You mean the FCC was actually trying before? /sarcasm
In all seriousness, we’ve poured billions of dollars into broadband expansion efforts since the early 2000s. Every single time it’s been largely hoovered up by Big Telecoms, failed to expand broadband/improve speeds/lower prices, and basically just gone right to their bottom line as a subsidy.
The solution all along has been funding municipal broadband as the baseline for private enterprise to compete against and surpass, but lobbyists have all but killed that dead up until the past ten years or so. You can’t treat broadband as a utility in legal language but not in practice, yet the USA seems perfectly fine with their status quo leaving them a laughing stock of the developed world.
I live in a place with a municipal telco, and let me tell you, it is night and day. The service is top notch, and it always just works to the point where on the rare occasion that they do have a problem, I spend hours trying to figure out what's wrong with my equipment, because it happens so rarely. It costs more than the local cable company, because the price you pay is actually what it costs to deliver the service, but there are never any wild swings in prices, and when service gets cheaper, you just get upgraded for no additional cost. I wish all companies worked this way.
Yea, Sonic.net is amazing. I definitely recommend them if you are in their service area. Sadly they have been really slow to expand their fiber service, probably because they started in San Francisco, home of the NIMBY.
Ziply Fiber is another ISP that I can wholeheartedly recommend. They have service all over the north west states of Oregon, Washington, and Idaho (though still not _everywhere_). Like Sonic.net, their quality of service is excellent. No congestion, very limited downtime, no bandwidth caps. They’ve even posted pictures of the accidents that have caused downtime. Most recently that was damaged battery banks that released enough hydrogen gas to shut down and evacuate a central office. Before that it was a drill that took out a fiber bundle; cue pictures of the engineers splicing all the individual fibers back together.
When Democratic Party controlled the federal government, they tried to. It is generally hard to do because it requires funding and passing laws. Another difficulty is that voters don’t reward Democratic Party for their efforts.
It is very easy for Republican Party to revert the changes. Moreover their donors reward the republican politicians for their efforts and their voters don’t punish them for their efforts.
+1 and more to this. Democrats are not good at all at letting voters know what they have done for them. In California that Republicans love to hate, we have clean air, free beaches, more protections from corporate predation, and so much more. Republicans are better at complaining and spreading fear, which sadly is a lower-energy and more effective method for getting votes.
We got an overview of those "efforts" a few months back and it was both comical and enraging to hear how much money and time was wasted. And this was a law+process crafted by Democrats, signed by a Democrat President, and presided over by a Democrat Administration.. they had all the pieces in place to accomplish their goals.
Jon Stewart has a good point there as well, saying “the Democrats think it can’t solve anything unless is solves everything”. Over the last decade or so they’ve weighed everything down with stupidity. Ezra has specific examples; that’s a really good interview.
Another good quote: “It’s almost as if they had designed this machine to make sure that people in rural areas _never_ get broadband…”
Ooh, Ezra: “This is the Biden administration’s process for it’s own bill. They wanted this to happen. This is how liberal government works now.”
Nonsense. It was working fine and following close to the expected schedule. Numerous states had submitted their proposals for how they would use the money, those had been reviewed and approved, and the states were accepting bids from companies to do the construction and starting to select and approve them.
Delaware, Louisiana, and Nevada had selected contractors and passed final federal review and were ready to start construction as soon as the funds were disbursed.
Another 40 had contractors selected and were just waiting for final federal review to complete and funds to be disbursed. If the program had not been changed by the Trump administration most of them would have had construction starting this year.
All large efforts like this take a few years to go from initial passing of the law that enables them to construction actually starting.
> Another difficulty is that voters don’t reward Democratic Party for their efforts.
It is worth noting that the Biden admin passed a $42B package of broadband rollout funds in 2021, and then connected ZERO people over the next few years.
This is because special interest groups saddled it with environmental, union, and DEI requirements that were so onerous that nobody was able to actually get/use the funds.
Yes, Democrats plan to do the right thing, but frankly they should be penalized for boondoggles like this. Until the party can sideline its special interest groups to get things done, it's not as simple as "Democrats good, GOP bad."
> When states use the federal subsidy to provide access in areas without internet, they no longer have incentives to confirm if the technology being deployed is appropriate for the area or if the customer base needs and wants what is being provided.
It's actually the opposite. When states use federal subsidies they have the freedom to consider the technology that is actually best for the area instead of just what is cheap enough for the state to afford without better help.
When they are limited to whatever is cheapest now they may have to pick something that meets the present needs of rural areas but will be inadequate in a few years.
This is also heavily described in _Abundance_. I'm not an expert, but it is beyond doubt that this program did not produce any kind of meaningful results, despite having a moon-landing-sized budget.
Even when it was created, Starlink was obviously the best option for quickly linking rural America to the Internet. That's obvious because that's EXACTLY HOW rural America has actually been linking itself to the Internet, despite it not being covered by the program. In ten years, that's probably still how they'll get online.
I am a Democrat, have been since I could vote, and refuse to vote red for as long as MAGA owns the GOP. It's still exhausting to watch my party flounder and give in to its worst instincts.
That was an illegal move done by Trump. Biden administration proposed regulations for the conditions of the funding and allow states to apply for it. The regulations went through the necessary process to become final and the states went through the work to fulfill the requirements and the fundings were approved to go out of the door based on the trenches.
Trump administration illegally impounded the funding. This has been disallowed for decades but sadly the republican political appointees at the supreme court is allowing him.
It doesn’t even matter, does it. It seems the democrats made a plan spanning multiple elections due to it being loaded with regulations and now Starlink can better serve rural areas at a fraction of the cost (for american taxpayers). Also the planned deployment was 2026 at the earliest?
I think you have your policy preferences and you have the idea that if starlink was allowed, it would magically take much less time to go through regulation process. I think this is not true, but I think we digress.
My point is that democrats don’t get credit and votes for improving internet for rural areas and republicans don’t get punished for destroying progress. The fact that you are criticizing democrats so harsh and think trump breaking the laws doesn’t matter just prove my points.
They don’t have to take that long. Seriously, when you build years and years of government reviews and challenges and approvals into the process of awarding funds you have only yourself to blame when the government works for four years and accomplishes nothing. Meanwhile a Starlink, a private company, just starts doing the work. They’ve spent 10 billion dollars of their own money and built a service that is hands down the best internet service available in any rural area of America. They expect to have 11 billion in revenue this year, just 6 years after the service first became available to paying customers.
The billions spent on rural broadband excluded Starlink as not technically feasible.
Many other billions have the same issues - I think no one knows how to actually hoover this the way the big co's do?
We've had much faster broadband happening because of commercial competition from scrappy startups and WISPS and fiber folks (think sonic)
I think something like 94% of RDOF/BEAD locations in california were defaulted (ie, awarded but customer actually never got service)?
It's crazy given the 100+ billion or so spent on USF / RDOF / BEAD / etc that they couldn't do $5b - $10b for something like starlink which at least in rural areas is able to serve folks pretty quickly and push hard on that for a bit. The unsubsidized commercial starklink services is already outcompeting the insanely subsidized buildouts (that cost insane amounts per person). Starlink was awarded the funds but then they were revoked.
In Japan it was competition that fixed this. The government did something (not sure what) but it was not "funding municipal broadband".
I'd say one thing to do is outlaw the contracts that let governments only approve a single provider.
It was Softbank that brought the competition. In 2001 they started offering 1meg connections for $20 a month. The competition cost 10x. Softbank had aggressive marketing too, putting up booths at neighborhood train stations where they could sign you up and hand you a device on your way home.
The competition lowered their prices to match within a couple of months. Softbank doubled the speed to 2meg, same price. The same pattern. The competition matched price within a few weeks. Softbank raised their speed to 8meg. repeat.
Hitting the most populous 10% of counties will get you a good way towards universal coverage (LA county alone as >9M people, which would rank #12 as a state population).
Competition is great except when it isn't. Rural customers get screwed because they're not worth the effort. You'll spend more money running fiber to them than you'll ever get back, and that's before you take competition into consideration.
Back when we had the AT&T monopoly the government forced them to serve rural customers. It was subsided by a small fee on every phone bill. That carried over to the baby Bells because they were local monopolies. The problem is, no one has land lines anymore and no companies are willing to take on the last mile problem without a hefty incentive. Being the broadband monopoly is part of that incentive.
I argued for years that the AT&T breakup was done in the stupidest way possible. Put the natural monopoly - the last mile - on its own and regulate the hell out of it like we did AT&T. Then allow competition to run the service on top of those connections. We started in that direction but lost our way somewhere.
Yea, the breakup of AT&T was stupid. Even worse, it just allowed all of the baby bells to be bought up one by one by a really badly run Canadian company. Now rather a lot of that money started leaving the country and being gobbled up in a foreign bankruptcy and pension scandal and stock market crash. Yay. (Ok, not all of that was entirely foreseeable.)
But to get back to the original topic, I was referring specifically to the local monopolies that so many cities have granted to cable companies which exclude not just other cable companies but certain types of internet services as well. I was specifically not talking about rural areas, where that’s not a problem. Not many cities out there in the countryside.
If you want to see fiber installed everywhere, make the FCC annul all of the local monopolies and speed up all of the local permitting processes. Just have the FCC rule that no local permitting process for installing internet service in existing conduits or on existing poles may take longer than a fixed period of time (I dunno, a month or two should suffice), otherwise the permits are automatically granted. If the permit applications don’t meet your local rules then you need to identify the problem and reject the application quickly, not drag things out for years. Something along those lines.
Subsidies to speed up rural deployment of real broadband are a separate topic.
It is not irrelevant, who is going to pay $30,000/mile to run fiber out to rural homes when the subscription fees for a lot of rural properties will never pay for the infrastructure? The US is massive and full of sparsely populated rural areas, all of which will ignored by capitalists. Starlink is then their only option.
I’m all for getting rid of local monopolies that aren’t municipally owned.
Again, this is irrelevant to the conversation we were having about price competition. Competition solves the problem of high prices automatically.
> The US is massive and full of sparsely populated rural areas, all of which will ignored by capitalists. Starlink is then their only option.
Ironically, Elon Musk is a capitalist who is making bank by serving the sparsely populated rural areas. He is certainly not ignoring them. You can get Starlink service in the middle of the Pacific Ocean, and if there is any more sparsely populated place on Earth then I don’t know about it.
At least have a coherent argument before you complain that competition can’t solve a problem.
American exceptionalism is odd, in that sometimes it is "we are better than everyone else", but sometimes it is "we can't have the nice things that all other rich countries have, because [nonsense reason]".
A popular one is 'density', and, okay, maybe this is somewhat true if you're talking about, say, Wyoming (though I think not as true as people often think), but California, for instance, has an only slightly lower population density than _France_, and at that point "we can't have proper transport/telecoms/whatever because density" is just an excuse, and not a convincing one.
The reason we can't have whatever nice things you want is because California doesn't want to spend the money on that nice thing, and it has to maintain a budget, unlike nations. Including a $40 Billion project on a budget makes many, many other things go away. Even if it's just temporarily to help pay for the construction of the service, the point stands.
So imagine if France couldn't go into debt - only the EU can. France wants a giga-train suddenly, so they ask for it. The current leader of the EU isn't a fan of more trains, so he turns it down. France goes back to its people and says "we can build the giga-train if we do XYZ", and people vote based on whether or not they want XYZ or the giga-train more.
I think it's possible you just might want things different from what others want. That study a while back which showed most Americans want cheap public transit so that everyone else gets off the road and gives them more space lives rent free in my head. Nobody wants these stupid trains.
I mean, yeah, the US having systemic problems which make capital investment difficult, or at least extremely slow, is _part_ of it, but there's also just a tendency to a complacent "we have these problems because we are _special_ and they can't be fixed" attitude amongst many Americans (I say this as an outside observer).
california is ~160,000mi^2 and france is ~244,000mi^2 (423 megameters squared, 632 megameters squared, metric), the population density, respectively, is 251/mi^2 and 281/mi^2. You're comparing an entire country to a single state (2% of the US states, in fact.)
I like to point stuff like this out whenever someone compares an entire EU country to some US state.
note: i edited the france density, as i did accidentally transcribe the wrong value. France is slightly denser per sqkm.
Yep. As an american I keep giving WTFs every time media discuss potential solutions to problems without even considering how other places solved them. It's like other countries don't exist, or hide their secrets.
Not just broadband, but same with healthcare, homelessness, gerrymandering etc. Just copy-paste little by little.
I was thinking the same thing until I learned about the high speed train plans California had (and still have?) and just how much of a disaster it is. It’s a red tape issue and nobody at the policy level seems to care to fix it.
It really is. I keep wondering when some tiny bit of humility will show up, but it is increasingly like asking Russians to have some humility. Not likely, and sometimes, not possible.
Rural broadband mandates were just welfare for the big carriers. There was never intended to be any results or accountability. Consumers have been paying into the fund since 1997 and for that we got richer carriers and a pocketful of bupkiss.
Municipal broadband is such a great idea - but it's as hard, if not harder, to convince a city council to pursue this over adding a simple bicycle lane to a small street downtown.
Cities know how to build roads. They don't know shit about operating high speed broadband. My taxes keep going up because the city keeps taking on more and more responsibilities that (a) they don't know how to provide and (b) isn't really their responsibility to provide.
The rural telco situation is pretty good. In fact, better in many cases than what you get in larger cities.
The issue is there's some perverse incentives at play with the grant monies. For brand new neighborhoods you can expect state of the art internet being available. However, after it's installed, you can expect it to be simply locked in place with no upgrades.
A major problem comes down to easements. If there's a utility pole nearby then some federal regulations make it super easy for a utility to just add their line onto the existing pole. There are access guarantees.
However, if no pole exists then you are looking at buried lines. That means each time you want to cross someone's driveway or property you need to reach out to the existing owner to negotiate access. Some people are easy to work with, others will just flat out deny access. (The BLM, for example, is notoriously hard to work with for placing underground utilities. As are a few big churches).
The US, frankly, has the wrong model. Leaving everything up to private companies is what creates all these problems. The companies have absolutely no way to guarantee a route for service and they have little skin in the game to run an update.
At a bare minimum, the better model is to make utilities public (at very least the lines themselves.) That gives the government a powerful ability, the ability to claim eminent domain to cram through any improvement they need. You could still have private ISPs, but you can relegate their roles to just running the equipment (switches, routers, etc) rather than doing all the additional line management work.
The US doesn't generally do new things anymore. We just shuffle money around into peoples' pockets. So we have plenty of societal wealth to do something like this, even given how large and rural many areas are, but we just don't. Because no one gets reelected because they rolled out bandwidth. They get reelected based on however their party is faring in the battle of the spectacle going on nationally.
Depends on the place. Live in rural Georgia (Carroll County). Within the last two years have had Spectrum fiber and Carroll EMC fiber run by the house. First is underground and second is on pole.
To be fair, there will be plenty of people using EU funded fiber over here, to complain about how the EU does nothing and is pure evil, and that the country should leave the union.
Same deal in Prince Edward Island. Very rural, near the coast, cows across the road, but gigabit fibre from Xplorenet who took advantage of government funding.
I'm in the UK on a 900Mbps symmetric for £40 a month.
Turns out regulation can be a good thing - first of all, OpenReach (the infrastructure part of BT that was split up post privatisation) are regulated in how much they can charge other providers for last mile FTTP/C[1]
But then there's also Access to Infrastructure regulation[2], which means duct and pole access isn't up to the telco's (and after all, existing infra was publicly funded, and new infra is often paid for by housing developers). This means I have 2 providers of FTTP to my property - CityFibre and OpenReach (and a third offering DOCSIS cable in Virgin Media).
These combined have led to a thriving market with genuine competition. Whilst most ISP's use the two providers with the most footprint (OpenReach and CityFibre) they compete on things like technical chops (AAISP), Price (the PlusNets of the world) and a mix of the two (Zen).
Regulation, when well designed, leads to more competitive markets.
Well, there's also cultural reasons. If the government of Japan entrusts you to self regulate on the basis of open access and no restrictions a Japanese business will understand the weight of expectations that have been placed on them.
A US telco would pop the champagne and then plan how best to screw their customers, safe in the knowledge that they have few competitors after they got local governments to ban municipal broadband.
I live in the second largest city in the US. In the distance I can see a switching station that carries a large amount of telecom and internet traffic.
Should be easy to get fiber no? Turns out they want a couple thousand dollars from me to run fiber up my street. Coax it is!
coax was $5,000 a mile, here, 5-6 mile run (i forget), and they said "there's no guarantee it does anything except TV" - that is, no internet.
fiber loop from cogent or centurylink was $15,000. even though they have a loop a half mile up the road. Dark fiber is probably cheaper, but i cannot remember the pricing to run my own fiber via some legit company. i had DSL (6mbit), switched to my own CPE with cellular wireless, then their CPE after they stopped selling the SIM plan i was using, then starlink after they stopped selling their fixed wireless for some newer one that isn't supported anywhere but metro areas.
