [4]—unless I’m being silly, the table seems to have the US far behind (from your list) S. Korea, Belgium, Austria, and several other European states.

[bhupy]

You are looking at the Mobile list.

Look at the right side table — we are talking about fixed broadband infrastructure (fiber), not Mobile.

[heraclius]

Ah yes, quite right. I thought the colours suggested that I was in fact already on the fixed broadband list.

[mmm_grayons]

Symmetric gigabit is under $80/mo from AT&T where I live. That's not only good new infrastructure but also significantly cheaper than the old cable stuff. Maybe were the space less regulated ISPs would make a greater investment, especially were it easier to start one's own. If the WISP idea ever gets really good that would solve most of the situation.

[FlyMoreRockets]

I live in a rural area and also pay $80/mo for internet. This gets me a 160 KB/s unmetered connection, which has been the best available for about a decade, unless I want to spend a lot of money or suffer cripling bandwidth caps. Millions of dollars have been awarded each year during this time to deploy rural broadband in the state. Areas serviced by small, local ISPs have fiber to the door, but unfortunately, I'm serviced by a big, nationwide ISP thata just sucks up all that money and never delivers. I can hardly wait for Starlink or one of the proposed competing satellite constellations to come online.

[FlyMoreRockets]

Also: during the 90's, Wired published a fiber map, and at that time, there were seven different fiber networks in place within a quarter mile from my house. For over a decade ISDN has been a little over a thousand feet away, but the provider won't connect without charging several thousands of dollars to run line and install a repeater, despite the tens of millions of dollars the state and feds have given them to roll out rural broadband.

[colejohnson66]

> Maybe were the space less regulated ISPs would make a greater investment, especially were it easier to start one's own.

I’m genuinely curious: what regulations are hampering investment? Cause it’s not net neutrality (ISPs have said to their investors that it doesn’t)

Seriously, the governments have taken almost a hands off approach to the internet, and it’s now evident that it’s not working.

[rayiner]

The government does not take a "hands off approach to the Internet." It might seem that way, because the federal government does that. But when you're talking about building physical things in specific places, that's largely the domain of cities and states, and accordingly most of the relevant regulation is at the state and local level.

State-and-local permitting is one huge morass. When I had fiber installed to my house, Comcast had to get permits to hang fiber on a utility poll. Then there were different permits for diverting traffic temporarily while hanging the fiber. Then there were other permits for trenching. The whole process took months. And there is a skilled worker there that has to shepard the process through the bureaucracy.

Other regulations prevent offering television service (a key revenue source) without commitments to build a certain footprint. That means you can't start an ISP with a "minimal viable product" and grow from there. You have to be willing to commit huge amounts of money up front before you see a cent of revenue.

It's very illustrative to look at the deals Google Fiber struck with municipalities, because that shows you where are the real pressure points. Two features appeared in almost every accepted proposal for a Fiber city: (1) one-shot, fast-track permitting; and (2) no build-out requirements.

[nickff]

The most common issue is that municipal governments treat service providers as a revenue source. They charge high fees per utility pole, foot of trench, or resident serviced. These fees are then passed on to the user, who thinks the provider is responsible for the high costs.

[vlan0]

>Symmetric gigabit is under $80/mo from AT&T where I live.

That's nice. It's apparent that you live in or near a large metropolotian area. But we're talking about places like Vermont, Upstate NY, and rural Ohio that currently have DSL, or 3G, as their only option.

>That's not only good new infrastructure but also significantly cheaper than the old cable stuff.

Not sure where you get your figures from, but fiber optic is more expensive per foot than copper. In fact, the FTTH installs use some of the most expensive fiber to limit potential issues. On top of that, rural areas are especially difficult because the population density just might mean miles between homes. So, this isn't as much about regulation as it is for-profit companies not willing to take a loss to serve the few.

[mindslight]

> Symmetric gigabit is under $80/mo from AT&T where I live.

One datapoint without even stating where you are is meaningless. In general, urban and well off suburban areas have enough resources and density for several competing providers that will actually maintain infrastructure. Meanwhile "competition" in rural areas consists of aging coax versus aging copper. We've had over a decade of private companies promising to build out modern infrastructure if given public money, with few results. Only when rural areas build out municipal fiber do they get reliable gigabit with no customer service hassle.

