Longmont, just south of Fort Collins, was the first city in Colorado to build out a municipal ISP, and it’s inspired surrounding cities to do the same. It’s where I live and it was a not-insignificant reason why I chose to move here. So far it has been a resounding success for the city. The buildout completed on time and adoption rates are higher than initially expected (the city planned for 37% but the last number I saw was around 54%). In my experience, the service has been so good as to be totally invisible. And I know my $50/mo rate will never rise. I wish Fort Collins the same experience. The fact that they both cities have municipal electric service will help this significantly.
In general I agree, but in practice certain nominal prices can be monotonically decreasing due to improved technology lowering prices faster than inflation raises them.
For instance, in terms of $/GB, data storage costs have been strictly decreasing in nominal terms even with the existence of inflation.
That said, I am not sure how much innovation is going into the costs of maintaining a fiber network, so it’s possible that it won’t decrease as fast as 2-3% a year on average. I would be really interested if someone had those numbers!
APM Marketplace [1] just had a researcher on the program the other day sharing that most Americans don’t use more than 20Mbps of bandwidth, and paying for anything above that as a typical internet user is throwing your money away.
Power users, businesses, researchers, and others might expect more, but not the majority of residential users. Muni fiber investments will last quite a while.
One reason for getting a faster link is that most broadband connections in the US are not symmetric.
Comcast in my area doesn't even make claims of upload speed anymore, but my recollection of when they used to is that it was about 10% of the download speed. If you want more than 2 Mbps upload, that 20 Mbps plan wouldn't cut it.
Part of the reason for asymmetric cable modem (DOCSIS) speeds is due to the coaxial medium being essentially half duplex. Operators split the RF spectrum into a lower frequency range for upstream, and a higher frequency range for downstream. Increasing the width of the upstream range would eat into the available downstream bandwidth, and also has other implications in terms of power, noise and overall network design.
There is a full duplex version of DOCSIS, but its adoption faces some challenges due to how operators have to re-architect their physical networks to support it.
There is also currently a push to extend the downstream spectrum up to higher frequencies, enabling both more downstream bandwidth and also the ability to move the upstream split higher, therefore increasing upstream bandwidth.
(Disclaimer: I work with a client who develops products in this space.)
This. Cox basically always provides an upload bandwidth of 10x less than the download (except their gigabit plan, which is 1000/30, which is bullshit). So I pay for 300/30, because I upload things. Man I wish I could get 1000/1000 where I live...
I don't disagree that asymmetric speeds are a problem, but I think there's this bubble where tech professionals think that once everyone has gigabit up and down, your average Joe is going to be demanding 10 gigabit and up, and the data isn't proving that out. People are happy to stream a couple of views of Netflix at once and that's it (even if you're thinking everyone is going to do live syncing of block volumes to the cloud; everyone might use iCloud for their Mac backups once Apple supports it natively for Time Machine, but even then deduplication and off peak transfers will do the heavy lifting).
Agreed with that, my connection is plenty fast. It would be nice to have a faster one, but not enough that I'm going to pay more for it.
I think I'm around 50 Mbps and it could be used much more heavily, but it doesn't make any sense on YouTube's end to send me a 50 megabit video stream. It's a the storage and upload costs on the provider side limiting that.
Even if you pay extra for Netflix's 4K support, that only needs 25 megabits. It's compressed to hell compared to a 4K bluray, but if we all had faster internet I don't know if that would change. It's lot harder to market the actual visual quality of a video compared to talking about how many pixels it has.
Maybe we'll see whole new services that need the bandwidth, but most stuff that sends a lot of data could be downloaded in advance if you don't mind waiting 5 minutes instead of having it instantly accessible. How many billion dollars are we willing to spend to build out faster networks to avoid that? What's the selling point for very fast streams, we go back to cloud mainframes and all of our devices are dumb terminals that stream our games and the whole OS from an AWS datacenter?
Why is asymmetric speeds a problem? I don't think there is anything inherently wrong with having multiple types of service for different use cases. That is true with other utilities (electric, water, gas) and is part of general infrastructure planning in those areas.
Of course the actual pricing, availability, etc are also important but I don't think there needs to be a one-size-fits-all model for Internet bandwidth.
I get that. I'm not saying that there aren't different use cases, I'm saying that it is reasonable that there should be different price points for the different use cases.
Even if the physical plant supported symmetric bandwidth, I would expect different pricing plans for different utilization patterns.
I purposely limit my usage because my connection goes to shit when it's over used. My family all understands this as well. If I had better connection id be consuming way more bandwidth.