There's a tree interfering with satellites at my location, and it's still 20x faster and better than anything before; except my own CPE solution, starlink is only 2x as good as that.
In the US fiber has started to show up in very rural areas now as well. My parents got it last year and I never thought that would happen in my hometown.
It's some small company I never heard of before, back when I lived there Verizon got a ton of government money to do this and just never did, everyone was pissed about that.
I'm still pissed about Verizon pocketing billions in subsidies and not finishing their work. My area was next up for fiber installation before Verizon stopped, and so I'm stuck with Comcast or DSL. Comcast claims to offer symmetrical GiB up/down, but I've yet to get that. Now I can't even get in touch with a human for support without spending 30 minutes with a bot.
Weird opposite experience. I’m suuuper rural NY (what others have described as ‘American Siberia’) and for some reason Verizon quietly started building out their fiber network in the past several years again, something even I didn’t anticipate happening.
I always knew they had Syracuse and Rochester as major FIOS markets, but I was blown away to see the trucks all the way this far north and on my very road. Last I’d known they’d said they had stopped all buildout. Next thing I know, they pick back up their pace, then pick up the homework they let Frontier finish for them with the Frontier FIOS buildout, too.
Our nearest town, Durham Kansas (population 100) got buried fiber optic recently. The company is actually planning/hoping to expand to the rural area around town as well, so maybe we'll get it as we're only two miles out of town. They do offer 1gig symmetric but I think it's $200/month. I'd just like the reliability and low latency of fiber vs our current WISP.
It's a huge difference isn't it? When I used to stay here in the past, the DSL was awful. It made it very annoying to work online. Now, it's as good as back in Seattle.
Oh lol there wasn't even DSL, everyone wanted that. The options were dialup (over ISDN for a while which was one of the best options until that went away), geosynchronous satellite, or LTE if you're even near a tower.
That's why I'm so surprised they have fiber all of a sudden.
Starlink is now kind of a global baseline that ISPs have to compete against regardless of location. In rural areas it's very reasonable to expect 200-400mbps downloads and 20-40mbps uploads for $120/month. A bit pricey but it's a level of service that DSL and GEO sats can't even think about matching, so companies have to build cable, fiber, or 5G towers if they want to have a comparable offering.
I was taking a train from somewhere in Bavaria to Austria. More than half of the ride the train lady was chilling next to us because "there is no internet, I cannot check tickets" the whole ride you could see houses out of the window. People are actually living there. This was like 6 years ago but I am still baffled by this.
In Switzerland it's very unlikely you find a mountain or road without 4g
"Today passengers enjoy at least 200 Mbit/s on 99 per cent of the 7 800 kilometres of main lines and even 300 Mbit/s or more on 95 per cent. Secondary lines also saw a transformation. Coverage of 100 Mbit/s rose from under 83 per cent to over 96 per cent in just three years."
However:
"Yet coverage is only half the story. Mobile signals must penetrate each carriage’s interior if passengers are to make calls or stream without interruption. Many modern trains are fitted with factory-installed windows engineered for signal permeability."
And of course many train's aren't fitted with windows like that (and operators are trying to retrofit them with microcells or in other ways).
Which self own? Besides the one making your industry dependent on energy from your military opponent? Besides the one putting all your eggs in selling diesel engines when China and the US were betting on computer driven battery powered EVs? Besides the one where you open your borders to unvetted illegal immigrants leading to a rise in crime, terror attacks and right wing extremism all over Europe?
I could have fiber, but I see no point in upgrading from DSL. Why? It works. In fact, I could have upgraded to 250 Mbit DSL years and years ago, but 100 Mbit is fine. Why bother spending money or time making changes to my home's critical infrastructure. I doubt I'm the only one.
And then there's the millions of boomers around whose only device is a smartphone and they never exceed the 10 or 20 or 30 GB they get on a mobile contract that's less than any DSL or fiber contract. Good luck selling them 900 Mbit symmetric links.
In fact I'm sure there must be hundreds of thousands who still pay for DSL that's essentially unused because their phone lost the Wifi credentials and there's no grandkid around to notice it. Their house will be upgraded to fiber when it gets resold because it ticks a box for the real estate agent.
> I could have fiber, but I see no point in upgrading from DSL. Why? It works. In fact, I could have upgraded to 250 Mbit DSL years and years ago, but 100 Mbit is fine. Why bother spending money or time making changes to my home's critical infrastructure. I doubt I'm the only one.
We get 250 Mbit into our house and our house has Ethernet cables going through the walls to different floors but the cables are all limited 100 Mbit and it is tilting me of the face of the planet... :(
Replacing them isn't much of an option, just maybe running other cables around the walls, which isn't the nicest option when the existing cables are all nicely not visible.
>I could have fiber, but I see no point in upgrading from DSL. Why? It works.
How do you explain to a caveman that a living in a house is nicer than living in a cave?
>Why bother spending money or time making changes to my home's critical infrastructure. I doubt I'm the only one.
You don't have to pay, In Eastern Europe telcos pay themselves to wire fiber to your door in hopes you'll sign up with them. It's something you don't have in Germany, it's called free market competition. The German mind just can't comprehend having competitive companies that serve the consumers and not sclerotic state supported monopolies that only support the shareholders by robbing the consumers. Why are you guys like this?
>And then there's the millions of boomers
Of course, the way to move forward technologically is to cater tech to boomers habits. That's why Germany is a SW innovation powerhouse.
> You don't have to pay, In Eastern Europe telcos pay themselves to wire fiber to your door in hopes you'll sign up with them.
It's wired, I just have to sign up. But 1000 Mbit is more expensive than the 100 Mbit I have now. I just don't see the point. If that makes me caveman, so be it. Work on your manners.
No, they didn’t. They decided not to raise the definition of broadband from 100×20 Mbps up to 1Gps.
And that’s simply because 100Mbps is actually a lot of bandwidth. We just don’t have any killer applications that need more than 20Mbps. 4K streaming on Netflix, to pick one salient example, only requires 16Mbps. At 100Mbps you can download a 50GB game in just 80 minutes. Sure, it would be nice to have 1Gbps and download the game in 8 minutes, but the definition is about _minimum requirements_, not things that are _nice to have_.
As such the definition we have is actually a good one! If every American had broadband according to this definition we will have actually made real progress. Nobody would be stuck on DSL any more, let alone dialup.
The FCC is still giving grants to anyone who is installing actual broadband in unserved areas, which means anyone installing cable or fiber in areas that don’t have them. It’s just not requiring that every customer have 1Gbps service in order to qualify.
> No, they didn’t. They decided not to raise the definition of broadband from 100×20 Mbps up to 1Gps.
No, the 100/20Mbps definition was only adopted last year and is being kept for now (though you'll never guess who voted against adopting it and wanted to keep 25/3Mbps!).
The 1000/500Mbps definition that Carr's FCC is trying to eliminate was a long term goal, something that the US should strive for eventually, and therefore federal funding should preferentially go to solutions that can or could eventually provide those speeds.
Carr's proposed rules[1] go well beyond just the definition of high speed broadband, though:
- He wants a stricter reading of the statute that requires a report on the deployment of broadband, reading "whether advanced telecommunications capability is being deployed to all Americans in a reasonable and timely fashion" as strictly referring to anyone making "incremental progress" on deployment, so the reports will shift from reporting coverage where broadband has actually been deployed to calling areas covered if anywhere in an area is working on deploying broadband there
- He also doesn't think "reasonable" access to broadband should include a consideration of the price of using that broadband, so wants to stop collecting pricing information. So long as someone is making incremental progress in an area that would make 100/20Mbps internet available to you at some price, you will now count as covered
- Other proposed changes around mobile speeds and school broadband I haven't been following.
Incidentally, even the 100/20Mbps definition seems at least potentially in danger, as the proposal is requesting comment on whether that's actually the definition they should be using in the future.
> The 1000/500Mbps definition that Carr's FCC is trying to eliminate was a long term goal, something that the US should strive for eventually, and therefore federal funding should preferentially go to solutions that can or could eventually provide those speeds.
Yes, that’s true. But don’t ignore their justification: the laws as passed by Congress say nothing of long term goals other than bringing everyone up to some minimum threshold. The FCC gets to define that threshold, but making up new goals on top of that is not actually supported by the law. Plus, with so much of the country installing fiber it’s already just happening automatically.
> He wants a stricter reading of the statute that requires a report on the deployment of broadband, reading "whether advanced telecommunications capability is being deployed to all Americans in a reasonable and timely fashion" as strictly referring to anyone making "incremental progress" on deployment, so the reports will shift from reporting coverage where broadband has actually been deployed to calling areas covered if anywhere in an area is working on deploying broadband there
You almost have it right. They want to actually measure improvements in access rather than just the number of people with access. Sure, if the number of people with access goes up then the situation is getting better. But what if all of the improvement is because people are moving to denser areas to get better internet access? It makes complete sense to measure the actual geography of access in addition to just the number of people who can access broadband.
The report states clearly that they intend to base their analysis on the FCC’s Broadband Data Collection <https://www.fcc.gov/BroadbandData>, aka the National Broadband Map. They just want the report to measure improvements to the number of addresses served not just the number of people served.
They also point out that while the National Broadband Map has addresses of every household, it doesn’t actually have any information about the population living at those addresses. It still uses Census Districts to estimate that, assuming that every address in a district houses the same number of people. They’re going to keep using that same estimation method, probably for the foreseeable future.
> Incidentally, even the 100/20Mbps definition seems at least potentially in danger, as the proposal is requesting comment on whether that's actually the definition they should be using in the future.
As I’ve demonstrated, there is significant headroom in the current definition. Would someone with 40×10Mbps DSL (provided using two phone lines) be significantly disadvantaged over someone with 100×20Mbps cable? They’ll still be able to stream 4K television and send HD video. Few ISPs offer this service, but I used to get it from Sonic.net more than a decade ago. I had to jump through some ridiculous hoops to get two actual real functioning telephone lines wired into the same apartment, but it was great. Even 25×3Mbps enables you to stream 4K television and participate in video conferences. The constraints on the quality of that video conference are real, but are they enough to handicap people? Yea, I think so, but it’s clearly a judgement call.
Furthermore there’s no indication that they actually want to change that definition immediately. Agencies are more or less required to invite comment from the public on all of their regulatory decisions:
> IV. A. 18. Broadband Speeds on Which to Report. The discussion in the 2024 Report focused on the availability of fixed broadband at speeds of 100/20 Mbps.⁴³ We propose to again focus our service availability discussion on fixed broadband at speeds of 100/20 Mbps and seek comment on this proposal.
the USDA, which gives broadband grants, considers an area "served by broadband" if there is a 2 megabit provider in the jerrymandered area. This includes cellular, even if using a fixed or hotspot link would be cost prohibitive.
Look at the coverage maps for rural areas with trees, and you'll see an interesting pattern. They used multi-pathing to their jerrymandering benefit. checkerboard pattern means the USDA says "no grant for you!"
i have no idea when or if the definition changed to 20mbit. there are people here still served by 6mbit - on a good day.
That used to be the case, but it isn’t any more. Check out the new broadband map: <https://broadbandmap.fcc.gov/home>; it goes by exact addresses rather than census districts or whatever it used to be.
Please re-read what i said. i said USDA, not FCC. As i tried to start a WISP and could not get any federal funding for it because of the gerrymandering. this was within the last 10 years.
It turns out the USDA changed it. February. of 2024. not 2015 (or whatever). So my information was good until about 18 months ago.
I know it often seems like i have no idea what i am talking about, but i very specifically said USDA.
note: they also changed it so cellular no longer counts, amazing:
> Sufficient access to broadband, as defined in the NOFO published on February 21, 2024, is 25 megabits per second (Mbps) downstream and 3 Mbps upstream (25/3). Mobile/cellular and satellite services will not be considered in making the determination of sufficient access to broadband. Additionally, areas with current broadband service from unlicensed wireless facilities, or have an enforceable commitment associated with unlicensed wireless facilities, are eligible for funding under this notice.
here's the old section (from 2018, you're not paying me to prove exactly the numbers i said, which are older by a year or two):
§ 1738.53 Eligible service area.
(a) A service area may be eligible for assistance as follows:
(1) For loan and loan/grant combinations, the proposed funded service area is completely contained within a rural area. For loan guarantee applications, the proposed funded service area must be contained within an area with a population of 50,000 or less, as defined in 7 U.S.C. 1991(a)(13);
(2) For loan/grant combinations, at least 90 percent of the households in the proposed service area must not have access to broadband service. For loans and loan guarantees, at least 50 percent of the households in the proposed service area must not have access to broadband service;
(3) No part of the proposed funded service area has three or more *incumbent service providers*; and
(4) No part of the proposed funded service area overlaps with the service area of current RUS borrowers or grantees with outstanding obligations. Notwithstanding, after October 1, 2020, the service areas of grantees that are providing service that is *less than 10 Mbps downstream or less than 1 Mbps upstream* will be considered unserved unless, at the time of the proposed application, the grantee has begun to construct broadband facilities that will meet the minimum acceptable level of service established in § 1738.55.
§ 1738.58; (c) Applicants will be asked to remove areas determined to be ineligible from their proposed funded service area. The application will then be evaluated based on what remains if the resultant service territory is de minimis in change. Otherwise, the Applicant will be requested to provide additional information to the Agency relating to the ineligible areas, such as updated pro forma financials. If the Applicant fails to respond, the application may be returned.
please note the starred parts, as well as 1738.58 (c) which explicitly says that it's gunna be gerrymandered.
Ok, apparently the USDA also provides their own separate grants for broadband deployment. How are their weird rules relevant to this discussion of the FCC’s grants program?
I agree that it would be really nice if it were symmetric. But remember, the definition is for a service that isn’t handicapping people, not for what would be nice to have. Maybe in 10 years we can make it symmetric.
No, I completely disagree. Remember that the whole point is to define the _minimum_ level of service at which people can reliably access “advanced telecommunications capability”. Here’s the definition:
> The term ‘advanced telecommunications capability’ is defined, without regard to any transmission media or technology, as high-speed, switched, broadband telecommunications capability that enables users to originate and receive high-quality voice, data, graphics, and video telecommunications using any technology.
By this definition, 4K video is a reasonable metric to judge whether a broadband service qualifies as an “advanced telecommunications capability” or not. Since the major streaming services all use less than 20Mbps to serve up 4K television, 1000Mbps is clearly not necessary. It certainly is nice to have, but it’s not _necessary_.
I think that fiber access will continue to grow as long as we remove the regulatory barriers that hamper it. Permitting is a big one. Cities often require separate permits for every street, often for every block!
If you have to install anything, then installing fiber is the obvious choice. It has immense headroom for the future; fiber optic cables that carry a gigabit today could easily carry 100,000× that just by buying fancier equipment to install at both ends. The upper limit is currently hundreds of terabits per second, and it goes up every year. We’re nowhere near the theoretical limitations of optical communications technologies. I think that ultimately this means that fiber will win out over coax eventually. But coax has headroom still, and there’s no reason to tear it all out where it’s already in use.
We should increase competition by abolishing local monopolies, but that’s something we should do no matter what level of service we adopt as the definition of broadband.
Most streaming services have extremely bit-starved encodings for their 4k streams and fall far short of the quality that a proper 4k video has.
They are also doing their encode much slower than real-time so they can achieve higher compression rates.
The definition also says "originate and receive" so you need to look at both upload and download bandwidth.
4k also isn't high enough quality for many use cases like VR.
Households also contain multiple devices, so just a single video stream isn't sufficient.
I would argue that by that definition they should be setting the forward-looking threshold as sufficient to both upload and download multiple live 8k video streams simultaneously.
> The definition also says "originate and receive" so you need to look at both upload and download bandwidth.
Yes, I agree that a symmetric definition would be better. The definition we have is pretty good though. The 20Mbps of upload is enough for 4K video streaming. It would be nice to have some more head room so that more than one person in a household could stream at full resolution at the same time, but the market has already solved that problem. Every market where 100×20Mbps broadband is available also makes higher upload speeds available.
> 4k also isn't high enough quality for many use cases like VR.
This is not a good argument. VR relies on local rendering of 3D assets, not streaming video.
> Households also contain multiple devices, so just a single video stream isn't sufficient.
100Mbps has headroom for, let me calculate here, 100/16=6 whole 4K video streams. If your whole family needs more than six simultaneous 4K video streams to live then your family has problems.
> …sufficient to both upload and download multiple live 8k video streams simultaneously.
This is just dumb. 99% of computers can’t even play 8K video; they need specialized hardware just to decode 4K video in real time and the hardware cannot handle 8K video. The cameras for it don’t exist, and neither do the displays. And finally, the FCC definition of broadband is about raising the bandwidth floor, not raising the ceiling. The ceiling is already high enough by far; I can get 50Gbps service at my house if I wanted to pay for it.
But also 8K video would still use less than 100Mbps of bandwidth!
