> The funding is going to a mix of cable, fiber, and fixed wireless providers, plus SpaceX's Starlink satellite network.
I think cable providers should be left out. They were already gifted many billions of dollars in the past for this same basic purpose but they pocketed the money. They can just use the free money they've already been given. Instead, the funding should go to existing munipical fiber ISPs to expand their coverage. Give money to people with a proven track record of doing what you want, not people with a proven track record of keeping the money for themselves.
If the issue is link speed symmetry, then cable should be out of the running entirely. Even my phone offers better upload speed than cable in my neighborhood. Same for my parents' place across the country. New money should go to fiber, satellite, wireless, or nothing at all.
It's at least somewhat inherent to the technology. DOCSIS download channels are shared among subscribers; upload channels are allocated exclusively per-subscriber. This architecture provides economic incentives for the provider to provision asymmetric circuits rather than symmetric ones.
100 Mbps up requires expanding the uplink band from 5-42MHz to 85MHz. That requires changing outdoor components like amplifier and filters, and upgrading millions of cable TV boxes, splitters and modems inside customers' homes. The combined cost of hiring more technicians and buying boxes might approach the cost of upgrading to fiber. Altice USA decided to skip DOCSIS 3.1 and start dragging fiber.
Which is all stuff they are doing anyway. Cable modems, even when not integrated with wifi don't last more than a couple years before someone adds a couple more channels, or tweaks the spec. Etc, Docsis 3.1 modems were available in the local best buy for probably 5+ years before my cable company would even provision them in docsis 3.0 modes.
They might have some excuses with the amps, but again, those can be upgraded as they age out/die too. This was the excuse 20 years ago when the amps didn't allow any upstream signals through in many areas. If a cable company hasn't been preparing/slowly rolling out new equipment as soon as the specs are available that is their own fault. Like any other technology you either get on the train, or you get left behind.
Beyond that it's pulling from a finite pot. Every Hz you allocate to upstream is one less Hz you can allocate to downstream.
Fiber doesn't have this constraint. It's trivial (and common) to deploy a pair to each premise, but, even if you don't want to do that you can shine light in both directions on the same piece of glass and reasonably* run a terabit across a single fiber (not a pair).
This is partly incorrect. Near the beginning of using data over coax cables the network architecture design dedicated more frequencies to getting data to your house as opposed to you sending data out.
This was done because the initial network design was that they would send massive amounts of data to you as images and you would only send small amounts of data requesting paper view or other channel/media options.
All components through out the network rely on what came before.
Business costs more because it’s less oversubscribed than resi, also is intended to be more robust in terms of change windows, and depending on your provider may have some best effort SLAs.
Business class in airlines -- goes the same place as coach, but "the service is better". Actually what they've done ist o mindfully take away from coach to stimulate upgrades to business class.
I signed up for business internet just so I could get a static ip address. Other internet services like sonic will give you a static address if you ask.
With Starlink I think the competition will get more real for them.
In the rare markets with actual competition the shitty cable companies provide at least okay service. In Palo Alto I can get 1gbps down/35mbps up on Comcast for $89 (with 1TB data cap I have to pay $50 extra to remove).
That's not too bad.
They also offer a 2gbps down and up service called gigabit pro, but it requires installation and it's pretty expensive: https://www.reddit.com/r/HomeNetworking/comments/fs6un2/comc.... It's also hard to get because the reps on the phone are not well trained and don't know about it.
This is only because there are a few internet options here. Where I grew up there is one provider (Spectrum, was Time Warner, and Adelphia before that), the service is bad and the price is at least double.
I recently moved to SF and I'm in a newer building with Google Fiber Webpass, it's 1gbps down and up with no cap for $57/month - it's awesome.
The worst part about Comcast (ignoring the 35mbps up which is pretty bad) is everything around the service. Their website is extremely hostile. It's impossible to do anything online. Their pricing is heavily obscured. They push shitty TV packages I don't want or care about. It's hard to cancel without going through retention.
Google Webpass? I signed up online in 5min, connected my own equipment and was good to go.
Hopefully with real competition the old services will either get competitive or die. I don't really care which.
This. I think the really confusing thing is why so few new entrants have joined the US telecom market. There is masses of cheap capital to build these networks out.
Most of Europe seems to be moving to this kind of model:
A cable TV provider using DOCSIS or FTTH (and expanding their footprint)
Existing incumbent telco providing service via VDSL/FTTH (rapidly rolling out FTTH)
New "altnets" (as they are called here) doing new build FTTH (like webpass in the US, but they are not just aimed at apartment buildings).
This has led to really significant competition. In portugal I have access to at least 1 FTTH and DOCSIS3.1. In London the same, my last flat actually had the option of 2 seperate FTTH providers (not sharing infrastructure) plus DOCSIS based cable.
I think the altnets are the key to forcing investment by the otherwise duopolistic incumbent telcos and cable companies.
I think what the FCC should be looking at is how to allow altnets to roll out service quicker. There is so much state/municipal legislation that I imagine that is the main blocker of doing this at scale.
> With Starlink I think the competition will get more real for them.
It's entirely possible for them to just ignore starlink. Due to the nature of satellite internet there's only so high a density of customers that they can have. At maximum theoretical efficiency and distribution this is maybe 1% of US households. In cities it will be irrelevant (starlink will either raise prices or turn away customers), in rural areas it may serve as some competition, but infrastruture investment in those areas is naturally lower return than in cities, so they're not likely to have a strong incentive to improve their service there. Elon Musk has specifically said Starlink is not competition for the big telcos.
>They also offer a 2gbps down and up service called gigabit pro, but it requires installation and it's pretty expensive: https://www.reddit.com/r/HomeNetworking/comments/fs6un2/comc.... It's also hard to get because the reps on the phone are not well trained and don't know about it.
This is how you know they arent really concerned about competition. This level of service is probably not quite a profitable because they likely cant oversubscribe their GPON nodes as aggressively with customers on that tier. The only other explanation is they dont care they are leaving money on the table.
In my area comcast is $89 for 200/10. And I live in an area with a significantly lower COL than Palo Alto.
I have to monitor my network for bandwidth usage and recently had to revert back to sneakernet for offsite backups because i was routinely approaching the data cap.
The only other option is CenturyLink DSL at like 5Mbps/768Kbps...
Starlink is available and frankly im considering it out of spite. I canceled cable and have the rare unbundled comcast plan. I could easily run dual ISP's for a while and be around what I was paying for cable TV and if starlink pan's out....i can just axe comcast.
But a 40ms first hop may not be tenable. Not sure.
I definitely dont live in a large city. But its not really the boonies either.
Yeah - even regular 1gbps down/35mbps up gigabit was hard to get when I first signed up.
I called and asked the guy if it was available, he said not in my area and they'd contact me when it was. This seemed wrong since I'd read it was available (and I guess he was just lying about contacting me?).
I called back and got a new guy, he said it was available but 'installation' was $300. This seemed wrong.
I tried a third time, this guy said I was confused and the word 'gigabit' didn't exist - I must be talking about the 'gigabyte data cap'. I begged him to click around his screen and look for gigabit service. He did and finally found it, "first I've seen this". I asked this guy about gigabit pro and he said it didn't exist.
I'd since switched to using comcast twitter to get someone at least moderately competent when I want something (they know about gigabit pro). Thankfully now I don't have to use them at all.
They are insanely difficult to buy stuff from.