[AnIdiotOnTheNet]

> Symmetric gigabit is under $80/mo from AT&T where I live

They advertise a similar deal in my city. It isn't actually available anywhere except for a tiny patch of rich-people-land of course.

[Red_Leaves_Flyy]

I have comcast, signed up for a $70 dollar 80/10 package, now it's 150/10 and 120/mo. I have my own router and basic cable. It would cost more to drop basic cable. I can't even downgrade to save money. Comcast has no competition above 100mbps so they can get away with charging 150 for a 1gbps asym connection that has a 1tb cap. Insane. Just slow me to 3g speeds during surges. The actual cost of a bit traveling through and out their network is infinitely close to 0.

[edlucas]

Burlington Telecom provided excellent service as a public utility. Symmetric gigabit for $70 and 150 Mbps for $55.

[thatfrenchguy]

$80 is ridiculously expensive

[Glyptodon]

In my area it's $88/month for 150Mbps down/10Mbps up.

I have been considering downgrading as one third the speed is around half the price I think. :/

[jtr1]

>Proper role of government is to actively encourage competition, not destroy market by subsidised public service.

Just highlighting this because while it functions as an unquestioned prior for a lot of us in the tech community, I think it conflates a means with an end. Competition is generally preferable because it builds in the enforcement of feedback between the preferences of consumers and capacities of producers of a good. Where that feedback is tight, it promotes the welfare of both. Common welfare is the end, competition is the means.

[eropple]

> But there is no such problem with fiber.

Have you ever lived in a neighborhood with ongoing FTTH, or even FTTC, installation? I assure you that there is certainly a problem with having that done multiple times. That's just the making-lives-shitty part and doesn't even get to capital costs, which ISPs just straight-up won't spend; go look at the ongoing embarrassment of "hey Verizon, when's FiOS coming to X?". (The answer, dear reader, is "never".)

Public fiber is better, but there should be, as with most necessary-for-modern-life services, a public option that puts a ceiling on what a private company can charge for service. If they can extract efficiencies to be cheaper, sure, great.

[toast0]

> Have you ever lived in a neighborhood with ongoing FTTH, or even FTTC, installation? I assure you that there is certainly a problem with having that done multiple times. That's just the making-lives-shitty part

I was around when AT&T ran the fiber in my old neighborhood. One day, they put pulleys on the poles. Another day, they pulled the fiber over the pulleys. The worst part was waiting for service to become available even though the neighborhood equipment was installed.

The couple of days with lots of trucks in the neighborhood was somewhat disruptive, but not that much more than trash day with a couple garbage trucks driving slowly through the neighborhood.

That still doesn't make it likely to get competition this way. AT&T seemingly only installed fiber because Google had announced they were going to, but Google never actually did, because it was too expensive (capital cost + regulatory requirements + pole access)

[eropple]

That sounds comparatively nice. I was living in a place where everything was trunked underground. R I P.

[toast0]

Underground utilities make maintenance and construction terribly expensive in dollars, time, and hassle.

If you're in an area where conditions are frequently bringing down poles or wires, it can make sense, but a lot of places want underground utilities for cosmetic reasons --- it's a lot of expense to give you a less flexible system.

[toomuchtodo]

That's why you use Dig Once. If you dig, you plan and coordinate, and you make efficient use of the time sink you're putting into laying utilities in the ground.

https://www.csg.org/pubs/capitolideas/enews/cs41_1.aspx

https://staticshare.america.gov/uploads/2016/04/6.-GCI-Dig-O...

https://arstechnica.com/tech-policy/2018/03/dig-once-rule-re...

[ardy42]

You could probably have something close to best-of-both worlds if they run underground conduit in the utility right of way.

[colejohnson66]

> Proper role of government is to actively encourage competition, not destroy market by subsidised public service.

I’m of the opinion that if the free market isn’t solving the problem, then the government needs to step in. It’s been repeatedly shown that, even after being given subsidies, ISPs will not spend it on upgrading their networks. Instead, they spend it on raising the salaries of executives and stock buybacks.

Maybe if a competitor were to come in and offer symmetrical gigabit fiber to the home, the ISPs would improve?