I disagree. I would argue that most people just don’t produce enough data to have a huge need for faster uploads. You would have to produce more than 200 gigabytes of data per day before your backups would take longer than a day to finish:
$ units 20Mbps*24hrs gigabytes
* 216
If you’re an entertainer then you might record more than 200GB of raw video per day, but most people aren’t. You wouldn’t even need a faster connection to _become_ an entertainer, even if you soon wanted one. You could stream 4K video of your antics all day and it would be less than 200GB.
And remember that the definition is about the _minimum_ bandwidth necessary to participate in society, not what you need to be at the peak of your entertainment career. People who do need higher speeds can and will pay extra for them; the FCC definition is not about limiting what products are available. It doesn’t even require anyone to have 100×20Mbps service. The FCC is just trying to get us to a point where we can say that 100% of Americans have _access_ to that level of service, even if they have _subscribed_ to a lower level of service to save money. Since 45 million Americans don’t even have access to 100×20Mbps service we’re still pretty far away from that.
Average household is a family, who arent streaming 4k video or uploading huge backups like i do, and 20Mbps would be fine for me. They typically watch netflix and youtube and doomscroll.
> At 100Mbps you can download a 50GB game in just 80 minutes.
100 Mbps is 100x as slow as my internet connection. I would really rather download a giant game in less than a minute, or even a few minutes, than more than an hour.
Of course you want it to be faster rather than slower. But waiting an hour to play your new game is not a handicap. Not compared to dialup, where that same download would take more than 80 days, if it finishes successfully at all:
Actually it’s mostly because we’ve spent the last few decades turning math into compression. Starting with MP3s in 1991 there’s a whole history of amazing improvements to audio and video codecs. We’ve made extremely rapid progress even though the process has frequently been weighed down by ridiculous software patents.
But that said, there’s nothing stopping video streaming services from offering higher bitrates if they think it’s worth it. After all, the FCC standard for broadband is 100Mbps, not 16Mbps. That’s a lot of headroom if you want better quality.
It really is mind blowing to see how much smaller a file with newer compression can be. Had to convert some videos to a codec that would work on my kids chromebook recently. What was a 300 meg video ballooned up to 2 gigs. One of the videos was 130 megs that jumped up to 3.2 gigs. Time lapse video, so I'm sure it had some of the sweet spots for compression. Still, was dramatic how much a modern codec can get that down.
> We just don’t have any killer applications that need more than 20Mbps.
Sure we do, but large ISPs demand their cut.
Backups. File transfers. Large games. Live 4K video chat. Language models. CAD models. Cloud-based spyware, like smarthome/car/phone/whatever telemetry, or security cameras.
> 4K streaming on Netflix, to pick one salient example, only requires 16Mbps.
So only the living room gets 4K, but family members can't all watch their own 4K streams in their own rooms? Bobby talks during the show, Sally giggles at everything, and Timmy wants everything to be dark while mom & dad want the whole room bright and quiet. Thanks, ISP, for forcing the family together!
Remember, we are talking about minimums here, the least common denominator. At 100×20Mbps you are not handicapped like you are if you are stuck on dialup.
Live 4K video chat needs less than 20×20Mbps, but remember also that most people don't have 4K televisions or 4K cameras. Even gamers mostly don’t have 4K monitors! The most recent Steam hardware survey <https://store.steampowered.com/hwsurvey/Steam-Hardware-Softw...> shows that just 5% of gamers have a 4K or better display. 55% have 1920×1080. I don’t have similar statistics for televisions or webcams.
The other uses you list require even less bandwidth. Nothing about a language model requires bandwidth. You either interact with it remotely, at a few kilobits per second, or you run it locally using no bandwidth at all. Large games might take an hour to download instead of minutes. Waiting for an hour is not going to handicap you. Most people shouldn’t even pay extra for that. Backups need some bandwidth, but only in proportion to the amount of data you have generated. Most people don’t create hundreds of gigabytes of new files that need to be backed up remotely; the largest files most people create are photographs and videos of family events, vacations, etc. These files can be backed up with no difficulty at 20Mbps.
> So only the living room gets 4K, but family members can't all watch their own 4K streams in their own rooms?
Pay attention and don’t be an idiot! The FCC definition of broadband is 100Mbps down and 20Mbps up. Netflix tops out at 16Mbps down, so those 100Mbps can supply 100/16=6 whole 4K television streams easily. If your family of six people is sitting in six different rooms watching six different television shows then your family has a problem. That problem is not a lack of bandwidth.
Another side of the coin is no one will ever serve you content at 1gbps. It just won’t happen. I get 500mb down on speedtests and netflix et al regularly throttle me to like 144p quality with constant buffering. I download a game on steam and it throttles to nothing for hours after making some initial good progress.
It’s quite rare for the average website to actually hit 1Gbps, that’s certainly true. Nobody makes webpages big enough to hit that speed, or even for that top speed to matter.
But I can download games from Steam at approximately full speed, for example. I subscribe to Ziply Fiber, and they certainly don’t throttle their users or oversubscribe their bandwidth. That said, there are other factors at play as well: can you even maintain 1Gbps to your _hard drive_? Can your computer decompress the downloaded data as fast as it comes in? Steam will slow down the download to match the speed at which it can decompress the data and write it to your disk.
Fiber just came through my area. They offer up to 3Gbps for less than I was paying for Comcast ~500Mbps asymmetric and for more money I can get 5Gbps... but I just signed up for the 500Mbps symmetric and pocket the difference monthly, because what the hell am I going to do with even 1Gbps? My Wifi can't 5Gbps, and all but two network devices in my house use Wifi to get to the internet. My NVMes can nominally do it, but it takes everything firing on all cylinders to actually achieve that. I've still got some spinning rust that is pretty full up at even the 500Mbps. I do run backups to AWS, but that runs in the nighttime anyhow and could still finish a complete non-incremental backup in 4-5 hours at full speed, and I have incrementals anyhow. Sure, the game per month I download from Steam would be ready in 4 minutes instead of 8, but, seriously, how much am I willing to pay for those four minutes? It's not like I'm staring at the progress bar at that point anyhow.
500Mbps is already enough for me to tailscale my house network up and have every single member of my family accessing the house Jellyfin server remotely simultaneously, which is not a realistic amount of load.
100Mbps down is still plenty for most people. 20Mbps up is definitely making some things annoying but most people will still be fine. It's a fine definition of minimum service for "broadband".
even spinning rust should be able to handle ~700 MBit. an SSD is generally on the order of 3 gigabits - even cheap ones can manage 300megabytes per second, which is on the order of 3Gb/s.
Sure, but remember that the disk bandwidth is the _uncompressed_ data. By definition it’ll be at least 2× larger than the compressed data you downloaded, if not more. Steam downloads are commonly disk or cpu limited on 1Gbps connections.
i'll concede you're probably right. In the late 1990s i had a Promise ATA 100mbit IDE card, and i swore that someday, i would have an IDE connection to the internet. Now, my CPE is called a terminal, but it's some proprietary cable from the terminal to the router. And my upstream isn't quite 100mbit (on starlink; it's about 40-50. On at&t 4g LTE it was 30-65mbit/s upstream).
so i'm still not at IDE speeds, yet. maybe on my cellphone in the metro i get that speed, though...
It is really ridiculous how fast our connections are these days, when you stop and actually think about it. Also pretty ridiculous how slow they used to be!
Interesting. Are ISPs known to throttle steam or something. I’ve noticed that steam almost never downloads at the same speed of get doing a speed test. I’ve noticed it many times through out my life, though admittedly I’ve been stuck either way the same ISP across many states.
Some ISP throttle sustained download, which explains having high test speeds and initially fast downloads that slow down with time. So it might not be Steam per se that they’re throttling, though it’s not impossible.
Steam also extracts while it's downloading and will slow download speeds if there is a healthy buffer of things to extract. If you're downloading a game with thousands of small files to an HDD or even just a cheap SSD chances are you'll be throttled by your own computer's throughput rather than your Internet connection.
With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money. That said, I've always felt satellite internet to be yucky but I'm used to "satellite internet" adding 1000ms to your latency and uploads that actually run over your telephone line.
I am curious if good satellite internet will lead to an exodus of people from cities and subdivisions. I actually live in my current house because the place I wanted to build on didn't have any form of internet access.
Also, I'm not really aware of anyone that is truly without "internet access" as cell phones have basically filled that area in years ago.
> With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money.
That was the same excuse used when nobody wanted to wire rural America for electricity. It was a bad take then and a worse take now.
Wiring those homes with single-mode fiber once will provide modern broadband for at least the next 50 years, if not longer.
Satellite is not and will never be a replacement for fiber.
> Satellite is not and will never be a replacement for fiber.
Plenty of suburban homes don't have fiber availability and are just fine. I'd rather have 3 coax companies competing for my business than 1 fiber option. I just really don't care that much about speeds above what coax (and Starlink) have to offer. Honestly, I'm on the lowest tier offered by my cable company, and I'd go lower if it would cut my monthly bill by a commensurate amount.
> I'd rather have 3 coax companies competing for my business than 1 fiber option.
And I’d rather have the city or county pull a single fiber back to a pop where an ISP can compete for my business because it’s absolutely absurd to have multiple companies pulling the same cable to a single address and using that last mile as a moat.
It's not just the limitations of satellite tech or quantity. There's also just the fundamental limit of RF in shared airspace. You run into bandwidth limits due to interference even without overcrowding low earth orbit with a satellite network. When you're running signals over a wire/fiber, your signals are confined and interference is managed relatively trivially.
> Huh? People used satellites as an excuse to not wire rural America for electricity?
Obviously that’s what I meant. And I didn’t mean people used the excuse of “we shouldn’t be pulling cable to all these houses when X is good enough”.
There were endless excuses to not electrify rural America including “they don’t need it”. It was eventually solved through co-ops.
That’s exactly how most rural areas are trying to solve fiber, but of course they get to fight the combination of folks like you that think “satellite is good enough” (it isn’t), and legacy ISPs suing them to slow or stop deployment.
Having spend several years under the thumb of a rural wireless internet service provider, the thought of all my neighbors sucking up shared Starlink bandwidth to stream Netflix in the evenings makes me nauseous. How is it a reasonable expectation that Starlink can really support entire rural areas?
Rural areas in the US have enjoyed disproportionate communications (and infrastructure in general) subsidies since the founding of the USPS. More recently, the Universal Service Fund has spent around $5-8B per year since the 90s, first on phone service then adding broadband. And there are more recent broadband efforts for broadband in the tens of $Bs.
How effective have these efforts been at getting broadband access in rural communities? And comparatively, how effective have they been at giving a revenue source for telcos that they haven't had to really deliver on promises for?
1. 25/2 DSL for $140/month (12 month contract with $40/mo seasonal disconnect)
2. 5G (60/0.03 which was 100/20 last summer until the nearby tower became overloaded).
3. LTE (see above) which I routinely get ~100/30 but which Verizon, at least, will not allow their boxes to use if it can connect via 5G, regardless of usability of the connection.
4. Fiber (100/100) at $89.95 (12m contract with $25/month seasonal disconnect). The fiber has higher speeds, but I didn't price them out. The costs were originally listed at $34.95 before they ran the lines, but they have upped them to $89.95 now that they are run to my lake home.
5. Starlink (supposedly 150/20 for Residential Lite which is available in my area) at $80/month on a month-to-month with purchase of a dish (I got mine refurbished for $135). I am routinely seeing ~400/40 with 25ms ping even though I shouldn't be. That said, the speeds are more variable (low as 120/7).
---
I would LOVE for there to be reasonably priced and very stable Internet for month-to-month, but there just isn't except for Starlink--at this point.
My cabin (I don't live there full time currently) has starlink's $10/month plan where you pay by the gigabyte. LTE isn't even an option because the topography kills terrestrial radio.
I don't know how you think $89.95 for 100/100 fiber isn't reasonable, unless you're the type of person who believes the introductory price at comcast is the real price.
The infrastructure has to be maintained year round, not just the summer months when lake houses are full. I don't think it's reasonable to expect affordable month-to-month pricing.
I explained below how it is defined by me as rural; but, yes, it's rural. In fact, my primary home JUST HAD FIBER RUN in my neighborhood last month and I get 1 bar of VZW. So; if anything, rural is ahead of major metro areas (my suburb of Minneapolis/St Paul has just under 90K residents). In fact, DSL from CenturyLink or whatever they call themselves today is 6mbit.
It's in Central MN. It is 11 miles, straight line, from a town of 2200 and 15 miles from a town of 4400. The closest town is 8 miles as the crow flies of 700 and only has a gas station and a couple of bars. There's a bar and volunteer fire department 4 miles as the crow flies that has a population of 48.
Believe me: this is rural. Until investment from the federal government and/or forced lower-bound limits on 'broadband' back about 6 years ago, the only option was 768/300 DSL.
I'm in a small town (5000 people) on the Utah/Arizona border. The internet options are better, cheaper, and more reliable than I had in old town Mountain View, Palo Alto, or Sunnyvale. I go with the symmetric 1 gig with static IP, but symmetric 8 gig is an option as well. Silicon Valley's home internet options just suck.
> I am curious if good satellite internet will lead to an exodus of people from cities and subdivisions
You seem to be imprinting wireline capacity onto satellite. This will always be an error. A core facet of satellite is it's inherent limited capacity. This has always been the case and no tech is on the horizon that will change this.
What makes satellite have an inherent limited capacity? Like there's no real physics limits on how big you can make a satellite. More reception power = more signal in the same frequency range.
> Like there's no real physics limits on how big you can make a satellite.
There absolutely are. Launch weight, heat dissipation, energy storage and output are a few of the engineering boundaries that must be respected. You can't bigger one without impacting all the others.
It also has to fit in the launch vehicle with whoever is sharing a ride with it.
Our theoretically ginormous satellite might not much improve capacity (in the cell it's servicing) due to limitation in the assigned spectrum. You can only divide up time and frequencies so far.
Whatever you're doing here, you will need to duplicate that to cover the nearby cells. Then multiply that by a region and that by a country and that by a planet. It takes a lot of satellites to cover a lot of area from a low-ish altitude.
Satellite internet has its place but it's never going to be dominant. Any area with high density is better served by wires, and any area with medium density is better served by a combination of wires and cellular internet. Satellite only make economic sense for the truly remote; the astronomical and recurring infrastructure costs are just killer (Starlink satellites last for 5 years, buried fiber lasts for 50 at least (while providing better latency and throughput)).
> Also, I'm not really aware of anyone that is truly without "internet access" as cell phones have basically filled that area in years ago
Unfortunately very much not true. The US is a very large and sparsely populated place. As far as cell and broadband service goes, much of West Virginia, huge chunks of PA, upstate NY, and Maine all have limited or no cell service. In the Midwest the communication infrastructure is often delivered by community co-ops, not corporations, that also deliver electric power. And dont get me started on the vast Western states.
The US has, frankly, an absolutely ludicrous patchwork of providers and infrastructure for telephone, power, and broadband. It looks reasonably peachy when you drive on the Interstate but get off there and youll see how limited it is.
Sat internet would really help people get from zero to 1, but that said, its still kinda crappy compared to fiber. Having the bird in range for 15 minutes until you have to handover to the next bird is not conducive to smooth, glitch free Internet.
> With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money.
Couldn't the same be said of expanding the electrical grid, when solar panels and batteries are an option?
> cell phones have basically filled that area in years ago
That doesn't work well in mountainous areas, which are often rural. There are three towers within a couple of miles of my house, but nothing close to line of sight to any of them. ~1/4 mile away I can get full service, but I have none at all at home.
Luckily Comcast / Xfinity rewired our little town a few years ago so we don't have to rely on satellite (not great during the heavy snowstorms and thunderstorms we get). I'd love to use cellular as a backup as we have 3~20 full days of blackouts per year (most of us have generators and/or batteries) and Xfinity goes down 80-90 minutes after a blackout starts.
Is internet access really the main determining factor for where people want to live? I know some people really dislike cities, but there are plenty of reasons to live in one if noise and crowds aren't dealbreakers.
Cell phones general do not work in rural areas unless you are very lucky or go climb the highest peak that happens to be in direct line of sight of a tower.
Upon reading the latter portion of your comment, a potential story for XKCD comic immediately sprang to mind...something along the lines of a tech person or team invent faster-than-lightspeed technology, but abandon their work on a spacecraft warp-style engines to improve satellite internet....because its just so much easier to make tons of money doing barely the minimum within the ISP monopoly business. lol :-D
> With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money.
A number of replies here explain how deploying wireline is absolutely a good expenditure of money.
Once you have read them, will you then understand?
> With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money. That said, I've always felt satellite internet to be yucky but I'm used to "satellite internet" adding 1000ms to your latency and uploads that actually run over your telephone line.
Depends on what you declare a "good expenditure of money".