During all those calls they keep pushing TV packages and get upset when I say no thank you. When I called to cancel they kept pushing promotional rates and asking me not to cancel. Even though my new service is less than half price, uncapped and has 30x more upload.
All that said, when the service itself is running it at least works pretty much all the time without issue.
That’s my experience as well. Uptime is decent. Just hope an outage is widespread and not at your ground block.
Because god help you if you need to reach out. It will be frustrating and painful and there is no escalation option.
Honestly I get better results just going through the Better Business Bureau. It literally solved my last line down issue faster that way that the empty promises from their call center robots
The only option I have is Spectrum with an absolutely crappy upload speed (currently 6MB). Sadly this neighborhood never got any other option. Starlink might finally provide some competition.
I’m in the same boat with spectrum, 9Mbps up, which works out to 1.44MB up in real world example. It takes me upwards of a full day to upload a 30 minute 4K YouTube video! It’s shockingly bad. I called to upgrade and the only option is business class for $200 per month offering 35Mbps up.
StarLink beta welcome email just arrived and I’m counting the days till I cancel Spectrum completely
Including cable providers is fine. The issue is how this money is structured. It needs to be a loan where the debt is forgiven once you reach certain milestones. The issue with the money earlier was there wasn't any teeth when it came to enforcing what was to be done with the funds.
Make the loan be equally competitive - any company can apply for any amount of money, and a small amount of the loan is forgiven per user they bring above the 100/100 mark (with possibly a bonus to rural users). Any backbone infrastructure created with these loans needs to be peerable as well.
>They were already gifted many billions of dollars in the past for this same basic purpose but they pocketed the money. They can just use the free money they've already been given.
I mean they could but they wont, and there was no reason to with Ajit Pai running the place. I daresay they'll have to be sued into compliance to do it now. Not that I object to that: they need to refund the taxpayers' money or do what they promised with it.
They’ll just claim something something 5G and will be back in the feds good graces. Honestly the cable industry is a monopoly and has grown beyond government oversight. This needs to change but we don’t have any other option than nullifying anti-municipal fiber laws and start laying down new lines.
Most of America doesn't have true (mm wave) 5g. About 1/3 of America lacks "5g" entirely. Assuming one does have access it's not financially prudent for most households to buy into 5g when the caps are as low as 100gb/mo.
Nitpick: mm-wave is just one of the dozen things that are in the 5G cornucopia of random technologies, it's not more or less "true 5G" than eg 5G edge computing.
I specified mm wave because regular 5g is about as good as 4g making the distinction worthless in practice. To actually benefit from 5g and justify the costs you'll need to be in one of the few cities with mm wave and have reasonable data usage agreements, which virtually no America can say they have both AFAIK.
In America 4>5 so we’ve appropriated “5G” to mean the mm-wave wireless transmission.
There are still large pockets of the US on data caps as low as 10gb. Throttled connections. Asymmetric networks. All so we can be nickel and dimed to death on fees, privilege, bundles, and in some cases double paying for content.
It's funny to me, reading "as low as 10gb". I used 45MB of data last month, and 9 of that was the Android OS. (And, yes, 4.33GB on wifi, of which 186MB was the OS). I know other people use their phones way more than I do, it's just a different world to me, where data limits are a thing to think about.
My neighborhood the Verizon 5G is pathetic, its slower than 4G and that is still slower than my terrible Spectrum wired internet. I only live in the middle of a 5M metro area.
We have either DSL or Comcast. 300/10mbps costs $100. Total rip-off but that’s the price of our county signing a 10 year exclusive deal with them. Now it’s expired but nobody is going to bring business in.
What "infrastructure" means needs to be nailed down before this can be discussed.
Internet service is seen as something in the same league as roads, power, water, and other utilities. Things that require public rights of way, extensive digging, cooperations among many jurisdictions or land owners, and special rules to make it all possible.
Clothing, food, and construction are all businesses that don't require that level of centralized coordination. Anyone can open their own bakery, or buy a farm. Anyone can set up sewing machines and purchase fabric, or get a contractors' license and start nailing sticks of wood together. Government still sets standards, of course. But these businesses don't have inherent monopolies the same way infrastructure businesses do.
Those things all have much more redundancy than comms. The US has three mobile internet providers (there's also MVNO's, but they use the same physical infrastructure as the big three), and any given home will usually only have access to two or three landline ISP's in terms of infra (e.g. cable, DSL, fiber).
In contrast, food can be shipped in to a city from innumerable farms, ditto for clothing. Home construction, there's lot of different home builders around, it's not a natural monopoly like communication.
For food production, I don't really have to trust any particular two or three farming companies -- if they fuck up, food can be shipped in from elsewhere. Whereas if my local ISP's are exploitative or simply incompetent, what's my recourse?
No one has monopoly control of the infrastructure in those industries. Anyone can buy and farm land, textile plants, etc. Its not possible to simply lay more conduit and compete.
I don't see an inherent problem with for-profit companies owning infrastructure, but it needs to not be monopolized and it needs to not be owned by a bunch of public market shareholders and hedge funds who don't even use the product. Utility monopolies and public companies rarely put customers first.
Cable isn't the only problem, companies like AT&T are too.
I don't think the money has to go to municipal fiber though. I'm currently in a rural city running 1Gbps/1Gbps with no data caps because of a small company got extra funding.
> They were already gifted many billions of dollars in the past for this same basic purpose but they pocketed the money.
Very little government money has been spent on broadband. Federal funding for broadband dates back only to the Obama administration. And cable companies have largely not received this money. You’re thinking of something else.
I think the Telecommunications Act of 1996 was supposed to help improve Internet access through its Universal Service provision. Of course all the telecoms pocketed the money and didn't do shit.
In my state there was USAC funding that went unawarded year over year, and for all I know still is. Carriers just don’t find it worth their while in many cases especially if they don’t need to worry about any threats to the captive markets they already serve.
> AT&T and CenturyLink accepted all funding for all states for which they accepted CAF funding. For AT&T, seventh year CAF funding totals approximately $427 million for 18 states. For CenturyLink, the total seventh year funding totals approximately $503 million for 33 states.
Over 7 years, AT&T has received half a billion dollars from the feds. That’s not a lot of money for building fiber. In rural high cost areas, where that money is supposed to be targeted, that might be like 20,000 homes. AT&T spends over $10 billion per year on wireline upgrades and maintenance: https://www.fiercetelecom.com/telecom/at-t-dedicates-40-40b-... ($16 billion in 2017). Over that 7 year period you’re talking $100+ billion in investment from AT&T versus $0.5 billion from the feds. Do you see why that money hasn’t moved the needle at all?
To my knowledge, Comcast has never received universal service fund (now called the connect America fund) money.
We can argue about whether universal service money has been used efficiently. But the idea that companies like Comcast received billions of dollars from the feds to build fiber everywhere is a very Trumpian misrepresentation (taking something with a tiny grain of truth and spinning it into a tall tale).
“On October 27, 2011, the FCC approved a six-year transfer process that would transition the money from the Universal Service Fund High-Cost Program into the new $4.5 billion a year Connect America Fund, effectively putting an end to the USF High-Cost Fund by 2018.“
“In March 2014, the FCC approved "Phase II" of the transition to the Connect America Fund, adding $1.8 billion a year in funding.”