Google Fiber was an attempt at that. But even with the weight of Google behind it, starting a new ISP proved too difficult, and they failed.

The barrier to entry is too high and it’s no longer a free market. That’s when the government should step in whether it’s regulation or providing a government run ISP.

[charwalker]

Fiber is identical as water pipes and railroads in that once it is laid down it is hugely expensive and unnecessary to run again. If it is then owned and controlled by a private entity then it will be exploited, expanded only for profit, and only maintained when totally broken. If it is owned, operated, and maintained by a municipal that has proper funding and governance it can work very well even if it is slower to deploy new technologies along the same lines as laying down new roads.

Separating anything that can operate like a monopoly is also the role of government, at least in terms of protecting markets and competition, and if the market then demonstrates it will opt for profit over availability then the service needs to be seize and operated or at least very highly regulated by the government just like water, power, and other utilities. There are many examples of this working when expansion or availability was not profitable and therefore not an option for a private company.

Basically, if it operates like a utility in requires high regulation to insure access and failing that requires government management and ownership to operate effectively. We tried giving billions to ISPs and it didn't go to expanding service. The market wants profit, not equal access.

https://www.vice.com/en_us/article/ywkn4b/study-throwing-tax...

https://www.eff.org/deeplinks/2020/04/frontiers-bankruptcy-r...

https://www.broadbandtechreport.com/fiber/ftth/article/16446...

http://www.qlife.net/sites/default/files/imported/broadbandr...

[objektif]

Providing fast internet access at very affordable prices is destroying market?

[tehjoker]

> Proper role of government is to actively encourage competition, not destroy market by subsidised public service.

This is neoliberal ideology. It's possible to build good public sector services as shown by the rest of the world.

[rayiner]

> The unemployed can be trained to run fiber,

I feel like this is a bit arrogant. This stuff is skilled work. A fiber splicer is a $5,000+ piece of equipment. There are electrical, optical, environmental, and safety considerations all in play. It's not something you can just teach people to do whose other employment options are flipping burgers.

A couple of years ago, I got to watch the whole fiber installation process in action when I had Comcast Gigabit Pro installed in my house. Leaving aside the months it took to get county permitting, it took a team of guys a whole day to run it from the nearest splice point 1,700 feet to my neighborhood. Another half day to run it 200 feet to the pole by my house. Another half day to trench it under my driveway and into my basement. And another half day to install and configure the CPE.

For those people who own homes: Think of how much money it costs to get a team of skilled laborers to spend a couple of days doing anything around your house. I just had a new heat pump condenser and air handler installed. I could buy the parts online for well under $3,000, but with install it came to $6,700. Almost $4,000 for two guys to complete work that took a few hours. I bought marble countertops that cost $2,300 for a slab, and another $2,000 for a day's worth of work to do finishing and installation. Tomorrow, I'm going to spend close to $500 in labor for an hour of work to get a sump pump replaced. Renovating a bathroom can involve $5,000 in materials costs and $10,000 in labor costs for something that takes someone a few days. I suspect my fiber install--probably six man days--cost Comcast close to $10,000.

[toomuchtodo]

It it my opinion that to believe in my fellow human is not arrogant. Labor is not fungible ("lump of labor fallacy"), but people are versatile. We should expect that people can be trained into skilled roles, with alternate paths for those who don't have the aptitude. We have trillions of dollars of infrastructure work to be done in this country [1] [2], and we should absolutely expect our citizens to be able to perform that work if provided the means to do so.

Disclaimer: I have run and spliced fiber as a hobbyist for those starting rural ISPs, and also know where you stand on private venture vs community/muni/non-profit broadband. I'm always interested in your constructive criticism (based on your industry experience), as it helps me understand ways to drive down the costs of non-profit broadband initiatives.

[1] https://en.wikipedia.org/wiki/Green_New_Deal

[2] https://www.infrastructurereportcard.org/

[rayiner]

When I say it's arrogant I mean we too readily accept that skilled labor doesn't require specialized aptitude for the job. We readily accept that we can't easily train unemployed low-skill workers to be programmers or lawyers, but posit that we can do the same when it comes to installing fiber networks or fitting pipes.