The thing about running fiber is that it's not particularly expensive to do. dig trench, install infrastructure, rebury it. Hell, if you have a utility pole, you can omit steps 1 and 3, so long as you're okay with the line being exposed to the elements.
Launching satellites is very expensive to do. There are very high barriers to entry that prevent meaningful competition from occurring for satellite internet providers, which is why HughesNet was crap for all of those years. It took NASA funding a commercial crew program to get a rocket built that could put payload in LEO for "cheap", and there's only two or three options that can reasonably be expected to operate that way.
I'd rather have a lot of competition for installing and operating fiber infrastructure than having an effective duopoly for running performant satellite internet.
I'm under no delusions that will occur in rural areas under the present regime in DC.
As for an exodus from cities and suburbs, no. There's no amount of internet speed that would make me move to a place where there's reduced services and infrastructure in every other category, like schools, hospitals (this is a big one), water, sewer, transit, etc. and there are millions more like me.
> There's no amount of internet speed that would make me move to a place where there's reduced services and infrastructure in every other category, like schools, hospitals (this is a big one), water, sewer, transit, etc. and there are millions more like me.
I moved to "rural lite", and have a well and septic. That part isn't so hard to live with; although no water for a couple days when our well pump went out wasn't fun.
The hard part IMHO, is that everything takes so long to get to. We've got a good small town that's a couple minutes away, but if I need to go to a big box store, that's 30-45 minutes, each way. I like it here, and I have a 8 acre parcel, which is incompatible with suburbia, but it's an adjustment.
Utility poles may exist, but they may be structurally inadequate so the fiber installer may have to wait years for the pole owner to get around to replacing a string of poles before the fiber can be hung.
The poles may have been placed along the back property line without an easy way for a new utility to get in there without nicely asking each household on the street to let them into their back yard.
Spend a little time reading Sonic's "Access" forum. There's a very real have-vs-have-not dynamic going on there because there are areas where Sonic just can't deploy fiber any time soon that are just blocks away from people with service. Estimated service dates keep slipping when they run into unknown unknowns. There is significant engineering and construction work required in each and every neighborhood.
That have-vs-have-not dynamic creates niches for satellite internet service -- rather than having to wait until all the fiber ducks are in a row, you can just pick up a kit at a hardware store or get it delivered, plug it in, point it at the sky, and you're done.
Yes. This is also why people are talking about doing space solar power beaming. The cost of regulation and access has gotten so high that it's actually getting close to becoming cheaper to launch those solar panels into space instead of just installing them on the ground.
> The thing about running fiber is that it's not particularly expensive to do. dig trench, install infrastructure, rebury it. Hell, if you have a utility pole, you can omit steps 1 and 3, so long as you're okay with the line being exposed to the elements.
Uhhhh.... it can cost several hundred dollars just to dig a trench across my yard. If the plan involves providing broadband to rural farmhouses that might be miles apart, that could be tens of thousands of dollars per customer.
Obviously on utility poles (where they exist) is going to be much cheaper, but you still need an ISP to build and operate the thing and set up things like repeater boxes and deal with outages.
> Also, I'm not really aware of anyone that is truly without "internet access" as cell phones have basically filled that area in years ago.
I mean, this is basically it. For paying bills, doing school work, messaging, getting news, etc the options these days are affordable and very convenient. Even video calling and streaming videos doesn't really affect my data plan too much.
The only advantage of a hardline to your house is if you are going to do copious amounts of high-def streaming or you need a very stable connection. More of a luxury/hobby resource.
>With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money.
There are two perspectives on this. From the user's standpoint, when I finally wired up my house and we were no longer using wifi, all the streaming video was just flawless, no more stuttering. And I was only sharing that air with devices in my own house and the two closest neighbors. I wouldn't want to have to share it with all of North America.
The other perspective is from the point of view of the ISPs themselves (the ones not Starlink). Can they pay off the cost of the infrastructure, and then make a profit? And it turns out that they can. Just not a quick one. If they can't profit on it this quarter, many see no incentive. Someone needs to light a fire under their asses.
These are both obvious, and if you "don't see how", maybe you need to look.
This is hilarious. Just 3 weeks ago I signed up for 10 Gbps symmetrical fibre to the home with XGS-PON, and it costs me US$23/month (naturally, this is not in the US).
The only US argument I hear with respect to fibre broadband internet, high-speed rail, good healthcare, and more is 'we're too big!'. At some point it starts being genuinely funny.
I wish. Starlink has gotten almost zero money from the government (relative to their revenue). If Starlink got even a fraction of the money that fixed broadband companies got they could absolute explode he size of the service.
It seems like an increased executive mandate on the telcos is an odd thing to cheer for, and im not convinced a mandate from FCC is really necessary any more. The fiber installation (networking, boring, marketing) industry has dramatically expanded, and I would expect telcos to continue installation of fiber to new markets since it's so much cheaper to manage than copper. with the ubiquity of 5g cell networks, the amount of fiber extended in the last decade, and the vast improvements of sat-net... is there a need for every property in the US to receive a fiber connection?
FCC isnt giving on up broadband, its cancelling its mandate for expansion, and likely going to delete the cash payouts to telcos as well? The FCC never delivered on AFFORDABLE... telcos have raked it in on that front.
I can get 1Gbps up/down in a remote area (close to the fields) in a mid-sized city in Brazil with _zero_ government subsidies. The energy company that owns the poles was required to offer it to anyone that wanted to rent space on them and a dozen different fiber-only ISP's came to life.
I understand the US has a lot of pre-existing infrastructure and it's hard to move on (sunk costs) but.. the situation is just insane. If the government just stopped getting bribed and allowed competition, it seems this situation would be solved in a couple of years.
I have family in very rural east texas. They have 1Gps bidirectional at the hands of EasTex Co-Op spending federal dollars to actually lay fiber across their service area.
Lots of them took the money and ran, some jammed it into real infrastructure.
> Even under an ideal situation where Trump authoritarianism is conquered and some sort of sensible alternative takes office, restoring oversight of companies like Comcast and AT&T — both bone-grafted to our domestic surveillance networks — is never going to be a priority in a Congress that’s now too corrupt to function, under a broken court system that treats corporate power as an unimpeachable deity.
Well, that's not the ideal situation. The ideal situation is a wholesale countervailing takeover of power that destroys the broken congress and courts, replaces them with a genuinely responsive government, conducts a large-scale seizure of the assets of the entities and individuals who have profited from this debacle, throws many of them in jail for decades if not life, and then begins to use those assets to rebuild things on a more equitable footing. Now that's less likely than the scenario described in the quote, but the scenario described in the quote is far from "ideal".
With everything that has been happening maybe that is a good thing. We don't need even more people becoming radicalized, becoming flat earthers or pushing raw milk in public schools.
These problems may not be new but they have reached levels this country hasn't seen in decades and there is no sign of slowing down. The availability of cheap AI on the internet isn't going improve anything anytime soon.
When I bought my home in 2009 (suburbia in major metro area) - AT&T was offering a whopping 6mbps DSL connection for wired internet. At the end of 2024, they were still only offering 6 mbps DSL.
This year - they rolled out fiber. So, it only took 16 years - maybe you'll get lucky.
It's always been available - you just have to pay for the install.
Get a quote from your local ISP and you'll realize why it took 16 years to bring fiber to your home. I received a near $100k quote to come across the street in a former building... it gets stupid expensive stupid fast.
Depends where you are and where the line needs to be run. Shutting down a busy city road? Gonna be expensive...
At our current building, we just had fiber run from an in-ground vault already on premises (but not serving the building) into our MPOE - cost about $18k to go less than 100 yards... majority of-which didn't require boring.
My original point was, people want fast internet out in the boonies but aren't willing to pay for it. Few residential customers are going to shell out $600+ a month for fast internet... but that's what's required for the ISP to recoup their construction costs in anything resembling a reasonable time period.
We will pay for it over the course of the years of paying for the service. Plus, the economic benefits of enabling remote production/consumption of digital services probably outweighs the cost by an order of magnitude or two.
> We will pay for it over the course of the years of paying for the service.
Yes, at like $600+ a month over 6 years... you can do the math, it's nasty.
> Plus, the economic benefits of enabling remote production/consumption of digital services probably outweighs the cost by an order of magnitude or two
ISP's aren't a charity service. Even your municipality cannot afford to run fiber to the boonies.
Okay, whatever about ADSL2+, why on earth would anyone still be doing ADSL1? Are they buying the DSLAMs off eBay? Like, I'm legitimately surprised that they can still get the equipment; you'd think it'd be cheaper just to go to ADSL2+.
10 years ago I had ADSL2+ at home (basically due to planning issues; traditionally, urban phone stuff had been hidden underground, and the telecom was having trouble getting planning for VDSL street cabs), and that felt like living in the past back then. ADSL1, today, is absurd.
There's finally competition in the US home broadband market...
*Fixed Wireless players*
T-Mobile, Verizon and ATT (tho later two isn't widely available)
*Low Orbit Satellite Internet*
Starlink
*Home Wired Connections*
Comcast, Charter, etc
I've been bouncing my cell phone service and home broadband around every year per the all the good promotions. Recently on a $40 a month Comcast home broadband (300Mbps) with one free line of unlimited data cell/data service promo. Previously was paying $100 to T-Mobile for Unlimited cell/data service and their fixed wireless service.
We are much better nowadays for affordability then a few years ago!
I though this article would be interesting but it appears to mostly be a unhinged rant by the author to vent about Trump's politics. It was not particularly good educational reading.
Contrarian opinion: What exactly is the point of "fast" internet?
Most people use the internet for entertainment, people can survive watching Netflix at 1080p/720p, it isn't debilitating.
If I'm wrong in my assessment, give me use cases where you require fast internet in a rural area.
If the question is cost, I don't see how laying fiber and equipment for hundreds, if not, thousands of miles is a solution to reducing cost (unless it's subsidized by the government).
Spending billions on mostly "entertainment" is a waste.
> Most people use the internet for entertainment, people can survive watching Netflix at 1080p/720p, it isn't debilitating. If I'm wrong in my assessment, give me use cases where you require fast internet in a rural area.
Lots of desire for more automation and data surrounding agriculture, for starters.
More generally, people work through internet applications. If you want to check your email, type up a document, have Zoom calls, etc., you need to do so through the internet. If you don't have fast internet, you can count on being less productive.
That still would be faster than dialup, older generations of satellite internet, and some tiers of DSL service can provide, and there are still areas where those are the options.
2. How large is each "diagnostic medical image"? I can't imagine this being in the order of TB
a. An existing 5G/LTE or satellite internet is perfectly serviceable for even GBs of medical images
b. Let's assume the images are in the order of TBs, does this justify spending billions for one person (or tens) in a "rural" area?
People were perfectly fine with 56k when most of the Internet was just text. Stalling the internet at that speed could have been argued for by saying anyone who wants DSL/Cable at home just wants it for Napster.
Fast Internet speeds allow for unthought of innovations. If we get up to terabyte speeds, maybe nobody cares about watching Netflix if we can now have holodecks that become fully immersible experiences that allow for educational training and, to your chagrin, entertainment as well. Maybe LLMs are basically free because everyone can just have all of human content constantly updated.
Letting speeds stall stifles innovation and let's Netflix just continually up their prices for the same 720p content you refer to.
> maybe nobody cares about watching Netflix if we can now have holodecks that become fully immersible experiences that allow for educational training and, to your chagrin, entertainment as well
I don't see why billions of dollars should be spent, for say, less than a low single digit percentage of the rural population who want to stick a machine to their head for educational training.
It's infinitely cheaper, a more sane and healthier decision to move to the location where they can sit in front of the instructor, if they value this education.
If we're going full reductio ad absurdum and taking snipes instead of conversation, then maybe the appropriate response is: You can be poor all you want, but don't expect someone else to subsidize your life.
My comment isn't really that absurd in the context of this whole thread and its insistence that decent quality internet service is something only the wealthy and/or those who live in urban areas should have access to.
You mean the FCC was actually trying before? /sarcasm
In all seriousness, we’ve poured billions of dollars into broadband expansion efforts since the early 2000s. Every single time it’s been largely hoovered up by Big Telecoms, failed to expand broadband/improve speeds/lower prices, and basically just gone right to their bottom line as a subsidy.
The solution all along has been funding municipal broadband as the baseline for private enterprise to compete against and surpass, but lobbyists have all but killed that dead up until the past ten years or so. You can’t treat broadband as a utility in legal language but not in practice, yet the USA seems perfectly fine with their status quo leaving them a laughing stock of the developed world.
I live in a place with a municipal telco, and let me tell you, it is night and day. The service is top notch, and it always just works to the point where on the rare occasion that they do have a problem, I spend hours trying to figure out what's wrong with my equipment, because it happens so rarely. It costs more than the local cable company, because the price you pay is actually what it costs to deliver the service, but there are never any wild swings in prices, and when service gets cheaper, you just get upgraded for no additional cost. I wish all companies worked this way.
Sonic in the Bay Area is similar and provides amazing quality and customer service. They also support net neutrality. I hate the big telcos.
Yea, Sonic.net is amazing. I definitely recommend them if you are in their service area. Sadly they have been really slow to expand their fiber service, probably because they started in San Francisco, home of the NIMBY.
Ziply Fiber is another ISP that I can wholeheartedly recommend. They have service all over the north west states of Oregon, Washington, and Idaho (though still not _everywhere_). Like Sonic.net, their quality of service is excellent. No congestion, very limited downtime, no bandwidth caps. They’ve even posted pictures of the accidents that have caused downtime. Most recently that was damaged battery banks that released enough hydrogen gas to shut down and evacuate a central office. Before that it was a drill that took out a fiber bundle; cue pictures of the engineers splicing all the individual fibers back together.
We need more local ISPs, municipal or otherwise.
Also community/municipal owned broadband is illegal in many places thanks to lobbying from the big telcos: https://broadbandnow.com/report/municipal-broadband-roadbloc...
This is almost the greater crime. Federal action on this should protect municipal wifi. Instead we get... this.
Yes.
When Democratic Party controlled the federal government, they tried to. It is generally hard to do because it requires funding and passing laws. Another difficulty is that voters don’t reward Democratic Party for their efforts.
It is very easy for Republican Party to revert the changes. Moreover their donors reward the republican politicians for their efforts and their voters don’t punish them for their efforts.
+1 and more to this. Democrats are not good at all at letting voters know what they have done for them. In California that Republicans love to hate, we have clean air, free beaches, more protections from corporate predation, and so much more. Republicans are better at complaining and spreading fear, which sadly is a lower-energy and more effective method for getting votes.
We got an overview of those "efforts" a few months back and it was both comical and enraging to hear how much money and time was wasted. And this was a law+process crafted by Democrats, signed by a Democrat President, and presided over by a Democrat Administration.. they had all the pieces in place to accomplish their goals.
Ref: https://youtu.be/NcZxaFfxloo?t=990
When you have tons of paperwork and ZERO deployments (not exaggerating), there's nothing to revert.. just a lot to be disappointed by.
Jon Stewart has a good point there as well, saying “the Democrats think it can’t solve anything unless is solves everything”. Over the last decade or so they’ve weighed everything down with stupidity. Ezra has specific examples; that’s a really good interview.
Another good quote: “It’s almost as if they had designed this machine to make sure that people in rural areas _never_ get broadband…”
Ooh, Ezra: “This is the Biden administration’s process for it’s own bill. They wanted this to happen. This is how liberal government works now.”
Nonsense. It was working fine and following close to the expected schedule. Numerous states had submitted their proposals for how they would use the money, those had been reviewed and approved, and the states were accepting bids from companies to do the construction and starting to select and approve them.
Delaware, Louisiana, and Nevada had selected contractors and passed final federal review and were ready to start construction as soon as the funds were disbursed.
Another 40 had contractors selected and were just waiting for final federal review to complete and funds to be disbursed. If the program had not been changed by the Trump administration most of them would have had construction starting this year.
All large efforts like this take a few years to go from initial passing of the law that enables them to construction actually starting.
Bug: Closed. Works as designed.
> Another difficulty is that voters don’t reward Democratic Party for their efforts.
It is worth noting that the Biden admin passed a $42B package of broadband rollout funds in 2021, and then connected ZERO people over the next few years.
This is because special interest groups saddled it with environmental, union, and DEI requirements that were so onerous that nobody was able to actually get/use the funds.
Yes, Democrats plan to do the right thing, but frankly they should be penalized for boondoggles like this. Until the party can sideline its special interest groups to get things done, it's not as simple as "Democrats good, GOP bad."
[1] https://www.washingtonpolicy.org/publications/detail/the-42-...
The arguments at that link are questionable. E.g.
> When states use the federal subsidy to provide access in areas without internet, they no longer have incentives to confirm if the technology being deployed is appropriate for the area or if the customer base needs and wants what is being provided.
It's actually the opposite. When states use federal subsidies they have the freedom to consider the technology that is actually best for the area instead of just what is cheap enough for the state to afford without better help.