And I can't even get a faster more pricy connection because my legacy contract gives me upload speeds I don't get till I add 50% to my bill
Exactly this. I work at a research and education network (a regional ISP), and we are asked to perform upgrades and new deployments constantly. If it is just residential use, the upgrade path will be more predictable. However, business and municipal users can scale quite unpredictably and the 10GE links allocated in 2019 might need to be 100GE in 2025.
It depends. If you're constantly doing new customer setup/teardown, yeah that'll cost man-hours. But once your network is up and running it really doesn't take much work to keep it running.
My ISP is a very small local outfit. They have a "few thousand" customers... and the whole thing is run by literally two dudes, doing everything from netops to hw installs to billing and support. Yeah I'll have a few hour outage once every few months, but overall the service is great and it really doesn't take that much effort to keep things ticking.
They don't have any written material, this is just from talking to the guys. Turns out starting an ISP isn't hard if you can lease last mile from one of your local big boys (and here, by law, they have to make last mile available for lease at a reasonable price).
Basically rent a quarter cab in a colo, get a router (or two), a switch (or two), and a server (or two). Negotiate with last mile provider and a pair of upstreams. Get your DC to x-connect you to your upstreams and last mile provider. Do some marketing (elevator ads in residential apt buildings are apparently basically gold). Buy a pallet of cable modems, spend a few weeks configuring your crap... $50k later, you have an ISP, and you can offer service at 50% of the big boys' rates while still raking in a substantial profit.
> data costs for bandwidth costs hasn't gone down at the same rate
For clarification, at the business level data has always been charged by capacity and not volume. And the costs have fallen at double digit rates since they've been measured.
Explicitly the reason for automating things, esp. since network connections don't require a ton of labor once provisioned.
Orchestration engines help to do the former (Ansible, Salt, Puppet, whatever), and they usually tie into monitoring systems like Nagios, ScienceLogic, or SolarWinds, and can trip, and then launch, remediation efforts automatically.
Isn't this article and the discussion precisely about the reason why costs for bandwidth haven't gone down? That is, the infrastructure providers are keeping the costs high because they can.
Well, Romania is known for cheap internet prices, but for reference, here the largest _private_ operator is offering 1Gbps for $9.33/mo, taxes included (19% VAT).
I kinda' doubt that $50/mo is really unsustainable, especially once the infrastructure is already built.
You can't really compare sticker prices here. US percapita GDP is about 6 times higher. The average salary is about 8 times higher. So as a proportion of those things, cost is fairly similar.
GDP doesn't matter here. Average salary might... presuming employees that maintain the system are paid average salary, AND the cost is overwhelmingly in the "salaries" category - which I sorta' doubt.
You don't have to take my word for it. There's recently been a push for expansion of Romania's broadband network to more rural areas. At an investment of about $110 million US, it will reach an expected 150,000 additional households, for a cost around $730 per household. Compare that to the US, where the estimate is closer to $3000 and as high as $8000 per household, a 7 to 11 fold increase over Romania. [0]
Nearly all aspects of infrastructure build out are heavily weighted towards labor costs. A 2009 rule in the US required large projects use union labor, and just that marginal increase in labor costs increased infrastructure costs by around 15%. Anecdotally, I had a home addition some years back. Materials costs about 20%, labor about 80%.
Also once built, the capital expenditures for something like broadband drop to negligible rates while the labor cost of maintenance & administration just keeps on going, maybe not as high as the initial build but you need new & different workers too-- sales reps, account managers, etc. Given that prices tend towards the marginal cost of production over time, in this case being mostly labor, it makes perfect sense that a country with 1/8 the labor costs could offer a service at 1/5 the cost in the US, the difference in those proportions accounting for the marginal capital costs of maintenance.
GDP per capita is a proxy for local price level, and it absolutely matters. Running a fiber service is overwhelmingly a matter of labor costs. Everyone from employees who deal with permitting and approvals to guys who run the cable to network operators to customer support representatives. The stuff that’s made in China—the fiber and the equipment, is a small part of those costs.
No, this is incorrect. GDP is a proxy for production in nominal terms. If it were just a proxy for local price level, then every place would have the same real income and would differ solely by relative prices. That the U.S. has a a GDP/capita that is 4x times some other nation does not mean that prices are 4x lower in the other nation.
GDP is a measure of production in nominal terms. It’s a rough proxy for labor prices (the kind of price I was talking about). That’s because a dominant component of GDP is personal incomes.
It's a rough proxy of production. Even assuming constant labor shares and working hours (which are not all constant), wages are higher in A than in B because workers produce more per hour in A than they do in B, and then there is a relative price effect. But it's not all a relative price effect, it's not a proxy for the relative price effect, and for large multiples, productivity dominates relative price effects for most occupations.