What’s important to realize is these subsides ends up providing paying customers to these companies. As such they don’t need to cover 100% of the cost of connections only the subset of that cost that wouldn’t be covered by monthly fees.
What's misleading about the numbers? I cited actual distributed money over 7 years for one carrier (AT&T). You're citing proposed amounts to be distributed for the whole program. The program has hit numerous delays, and most of that money has not been distributed. Regarding Phase II, for example: https://techcrunch.com/2018/02/02/fcc-gets-ready-to-kick-off... ("The fund was proposed back in 2013, but for various reasons never saw the light of day.").
It doesn't matter if the subsidies cover 100% of the costs or not--the point is that the funding is a tiny fraction of what the industry already spends, and isn't enough to move the needle outside a small number of very high-cost rural places. The replacement for the connect America Fund, the Rural Broadband Opportunity Fund, plans to distribute $20 billion over a decade: https://thecounter.org/fcc-20-billion-dollar-rural-broadband.... (Again, plans.) Wireline providers invest about $40 billion annually--or around $400 billion over the next decade.
The notion that subsidies are more than a drop in the bucket (and should be expected to significantly move the needle on broadband deployment) is just false.
The fact you’re talking about a subset of total subsidies to all infrastructure spending by these companies. This wasn’t supposed to be new money, it was renaming an existing telecom subsidies as something else.
In the associated April 2014 Connect America Report and Order, the Commission decided that a competitive ETC that is awarded “support through the competitive bidding process will cease to receive legacy phase-down support for those specific areas upon commencement of Connect America Phase II support.”12 As of July 1, 2014, consistent with the USF/ICC Transformation Order, the Commission froze fixed competitive ETCs’ legacy support at 60%...
Also, wireline providers don’t invest $40 billion per year in extending their networks to new customers. The majority of that money is spent on maintaining and upgrading existing equipment as part of their normal operating costs. It’s like complaining they can’t simply get money without doing anything.
Whatever conclusions you reach from the points made in that thread, there is an important additional point here. That thread is talking about telephone companies. OP is talking about cable.
We tend to think of both of them as just “broadband” today, but the regulatory structure is totally different. The central premise of the notion that telephone companies received “billions of dollars from the government” is that the government lifted rate regulation on telephone companies in the 1990s. Rate regulation told the companies what they could charge for local phone service. The argument is that phone companies made $200 billion more than they would have had had those regulations remained in place.
I disagree with that assertion for phone companies, as addressed in the thread above. But even on its own merits that argument is not applicable to cable companies. Cable companies weren’t rate regulated (except that in some places there is a regulated minimum television service they must provide at a government-specified price). There was no government action in the 1990s that contemplated subsidies to create incentives for cable companies to build fiber.
Imagine how good our network technology policy would be if the Senate weren't run by septagenarians and octogenarian.
I fear that data caps for home internet will become standard. There is no technological reason for these caps. For wireless, they could be defended by arguing that companies only get to use so much spectrum because of FCC laws. But the only limit for home internet is the ISP's willingness to deploy equipment and install cable/fiber runs. That's exactly what customers are paying them to do.
I dunno there's a handful of millennial politicians in congress and they're hit-or-miss.
I would focus less on the age of the Senators and more on the fact that your median voter is 50 years old. And your median primary voter is 58 years old. They also consume significantly more television (and less Internet) than people under 40 (Millennials and Gen-Z).
There's only one senator under 40 and he's pretty supportive of an open internet.
AOC seems like she understands the massive reach and power of the internet...she chided Dems for under-spending on online ads and went after Zuckerberg for not deleting false content. She's not even a technologist and she "gets" technology more than many who have much more experience. Don't get me wrong, I'm not particularly in awe of her or anyone else in Congress...just clarifying that it does indeed seem that the young ones understand technology's impact on society far better than the old ones.
Most people under 65 in our rural area are dropping ATT u-verse ($200+ per month) and keeping only DSL + Roku + a few select channels because of cost. Cable television is too expensive.
Flatrate models usually means that a large majority of users actually covers the costs for a the minority high consumers. It is true that the marginal cost for more transmitted data is almost zero, but the overhead costs of running (and expanding) networks are certainly not. Cost has to be covered and you could very well argue that it makes sense that the large consumers pay more.
Besides maybe above the 98th percentile, data peering costs are very low that the physical infrastructure dominates the cost of service for residential users.
Since residential connects are over subscribed anyway, it is trivial to make users throttle during peak demand based on their current/historical usage.
If you use Cloudflare's costs and peering:transit ration from 2014 (which is almost certainly cheaper now); 1Mbps/month costs $8. [1]
1 Mbps/month is ~329GBytes. A 1TB data cap would be just ~3.1 Mbps/month costing $24.8.
I'd estimate that a last mile ISP could probably get an effective rate of half or less since connecting to end users/consumers is an advantage. And in a lot of cases, the same ISP offering residential service is also selling transit to other networks so in a way the data is already paid for by the time it reaches your local ISP.
I checked Comcast's 10k report and their margin for the segment including broadband is over 40%, which is crazy but about what you would expect from a de facto monopoly. I don't think forcing them to invest in capex to benefit customers is asking so much when they are making those kinds of margins.
Those margins are misleading because they're _multi_ service operators, and accounting standards require that you can only list direct costs.
Revenue is easy: how much did you take in for video? phone? Internet access?
Costs are harder because you can only include business line direct costs. Since the cable plant is used by voice, video and data services it's not a direct cost of any of them. Same thing with the service vehicle fleet, call centers, etc. Most things get saddled in "administration" categories and obscure the true cost of providing the service. As a company overall, their margins been hovering around 8-12%.
They report all of the services using the cable as a single segment of their business. Of course they are attributing operating costs to that segment.
Like 40% of the segment cost is "programming" (TV), so the internet part of the service likely has even better operating margin than the segment overall (basically, slightly higher revenue than TV with considerably lower costs).
If the network is more heavily used, it will require more capacity where it connects to other parts of the network or other networks. More ports and more wires means more failures of ports and wires and more diagnosis of ports and wires and more replacement of ports and wires.
Or, if capacity is not increased to meet demand, you need more people at the callcenter to ignore customer complaints.
That sounds like a fixed cost to build the network, not the cost of running the network. I agree, building a higher capacity network is more expensive, but how does the cost of running a network change with utilization? If a network is underutilized does it somehow cost less to operate?
If it's underutilized, you can save money by delaying replacement.
Let's keep things simple and say in one interconnect you'e got 4x 10G ports. If your utilization peaks at 5 Gbps, you way overbuilt that connection, but if one port breaks, you don't need to fix it until another port breaks; you can wait until three out of four ports are broken if you feel lucky. That delays hardware replacement and eliminates service trips and probably reduces technician time. If that connection peaks at 35 Gbps, you should fix any issues ASAP. (And start planning for capacity increases).
Of course, if the connection is to another network, they may bill for usage too; although I'm pretty sure US dominant ISPs are not paying for usage on most of their connections.
When a link reaches 50% it's time to upgrade. Heavier utilization and faster traffic growth means shorter intervals between capital expenditures. $10 per 50GB is a monopoly windfall but it does theoretically hold down demand. People don't BitTorrent 24/7 or leave video chat running while they sleep if there's punitive overage fees.
Well just think about the backhaul or middle mile costs - the more the network is utilized, the higher the middle mile cost commitments will be. There’s a direct relationship there, assuming your oversubscription targets stay basically constant.