When they are limited to whatever is cheapest now they may have to pick something that meets the present needs of rural areas but will be inadequate in a few years.
This is also heavily described in _Abundance_. I'm not an expert, but it is beyond doubt that this program did not produce any kind of meaningful results, despite having a moon-landing-sized budget.
Even when it was created, Starlink was obviously the best option for quickly linking rural America to the Internet. That's obvious because that's EXACTLY HOW rural America has actually been linking itself to the Internet, despite it not being covered by the program. In ten years, that's probably still how they'll get online.
I am a Democrat, have been since I could vote, and refuse to vote red for as long as MAGA owns the GOP. It's still exhausting to watch my party flounder and give in to its worst instincts.
That was an illegal move done by Trump. Biden administration proposed regulations for the conditions of the funding and allow states to apply for it. The regulations went through the necessary process to become final and the states went through the work to fulfill the requirements and the fundings were approved to go out of the door based on the trenches.
Trump administration illegally impounded the funding. This has been disallowed for decades but sadly the republican political appointees at the supreme court is allowing him.
It doesn’t even matter, does it. It seems the democrats made a plan spanning multiple elections due to it being loaded with regulations and now Starlink can better serve rural areas at a fraction of the cost (for american taxpayers). Also the planned deployment was 2026 at the earliest?
I think you have your policy preferences and you have the idea that if starlink was allowed, it would magically take much less time to go through regulation process. I think this is not true, but I think we digress.
My point is that democrats don’t get credit and votes for improving internet for rural areas and republicans don’t get punished for destroying progress. The fact that you are criticizing democrats so harsh and think trump breaking the laws doesn’t matter just prove my points.
Plans to actually do things take time.
They don’t have to take that long. Seriously, when you build years and years of government reviews and challenges and approvals into the process of awarding funds you have only yourself to blame when the government works for four years and accomplishes nothing. Meanwhile a Starlink, a private company, just starts doing the work. They’ve spent 10 billion dollars of their own money and built a service that is hands down the best internet service available in any rural area of America. They expect to have 11 billion in revenue this year, just 6 years after the service first became available to paying customers.
The billions spent on rural broadband excluded Starlink as not technically feasible.
Many other billions have the same issues - I think no one knows how to actually hoover this the way the big co's do?
We've had much faster broadband happening because of commercial competition from scrappy startups and WISPS and fiber folks (think sonic)
I think something like 94% of RDOF/BEAD locations in california were defaulted (ie, awarded but customer actually never got service)?
It's crazy given the 100+ billion or so spent on USF / RDOF / BEAD / etc that they couldn't do $5b - $10b for something like starlink which at least in rural areas is able to serve folks pretty quickly and push hard on that for a bit. The unsubsidized commercial starklink services is already outcompeting the insanely subsidized buildouts (that cost insane amounts per person). Starlink was awarded the funds but then they were revoked.
Starlink didn't get RDOF funding because RDOF required a minimum of 100/20 Mbps, and Starlink failed to meet that in too many rural locations.
In Japan it was competition that fixed this. The government did something (not sure what) but it was not "funding municipal broadband".
I'd say one thing to do is outlaw the contracts that let governments only approve a single provider.
It was Softbank that brought the competition. In 2001 they started offering 1meg connections for $20 a month. The competition cost 10x. Softbank had aggressive marketing too, putting up booths at neighborhood train stations where they could sign you up and hand you a device on your way home.
The competition lowered their prices to match within a couple of months. Softbank doubled the speed to 2meg, same price. The same pattern. The competition matched price within a few weeks. Softbank raised their speed to 8meg. repeat.
AFAICT it all finally settled (25 years later). Currently $40 for 1gig fiber or $60 for 10gig fiber or $50 for https://www.softbank.jp/internet/sbhikari/
And NTT is competitive https://flets.com/
It’s probably important to mention that Japan is half the size of Texas, or roughly 5% of the area of the lower 48 states.
Japan has a population density that is 8 times higher than the lower 48 states.
> It’s probably important to mention that Japan is half the size of Texas, or roughly 5% of the area of the lower 48 states.
The US population is fairly concentrated. Various metrics to make the point: 40% of the US population lives in coastal counties:
* https://oceanservice.noaa.gov/facts/population.html
About two-thirds of the population lives with-in 100 miles of the border:
* https://www.aclu.org/know-your-rights/border-zone
The top 143 counties (out of 3244) contain half the population:
* https://en.wikipedia.org/wiki/County_(United_States)
Hitting the most populous 10% of counties will get you a good way towards universal coverage (LA county alone as >9M people, which would rank #12 as a state population).
Irrelevant. Competition fixes everything, and local monopolies prevent that competition.
Competition is great except when it isn't. Rural customers get screwed because they're not worth the effort. You'll spend more money running fiber to them than you'll ever get back, and that's before you take competition into consideration.
Back when we had the AT&T monopoly the government forced them to serve rural customers. It was subsided by a small fee on every phone bill. That carried over to the baby Bells because they were local monopolies. The problem is, no one has land lines anymore and no companies are willing to take on the last mile problem without a hefty incentive. Being the broadband monopoly is part of that incentive.
I argued for years that the AT&T breakup was done in the stupidest way possible. Put the natural monopoly - the last mile - on its own and regulate the hell out of it like we did AT&T. Then allow competition to run the service on top of those connections. We started in that direction but lost our way somewhere.
Yea, the breakup of AT&T was stupid. Even worse, it just allowed all of the baby bells to be bought up one by one by a really badly run Canadian company. Now rather a lot of that money started leaving the country and being gobbled up in a foreign bankruptcy and pension scandal and stock market crash. Yay. (Ok, not all of that was entirely foreseeable.)
But to get back to the original topic, I was referring specifically to the local monopolies that so many cities have granted to cable companies which exclude not just other cable companies but certain types of internet services as well. I was specifically not talking about rural areas, where that’s not a problem. Not many cities out there in the countryside.
If you want to see fiber installed everywhere, make the FCC annul all of the local monopolies and speed up all of the local permitting processes. Just have the FCC rule that no local permitting process for installing internet service in existing conduits or on existing poles may take longer than a fixed period of time (I dunno, a month or two should suffice), otherwise the permits are automatically granted. If the permit applications don’t meet your local rules then you need to identify the problem and reject the application quickly, not drag things out for years. Something along those lines.
Subsidies to speed up rural deployment of real broadband are a separate topic.
It is not irrelevant, who is going to pay $30,000/mile to run fiber out to rural homes when the subscription fees for a lot of rural properties will never pay for the infrastructure? The US is massive and full of sparsely populated rural areas, all of which will ignored by capitalists. Starlink is then their only option.
I’m all for getting rid of local monopolies that aren’t municipally owned.
Again, this is irrelevant to the conversation we were having about price competition. Competition solves the problem of high prices automatically.
> The US is massive and full of sparsely populated rural areas, all of which will ignored by capitalists. Starlink is then their only option.
Ironically, Elon Musk is a capitalist who is making bank by serving the sparsely populated rural areas. He is certainly not ignoring them. You can get Starlink service in the middle of the Pacific Ocean, and if there is any more sparsely populated place on Earth then I don’t know about it.
At least have a coherent argument before you complain that competition can’t solve a problem.
USA believes it's better than everyone else, it's very inward-looking. It's been quite striking.
American exceptionalism is odd, in that sometimes it is "we are better than everyone else", but sometimes it is "we can't have the nice things that all other rich countries have, because [nonsense reason]".
A popular one is 'density', and, okay, maybe this is somewhat true if you're talking about, say, Wyoming (though I think not as true as people often think), but California, for instance, has an only slightly lower population density than _France_, and at that point "we can't have proper transport/telecoms/whatever because density" is just an excuse, and not a convincing one.
The reason we can't have whatever nice things you want is because California doesn't want to spend the money on that nice thing, and it has to maintain a budget, unlike nations. Including a $40 Billion project on a budget makes many, many other things go away. Even if it's just temporarily to help pay for the construction of the service, the point stands.
So imagine if France couldn't go into debt - only the EU can. France wants a giga-train suddenly, so they ask for it. The current leader of the EU isn't a fan of more trains, so he turns it down. France goes back to its people and says "we can build the giga-train if we do XYZ", and people vote based on whether or not they want XYZ or the giga-train more.
I think it's possible you just might want things different from what others want. That study a while back which showed most Americans want cheap public transit so that everyone else gets off the road and gives them more space lives rent free in my head. Nobody wants these stupid trains.
I mean, yeah, the US having systemic problems which make capital investment difficult, or at least extremely slow, is _part_ of it, but there's also just a tendency to a complacent "we have these problems because we are _special_ and they can't be fixed" attitude amongst many Americans (I say this as an outside observer).
There are vast parts of California that are pretty much empty. However, the lack of competition in broadband at the consumer level does suck.
I mean, same goes for France, tho.
What goes for France?
I have a 5gbit down, 900mbit up connection with 200 tv channels a d a landline for 30€ over fiber. And I'm not even in a big city
In the US I had spectrum at 80$ a month for just internet at like 20mbit down or something...
california is ~160,000mi^2 and france is ~244,000mi^2 (423 megameters squared, 632 megameters squared, metric), the population density, respectively, is 251/mi^2 and 281/mi^2. You're comparing an entire country to a single state (2% of the US states, in fact.)
I like to point stuff like this out whenever someone compares an entire EU country to some US state.
note: i edited the france density, as i did accidentally transcribe the wrong value. France is slightly denser per sqkm.
That particular US state has a GDP larger than that entire EU country.
well, 400 billion less (USD, i think), but like 2/3rds the population. California $4.1trillion 39mm pop; france $4.5trillion 66mm/68mm pop.
Are you sure about your numbers?
It seems like you're mixing density units.
France has 66M people for 547k km^2, which is 122/km^2 , California has 39M for 403k km^2, which is 97/km^2.
Yep. As an american I keep giving WTFs every time media discuss potential solutions to problems without even considering how other places solved them. It's like other countries don't exist, or hide their secrets.
Not just broadband, but same with healthcare, homelessness, gerrymandering etc. Just copy-paste little by little.
As an HN viewer I keep giving WTFs every time HNers discuss potential solutions to problems without even considering how other places solved them.
It's not just an american media problem. Plenty of people here calling for municipal broadband ignoring what worked in other countries.
I was thinking the same thing until I learned about the high speed train plans California had (and still have?) and just how much of a disaster it is. It’s a red tape issue and nobody at the policy level seems to care to fix it.
It really is. I keep wondering when some tiny bit of humility will show up, but it is increasingly like asking Russians to have some humility. Not likely, and sometimes, not possible.
There is too much religion in the states for the people to be humble.
Sorry can't hear you over my air conditioner.
Rural broadband mandates were just welfare for the big carriers. There was never intended to be any results or accountability. Consumers have been paying into the fund since 1997 and for that we got richer carriers and a pocketful of bupkiss.
Municipal broadband is such a great idea - but it's as hard, if not harder, to convince a city council to pursue this over adding a simple bicycle lane to a small street downtown.
Cities know how to build roads. They don't know shit about operating high speed broadband. My taxes keep going up because the city keeps taking on more and more responsibilities that (a) they don't know how to provide and (b) isn't really their responsibility to provide.
Many states forbid municipalities from offering broadband without approval from the legislature.
If I've learned anything from the last decade of politics, the proper solution to that nonsense is to give them the middle finger and do it anyway.
If you're planning on moving, this list might influence where you want to go:
https://broadbandnow.com/municipal-providers
AFAIK most of the RDOF money was awarded to smaller ISPs and there are supposed to be penalties for not delivering.
Can’t penalize a company that disappears.
Other countries have private enterprises rolling out broadband so I think the difference here is good old fashioned corruption.
The rural telco situation is pretty good. In fact, better in many cases than what you get in larger cities.
The issue is there's some perverse incentives at play with the grant monies. For brand new neighborhoods you can expect state of the art internet being available. However, after it's installed, you can expect it to be simply locked in place with no upgrades.
A major problem comes down to easements. If there's a utility pole nearby then some federal regulations make it super easy for a utility to just add their line onto the existing pole. There are access guarantees.
However, if no pole exists then you are looking at buried lines. That means each time you want to cross someone's driveway or property you need to reach out to the existing owner to negotiate access. Some people are easy to work with, others will just flat out deny access. (The BLM, for example, is notoriously hard to work with for placing underground utilities. As are a few big churches).
The US, frankly, has the wrong model. Leaving everything up to private companies is what creates all these problems. The companies have absolutely no way to guarantee a route for service and they have little skin in the game to run an update.
At a bare minimum, the better model is to make utilities public (at very least the lines themselves.) That gives the government a powerful ability, the ability to claim eminent domain to cram through any improvement they need. You could still have private ISPs, but you can relegate their roles to just running the equipment (switches, routers, etc) rather than doing all the additional line management work.
The lobbyists shutting down any attempts at municipal broadband development for so many years is such a catastrophe.
1 Timothy 6:10: "For the love of money is the root of all evil []"
I am currently on a farm in a very rural area, in a central EU country.
900mbps symmetric, $25/month fiber. The fiber run was subsidized/possibly entirely funded by EU money.
> EU support to rural revitalisation through broadband roll-out and smart solutions
https://digital-strategy.ec.europa.eu/en/news/eu-support-rur...
The US doesn't generally do new things anymore. We just shuffle money around into peoples' pockets. So we have plenty of societal wealth to do something like this, even given how large and rural many areas are, but we just don't. Because no one gets reelected because they rolled out bandwidth. They get reelected based on however their party is faring in the battle of the spectacle going on nationally.
Depends on the place. Live in rural Georgia (Carroll County). Within the last two years have had Spectrum fiber and Carroll EMC fiber run by the house. First is underground and second is on pole.
Have you ever wondered why each provider needs their own wire?
You could have the wire be owned by a utility and let companies compete with services?
I live in an insanely rural no-stop-light town, and they just ran 5gbps fiber outside my door. I already have 1gbps from another company.
Can't justify the 5gbps, though - it's like $170/mo + taxes.
To be fair, there will be plenty of people using EU funded fiber over here, to complain about how the EU does nothing and is pure evil, and that the country should leave the union.
I don’t think this is even remotely true.
Feel free to take a ride in the high speed cross country rail to argue about it with me to my face.
Oh...
Same deal in Prince Edward Island. Very rural, near the coast, cows across the road, but gigabit fibre from Xplorenet who took advantage of government funding.
I'm in the UK on a 900Mbps symmetric for £40 a month.
Turns out regulation can be a good thing - first of all, OpenReach (the infrastructure part of BT that was split up post privatisation) are regulated in how much they can charge other providers for last mile FTTP/C[1]
But then there's also Access to Infrastructure regulation[2], which means duct and pole access isn't up to the telco's (and after all, existing infra was publicly funded, and new infra is often paid for by housing developers). This means I have 2 providers of FTTP to my property - CityFibre and OpenReach (and a third offering DOCSIS cable in Virgin Media).
These combined have led to a thriving market with genuine competition. Whilst most ISP's use the two providers with the most footprint (OpenReach and CityFibre) they compete on things like technical chops (AAISP), Price (the PlusNets of the world) and a mix of the two (Zen).
Regulation, when well designed, leads to more competitive markets.
[1] https://www.ofcom.org.uk/phones-and-broadband/telecoms-infra... [2] https://www.legislation.gov.uk/uksi/2016/700/contents
Japan has no regulation and you can get 10 Gbps for the same price or lower. The main reason is it skipped ADSL completely and went staight to fiber.
Well, there's also cultural reasons. If the government of Japan entrusts you to self regulate on the basis of open access and no restrictions a Japanese business will understand the weight of expectations that have been placed on them.
A US telco would pop the champagne and then plan how best to screw their customers, safe in the knowledge that they have few competitors after they got local governments to ban municipal broadband.
I live in the second largest city in the US. In the distance I can see a switching station that carries a large amount of telecom and internet traffic.
Should be easy to get fiber no? Turns out they want a couple thousand dollars from me to run fiber up my street. Coax it is!
coax was $5,000 a mile, here, 5-6 mile run (i forget), and they said "there's no guarantee it does anything except TV" - that is, no internet.
fiber loop from cogent or centurylink was $15,000. even though they have a loop a half mile up the road. Dark fiber is probably cheaper, but i cannot remember the pricing to run my own fiber via some legit company. i had DSL (6mbit), switched to my own CPE with cellular wireless, then their CPE after they stopped selling the SIM plan i was using, then starlink after they stopped selling their fixed wireless for some newer one that isn't supported anywhere but metro areas.
There's a tree interfering with satellites at my location, and it's still 20x faster and better than anything before; except my own CPE solution, starlink is only 2x as good as that.
In the US fiber has started to show up in very rural areas now as well. My parents got it last year and I never thought that would happen in my hometown.
It's some small company I never heard of before, back when I lived there Verizon got a ton of government money to do this and just never did, everyone was pissed about that.