I assume we disagree on this point, but I think it's accurate to say that most people lack the patience, attention to detail, and spatial, mathematical, planning, and problem-solving skills to do a lot of jobs that we assume people "can be trained to do." I don't think that Verizon pays service techs $40/hour + benefits out of generosity. My neighbor works at a local brewery. He sees a lot of guys (some with college degrees) that can't master "move vat A and load it into equipment B, but watch out for X, Y and Z."

I've spent my whole life getting roped into helping people with IT. Most of the time, it's a matter of "read the instructions completely, then click 'Next' until the installation is completed." Some surely could be taught to do it themselves. But I would not be surprised to find many for whom simply "read the instructions completely before doing anything" is a barrier, or who just cannot grasp the process of: "try X, observe results, modify approach, try Y."

[toomuchtodo]

I think this is a fair point, but you find this quandary in hiring for most roles: "Does this person give enough of a shit about the job to at least try to get it right?" The economy is still somewhat functioning, so I assume the answer to that is yes, you can find people who can mostly do the job they're in.

This is not to say you don't need education so workers know how to do the job, effective leadership for those folks, as well as quality assurance and other mechanisms to ensure the work being spec'd is done appropriately. I put forth that there is a middle ground between the nihilistic idea that "these people can only flip burgers" and the naive assumption that anyone can be a brain surgeon and policy put in place around such aspirations.

> But I would not be surprised to find many for whom simply "read the instructions completely before doing anything" is a barrier, or who just cannot grasp the process of: "try X, observe results, modify approach, try Y."

I agree an early filter must be used to route these folks into harmless roles or simply benefit programs. Such is the struggle of searching for and retaining talent. People are hard.

[vtthrowaway2519]

>A couple of years ago, I got to watch the whole fiber installation process in action when I had Comcast Gigabit Pro installed in my house. Leaving aside the months it took to get county permitting, it took a team of guys a whole day to run it from the nearest splice point 1,700 feet to my neighborhood. Another half day to run it 200 feet to the pole by my house. Another half day to trench it under my driveway and into my basement. And another half day to install and configure the CPE.

That's an absurd amount of time. There must have been other considerations which made it unusual. When WCVT brought fiber to our area in Vermont it took maybe 5-6 hours to run the 1600 feet or so of our side road, and an extra hour for one house on the way with its own few hundred feet, using their larger scale trencher system + an air blaster if they needed to do a horizontal direct bore which could do 25-50' very quickly. Dig Safe had been by a day or two before just like for any normal ground work anyone would do, so all underground cables/pipes were already painted. The fiber ended up right at our place where we put it through a hole I drilled in the foundation. It wasn't any special ISP team either they had one of the local construction companies take care of the bulk part. Then one tech of theirs drove in, maybe 30 minutes tops to get that spliced and run to my rack. Then like, 5 minutes for the GPON CPE, max? They'd obviously done a proper job of calculating optical power attenuation ahead of time so no issues there. Seriously a -half a day- to install and configure CPE? That makes no sense at all, so it must be some Comcast thing, or something went wrong. GPON is binary, it's either within acceptable optical power range or not.

If you're counting anything LAN side beyond the fiber end point you shouldn't because that is completely independent from the WAN drop and represents an arbitrary amount of time that will be different for every customer.

[rayiner]

In Maryland, all work in rights of way must be done by a company specially authorized by the state to perform utility-related construction. Then each county has licensing and permitting processes for work "related" to telecom utilities. Additionally, our utilities are above ground, and our neighborhood is too tight to allow use of a trencher, so all that work was done by hand. This is just how much of the northeast/mid-atlantic is. It's a double-edged sword: these are relatively high-density places where a single fiber cable can serve a lot of houses. But at the same time, they tend to be already built up (my subdivision was laid out in 1920), dense with narrow access ways, and in counties/states with lots of bureaucracy. (I am currently working on a "vegetation management plan" so I can clear the weeds from my backyard.)

For one-off fiber installs, Comcast uses active Ethernet instead of GPON, so maybe that's why the CPE installation took so long. Although, to be fair, I'm counting the time it took to drill a hole in the foundation as part of the half day of CPE install and the time to configure the router/access point--neither of which your average customer is qualified to do themselves.