> Imagine how good our network technology policy would be if the Senate weren't run by septagenarians and octogenarian.
The ages of the senators proposing this are 76 (Joe Manchin), 76 (Angus King), 65 (Rob Portman), and 56 (Michael Bennet). 3 of the 4 are older than the median Senator age.
None of the 4 are in the youngest 25%.
Image how good our opinions of the Senate would be if we based them in facts.
Nearly every single bill in Congress is introduced by non-chairmen, so this is an odd misdirection. Bills in both the entire session and in each subcommittee are voted on by members.
And again, this demonstrates that mocking old people as somehow the problem is factually unsupportable by the facts.
To be fair, it is not the senators themselves who come up with policy proposals. Their staff do most of the leg work, and the staff I have personally met were all pretty young and seemed pretty comfortable with technology (in fairness, I was meeting them specifically to talk about a technology-related bill and they were probably chosen for their expertise).
As for data caps, those are basically unnecessary in a 5g world and were never necessary for wireline service. The necessary of caps prior to 5g is debatable. Data caps were always just an excuse for ISPs to charge more.
Are millennials any different? Outside tech, I don’t know anybody who cares about broadband as an aside, except when pushing a different political angle such as inequality.
Yeah I think in many cases it’s taken for granted by the youngest generation. My youngest nephews have never known a time when you were unable to see live video of the person on the other end of a phone call. My cell phone (on cell net) just clocked 82/7Mbps. I still remember getting my first cable modem at a whopping 1Mbps nominal dl. XD
I genuinely wonder about the effects of later generations growing up in these conditions. It feels ripe for learned helplessness. "Well it's just always been bad" is a common, if ineffective response.
People who have terrible internet care quite a bit, especially if they use it a lot. Many, many people I know living in rural areas of the US get 3-6mbps max. The only hardwired option is DSL, since cable companies often don’t service these areas. Everyone who uses internet a lot for things like streaming is very frustrated by the situation, and that’s often younger generations.
Data caps should be simply illegal on anti-trust grounds for incumbent ISPs most which also own video / TV businesses. You don't even need to invent any new laws for that, just make competition law actually functional and not toothless.
There is one reason for some caps, which is that you'll always get oversubscription in the cities, and the caps could reflect that to prevent 100% utilisation. They don't have to be so small that people actually think about them. Basically set it at 1TB for residential for the lowest plan and forget about it.
I don't think you realize how fast 1tb of data goes when you have a family of 5 all doing zoom calls and remote learning all day, then streaming video (youtube, live tv, netflix) at night.
That's why I mentioned lowest plan. Sure, bump it up for higher levels. But if you want 100h of 4k content a month... just pay for the higher quality to subsidize network development which can handle it?
This is totally out of the minimum requirement area.
The 4K content is probably coming from a rack in the local ISP's DC or nearby colo in a peering agreement with the video streaming operator, and barely costing the ISP anything.
The place where ISPs skimp is between the home and the DC where the monopoly is.
Yep. I used like 2TB last month. And that‘s with 2 people in the household. Granted, I‘m certainly a fringe user. But with 4K streaming, YouTube/Twitch, PS5 downloads, Steam, server stuff, usenet, video conferencing, WFM, etc. it kind of stacks up.
Lets try and suss out some of the multiple issues with US internet:
-regulating ISPs as a free market is, to be hyperbolic, a disaster. there is effectively a monopoly in most parts of the country.
-25/3 Mbps is a very reasonable speed for most people. the issue is that its provided on a best effort basis that is never maintained for long, if at all. netflix says thats enough to stream UHD[0]
-FTA: "Those data points likely undercount the number of unserved Americans because the FCC lets ISPs count an entire census block as served even if it can serve just one home in the block" hilarious - what else is there to say.
-FTA: "The senators are also frustrated by differing standards across agencies. [...] the FCC defines [...] Alternatively, the US Department of Agriculture (USDA) defines" why is this the federal government's job again? the internet is clearly not an enumerated power. is this the interstate commerce clause that keeps on giving in action again? i think decentralizing federal power would help alleviate a lot of the existential dread vis a vi corruption that many americans are feeling.
all in all, this reeks of regulatory capture to me. if we're gonna have a serious discussion about internet infrastructure, i think the starting point is how we classify ISPs and where in the city-state-federal hierarchy the responsibility lies. changing a 25 to a 100 is a cool bullet on a resume, but i dont think this is the low handing fruit here.
> 25/3 Mbps is a very reasonable speed for most people. the issue is that its provided on a best effort basis that is never maintained for long, if at all. netflix says thats enough to stream UHD[0]
Please expand upon your point. Why exactly is 25/3 reasonable for most people? I'm not necessarily disagreeing; indeed, this would be a god-level speed compared to what's currently standard for most of Latin America. However, your point about Netflix is irrelevant because not every use case can be buffered. Also, many use cases in our new work-from-home reality require better up speed than 3Mbps with bad ping and high jitter.
> your point about Netflix is irrelevant because not every use case can be buffered
nothing special about netflix, just a service that most people use. even their highest quality offering is streamable if we actually got advertised speeds. just trying to say that 25 isnt a meaningful cap for most people.
> require better up speed than 3Mbps with bad ping and high jitter
rejectfinite makes this point downthread and i agree with both of you. i think theres a lot more bang for your buck to be had by reducing quality problems and widening the upload pipe.
100Mbps download so that multiple people can simultaneously stream 4k video is not a sane default.
> 100Mbps download so that multiple people can simultaneously stream 4k video is not a sane default.
Remote work has become very much A Thing since the COVID lockdown. Solid upload/download bandwidth isn't about streaming multiple 4K movies. It's about productivity where a sketchy 3Mbps connection is a hurdle.
> 100Mbps download so that multiple people can simultaneously stream 4k video is not a sane default.
It is if you don't want to update the minimum every 5 years. 100 isn't even high enough, IMO. It should be at least 500 down and at least 100 up to be safe for another decade, maybe 2.
If the pandemic has shown most of the world anything, it's that people need internet with high speeds. With the increasing gap between rich and poor, there are more people in flat shares too (per flat and total flat shares). Houses are bought later, if at all.
> -25/3 Mbps is a very reasonable speed for most people. the issue is that its provided on a best effort basis that is never maintained for long, if at all. netflix says thats enough to stream UHD[0]
Even ignoring the "more than one person" issue, 3Mbps makes it very difficult to do things like file transfers and backups and accessing home data from elsewhere.
25/10 per person is a much better minimum bar than 25/3
> all in all, this reeks of regulatory capture to me.
Why? How does this favor entrenched players over newcomers?
The companies I see being disadvantaged by this are incumbent cable companies that need to update equipment to let more frequencies be shifted from download to upload. But that's not regulatory capture by ISPs and should have good effects. And this doesn't seem like some kind of incumbent ISP vs. incumbent ISP regulatory capture either.
> -regulating ISPs as a free market is, to be hyperbolic, a disaster. there is effectively a monopoly in most parts of the country.
> all in all, this reeks of regulatory capture to me. if we're gonna have a serious discussion about internet infrastructure, i think the starting point is how we classify ISPs and where in the city-state-federal hierarchy the responsibility lies. changing a 26 to a 100 is a cool bullet on a resume, but i dont think this is the low handing fruit here.