When you are asking "how much would it cost to do something in B than in A", workers in B might be paid 4x more than in A, but you may also need 1/4 of the workers to do the same job. So the issue isn't wages, it's output per unit work, primarily because of accumulated human and physical capital, as well as things like infrastructure, legal system, needing to pay bribes or efficiency wages, etc.
I don't have the 1Gbps connection, I took the cheaper 400Mbps one. But they're fairly good at actually offering the promised bandwidth. Again, internet is one of the very few top-notch things here, better than most of the world.
What is the effective speed you can get because of bandwidth throttling done by remote servers? Can you get close to 400Mbps to cloud storage for example? Currently I have 150Mbps and been wondering if increasing it would be worth the extra cost.
It really depends on the cloud storage provider, and once you have gigabit, you can tell who's being cheap with their CDN/local POP. Some downloads scream along and gigabit is worth it. It's ridiculous to me that local disk can be the bottleneck. (This is with a spinning disk. An SSD, less so.) Others downloads... don't.
My first reaction was to say that's wrong. Maybe it's just new, but last time I checked, a couple of years ago, there's no way to get a decent land line under 40€. The trick was that the "cuota de línea" is hidden from the listed price. Most of the land lines are owned by Movistar that rents them to Vodafone thieves and other ISPs that consider it an independent expense, as if it's not just a regular cost. Another possibility are time-limited offers.
Now it's my landlord who pays so I don't care, but still would like to know where did you find that price.
Thanks, I had Jazztel years ago and it worked well and they were cheap, but that's better. "Base imponible" suggests that VAT still must be applied though.
That's not how it works. Bandwidth is a shared resource and most people aren't using 1gpbs every second of the day. Profit for telecom providers isn't on a variable basis and therefore unit economics make much less sense.
with 3gb of data - maybe I'm a weird consumer, but I don't get it what's with this insanely low data caps... while in France I had a 100gb / month 4G data plan for ~20EUR, and couple times I even exceeded those 100 gb!
how/what are most people using internet for that they tolerate such small caps?!
My phone is always on WiFi unless I am in my car driving from place to place. I use under 2GB a month, closer to 1GB most months. I can't even comprehend how you would use 100GB of data on a mobile device in a month. My primary internet consumption is sitting at a computer all day long, or a laptop on the couch. The mobile data is just for traffic, slack, email, and news. All my podcasts are downloaded on WiFi, all my music is as well. There is just no reason for me to use any real amount of data while not on WiFi.
You live an urban lifestyle where you commute and travel by train and you don't stay in your suburban palace most of the time when you're not at work. Then you look at youtube videos during that time.
uhm, yeah, the sim card was in a mobile "pocket router" and shared to phone(s) + tablet + laptop, and I'm sometimes watching multiple streaming videos at the same time as a way of mixing music (Bach + heavy metal instrumental rocks :P) without bothering to turn video off or to low qual even if I'm not watching it
Unless you're consuming a lot of media, multiple GB is a lot. I have a super cheap plan with 500MB/month. It might sound impossible, but really all I have to do is turn off background mobile data for most apps, which had no business constantly sucking bytes anyway. Bonus, now my battery life is better.
If you have no WiFi access it's a somewhat different story, but WiFi is ubiquitous.
I know a few people in France who actually ditched their landline contract, they had full 4G coverage but were still at 8mbps with their landline. So now they're doing tethering for the complete house instead. The only thing they can't do is gaming because the ping is atrocious.
It's also cheaper, a 4G contract with a 100GB cap is about 15€ per month, whereas my FTTH connection at 900Mbps down / 400Mbps up cost me 40€ per month.
None. Those are private companies that do this. It's the same in Germany. I have a 7 euros/month unlimited calls cell plan but only 500 mb data, which is enough for me. I pay around 25 Euros month for Internet at home. It's fast enough for Steam/Streaming, I don't even know the speed because I stopped caring a while ago.
In Russia, 1gbps may be from $3 to $10 per month depending on the place, and it works as such at least throughout Russia maybe falling to 200-300mbps to far off places like the U.S.
In Lithuania, i paid 15 EUR a month for 500/500 and it worked perfectly too.
U.S. prices are super high - mostly a consequence of most people living in detached houses - long distances and high complexity of wiring. In Russia and Lithuania, most live in Soviet-style apartment blocks - high user density.
IP transit rates drop by like 30% per year[1], you have to assume transit is a large portion of expenses. Further, all the hardware an ISP needs get cheaper year after year. There's no reason prices should ever need to rise, apart from greed.
The $50/mo rate is for charter members who signed up within the first 3 months of it being available. New customers have to pay $70/mo, which drops to $60/mo after the first year. So not everyone is paying the $50/mo rate.
Not if they also fix the bandwidth allowance. Over time, I would expect the cost of provisioning a 1 Gbps link to drop faster than that $50 loses purchasing power.