[washadjeffmad]

A four year degree shows a man can be trained.

After guiding in the low hundreds of employees from a place of no experience to being competent in technical roles, I've learned that the outcomes of training are a test of management and leadership. People you might avoid in passing on the street, whose life circumstances you might disdain, who have succeeded nowhere else can be trained to do immensely complicated and valuable things and make a career out of it.

If you don't want anecdata, then look no further than the success of the US military. You might have fewer teeth than fingers, but they'll find a place for you.

[rayiner]

Four year college graduates are selected from among a pool of people who did well in high school. Likewise, I suspect the employees you work with were not hired from among the ranks of the chronically unemployed. The US military is not a good counter-example. Members of the armed forces, both at the enlisted and officer levels, are better educated than the population as a whole.

This is not a matter of “disdaining” anyone’s “life circumstances.” It’s a matter of acknowledging that infrastructure work requires skills that not everyone has, just like programming does.

[fibermoney]

> A fiber splicer is a $5,000+ piece of equipment.

Nah. A fusion splicer costs $900. You can train a monkey to splice fiber. I give you exhibit A: me, the monkey.

I had no previous formal training in fiber optics nor splicing before I got my first fusion splicer. I read the instructions and I started splicing.

[Red_Leaves_Flyy]

Like most manual construction jobs, they can be done by anyone, with training. The hard, expensive, time consuming, and difficult part is finding people that give enough of a fuck to do a decent job consistently without constant direct supervision. Once you've got good entry level staff then you need to train them. Most people don't go into construction to be a teacher, so that's a whole other problem to solve.

Likewise, all else being equal, I'd flip burgers over any construction job. I've met far too many broken people that were ate up and spit out by construction companies in their forties to spend the rest of their lives on Medicaid and disability to risk my future working ability for a temporary, seasonal, small pay bump.

[jessaustin]

I don't know anything about Windstream, but Frontier is another firm (others include FairPoint and Hawaiian Telcom) that tried to find a business in Verizon's trash bin. In every case, poor service and bankruptcy was what they found instead. Regulators knew they shouldn't let VZN divest from these customer bases, so they imposed some fig-leaf requirements that didn't accomplish anything.

[taborj]

> but Frontier is another firm (others include FairPoint and Hawaiian Telcom) that tried to find a business in Verizon's trash bin

Can confirm; now our ISP/telco is Ziply Fiber, whoever they are.

[Wistar]

Ziply is small, headquartered in Kirkland, WA. So far, they have managed to get local TV station KIRO 7 back on their lineup after 2 years of Frontier refusing to pay what KIRO's owners, Cox Communications, were asking.

The monthly bill has gotten totally out of hand, though.

[evandev]

The problem with treating internet as a utility, is a lot of the homes in Vermont have their own water well and septic tank and don't have water, sewer as a utility.

[toomuchtodo]

I assume those homes have power? Power is a utility.

Fun side note: Green Mountain Power, the largest power utility in Vermont, is working with Tesla on distributed energy storage and grid management to lower power costs and for resiliency purposes (to reduce power outages in the winter due to extreme winter weather conditions). This is how a utility is supposed to work!

[vtthrowaway2519]

I've got that, a couple of subsidized Tesla Powerwalls where the deal is they get to draw up to 50% of the juice during high load times. It's a pretty reasonable deal. From the customer point of view it definitely can take the place of a generator, GMP is pretty fast about bringing power back up usually so a few PW2s is pretty good even without solar. And unlike a generator PWs don't have regular testing/maintenance/consumables/noise/pollution. On GMP's side though the big point is to offset having to buy very expensive peak load power. It's already saved them a ton of money, old press piece for example:

https://greenmountainpower.com/2018/07/24/whoa-heatwave-savi...

I'll be a little curious if usage patterns are different this summer with more people home, but it's worked out well for them under typical conditions when there are big spikes in the afternoon/evening as people get back and turn on AC or the like.

[evandev]

Hmm, good point.

I'm about to move to Vermont and Green Mountain Power will be my power utility. It's good to hear this!

[toomuchtodo]

I wish you the best. Vermont is beautiful, and a lovely place to call home.