I'm still pissed about Verizon pocketing billions in subsidies and not finishing their work. My area was next up for fiber installation before Verizon stopped, and so I'm stuck with Comcast or DSL. Comcast claims to offer symmetrical GiB up/down, but I've yet to get that. Now I can't even get in touch with a human for support without spending 30 minutes with a bot.
Weird opposite experience. I’m suuuper rural NY (what others have described as ‘American Siberia’) and for some reason Verizon quietly started building out their fiber network in the past several years again, something even I didn’t anticipate happening.
I always knew they had Syracuse and Rochester as major FIOS markets, but I was blown away to see the trucks all the way this far north and on my very road. Last I’d known they’d said they had stopped all buildout. Next thing I know, they pick back up their pace, then pick up the homework they let Frontier finish for them with the Frontier FIOS buildout, too.
Our nearest town, Durham Kansas (population 100) got buried fiber optic recently. The company is actually planning/hoping to expand to the rural area around town as well, so maybe we'll get it as we're only two miles out of town. They do offer 1gig symmetric but I think it's $200/month. I'd just like the reliability and low latency of fiber vs our current WISP.
It's a huge difference isn't it? When I used to stay here in the past, the DSL was awful. It made it very annoying to work online. Now, it's as good as back in Seattle.
Oh lol there wasn't even DSL, everyone wanted that. The options were dialup (over ISDN for a while which was one of the best options until that went away), geosynchronous satellite, or LTE if you're even near a tower.
That's why I'm so surprised they have fiber all of a sudden.
Starlink is now kind of a global baseline that ISPs have to compete against regardless of location. In rural areas it's very reasonable to expect 200-400mbps downloads and 20-40mbps uploads for $120/month. A bit pricey but it's a level of service that DSL and GEO sats can't even think about matching, so companies have to build cable, fiber, or 5G towers if they want to have a comparable offering.
In Vermont you can get GB fiber to the house but it will cost you $90/month after the intro period. Vendor: Fidium.
My parents for fiber in a rural area too. It's roughly $100/mo for something like 20/20.
Which country? In Germany and Austria you still have DSL cable because the telco monopolies are golf buddies with the politicians.
I was taking a train from somewhere in Bavaria to Austria. More than half of the ride the train lady was chilling next to us because "there is no internet, I cannot check tickets" the whole ride you could see houses out of the window. People are actually living there. This was like 6 years ago but I am still baffled by this.
In Switzerland it's very unlikely you find a mountain or road without 4g
"Germany's 5G coverage was 99% in 2024, slightly exceeding the EU average."
https://www.heise.de/en/news/EU-digitization-report-Fiber-op...
"Today passengers enjoy at least 200 Mbit/s on 99 per cent of the 7 800 kilometres of main lines and even 300 Mbit/s or more on 95 per cent. Secondary lines also saw a transformation. Coverage of 100 Mbit/s rose from under 83 per cent to over 96 per cent in just three years."
However:
"Yet coverage is only half the story. Mobile signals must penetrate each carriage’s interior if passengers are to make calls or stream without interruption. Many modern trains are fitted with factory-installed windows engineered for signal permeability."
And of course many train's aren't fitted with windows like that (and operators are trying to retrofit them with microcells or in other ways).
https://www.connectivity.technology/2025/05/seamless-5g-conn...
I hear complaints about German internet connections all the time, and it blows my mind that this could still be the case. What an utter self-own.
I am currently in Poland, the very SW corner. Close enough to get both German and Czech radio in the car.
>What an utter self-own.
Which self own? Besides the one making your industry dependent on energy from your military opponent? Besides the one putting all your eggs in selling diesel engines when China and the US were betting on computer driven battery powered EVs? Besides the one where you open your borders to unvetted illegal immigrants leading to a rise in crime, terror attacks and right wing extremism all over Europe?
Because I lost track.
> Which self own? Besides the one making your industry dependent on energy from your military opponent?
Yeah, that too I suppose. How could Schroeder have known? :/
I could have fiber, but I see no point in upgrading from DSL. Why? It works. In fact, I could have upgraded to 250 Mbit DSL years and years ago, but 100 Mbit is fine. Why bother spending money or time making changes to my home's critical infrastructure. I doubt I'm the only one.
And then there's the millions of boomers around whose only device is a smartphone and they never exceed the 10 or 20 or 30 GB they get on a mobile contract that's less than any DSL or fiber contract. Good luck selling them 900 Mbit symmetric links.
In fact I'm sure there must be hundreds of thousands who still pay for DSL that's essentially unused because their phone lost the Wifi credentials and there's no grandkid around to notice it. Their house will be upgraded to fiber when it gets resold because it ticks a box for the real estate agent.
> I could have fiber, but I see no point in upgrading from DSL. Why? It works. In fact, I could have upgraded to 250 Mbit DSL years and years ago, but 100 Mbit is fine. Why bother spending money or time making changes to my home's critical infrastructure. I doubt I'm the only one.
We get 250 Mbit into our house and our house has Ethernet cables going through the walls to different floors but the cables are all limited 100 Mbit and it is tilting me of the face of the planet... :(
Replacing them isn't much of an option, just maybe running other cables around the walls, which isn't the nicest option when the existing cables are all nicely not visible.
>I could have fiber, but I see no point in upgrading from DSL. Why? It works.
How do you explain to a caveman that a living in a house is nicer than living in a cave?
>Why bother spending money or time making changes to my home's critical infrastructure. I doubt I'm the only one.
You don't have to pay, In Eastern Europe telcos pay themselves to wire fiber to your door in hopes you'll sign up with them. It's something you don't have in Germany, it's called free market competition. The German mind just can't comprehend having competitive companies that serve the consumers and not sclerotic state supported monopolies that only support the shareholders by robbing the consumers. Why are you guys like this?
>And then there's the millions of boomers
Of course, the way to move forward technologically is to cater tech to boomers habits. That's why Germany is a SW innovation powerhouse.
> You don't have to pay, In Eastern Europe telcos pay themselves to wire fiber to your door in hopes you'll sign up with them.
It's wired, I just have to sign up. But 1000 Mbit is more expensive than the 100 Mbit I have now. I just don't see the point. If that makes me caveman, so be it. Work on your manners.
No, they didn’t. They decided not to raise the definition of broadband from 100×20 Mbps up to 1Gps.
And that’s simply because 100Mbps is actually a lot of bandwidth. We just don’t have any killer applications that need more than 20Mbps. 4K streaming on Netflix, to pick one salient example, only requires 16Mbps. At 100Mbps you can download a 50GB game in just 80 minutes. Sure, it would be nice to have 1Gbps and download the game in 8 minutes, but the definition is about _minimum requirements_, not things that are _nice to have_.
As such the definition we have is actually a good one! If every American had broadband according to this definition we will have actually made real progress. Nobody would be stuck on DSL any more, let alone dialup.
The FCC is still giving grants to anyone who is installing actual broadband in unserved areas, which means anyone installing cable or fiber in areas that don’t have them. It’s just not requiring that every customer have 1Gbps service in order to qualify.
> No, they didn’t. They decided not to raise the definition of broadband from 100×20 Mbps up to 1Gps.
No, the 100/20Mbps definition was only adopted last year and is being kept for now (though you'll never guess who voted against adopting it and wanted to keep 25/3Mbps!).
The 1000/500Mbps definition that Carr's FCC is trying to eliminate was a long term goal, something that the US should strive for eventually, and therefore federal funding should preferentially go to solutions that can or could eventually provide those speeds.
Carr's proposed rules[1] go well beyond just the definition of high speed broadband, though:
- He wants a stricter reading of the statute that requires a report on the deployment of broadband, reading "whether advanced telecommunications capability is being deployed to all Americans in a reasonable and timely fashion" as strictly referring to anyone making "incremental progress" on deployment, so the reports will shift from reporting coverage where broadband has actually been deployed to calling areas covered if anywhere in an area is working on deploying broadband there
- He also doesn't think "reasonable" access to broadband should include a consideration of the price of using that broadband, so wants to stop collecting pricing information. So long as someone is making incremental progress in an area that would make 100/20Mbps internet available to you at some price, you will now count as covered
- Other proposed changes around mobile speeds and school broadband I haven't been following.
Incidentally, even the 100/20Mbps definition seems at least potentially in danger, as the proposal is requesting comment on whether that's actually the definition they should be using in the future.
[1] https://docs.fcc.gov/public/attachments/DOC-413059A1.pdf
> The 1000/500Mbps definition that Carr's FCC is trying to eliminate was a long term goal, something that the US should strive for eventually, and therefore federal funding should preferentially go to solutions that can or could eventually provide those speeds.
Yes, that’s true. But don’t ignore their justification: the laws as passed by Congress say nothing of long term goals other than bringing everyone up to some minimum threshold. The FCC gets to define that threshold, but making up new goals on top of that is not actually supported by the law. Plus, with so much of the country installing fiber it’s already just happening automatically.
> He wants a stricter reading of the statute that requires a report on the deployment of broadband, reading "whether advanced telecommunications capability is being deployed to all Americans in a reasonable and timely fashion" as strictly referring to anyone making "incremental progress" on deployment, so the reports will shift from reporting coverage where broadband has actually been deployed to calling areas covered if anywhere in an area is working on deploying broadband there
You almost have it right. They want to actually measure improvements in access rather than just the number of people with access. Sure, if the number of people with access goes up then the situation is getting better. But what if all of the improvement is because people are moving to denser areas to get better internet access? It makes complete sense to measure the actual geography of access in addition to just the number of people who can access broadband.
The report states clearly that they intend to base their analysis on the FCC’s Broadband Data Collection <https://www.fcc.gov/BroadbandData>, aka the National Broadband Map. They just want the report to measure improvements to the number of addresses served not just the number of people served.
They also point out that while the National Broadband Map has addresses of every household, it doesn’t actually have any information about the population living at those addresses. It still uses Census Districts to estimate that, assuming that every address in a district houses the same number of people. They’re going to keep using that same estimation method, probably for the foreseeable future.
> Incidentally, even the 100/20Mbps definition seems at least potentially in danger, as the proposal is requesting comment on whether that's actually the definition they should be using in the future.
As I’ve demonstrated, there is significant headroom in the current definition. Would someone with 40×10Mbps DSL (provided using two phone lines) be significantly disadvantaged over someone with 100×20Mbps cable? They’ll still be able to stream 4K television and send HD video. Few ISPs offer this service, but I used to get it from Sonic.net more than a decade ago. I had to jump through some ridiculous hoops to get two actual real functioning telephone lines wired into the same apartment, but it was great. Even 25×3Mbps enables you to stream 4K television and participate in video conferences. The constraints on the quality of that video conference are real, but are they enough to handicap people? Yea, I think so, but it’s clearly a judgement call.
Furthermore there’s no indication that they actually want to change that definition immediately. Agencies are more or less required to invite comment from the public on all of their regulatory decisions:
> IV. A. 18. Broadband Speeds on Which to Report. The discussion in the 2024 Report focused on the availability of fixed broadband at speeds of 100/20 Mbps.⁴³ We propose to again focus our service availability discussion on fixed broadband at speeds of 100/20 Mbps and seek comment on this proposal.
the USDA, which gives broadband grants, considers an area "served by broadband" if there is a 2 megabit provider in the jerrymandered area. This includes cellular, even if using a fixed or hotspot link would be cost prohibitive.
Look at the coverage maps for rural areas with trees, and you'll see an interesting pattern. They used multi-pathing to their jerrymandering benefit. checkerboard pattern means the USDA says "no grant for you!"
i have no idea when or if the definition changed to 20mbit. there are people here still served by 6mbit - on a good day.
That used to be the case, but it isn’t any more. Check out the new broadband map: <https://broadbandmap.fcc.gov/home>; it goes by exact addresses rather than census districts or whatever it used to be.
The definition was changed to 100×20 Mbps last year: https://docs.fcc.gov/public/attachments/DOC-401205A1.pdf. 10 years before that it was set to 25×3Mbps.
Please re-read what i said. i said USDA, not FCC. As i tried to start a WISP and could not get any federal funding for it because of the gerrymandering. this was within the last 10 years.
It turns out the USDA changed it. February. of 2024. not 2015 (or whatever). So my information was good until about 18 months ago.
it is now 25/3 (what you are writing as 25*3). https://www.usda.gov/sustainability/infrastructure/broadband... under "mapping", the first question.
I know it often seems like i have no idea what i am talking about, but i very specifically said USDA.
note: they also changed it so cellular no longer counts, amazing:
> Sufficient access to broadband, as defined in the NOFO published on February 21, 2024, is 25 megabits per second (Mbps) downstream and 3 Mbps upstream (25/3). Mobile/cellular and satellite services will not be considered in making the determination of sufficient access to broadband. Additionally, areas with current broadband service from unlicensed wireless facilities, or have an enforceable commitment associated with unlicensed wireless facilities, are eligible for funding under this notice.
here's the old section (from 2018, you're not paying me to prove exactly the numbers i said, which are older by a year or two):
please note the starred parts, as well as 1738.58 (c) which explicitly says that it's gunna be gerrymandered.Ok, apparently the USDA also provides their own separate grants for broadband deployment. How are their weird rules relevant to this discussion of the FCC’s grants program?
> No, they didn’t. They decided not to raise the definition of broadband from 100×20 Mbps up to 1Gps.
it should be at least 100x100
I agree that it would be really nice if it were symmetric. But remember, the definition is for a service that isn’t handicapping people, not for what would be nice to have. Maybe in 10 years we can make it symmetric.
5 years ago it should've been symmetric gigabit already but yet here we are
No, I completely disagree. Remember that the whole point is to define the _minimum_ level of service at which people can reliably access “advanced telecommunications capability”. Here’s the definition:
> The term ‘advanced telecommunications capability’ is defined, without regard to any transmission media or technology, as high-speed, switched, broadband telecommunications capability that enables users to originate and receive high-quality voice, data, graphics, and video telecommunications using any technology.
By this definition, 4K video is a reasonable metric to judge whether a broadband service qualifies as an “advanced telecommunications capability” or not. Since the major streaming services all use less than 20Mbps to serve up 4K television, 1000Mbps is clearly not necessary. It certainly is nice to have, but it’s not _necessary_.
I think that fiber access will continue to grow as long as we remove the regulatory barriers that hamper it. Permitting is a big one. Cities often require separate permits for every street, often for every block!
If you have to install anything, then installing fiber is the obvious choice. It has immense headroom for the future; fiber optic cables that carry a gigabit today could easily carry 100,000× that just by buying fancier equipment to install at both ends. The upper limit is currently hundreds of terabits per second, and it goes up every year. We’re nowhere near the theoretical limitations of optical communications technologies. I think that ultimately this means that fiber will win out over coax eventually. But coax has headroom still, and there’s no reason to tear it all out where it’s already in use.
We should increase competition by abolishing local monopolies, but that’s something we should do no matter what level of service we adopt as the definition of broadband.
Most streaming services have extremely bit-starved encodings for their 4k streams and fall far short of the quality that a proper 4k video has.
They are also doing their encode much slower than real-time so they can achieve higher compression rates.
The definition also says "originate and receive" so you need to look at both upload and download bandwidth.
4k also isn't high enough quality for many use cases like VR.
Households also contain multiple devices, so just a single video stream isn't sufficient.
I would argue that by that definition they should be setting the forward-looking threshold as sufficient to both upload and download multiple live 8k video streams simultaneously.
> The definition also says "originate and receive" so you need to look at both upload and download bandwidth.
Yes, I agree that a symmetric definition would be better. The definition we have is pretty good though. The 20Mbps of upload is enough for 4K video streaming. It would be nice to have some more head room so that more than one person in a household could stream at full resolution at the same time, but the market has already solved that problem. Every market where 100×20Mbps broadband is available also makes higher upload speeds available.
> 4k also isn't high enough quality for many use cases like VR.
This is not a good argument. VR relies on local rendering of 3D assets, not streaming video.
> Households also contain multiple devices, so just a single video stream isn't sufficient.
100Mbps has headroom for, let me calculate here, 100/16=6 whole 4K video streams. If your whole family needs more than six simultaneous 4K video streams to live then your family has problems.
> …sufficient to both upload and download multiple live 8k video streams simultaneously.
This is just dumb. 99% of computers can’t even play 8K video; they need specialized hardware just to decode 4K video in real time and the hardware cannot handle 8K video. The cameras for it don’t exist, and neither do the displays. And finally, the FCC definition of broadband is about raising the bandwidth floor, not raising the ceiling. The ceiling is already high enough by far; I can get 50Gbps service at my house if I wanted to pay for it.
But also 8K video would still use less than 100Mbps of bandwidth!
20Mbps upload is fine for pretty much everything, other than uploading large files.
Sure, in a single person household with only a smart tv and laptop. Which is not the average person.