The distributed regulation is the biggest problem IMO. ISPs are generally granted explicit or effective monopolies by local governments. Local competition is difficult enough for a utility business, but city governments make it effectively impossible in many places. Then there's a few big players who are able to claim competition and get lenient regulation at the federal level.
> ISPs are generally granted explicit or effective monopolies by local governments
Granting ISPs explicit monopolies has been illegal since the 1992 cable act. Show me a franchise agreement that has an exclusivity provision. They’re public documents.
How do you explain that even Google Fiber (with near-unlimited capital) has failed to make any progress? If I remember correctly they were denied access to telephone poles that incumbents were using.
In fact the law requires access to telephone poles. In one state Google got into a fight with incumbents over a policy where Google’s installers would be allowed to move incumbent’s equipment already on the poles to make room for its own, instead of following an existing process for having the incumbents move their own equipment. That’s probably a good idea assuming indemnification is in place, but Google wasn’t being denied access to the poles.
On the other side of the scale, Google Fiber has had the carpet rolled out for it in many places, for example being given for example free leases to public property for installing fiber huts: https://techliberation.com/2012/08/07/what-google-fiber-says...
> Google received stunning regulatory concessions and incentives from local governments, including free access to virtually everything the city owns or controls: rights of way, central office space, power, interconnections with anchor institutions, marketing and direct mail, and office space for Google employees. City officials also expedited the permitting process and assigned staff specifically to help Google. One county even offered to allow Google to hang its wires on parts of utility poles – for free – that are usually off-limits to communications companies.
> But we believe Google Fiber’s most significant impact was to change the nature of relations between infrastructure providers and local authorities. Even after substantial deregulation in the 1980’s and 1990’s, and even as separate networks and technologies converged on a single internet-based standard, local governments continued to treat network providers as quasi-governmental public utilities, regulating their construction efforts and access to public rights of way with cumbersome procedures developed decades earlier.
> Google Fiber has started with three neighborhoods in Louisville--Portland, Newburg, and Strathmoor (better known as the Highlands, where all the young professionals hang out). Residents of those neighborhoods can start ordering the service today by going to google.com/fiber/louisville. There are two packages to choose from:
While focusing on higher income neighborhoods where more people are likely to sign up for service seems like a no brainer in fact it’s an explosive political in most cities. That was the exact reason Verizon wasn’t allowed to deploy FiOS to Baltimore, and was a major reason for delays in FiOS deployment for NYC.
Looking at the details of how Google approached Fiber in various cities is a roadmap for what’s holding back deployment across the country. The problem was not incumbent lobbying—cities like Baltimore were falling over themselves to get Google to build Fiber. But what Google demanded in every fiber city was the sort of thing it got in Kansas City: elimination of the red tape that drives up costs. It’s no surprise that Google ended up mainly building Fiber in a bunch of red state cities.
>-25/3 Mbps is a very reasonable speed for most people. the issue is that its provided on a best effort basis that is never maintained for long, if at all. netflix says thats enough to stream UHD[0]
Yes, for one person.
Having lived with a family fo 4 using ADSL, it was not fun when 3 people watched Netflix on their own screens, and having to update a game.
Or with the pandemic and WFH, you have maybe 4 people using DSL for Teams/Skype/Zoom meetings and trying to work/study at the same time.
My backup broadband is 25/3; speed tests come in at 35/3. On three weekdays in 2020, there was a major outage on my primary (gigabit) connection and our home network failed over to the backup. This is during COVID, with three family members on video calls simultaneously. We barely noticed each time. A solid, actual 3mbit up is perfectly adequate for home use if you don't go overboard.
> Having lived with a family fo 4 using ADSL, it was not fun when 3 people watched Netflix on their own screens, and having to update a game.
streaming netflix in HD instead of UHD is 5mbps. 5x3=15, leaving 10 free to download a game. pretty reasonable. the pain was likely that you didnt have a real 25meg pipe, or that there was latency or jitter. a true 25Mbps would have been enough.
> the pandemic and WFH
I think you could make a strong argument that 3meg up is no longer sufficient. we live in an era where duo, facetime, and conference call programs and ubiquitous.
i do not, however, think you could make the argument that changing the download Mbps target from 25 to 100 is the best way to spend billions of dollars in a market that no one would accuse of being saturated and highly competitive.
>-25/3 Mbps is a very reasonable speed for most people.
Except most household have more than 1 person. A family of four sharing 100Mbps isn't a lot.
I am not sure if 100Mbps upload is a realistic option in US. But I would at least expect 50Mbps to be minimum per household. DOSCIS 4.0 should be able to handle that.
What's the upload speed in this example? Symmetric 100Mb would be more than fine for 4 people. And just fine with 100/25. However, 100/5 will be stressful.
Personally I think getting 100Mbps defined as broadband upload is a little too harsh. Especially when US infrastructure are so far behind.
DOCSIS 4.0 was ( finally ) finalised last year. And could be pushed for upgrading. But that is just the last mile, Cable companies will still have to ensure fibre roll out. Look at UK. They had a half decent infrastructure with DSL and could have pushed G.Fast, instead they just bite the bullet and pushed FTTP to maximum.
I'm in the US and currently get 200/40 with DOCSIS 3. These speeds have been available for around 10 years here.
Although technically I was only upgraded from 100 to 200 for free a few years ago because a competitor came in since then with fiber who offers 200/200 at the same price. I haven't switched over because they cut me a deal and truthfully I don't upload enough for the extra amount to matter and I get single digit ping times out of my provider which I'm happy with.
25mbps down is "ok" for most tasks, but 3mbps up isn't sufficient for WFH. For interactive collaboration you want lowest latency possible and enough upload bandwidth to deliver crisp 720p footage, for eg. presentations, whiteboarding, webcam.
I have 50/20 in Australia on VDSL2, and it is just barely good enough for this. You're right about it not being consistent/maintained though. Some days I could collab remotely no worries, other days it was simply impossible.
I feel that ISPs are far from a free market. The regulation as it is works to ensure a monopoly or duopoly. I'm not saying a true free market is a solution, but it would be good if we stopped letting them pretend it's a "free market". It's also pretty funny how 100Mbps is presented as some laudable goal. The technology to give everyone 1Gbps+ is not new and any well-designed modern infrastructure will easily allow such speeds. 100Mbps as the goal tacitly admits that there is no prospect of actually developing well-designed and efficient infrastructure.
> 25/3 Mbps is a very reasonable speed for most people.
No, 3 megabits upload is not reasonable. If people have to work from home or school from home the 3 megabits upload quickly becomes too little for a google meet/teams video chat or screensharing.
It becomes even worse if multiple family members have to do it at the same time.
Elon himself has stated that Starlink is not intended to and should not be seen as a competitor to traditional terrestrial ISPs. It’s meant to supplement people in rural and possibly exurban areas, but it will not be breaking the monopolies that ATT, Comcast, etc hold over urban areas.
> So will Starlink be a good option for anyone in the United States? Not necessarily. Musk said there will be plenty of bandwidth in areas with low population densities and that there will be some customers in big cities. But he cautioned against expecting that everyone in a big city would be able to use Starlink.
...
> "I want to be clear, it's not like Starlink is some huge threat to telcos. I want to be super clear it is not," Musk said. "In fact, it will be helpful to telcos because Starlink will serve the hardest-to-serve customers that telcos otherwise have trouble doing with landlines or even with... cell towers."