Your $50/mo for 1 Gbps can therefore remain stable, and the ISP might someday offer 10 Gbps for $70/mo as an upgrade offer. Clearly, by those prices, as long as the per-customer costs are less than $47.78, that $50 more than covers the cost of the 1 Gbps bandwidth. Everyone who doesn't upgrade ends up subsidizing further build-out.
It isn't but the number of people who get that deal is limited. It was a limited time offer for people who signed up when the program was first getting started. I'm not worried about the funding level over time.
I grew up there, and just moved back. Is much nicer than 30 years ago, and legitimately nicer than boulder.
Fiber is great, and people forget how much of US infrastructure was communal in early electric buildouts as well.
Some of the objections seems to be reflexive anti-government sentiment. I haven't seen corporate services do well enough to convince me of the realistic improvements. And with most services negotiated on the national level (Netflix, apple, hbo, whatever) as direct buys, the last mile as a subsidy for cable negotiation seems wasteful.
There is almost no diversity in Colorado when compared with the large coastal cities. You may still be welcomed but don’t expect to see many Asians or Asian restaurants and stores.
You might want to check out the area between Boulder and Denver too. It's mostly never ending suburbia in a sense but at least you have both Boulder and Denver open to you for employment options if you live there. There's also some quaint places with smaller town feel like louisville.
Commuting from Longmont to Denver is possible but would be a lot of hours driving/on a bus. A rail link is planned but it is scheduled for something ridiculous like 2050 or sometime infinitely far in the future.
The Denver tech job scene is pretty strong. A lot of startups and a growing corporate outpost presence, as well as all of the "old tech" around the Tech Center.
Northern Colorado (Longmont, Ft. Collins) has traditionally been a center for hardware design and test equipment, but unfortunately most of these companies have shrunk overall and have reduced their Colorado presence.
I would say that in recent years the epicenter of startup software jobs in Colorado has shifted back from the brief Boulder startup run and moved into downtown Denver, while Boulder has become a center for corporate outposts (Google, Twitter, Amazon, Microsoft, Workday, NetApp, Splunk).
Agreed - I wish FoCo could attract more tech companies - it is a great city with highly educated people (esp with CSU here). But for now your best bet is remote work if you want to get paid a salary equal to the COL of FoCo (which is almost the same as Denver/Boulder). Sadly due to lack of competition in tech, the few tech companies here pay well below market because they can. You can literally move to Boulder/Den and get a 30-40% pay increase...
Theres still locations here that can't get fiber. The local hospital is owned by a church and they own apartments adjacent to the hospital. These apartments are surrounded by others that get fiber but the map still shows they aren't planned.
If you can't get municipal fiber where you live in Longmont (I'm a resident; hi!), you should most likely be complaining to your landlord, not the city. If you're not sure, call up Longmont Power & Communications and ask them what's up.
There are a few apartment complexes around Longmont that have resisted integration with the city's fiber network. My understanding is that the city has negotiated with each of them over the installation of fiber on their premises and that some of those negotiations didn't succeed, e.g. the apartment complex owners wanted to own and control the fiber that the city would pay to install.
But this rings true. My building has some sort of an arrangement with Comcast. It's only Comcast and AT&T that are there right now. They pretend to work with other ISPs I bring in but usually cite "something something safety" and prevent them from setting it up. I have given up for now since Comcast has got the price down to $50 for a 75Mbps line for me.
Point is, while I don't know how and can't confirm, I have a strong feeling that Comcast pays off landlords somehow.
If there is some geographical last mile type issue to bring internet to these buildings that are surrounded by fiber, it could be helpful to pitch the church/hospital with a wireless link from one of the fiber pops. I have brought service to medical and hospital buildings in this way before and it's always worked out well.
In the case described above, the topography is "completely flat", and the internet provider in question already provides electrical service to the premises.
That should mean they have public utility status, access to public poles etc. It would simplify such a solution. I've deployed such links in a flat scenario as well.
Came here to dispute that. Your rate may never rise but deficits will come out of taxes.
Being a public service the budget is publicly available.
For 2018, I count $11,930,874 in expenses and $12,420,323 in revenue (literally 99.9% from $50/mo service charge). That's a 3.9% margin. At that margin, I don't see how rates will never rise. The expenses will go up with inflation. If the revenue remains constant ...
That said, it wasn't readily obvious so I didn't dig deeply to see if some 2018 expenses were unusual one-time charges, or anything like that.
In Colorado here as well, but unfortunately unincorporated Jeffco. Really hope this begins to spread through Colorado, since right now our only option is Comcast.
If it does well for Fort Collins and Longmont, maybe other areas will try it as well.