I disagree. I would argue that most people just don’t produce enough data to have a huge need for faster uploads. You would have to produce more than 200 gigabytes of data per day before your backups would take longer than a day to finish:
If you’re an entertainer then you might record more than 200GB of raw video per day, but most people aren’t. You wouldn’t even need a faster connection to _become_ an entertainer, even if you soon wanted one. You could stream 4K video of your antics all day and it would be less than 200GB.And remember that the definition is about the _minimum_ bandwidth necessary to participate in society, not what you need to be at the peak of your entertainment career. People who do need higher speeds can and will pay extra for them; the FCC definition is not about limiting what products are available. It doesn’t even require anyone to have 100×20Mbps service. The FCC is just trying to get us to a point where we can say that 100% of Americans have _access_ to that level of service, even if they have _subscribed_ to a lower level of service to save money. Since 45 million Americans don’t even have access to 100×20Mbps service we’re still pretty far away from that.
> Sure, in a single person household with only a smart tv and laptop. Which is not the average person.
What is the average person/household? How many devices? What use cases (and therefore bandwidth) meet their needs?
Average household is a family, who arent streaming 4k video or uploading huge backups like i do, and 20Mbps would be fine for me. They typically watch netflix and youtube and doomscroll.
640k should be enough for anyone after all
Not for video conferencing.
> At 100Mbps you can download a 50GB game in just 80 minutes.
100 Mbps is 100x as slow as my internet connection. I would really rather download a giant game in less than a minute, or even a few minutes, than more than an hour.
Of course you want it to be faster rather than slower. But waiting an hour to play your new game is not a handicap. Not compared to dialup, where that same download would take more than 80 days, if it finishes successfully at all:
>4K streaming on Netflix, to pick one salient example, only requires 16Mbps.
Because they compress all their streams to hell to save a few cents and to fit through people's shitty internet connections.
Actually it’s mostly because we’ve spent the last few decades turning math into compression. Starting with MP3s in 1991 there’s a whole history of amazing improvements to audio and video codecs. We’ve made extremely rapid progress even though the process has frequently been weighed down by ridiculous software patents.
But that said, there’s nothing stopping video streaming services from offering higher bitrates if they think it’s worth it. After all, the FCC standard for broadband is 100Mbps, not 16Mbps. That’s a lot of headroom if you want better quality.
It really is mind blowing to see how much smaller a file with newer compression can be. Had to convert some videos to a codec that would work on my kids chromebook recently. What was a 300 meg video ballooned up to 2 gigs. One of the videos was 130 megs that jumped up to 3.2 gigs. Time lapse video, so I'm sure it had some of the sweet spots for compression. Still, was dramatic how much a modern codec can get that down.
> We just don’t have any killer applications that need more than 20Mbps.
Sure we do, but large ISPs demand their cut.
Backups. File transfers. Large games. Live 4K video chat. Language models. CAD models. Cloud-based spyware, like smarthome/car/phone/whatever telemetry, or security cameras.
> 4K streaming on Netflix, to pick one salient example, only requires 16Mbps.
So only the living room gets 4K, but family members can't all watch their own 4K streams in their own rooms? Bobby talks during the show, Sally giggles at everything, and Timmy wants everything to be dark while mom & dad want the whole room bright and quiet. Thanks, ISP, for forcing the family together!
Remember, we are talking about minimums here, the least common denominator. At 100×20Mbps you are not handicapped like you are if you are stuck on dialup.
Live 4K video chat needs less than 20×20Mbps, but remember also that most people don't have 4K televisions or 4K cameras. Even gamers mostly don’t have 4K monitors! The most recent Steam hardware survey <https://store.steampowered.com/hwsurvey/Steam-Hardware-Softw...> shows that just 5% of gamers have a 4K or better display. 55% have 1920×1080. I don’t have similar statistics for televisions or webcams.
The other uses you list require even less bandwidth. Nothing about a language model requires bandwidth. You either interact with it remotely, at a few kilobits per second, or you run it locally using no bandwidth at all. Large games might take an hour to download instead of minutes. Waiting for an hour is not going to handicap you. Most people shouldn’t even pay extra for that. Backups need some bandwidth, but only in proportion to the amount of data you have generated. Most people don’t create hundreds of gigabytes of new files that need to be backed up remotely; the largest files most people create are photographs and videos of family events, vacations, etc. These files can be backed up with no difficulty at 20Mbps.
> So only the living room gets 4K, but family members can't all watch their own 4K streams in their own rooms?
Pay attention and don’t be an idiot! The FCC definition of broadband is 100Mbps down and 20Mbps up. Netflix tops out at 16Mbps down, so those 100Mbps can supply 100/16=6 whole 4K television streams easily. If your family of six people is sitting in six different rooms watching six different television shows then your family has a problem. That problem is not a lack of bandwidth.
Another side of the coin is no one will ever serve you content at 1gbps. It just won’t happen. I get 500mb down on speedtests and netflix et al regularly throttle me to like 144p quality with constant buffering. I download a game on steam and it throttles to nothing for hours after making some initial good progress.
It’s quite rare for the average website to actually hit 1Gbps, that’s certainly true. Nobody makes webpages big enough to hit that speed, or even for that top speed to matter.
But I can download games from Steam at approximately full speed, for example. I subscribe to Ziply Fiber, and they certainly don’t throttle their users or oversubscribe their bandwidth. That said, there are other factors at play as well: can you even maintain 1Gbps to your _hard drive_? Can your computer decompress the downloaded data as fast as it comes in? Steam will slow down the download to match the speed at which it can decompress the data and write it to your disk.
Fiber just came through my area. They offer up to 3Gbps for less than I was paying for Comcast ~500Mbps asymmetric and for more money I can get 5Gbps... but I just signed up for the 500Mbps symmetric and pocket the difference monthly, because what the hell am I going to do with even 1Gbps? My Wifi can't 5Gbps, and all but two network devices in my house use Wifi to get to the internet. My NVMes can nominally do it, but it takes everything firing on all cylinders to actually achieve that. I've still got some spinning rust that is pretty full up at even the 500Mbps. I do run backups to AWS, but that runs in the nighttime anyhow and could still finish a complete non-incremental backup in 4-5 hours at full speed, and I have incrementals anyhow. Sure, the game per month I download from Steam would be ready in 4 minutes instead of 8, but, seriously, how much am I willing to pay for those four minutes? It's not like I'm staring at the progress bar at that point anyhow.
500Mbps is already enough for me to tailscale my house network up and have every single member of my family accessing the house Jellyfin server remotely simultaneously, which is not a realistic amount of load.
100Mbps down is still plenty for most people. 20Mbps up is definitely making some things annoying but most people will still be fine. It's a fine definition of minimum service for "broadband".
even spinning rust should be able to handle ~700 MBit. an SSD is generally on the order of 3 gigabits - even cheap ones can manage 300megabytes per second, which is on the order of 3Gb/s.
Sure, but remember that the disk bandwidth is the _uncompressed_ data. By definition it’ll be at least 2× larger than the compressed data you downloaded, if not more. Steam downloads are commonly disk or cpu limited on 1Gbps connections.
i'll concede you're probably right. In the late 1990s i had a Promise ATA 100mbit IDE card, and i swore that someday, i would have an IDE connection to the internet. Now, my CPE is called a terminal, but it's some proprietary cable from the terminal to the router. And my upstream isn't quite 100mbit (on starlink; it's about 40-50. On at&t 4g LTE it was 30-65mbit/s upstream).
so i'm still not at IDE speeds, yet. maybe on my cellphone in the metro i get that speed, though...
It is really ridiculous how fast our connections are these days, when you stop and actually think about it. Also pretty ridiculous how slow they used to be!
Get a better ISP. People will absolutely serve you content at >1Gbps; what this sounds like is your ISP gaming the speed tests.
(What does fast.com show you? This at least used to be somewhat harder for them to game.)
I manage to download entire Steam games at 1 gbps just fine. There might be something going on with your ISP.
Interesting. Are ISPs known to throttle steam or something. I’ve noticed that steam almost never downloads at the same speed of get doing a speed test. I’ve noticed it many times through out my life, though admittedly I’ve been stuck either way the same ISP across many states.
Steam downloads are often heavily cpu bound at higher speeds. I usually max out my 7800x3d at <1.5Gig.
Some ISP throttle sustained download, which explains having high test speeds and initially fast downloads that slow down with time. So it might not be Steam per se that they’re throttling, though it’s not impossible.
Steam also extracts while it's downloading and will slow download speeds if there is a healthy buffer of things to extract. If you're downloading a game with thousands of small files to an HDD or even just a cheap SSD chances are you'll be throttled by your own computer's throughput rather than your Internet connection.
I regularly get software to download at that speed.
Discussion (272 points, 15 days ago, 207 comments) https://news.ycombinator.com/item?id=44641464
With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money. That said, I've always felt satellite internet to be yucky but I'm used to "satellite internet" adding 1000ms to your latency and uploads that actually run over your telephone line.
I am curious if good satellite internet will lead to an exodus of people from cities and subdivisions. I actually live in my current house because the place I wanted to build on didn't have any form of internet access.
Also, I'm not really aware of anyone that is truly without "internet access" as cell phones have basically filled that area in years ago.
> With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money.
That was the same excuse used when nobody wanted to wire rural America for electricity. It was a bad take then and a worse take now.
Wiring those homes with single-mode fiber once will provide modern broadband for at least the next 50 years, if not longer.
Satellite is not and will never be a replacement for fiber.
> Satellite is not and will never be a replacement for fiber.
Plenty of suburban homes don't have fiber availability and are just fine. I'd rather have 3 coax companies competing for my business than 1 fiber option. I just really don't care that much about speeds above what coax (and Starlink) have to offer. Honestly, I'm on the lowest tier offered by my cable company, and I'd go lower if it would cut my monthly bill by a commensurate amount.
> I'd rather have 3 coax companies competing for my business than 1 fiber option.
And I’d rather have the city or county pull a single fiber back to a pop where an ISP can compete for my business because it’s absolutely absurd to have multiple companies pulling the same cable to a single address and using that last mile as a moat.
Huh? People used satellites as an excuse to not wire rural America for electricity?
Starlink might not be as fast as fiber, but it's more than good enough.
> Starlink might not be as fast as fiber, but it's more than good enough.
Users running up against oversubscription - this is a when not if condition. The capacity limits inherent to satellite tech are what they are.
ref: https://kagi.com/search?q=starlink+oversubscribed&dr=4&r=us&...
Between now and then, users' routine downloads run up against caps (eg:steam game packs).
It's not just the limitations of satellite tech or quantity. There's also just the fundamental limit of RF in shared airspace. You run into bandwidth limits due to interference even without overcrowding low earth orbit with a satellite network. When you're running signals over a wire/fiber, your signals are confined and interference is managed relatively trivially.
Search engine are not references, but how do you generate sharing links in Kagi?
There's a share link (the < symbol with nodes at the point and tips) right to the bottom and right of the search bar on a results page.
They were responding to:
> I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money.
In pure ROI, it's the same as running electrical service to rural areas. But still very important in general
> Huh? People used satellites as an excuse to not wire rural America for electricity?
Obviously that’s what I meant. And I didn’t mean people used the excuse of “we shouldn’t be pulling cable to all these houses when X is good enough”.
There were endless excuses to not electrify rural America including “they don’t need it”. It was eventually solved through co-ops.
That’s exactly how most rural areas are trying to solve fiber, but of course they get to fight the combination of folks like you that think “satellite is good enough” (it isn’t), and legacy ISPs suing them to slow or stop deployment.
Having spend several years under the thumb of a rural wireless internet service provider, the thought of all my neighbors sucking up shared Starlink bandwidth to stream Netflix in the evenings makes me nauseous. How is it a reasonable expectation that Starlink can really support entire rural areas?
It's not, but we've never cared anyways, so why start now with this?
> but we've never cared anyways
Rural areas in the US have enjoyed disproportionate communications (and infrastructure in general) subsidies since the founding of the USPS. More recently, the Universal Service Fund has spent around $5-8B per year since the 90s, first on phone service then adding broadband. And there are more recent broadband efforts for broadband in the tens of $Bs.
How effective have these efforts been at getting broadband access in rural communities? And comparatively, how effective have they been at giving a revenue source for telcos that they haven't had to really deliver on promises for?
It's not an expectation; it already happened. People love it.
> I'm not really aware of anyone that is truly without "internet access" as cell phones have basically filled that area in years ago.
Restated otherwise: You are not aware of the millions of Americans whose cell signal is weak to non-existent in places they routinely spend time at.
My rural cabin has access to:
1. 25/2 DSL for $140/month (12 month contract with $40/mo seasonal disconnect)
2. 5G (60/0.03 which was 100/20 last summer until the nearby tower became overloaded).
3. LTE (see above) which I routinely get ~100/30 but which Verizon, at least, will not allow their boxes to use if it can connect via 5G, regardless of usability of the connection.
4. Fiber (100/100) at $89.95 (12m contract with $25/month seasonal disconnect). The fiber has higher speeds, but I didn't price them out. The costs were originally listed at $34.95 before they ran the lines, but they have upped them to $89.95 now that they are run to my lake home.
5. Starlink (supposedly 150/20 for Residential Lite which is available in my area) at $80/month on a month-to-month with purchase of a dish (I got mine refurbished for $135). I am routinely seeing ~400/40 with 25ms ping even though I shouldn't be. That said, the speeds are more variable (low as 120/7).
---
I would LOVE for there to be reasonably priced and very stable Internet for month-to-month, but there just isn't except for Starlink--at this point.
My cabin (I don't live there full time currently) has starlink's $10/month plan where you pay by the gigabyte. LTE isn't even an option because the topography kills terrestrial radio.
Starlink completely changed rural internet. It's revolutionary.
I don't know how you think $89.95 for 100/100 fiber isn't reasonable, unless you're the type of person who believes the introductory price at comcast is the real price.
I pay 2/3rds of that for 10x the speed. 100/100 fiber is not that great.
We pay 2/3rds of that for 15/3 WISP service with variable latency and common outages. That's a great price!
The infrastructure has to be maintained year round, not just the summer months when lake houses are full. I don't think it's reasonable to expect affordable month-to-month pricing.
I know a guy who lives in rural West Virginia. He has fiber to his home, but the top speed is (IIRC) 10 Mbps down.
Is it like... cotton fiber?
Rural areas don't have anywhere near that many options.
DSL, fiber? Those do not exist if you are actually rural.
I explained below how it is defined by me as rural; but, yes, it's rural. In fact, my primary home JUST HAD FIBER RUN in my neighborhood last month and I get 1 bar of VZW. So; if anything, rural is ahead of major metro areas (my suburb of Minneapolis/St Paul has just under 90K residents). In fact, DSL from CenturyLink or whatever they call themselves today is 6mbit.
I am doubting the level of rural that has that level of connectivity.
Your options there are far better than what I had living adj to downtown mountainview not so many years ago.
It's in Central MN. It is 11 miles, straight line, from a town of 2200 and 15 miles from a town of 4400. The closest town is 8 miles as the crow flies of 700 and only has a gas station and a couple of bars. There's a bar and volunteer fire department 4 miles as the crow flies that has a population of 48.
Believe me: this is rural. Until investment from the federal government and/or forced lower-bound limits on 'broadband' back about 6 years ago, the only option was 768/300 DSL.
I'm in a small town (5000 people) on the Utah/Arizona border. The internet options are better, cheaper, and more reliable than I had in old town Mountain View, Palo Alto, or Sunnyvale. I go with the symmetric 1 gig with static IP, but symmetric 8 gig is an option as well. Silicon Valley's home internet options just suck.
> I am curious if good satellite internet will lead to an exodus of people from cities and subdivisions
You seem to be imprinting wireline capacity onto satellite. This will always be an error. A core facet of satellite is it's inherent limited capacity. This has always been the case and no tech is on the horizon that will change this.
What makes satellite have an inherent limited capacity? Like there's no real physics limits on how big you can make a satellite. More reception power = more signal in the same frequency range.
> Like there's no real physics limits on how big you can make a satellite.
There absolutely are. Launch weight, heat dissipation, energy storage and output are a few of the engineering boundaries that must be respected. You can't bigger one without impacting all the others.
It also has to fit in the launch vehicle with whoever is sharing a ride with it.
Our theoretically ginormous satellite might not much improve capacity (in the cell it's servicing) due to limitation in the assigned spectrum. You can only divide up time and frequencies so far.
Whatever you're doing here, you will need to duplicate that to cover the nearby cells. Then multiply that by a region and that by a country and that by a planet. It takes a lot of satellites to cover a lot of area from a low-ish altitude.
Satellite internet has its place but it's never going to be dominant. Any area with high density is better served by wires, and any area with medium density is better served by a combination of wires and cellular internet. Satellite only make economic sense for the truly remote; the astronomical and recurring infrastructure costs are just killer (Starlink satellites last for 5 years, buried fiber lasts for 50 at least (while providing better latency and throughput)).