> Starlink will likely serve the "3 or 4 percent hardest-to-reach customers for telcos" and "people who simply have no connectivity right now, or the connectivity is really bad," Musk said. "So I think it will be actually helpful and take a significant load off the traditional telcos."
I agree that Starlink will be a game changer for many people, but I just don't see how they can be expected to replace traditional terrestrial ISPs in major population centers.
Beaming internet from space solves a ton of problems for anyone out at the ends of rural copper, or using p2p systems, or where there simply isn't another alternative.
But using myself as an example, I live in a part of NYC where there is fiber internet in the street in front of my building - and my landlord has zero obligation to allow the 60-ish apartments in the building to access it. We're stuck on legacy cable, which has abysmal upload speeds and congestion issues.
The dense urban or suburban environments theoretically have more options, but thats not always reality... and I'm not sure that tens of millions of people replacing their local ISP with satellite is desirable, or even technically viable. New provider options are important, but making sure access to the existing options is equitable and reasonable (across all the various local governments, providers, middlemen, etc) is important too.
Not to mention there’s only so much rooftop space for satellite dishes. Dishes, especially the ones pointing up, tend to require clearing maintenance in northern latitudes. It doesn’t seem very practical for areas where people are stacked on top of each other.
What I hope to see from Starlink is some freedom for more knowledge workers to work remotely, as in more rural communities.
I live in Poland and I am BAFFLED at how bad internet in US, supposedly one of the most developed countries on Earth, is.
I have commercial (ie high priority), 600Mbit/120Mbit broadband with extremely stable (under 15ms, over 99% of th time) round trip to most of the Poland, for equivalent of 20 USD per month.
I know its not the same everywhere in Poland, but consider I also have three mobile internet plans. One of them is 4G limited to 20Mbit but unlimited monthly transfer, for 5 USD per month. This is also quite stable and allows me to comfortably work remotely from almost every place in Poland and serves me as a backup link.
(1) Poland is a LOT smaller than the United States. It's not much bigger than the state of Wyoming, yet it has more than 60x the population. That makes building infrastructure a lot cheaper and more profitable.
(2) The US had the Internet FIRST. That's not an advantage when it comes to infrastructure. For example, England had roads hundreds of years before the US did, so roads in the United States are quite a bit better. Rome had running water (of a sort) a long time before most anywhere else but I'll bet its water situation today is quite a bit worse on average than most of the United States.
1. I am tired of hearing this rehashed over and over again. Yes, overall density in US is lower than most countries in Europe, but even high density places like New Jersey still have plenty of spots with only one, overpriced, shitty provider. For example, I am paying $98/mo for 400/40mbps cable, and that's my only wired option. My dad's country side house at the foothills of the Romanian Carpathian mountains gets 900/450mbps FTTH service for $10/mo.
2. Back in the late 90s, both Romania and my NJ neighborhood were at the same point, twisted copper phone lines and cable TV service over coax cable. Yet 20+ years later, the differences are stark.
Don't get me started on cell phone service :)
That explanation is even more strange if you figure out nothing of the infrastructure from 20 years ago still exists that could explain the difference. Every single active or passive network element must have been changed when technology evolved, likely multiple times.
I have a better explanation, and it is just greedy companies that invest absolute minimum into infrastructure, oversubscribe their internet access and pioneered worst possible customer service just because the customers have no other option.
There are parts of the US that have similar or higher population density to poland or other european countries. There are no reason they cant get as good internet. Not all ISP's have to cover the entire country.
Tangential, but I remember a tour guide at the Colosseum saying they could flood the floor of the arena for “naval” shows in less time than it would take you to flush the toilet twice in many apartments in the city today. Unsure how accurate that was, but impressive description.
it’s not baffling. we decided to glamorize billionaires rather than the amazing country we could have had if we’d stopped at millionaires as the height of (economic) status and reasonable wealth accumulation. a billionaire simply cannot allocate capital as efficiently as a thousand millionaires, so we choke on our own collective greed.
I literally haven't been to the doctor in four years, because the last time I went in for an X-ray they charged me immediately, and then later sent a bill for some more charges on the same X-ray.
My insurance and medical provider are the same entity, which I thought was supposed to avoid this exact situation.
Now I just avoid any medical treatment unless it is completely unavoidable, because I have absolutely no trust that the costs won't just be suddenly upgraded afterwords.
And that’s with insurance. Imagine how it is without. Though I think most Americans could learn a lot about the importance of taking care of themselves better by going without insurance for a while. It tends to drive a healthier lifestyle IME.
How bad is it really? I get that there are rural and outlier areas but the average bandwidth is higher in the US than Poland. Maybe I just find it hard to relate, but I and most of the heavily populated areas of the North East have had symmetrical FTTH for 15 years now. We have Gigabit up and down for about $90 USD a month.
If you drive out of a populated area across the US, nearly all of it is very sparsely populated, and therefore has terrible internet. 18% of Americans live in these areas, and that’s 54 million people. That’s a lot of frustrated individuals! Even in more populated rural areas like Ohio, you might get cable (but not fiber) to the nearest village, but even a half mile out of town, it’s only DSL (6mbps down on a good day).
Average bandwidth isn't a good metric of internet access.
A community that has 1 home with 1Gb/s and 10 homes with 5Mb/s would have an average speed of about (1000 + (10*5))/11 = 95 Mb/s, which is accurate but misleading
An interesting parallel is how much water/sewer, electricity, and natural gas connections cost. Cities have been providing these for centuries now and it's generally about $30-$50 base cost per month to connect to the utilities. This is for physical infrastructure requiring continual maintenance and treatment plants and movement of physical goods.
Internet service is almost static once installed; fiber lasts for decades (barring visits from backhoes) and the networking equipment can be fairly centralized and doesn't cost a lot to keep the lights on. It's insane that ISPs charge the rates that they do. Multi-mode SFPs are $50, switches are low thousands for enterprise grade, a couple long range 100G SFPs are another thousand. $50/month * 48 ports in a neighborhood pays that off pretty fast.
Fiber can't be any harder to install than fluid-filled pipes or electrical cables.
Hi from the developed world.. just checked, the slowest available broadband here is 200Mpbs/20Mpbs. Default is 200/200, I and most people I know went with 1000/1000Mpbs.
Thanks for the response! I'm quite jealous. The lowest I've seen for symmetrical 1Gbps here is ~$65, and that's rare.
Any chance I could ask you a few questions about internet access in Denmark? I'm working towards some local improvements here in the U.S. and curious to get some more data points.
If you email me at hnPotterTheOtter ~ gmail it will go to my normal email address.
Also from DK. I have a 1000/1000 connection for approximately 20$ per month
There was talks about upgrading to 10gbit/10gbit for 30$, since the infrastructure can deliver that, but since consumer hardware is still stuck at 1gbit it was postponed until there's actually a market for it
Thanks for the response! I'm quite jealous. The lowest I've seen in the U.S. for symmetrical 1Gbps is ~$65, and that's rare. Even more rare is 10Gbps, and the cheapest I've seen is $250. The best I can get is 940/35, for $120/month :(
Any chance I could ask you a few questions about internet access in Denmark? I'm working towards some local improvements here in the U.S. and curious to get some more data points.
If you email me at hnPotterTheOtter ~ gmail it will go to my normal email address.
I hope you realize that the US is not uniform as a single entity, because I live here, and I can also say "hi from the developed world", given I have at least 2 providers here willing to give me symmetrical 1000Mbps up/down for $50-70 with no data caps (and I have been using one of them for the past 2 years and couldn't be happier). But this is far from being the case everywhere in the US.
It's condescending but there's truth to it. Internet speed in the US is a symptom of a larger issue that the country has with incentives towards good governance, which is more often seen in poorer countries.
I've had a symmetrical gigabit connection for the past year and a half and there's no way I'd go back to anything slower. The peace of mind from not having to worry about something in the background slowing down your connection and the ability to download digital games faster than the physical copies could be installed from the disc is just too valuable.
Improving internet for America outside of the cities means investing in the infrastructure to upgrade it. It also means outlawing the data cap scam and banning ISPs from overselling their bandwidth because high speed internet is useless if you can't actually use it. Its the same principle as banning junk health insurance plans that are useless when you actually need medical care: fraud should be illegal even if it "saves some people some money".
It looks that way to me (40 LEI ~= $9.81 USD). Personally I'm more impressed that they advertise a 200/100 Mbps minimum speed. I'm not sure I've ever seen a minimum speed listed by a US residential ISP. If you're very lucky they'll tell you the average speed to expect and not a theoretical maximum.
They also didn't say anything about data caps, though there could be something in the contracts. I don't read Romanian myself and the PDFs of the contracts are time-consuming to manually reformat and feed into Google Translate. (Text selection goes across the page, not down each column, so it interleaves the columns.) But those minimum speeds—which BTW are guaranteed in the contracts—and no data caps would imply an oversubscription rate of only 4-5x.
One question to ask would be whether they are getting a subsidy to provide service at those rates. The subscriber might not be paying the full cost, or at least not up front.
No data caps. No subsidy either. Or who knows, maybe sometimes in some remote villages (?)
Main reasons for such a good network and low prices: low income and piracy, which fueled a giant "duct-taped" Lan infrastructure some 20 years ago, with ISP's at every corner, which were then bought because the infrastructure was already there.
See the link in my other comment here for
a better explanation.
Nice to see. I would have hoped to see more of a mention around latency. A lot of tools are sensitive to this and a theoretically fat pipe is not useful if you have 100ms+ latencies.
Latency is the real killer. I used HughesNet satellite internet while travelling for a while and it has a 650ms round trip. Even sites built by talented and well compensated engineers would sometimes be completely unusable. For example, YoutubeTV used to download the video data in tiny little sub 1MB chunks, each from a seemingly separate CDN endpoint. The result was that each chunk needed a DNS roundtrip and then another one or two for SSL handshaking so it would take at a minimum 1.5 - 2 seconds to even start downloading the data. Needless to say, it was not usable.
Moral of the story - latency and roundtrips are the devil and some users have it extra bad, but if you make it work for them, it's going to be amazing for everyone else too.
So, it might be that the senators aren't pushing that because the FCC has already been on the ball for latency, and pushing to keep latencies down around 30-40ms
No 1Gbps should be the standard for wired and 300mbps for wireless and satellite. We’re not exactly leaping into the future with this effort — we are barely catching up. If the US wants to continue to lead the world we need high speed internet.
It seems to have become less common for providers to advertise upload speeds. I was helping a friend pick a plan from a non-comcast cable provider, and she had to call to ask about the upload speeds, as the information was not on the web.
It’s because they’re universally terrible, and don’t help sell the product.
If it’s slow uploading photos, support can sell them on upgrading to the next highest plan to ‘go faster’ - congrats your uplink speed is now 5mbs instead of 3mbs isn’t as impressive as ‘we upgraded you to 1gig!’.
The only folks who know to ask and care about upload speeds already know Comcast is terrible, and are going to try to figure any way to avoid them.
One great thing about higher upload is the ability to host your own content, be a tor node, release decentralized software platforms that can actually work.
Imagine if people could just buy a premade raspi that hosted their social media page, ddns, etc.
While we're at it, let's prevent ISPs from disallowing us to run a business on our home internet. Why should it matter to them how we use our bits? Sell us on a support plan and higher uptime, sure, but there's no reason we should be disallowed from private enterprise because we're not "paying for a business account".
Yep, ISP blacklist addresses as large enterprise businesses only, or Small/Medium Business only and do not permit residential internet installation even when the property is rezoned or has been converted to residential.
My ISP doesn’t prohibit doing business over the connection or hosting services for commercial gain, with the exception of reselling internet access - like using your broadband connection as a paid hotspot for anonymous customers is disallowed in ToS.
This would be 100 Mbps faster than the non-existent cable that stopped 4 miles from my dad’s home in Wisconsin. They went down a rural road 40 years ago hooking up houses who wanted and stopped when they reached a home that said no. They had wireless for a while but the company shut down. Starlink is his only real hope.
It's not hard to generate a large volume of data with today's digital cameras (video & still). Backing that up to the cloud at 10 Mbps takes forever. Oh you want to switch cloud backup providers now? It will be another year before your initial backup completes.
Then we lose ability to develop new services that requires uplink bandwidth. Even now, there are services that need such bandwidth like cloud storages, YouTube.
This is kind of the whole problem with the world right now in multiple areas. We have a delusion that we are in a post scarcity age. I watch modern youtube videos on my 14 year old thinkpad on 802.11b wifi without issue. It works fine, seriously.
Yeah, why would I try to do an online backup of my data in less than a day? Or transfer high res video files somewhere? Or (gasp) host something?
The current ISP setup is based on a consumer only model of the internet which benefits the big corps and stifles anyone trying to actually DO anything but consume.
AT&T fiber in San Jose was GPON which is a 2.5 G down/1.25G up, split to several households (where I was, it was sixish lots per pole, but some lots had several units, and I'm not sure if the fiber to the pole was split earlier). They sell 1g/1g service on that, and it's clearly oversubscribed heavier on upload than download. 2:1 down:up is way better than most DSL ratios, and I usually saw better up than down speeds.
Paying A$90 (~70 USD) for broadband based on HFC to freestanding house, 100Mbps (downstream) / 40Mbps (upstream), hit by outages from time to time as ISPs are simply NBN resellers, which in turn is a wrapper on top of Telstra's infrastructure. Will be keen to see how Starlink rollout goes down-under.
NBN has had an order of magnitude more outages in the last few years than ADSL2 ever had for me. Maybe it's limited to HFC, but I doubt it. One time I got an outage report for work being done, but the internet only went down when they left for the day and messed something up on their way out.
Isn't Telstra just an NBN reseller at this point? IIRC that was why they were planning to start pushing cellular home internet, as they own their cellular assets outright, whereas NBN bought the wireline assets and charges them per customer.
It's ok for video streaming and conference calls. And I guess that's the minimal threshold. But with more and more services and offerings in the cloud, especially iCloud that just happily backs-up data out of the box and OneDrive 1TB corporate plan it's not going to win any awards.
With the pandemic and the changes in work from home habits, if I was a small town mayor I'd be rolling out fiber right now like there's no tomorrow. Big upfront costs but it instantly boosts the value of every property covered and attracts a new demographic that's not tied to any specific geographic location. Even with folks doing part time work from home, the additional time spent in the town during work days means that commercial real-estate will appreciate as well.
Unfortunately, mayors haven't been seeing a lot of resources to work with this year. First priority will always be to stop the bleeding. And, of course, printing money isn't an option for them, so the opportunity will mostly be missed.