> Also, I'm not really aware of anyone that is truly without "internet access" as cell phones have basically filled that area in years ago
Unfortunately very much not true. The US is a very large and sparsely populated place. As far as cell and broadband service goes, much of West Virginia, huge chunks of PA, upstate NY, and Maine all have limited or no cell service. In the Midwest the communication infrastructure is often delivered by community co-ops, not corporations, that also deliver electric power. And dont get me started on the vast Western states.
The US has, frankly, an absolutely ludicrous patchwork of providers and infrastructure for telephone, power, and broadband. It looks reasonably peachy when you drive on the Interstate but get off there and youll see how limited it is.
Sat internet would really help people get from zero to 1, but that said, its still kinda crappy compared to fiber. Having the bird in range for 15 minutes until you have to handover to the next bird is not conducive to smooth, glitch free Internet.
> With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money.
Couldn't the same be said of expanding the electrical grid, when solar panels and batteries are an option?
Depending on where you are, that's an entirely valid option IMO... There are lots of houses with septic tanks without centralized sewers.
> cell phones have basically filled that area in years ago
That doesn't work well in mountainous areas, which are often rural. There are three towers within a couple of miles of my house, but nothing close to line of sight to any of them. ~1/4 mile away I can get full service, but I have none at all at home.
Luckily Comcast / Xfinity rewired our little town a few years ago so we don't have to rely on satellite (not great during the heavy snowstorms and thunderstorms we get). I'd love to use cellular as a backup as we have 3~20 full days of blackouts per year (most of us have generators and/or batteries) and Xfinity goes down 80-90 minutes after a blackout starts.
Is internet access really the main determining factor for where people want to live? I know some people really dislike cities, but there are plenty of reasons to live in one if noise and crowds aren't dealbreakers.
Cell phones general do not work in rural areas unless you are very lucky or go climb the highest peak that happens to be in direct line of sight of a tower.
Satellite has a lower cap on latency, until we figure out faster-than-lightspeed radio transmission (if that's even possible)
Upon reading the latter portion of your comment, a potential story for XKCD comic immediately sprang to mind...something along the lines of a tech person or team invent faster-than-lightspeed technology, but abandon their work on a spacecraft warp-style engines to improve satellite internet....because its just so much easier to make tons of money doing barely the minimum within the ISP monopoly business. lol :-D
> With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money.
A number of replies here explain how deploying wireline is absolutely a good expenditure of money.
Once you have read them, will you then understand?
> With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money. That said, I've always felt satellite internet to be yucky but I'm used to "satellite internet" adding 1000ms to your latency and uploads that actually run over your telephone line.
Depends on what you declare a "good expenditure of money".
The thing about running fiber is that it's not particularly expensive to do. dig trench, install infrastructure, rebury it. Hell, if you have a utility pole, you can omit steps 1 and 3, so long as you're okay with the line being exposed to the elements.
Launching satellites is very expensive to do. There are very high barriers to entry that prevent meaningful competition from occurring for satellite internet providers, which is why HughesNet was crap for all of those years. It took NASA funding a commercial crew program to get a rocket built that could put payload in LEO for "cheap", and there's only two or three options that can reasonably be expected to operate that way.
I'd rather have a lot of competition for installing and operating fiber infrastructure than having an effective duopoly for running performant satellite internet.
I'm under no delusions that will occur in rural areas under the present regime in DC.
As for an exodus from cities and suburbs, no. There's no amount of internet speed that would make me move to a place where there's reduced services and infrastructure in every other category, like schools, hospitals (this is a big one), water, sewer, transit, etc. and there are millions more like me.
> There's no amount of internet speed that would make me move to a place where there's reduced services and infrastructure in every other category, like schools, hospitals (this is a big one), water, sewer, transit, etc. and there are millions more like me.
I moved to "rural lite", and have a well and septic. That part isn't so hard to live with; although no water for a couple days when our well pump went out wasn't fun.
The hard part IMHO, is that everything takes so long to get to. We've got a good small town that's a couple minutes away, but if I need to go to a big box store, that's 30-45 minutes, each way. I like it here, and I have a 8 acre parcel, which is incompatible with suburbia, but it's an adjustment.
Utility poles may exist, but they may be structurally inadequate so the fiber installer may have to wait years for the pole owner to get around to replacing a string of poles before the fiber can be hung.
The poles may have been placed along the back property line without an easy way for a new utility to get in there without nicely asking each household on the street to let them into their back yard.
Spend a little time reading Sonic's "Access" forum. There's a very real have-vs-have-not dynamic going on there because there are areas where Sonic just can't deploy fiber any time soon that are just blocks away from people with service. Estimated service dates keep slipping when they run into unknown unknowns. There is significant engineering and construction work required in each and every neighborhood.
That have-vs-have-not dynamic creates niches for satellite internet service -- rather than having to wait until all the fiber ducks are in a row, you can just pick up a kit at a hardware store or get it delivered, plug it in, point it at the sky, and you're done.
Yes. This is also why people are talking about doing space solar power beaming. The cost of regulation and access has gotten so high that it's actually getting close to becoming cheaper to launch those solar panels into space instead of just installing them on the ground.
> The thing about running fiber is that it's not particularly expensive to do. dig trench, install infrastructure, rebury it. Hell, if you have a utility pole, you can omit steps 1 and 3, so long as you're okay with the line being exposed to the elements.
Uhhhh.... it can cost several hundred dollars just to dig a trench across my yard. If the plan involves providing broadband to rural farmhouses that might be miles apart, that could be tens of thousands of dollars per customer.
Obviously on utility poles (where they exist) is going to be much cheaper, but you still need an ISP to build and operate the thing and set up things like repeater boxes and deal with outages.
Median latency 21ms over the last 15 minutes, Starlink.
> Also, I'm not really aware of anyone that is truly without "internet access" as cell phones have basically filled that area in years ago.
I mean, this is basically it. For paying bills, doing school work, messaging, getting news, etc the options these days are affordable and very convenient. Even video calling and streaming videos doesn't really affect my data plan too much.
The only advantage of a hardline to your house is if you are going to do copious amounts of high-def streaming or you need a very stable connection. More of a luxury/hobby resource.
> Also, I'm not really aware of anyone that is truly without "internet access" as cell phones have basically filled that area in years ago.
Then you clearly don't know anyone who lives in a rural area, because much of the US still doesn't have good cell coverage.
>With satellite internet from starlink (and I think amazon is attempting their own version?) I don't see how running internet over wires to rural housing could ever be seen as a good expenditure of money.
There are two perspectives on this. From the user's standpoint, when I finally wired up my house and we were no longer using wifi, all the streaming video was just flawless, no more stuttering. And I was only sharing that air with devices in my own house and the two closest neighbors. I wouldn't want to have to share it with all of North America.
The other perspective is from the point of view of the ISPs themselves (the ones not Starlink). Can they pay off the cost of the infrastructure, and then make a profit? And it turns out that they can. Just not a quick one. If they can't profit on it this quarter, many see no incentive. Someone needs to light a fire under their asses.
These are both obvious, and if you "don't see how", maybe you need to look.
you could not pay me to use starlink
[dead]
This is hilarious. Just 3 weeks ago I signed up for 10 Gbps symmetrical fibre to the home with XGS-PON, and it costs me US$23/month (naturally, this is not in the US).
The only US argument I hear with respect to fibre broadband internet, high-speed rail, good healthcare, and more is 'we're too big!'. At some point it starts being genuinely funny.
Can we stop paying those fees for companies to expand broadband since it's not needed any more?
It's all getting redirected to Starlink
I wish. Starlink has gotten almost zero money from the government (relative to their revenue). If Starlink got even a fraction of the money that fixed broadband companies got they could absolute explode he size of the service.
It seems like an increased executive mandate on the telcos is an odd thing to cheer for, and im not convinced a mandate from FCC is really necessary any more. The fiber installation (networking, boring, marketing) industry has dramatically expanded, and I would expect telcos to continue installation of fiber to new markets since it's so much cheaper to manage than copper. with the ubiquity of 5g cell networks, the amount of fiber extended in the last decade, and the vast improvements of sat-net... is there a need for every property in the US to receive a fiber connection?
FCC isnt giving on up broadband, its cancelling its mandate for expansion, and likely going to delete the cash payouts to telcos as well? The FCC never delivered on AFFORDABLE... telcos have raked it in on that front.
In other words, seems like a non-event?
I can get 1Gbps up/down in a remote area (close to the fields) in a mid-sized city in Brazil with _zero_ government subsidies. The energy company that owns the poles was required to offer it to anyone that wanted to rent space on them and a dozen different fiber-only ISP's came to life.
I understand the US has a lot of pre-existing infrastructure and it's hard to move on (sunk costs) but.. the situation is just insane. If the government just stopped getting bribed and allowed competition, it seems this situation would be solved in a couple of years.
I have family in very rural east texas. They have 1Gps bidirectional at the hands of EasTex Co-Op spending federal dollars to actually lay fiber across their service area.
Lots of them took the money and ran, some jammed it into real infrastructure.
> Even under an ideal situation where Trump authoritarianism is conquered and some sort of sensible alternative takes office, restoring oversight of companies like Comcast and AT&T — both bone-grafted to our domestic surveillance networks — is never going to be a priority in a Congress that’s now too corrupt to function, under a broken court system that treats corporate power as an unimpeachable deity.
Well, that's not the ideal situation. The ideal situation is a wholesale countervailing takeover of power that destroys the broken congress and courts, replaces them with a genuinely responsive government, conducts a large-scale seizure of the assets of the entities and individuals who have profited from this debacle, throws many of them in jail for decades if not life, and then begins to use those assets to rebuild things on a more equitable footing. Now that's less likely than the scenario described in the quote, but the scenario described in the quote is far from "ideal".
That's fine as long as they unshackle the rules around municipal internet. Get out of our way and we'll build a way better system.
Instead it's rules for thee and not for shitty massive telecoms. Hypocrisy is too light a word.
With everything that has been happening maybe that is a good thing. We don't need even more people becoming radicalized, becoming flat earthers or pushing raw milk in public schools.
Talk radio reaches well past broadband. These problems aren’t new, they are just social problems from decades ago finally surfacing.
These problems may not be new but they have reached levels this country hasn't seen in decades and there is no sign of slowing down. The availability of cheap AI on the internet isn't going improve anything anytime soon.
The FCC wasn't trying before.
We need to put down a law to force network infrastructure companies to compete, like European nations, if we truly want change.
I would expect that we can get some press releases that U.S broadband is the fastest and most affordable in the world though.
Sooner manned trip to mars than break up the telco monopolies
Well, they weren't exactly doing a bang up job with it.
It hasn’t been working anyway.
It was just more corporate welfare.
News from 2 weeks ago, lots of discussion: https://news.ycombinator.com/item?id=44641464
Maybe in another 20 years my local monopoly will adopt ̶I̶P̶v̶6̶ a technology faster than 6mbps/768k ADSL1.
When I bought my home in 2009 (suburbia in major metro area) - AT&T was offering a whopping 6mbps DSL connection for wired internet. At the end of 2024, they were still only offering 6 mbps DSL.
This year - they rolled out fiber. So, it only took 16 years - maybe you'll get lucky.
It's always been available - you just have to pay for the install.
Get a quote from your local ISP and you'll realize why it took 16 years to bring fiber to your home. I received a near $100k quote to come across the street in a former building... it gets stupid expensive stupid fast.
A sewer line is on the order of $10k.
Fiber across a street should not be $100k.
Maybe municipal regulations make it harder than it seems. I’m curious.
Depends where you are and where the line needs to be run. Shutting down a busy city road? Gonna be expensive...
At our current building, we just had fiber run from an in-ground vault already on premises (but not serving the building) into our MPOE - cost about $18k to go less than 100 yards... majority of-which didn't require boring.
My original point was, people want fast internet out in the boonies but aren't willing to pay for it. Few residential customers are going to shell out $600+ a month for fast internet... but that's what's required for the ISP to recoup their construction costs in anything resembling a reasonable time period.
We will pay for it over the course of the years of paying for the service. Plus, the economic benefits of enabling remote production/consumption of digital services probably outweighs the cost by an order of magnitude or two.
> We will pay for it over the course of the years of paying for the service.
Yes, at like $600+ a month over 6 years... you can do the math, it's nasty.
> Plus, the economic benefits of enabling remote production/consumption of digital services probably outweighs the cost by an order of magnitude or two
ISP's aren't a charity service. Even your municipality cannot afford to run fiber to the boonies.
Okay, whatever about ADSL2+, why on earth would anyone still be doing ADSL1? Are they buying the DSLAMs off eBay? Like, I'm legitimately surprised that they can still get the equipment; you'd think it'd be cheaper just to go to ADSL2+.
10 years ago I had ADSL2+ at home (basically due to planning issues; traditionally, urban phone stuff had been hidden underground, and the telecom was having trouble getting planning for VDSL street cabs), and that felt like living in the past back then. ADSL1, today, is absurd.
There's finally competition in the US home broadband market...
*Fixed Wireless players*
T-Mobile, Verizon and ATT (tho later two isn't widely available)
*Low Orbit Satellite Internet*
Starlink
*Home Wired Connections*
Comcast, Charter, etc
I've been bouncing my cell phone service and home broadband around every year per the all the good promotions. Recently on a $40 a month Comcast home broadband (300Mbps) with one free line of unlimited data cell/data service promo. Previously was paying $100 to T-Mobile for Unlimited cell/data service and their fixed wireless service.
We are much better nowadays for affordability then a few years ago!
That's par for the course with this administration.
If you want to get inside the mind of "Why do this?" for the current administration all you really need to do is ask yourself 2 questions.
"Does the right consider this woke?"
and
"If I was a libertarian billionaire looking out for no one but myself, whatever the price to the public good, what would I decide?"
and you will be able to predict with 80% accuracy what they are going to do and their stance on any given situation.
now the regional broadband monopolies can become a global space monopoly
I though this article would be interesting but it appears to mostly be a unhinged rant by the author to vent about Trump's politics. It was not particularly good educational reading.
Contrarian opinion: What exactly is the point of "fast" internet?
Most people use the internet for entertainment, people can survive watching Netflix at 1080p/720p, it isn't debilitating. If I'm wrong in my assessment, give me use cases where you require fast internet in a rural area.
If the question is cost, I don't see how laying fiber and equipment for hundreds, if not, thousands of miles is a solution to reducing cost (unless it's subsidized by the government).
Spending billions on mostly "entertainment" is a waste.
> Most people use the internet for entertainment, people can survive watching Netflix at 1080p/720p, it isn't debilitating. If I'm wrong in my assessment, give me use cases where you require fast internet in a rural area.
Lots of desire for more automation and data surrounding agriculture, for starters.
More generally, people work through internet applications. If you want to check your email, type up a document, have Zoom calls, etc., you need to do so through the internet. If you don't have fast internet, you can count on being less productive.
None of those things you listed require internet faster than required to stream Netflix at 1080p.
That still would be faster than dialup, older generations of satellite internet, and some tiers of DSL service can provide, and there are still areas where those are the options.
Working from home when you're a doctor:
https://bbcmag.com/chattanooga-doctor-becomes-worlds-first-1...
1. Chattanooga, TN isn't exactly "rural"
2. How large is each "diagnostic medical image"? I can't imagine this being in the order of TB
(I accidentally replied to the parent comment, reposting here)
A single high-resolution pathology slide can be 20 GB. A busy pathologist can read up to 300 slides per day. That's 6 TB a day to download.
If you're next question is why would a busy pathologist live in a rural area, I have no idea.
A single high-resolution pathology slide can be 20 GB. A busy pathologist can read up to 300 slides per day. That's 6 TB a day to download.
People were perfectly fine with 56k when most of the Internet was just text. Stalling the internet at that speed could have been argued for by saying anyone who wants DSL/Cable at home just wants it for Napster.
Fast Internet speeds allow for unthought of innovations. If we get up to terabyte speeds, maybe nobody cares about watching Netflix if we can now have holodecks that become fully immersible experiences that allow for educational training and, to your chagrin, entertainment as well. Maybe LLMs are basically free because everyone can just have all of human content constantly updated.
Letting speeds stall stifles innovation and let's Netflix just continually up their prices for the same 720p content you refer to.
> maybe nobody cares about watching Netflix if we can now have holodecks that become fully immersible experiences that allow for educational training and, to your chagrin, entertainment as well
I don't see why billions of dollars should be spent, for say, less than a low single digit percentage of the rural population who want to stick a machine to their head for educational training.
It's infinitely cheaper, a more sane and healthier decision to move to the location where they can sit in front of the instructor, if they value this education.
So just stop being poor. Got it.
If we're going full reductio ad absurdum and taking snipes instead of conversation, then maybe the appropriate response is: You can be poor all you want, but don't expect someone else to subsidize your life.
My comment isn't really that absurd in the context of this whole thread and its insistence that decent quality internet service is something only the wealthy and/or those who live in urban areas should have access to.
> You can be poor all you want, but don't expect someone else to subsidize your life.
The rich and wealthy have their lifestyles heavily, heavily subsidized, so why not the poor as well?