Would people saturate their connection for a long duration of time streaming 4K or whatever? I know for me I am what's called 'extremely online', but I use the web intermittently, and don't saturate my connection streaming NSFW content for 12 hour intervals. When I need it, my 100Mbps connection is there for me, when I need to download an ISO or whatever, but that's the only time I use the fast connection. The rest is all low-bandwidth activity (surfing twitter, checking email etc).
Australia has been bee-lining towards faster internet for over a decard. (Which was somewhat sabotaged but changing the "All FTTP" to FTTN leaving most suburbs with substantially worse internet)
I just got upgraded to 1000/50... Yes that's 1 Gigabit down and 50 megabits up.
This can actually break if your download requires too much Request headers (eg Range requests) it will choke on the upload when downloading at full speed.
Somewhat related, if you would like to query any location in the U.S. and see internet speeds, the government has a site for it. [1] I am using that to decide where to move. Use the pull down menu to change from coordinates to address. You can put in a street address.
It’s a point to note that without Google Fiber bashing in traditional telecom, we’d still be at 10Mbps upload/download being deemed “blazing fast internet”.
Keep that in mind the next time you pay your monthly bill for AT&T/Time Warner/Comcast/Verizon. They could’ve given you gigabit speeds the entire time.
This legislation is dead if it doesn't prevent local munis from kowtowing to incumbant ISPs and thwarting municipal fiber. That's what we need: a ton more competition from the municipalities in this space given that I think broadband should be a public good like the highway system is.
I am genuinely perplexed reading the comments here about how bad US internet is. Knock on wood, I have basically never felt bandwidth or latency constrained in practical terms once the dial-up days ended.
I forget what plan I pay for with my cable provider but I just tested to 200 Megabit down and 18 up. That sounds low compared to numbers people throw around here but in reality my wife and I are able to Zoom or stream at the same time, browse in parallel and basically never feel any impact of latency. Basically, everything we want to do in real time works in real time and anything large I want to download (eg: linux ISOs) takes just a few seconds. Whatever numbers we are throwing around for comparison just don't seem to be practical concerns.
We're in NYC. With COVID, my coworkers scattered all over the US - east, west and south. Nobody put in a particular effort to look for places based on connectivity quality and yet everyone is able to WFH fine. Everyone moved to either be close to family, to a warmer client, or to skiing and the mountains. And yet somehow everyone ended up with great connectivity that allows us to seamlessly run a large and successful firm.
It's just funny to hear people complain about how bad things are when in reality everything works mainly fine. Like - really - what are you actually unable to do? (I am genuinely curious)
My internet bill is $38 in NYC by the way. RCN cable.
One final thing - it's nonsensical for people to say "I have X connectivity and I am in a Central European village." Obviously... the us installed its infrastructure before you did and it works well enough so we're not going to tear it out and replace it for no reason. You didn't have connectivity until much later so of course you're currently using newer technology. At some point we'll have to upgrade ours and leap ahead of you again.
Exactly, this conversation is about how terrible internet is when you aren’t in the most densely populated urban areas. Yes, it’s mostly fine in cities and even ok inside some small towns. It’s usually terrible when you live outside of the small-town city limits and not on a state highway.
> and it works well enough so we’re not going to tear it out
Your 200Mbps connection is definitely not on original infrastructure. For the 50 million plus people living with DSL connections which get barely 6mbps download on a good day, they would not agree that the older infra we have “works fine”. I know many people with this story because I’m from a part of the country like that — drive across America, and you’ll see that nearly everywhere is very sparsely populated. That covers a lot of people! You might be able to experience this yourself eventually if you go on a road trip and stay at a hotel not in a large city.
I don’t think it’s nonsensical to suggest that the US could also be building out a nationwide fiber network, and I think we should be striving towards that.
Strange how "broadband" and "high speed internet" can have a different meaning in various parts of the country. For ex: a 10 mbps dsl line is considered "high speed" in New Hampshire.
The article addresses the issues of multiple agencies with multiple definitions of broadband, specifically mentioning that the Dept of Agriculture has a definition of 10/1.
I live in Portland, OR and pay $30 for 240mbps down, 6mbps up. It’s part of “Comcast communities”, which is something the apartment complex signed on to for everyone here. A bit cheaper than normal, I suspect.
As someone whose website (and most of the websites I visit) has a relatively small footprint, 100/100Mbps would be more than comfortable for me to host and surf on
According to that map download speeds are nearly twice as fast in the US as they are in South Korea? I don't know where their data is coming from but that sounds awfully suspect.
I find terms such as “broadband” and “high definition” that need to be updated every so often to be incredibly useless.
Numbers and units exist for a reason. If one wish to speak of 100 Mib/s, then one should simply do so. — that way, I, and everyone else, knows exactly what is going on.
I ran into this on a government contract that wanted videos of certain things submitted in, per specification, "high definition digital video" okay 1080/60 it is.
Oh, no that's too big for them to email to their antiquated consultants, turns out they want 15MiB max video clips.
Well to do that I had to step down to 720/30. Which they then complained wasn't clear enough for their uses. Three rounds of RFI later I'm back as square one, they've defined HD as 1080/60 and they say they'll figure out how to distribute the clips. Just annoying as all get out when they could have started with clear language (numbers) to begin with.
Yes they should, but what will happen is websites and apps will get fatter and sloppier until the end result is slow, laggy performance.
It's a bit like doubling your salary and increasing your spending by more than double. That seems to be how website and app development behaves.
Imagine if bandwidth increased, latency decreased (approaching theoritical minimums), and websites/apps were built with a strict (but appropriately flexible) network budget.
Perhaps the best catchphrase here is "weakest link".
Even with 500+ mbps up and down, and pihole blocking a lot of ads, and noscript blocking some analytics and other trackers, a lot of web experiences are underwhelming.
Also, the recent fascinating post about how someone sped up GTA 5 online illustrates how the weakest link can be so weak that it ruins everything.
I didn't vote you down, but your comment didn't really add a lot of value. It's like whenever expanding roads come up, people say "but this encourages more traffic!" True, but not really the point.
But this is HN, and a lot of us are responsible for the state of web browsing! (Plus a lot are junior and have not considered these performance topics.) Absolutely everyone's connection to the internet should be fast and unfiltered, but if the supply side doesn't behave responsibly, the gains will be lost.
(edit-added)
And let's pause a moment to consider actual bandwidth needs. viop, video calls, and 1080 streaming all need less than 10mbps.
Online gaming usually benefits more from low latency than from bandwidth.
Honestly, only downloading files (app and OS updates) really benefits from bandwidth above 10mbps. And to be fair, now that phone, app, and OS updates are 1GB+, there is some value in increasing bandwidth.
Most day to day problems come from unreliable networks, where latency spikes, or from heavily throttled upload speeds which effectively tie up the overall connection.
And finally, the issue is not in forcing providers to offer reasonable performance, but instead to undo the last 20 years of corporate-government cozying which has been allowed to increase dramatically in the US. This has resulted in effective monopolies with high prices and low performance - and no alternatives.
I think cable providers should be left out. They were already gifted many billions of dollars in the past for this same basic purpose but they pocketed the money. They can just use the free money they've already been given. Instead, the funding should go to existing munipical fiber ISPs to expand their coverage. Give money to people with a proven track record of doing what you want, not people with a proven track record of keeping the money for themselves.
I'm on municipal fiber and it's amazing.