Basically every telecom network is built as fiber to the node not fiber to the premises. The nasty truth of all communication and transportation systems is that arteries are cheap and capillaries are expensive.
If I am building roads, it is WAY cheaper to just build highways than it is to build the small roads to everyone’s homes. When doing construction for telecom, the payback calculations are based upon “passings” which basically says “if I lay this cable, how many potential subscribers will I get?” which is a function of addressable users and sales modeling. Comcast used to believe that in a new market, competing against att only, they could win more than 50% of customers over just by connecting them to the Comcast network.
Please always remember, telecom, and transportation, are not technology businesses, they are real estate businesses. Telecom’s real estate are exclusive operation licenses, the most important of which being wireless spectrum, and the second most important of which being the places where they can exclusively tear up the earth to lay cable.
Circling back on my point at the beginning, if we laid fiber to every home, we would not need to upgrade the network connection itself for a long, long time. As it stands now, we have a lot of work to do if we want real high speed options in the future.
More fiber to the home, less fiber to distribution nodes!!
So we had this in the Florida Keys with AT&T who is the service provider of our area. I live 2 miles out on a peninsula of an island named Summerland, anyway I bought my house and AT&T assured me that I could get DSL but fiber would not be an option. I knew it was copper all the way to my home so I figured it would be the case.
Anyways, I buy said house AT&T comes out and tried to get DSL working, and the lines have been so patched and spliced that they cannot even get 1mbps, so they basically tell me I am screwed that it would not be cost effective to run new copper down a 2 mile stretch and that my only option would be to pay (or band my neighbors together to pay -- All 2 of them) for them to string fiber the whole way. I think it would be something like 70k.
2 Months later Hurricane Irma hits and rips down every poll down our street, and AT&T is forced to restring the whole street due to the fact that they are legally required to provide phone service to every customer in their "exclusive" area or they loose said exclusivity. Fortunately for me they opted to string it with fiber, and provided fiber to the home.
My point is, I don't see how allowing these "exclusive" coverage monopolies does anything but harm the consumer. I imagine a community based provider like the one in the article would be met with a host of legal challenges here.
> My point is, I don't see how allowing these "exclusive" coverage monopolies does anything but harm the consumer. I imagine a community based provider like the one in the article would be met with a host of legal challenges here.
I think the problem that was trying to solve is to avoid the Comcast poles, AT&T poles, Verizon poles, RCN poles and probably others, all strung up with their own networks of cables and wires.
It’s precisely due to the fact att destroyed competitive telecom equipment and infrastructure at every turn for the last 150 years that we have exclusivity agreements. Att has threatened to break domestic telecommunications over and over again if they couldn’t have monopolies (although now Comcast exercising those same monopolies is hurting att).
Funny parts is we actually have that here, the next island over (Ramrod Key), is exclusively Xfinity/Comcast and I believe Key West is split between AT&T, Comcast and Verizon (but most of it is underground there). I am not sure what the upper islands are but I would assume a mix of at least AT&T and Comcast.
Without the exclusivity agreement, it sounds like you wouldn’t have been reconnected. A community provider probably wouldn’t allocate 70k in capital to bring 3 (potential) customers online.
So you’re benefitting by the subsidy of users in denser areas being forced into exclusivity.
Doesn't 5G solve that problem? Lay fiber to a bunch of 5G nodes and then you can service hundreds of customers who only need to install a wireless modem to get fiber speeds and latency.
Oh god, I can't think of many things worse(network-wise) than having wireless internet at home, no matter how fancy the technology behind it is.
I'm out in the countryside, with wireless 4g lte data through a mvno. It's actually pretty decent compared to my other option, which was 4mbps max DSL. Multiplayer games are out, but streaming video at a decent quality isn't bad at all, certainly better than satellite internet.
On the flip side, our local provider got a grant to update their network to fiber (and a threat to lose exclusivity if they didn't improve the network). Oddly enough, they'll be charging $200 for gigabit, and offering lower speeds for more reasonable prices, even though they're installing fiber directly to our home next spring.
It is kinda shitty, but then again, it's just another tradeoff to make when living out here. Personally, I wouldn't give it up for anything compared to living in the city.
It can work for light users who don't really care about speed if they get something decent, otherwise is a very bad idea, not to mention that it's basically impossible to avoid shades or guarantee any coverage at all.
The ISP I currently work for (in Spain) is playing with this idea for sparsely populated areas, where we didn't lay FTTH yet and our customers are connected through DSL. It can potentially save a lot of money on deployment but the current BTS locations are not enough, we'd have to build at least 2x for decent coverage, update their links and so on. I don't know, I think we will just lay FTTH in the end. Dealing with saturation on wireless connections is very difficult sometimes.
Except we already gave telecoms billions of dollars to do exactly that, the last mile connections. They did a little bit then took the money and ran.
I have fiber directly to my house here in the Philippines. It's pretty much becoming the standard offering here now, with mobile data as the low-entry-cost alternative. Lack of pre-existing infrastructure has some interesting side effects.
To distill this down to a memorable heuristic, the cost of building out a municipal network of any kind is proportional to the mean linear distance of the street frontage.
This has interesting implications for civic planning. It is very much cheaper to put dwellings right on the frontage line, with zero-setbacks to adjacent properties, and then have huge backyards, than it is to center each building on its lot. That relatively small distance between the street/conduit/pole/pipe and one building is multiplied thousands of times across a municipality. An extra meter of frontage on a property at the root-end of a street is counted multiple times when calculating distances to properties closer to the leaf-end of the street.
Things that seem relatively minor for individual properties, like residential street widths, setbacks, minimum lot sizes, and maximum building heights, these have a huge impact on the cost of building municipal service networks. Those zoning rules established for aesthetic reasons literally create millions of dollars of economic impact. You just can't have cheap fiber to the premises if your premises can't be closer than 20' to the property line.
We have a local ISP [0] that builds FTTB, usually with 1000BASE-T to the individual units. Sometimes, if laying STP cable isn't easy and the customers don't need anything more than VDSL can provide, they replace the (L3?) switch in the basement with a DSLAM, but I think those are quite a bit more expensive than a switch with 10G SPF uplink and 1000BASE-T downlink (possibly aggregating a few 10G uplinks for large buildings, I'd guess).
(Germans tend to deploy shielded cable, seems to have something to do with both unifying deployments between known noisy RF-environments/eliminating problems with noisy RF-environments up front and with (electrical?) codes that don't discriminate against shielded twisted pair cabling.)
If you pay for the additional setup, they're happy to provide 10G/2G or even 10G/10G for 199/399 EUR/month. That apparently requires you to provide singlemode wiring from the basement to the units, but to be honest, that's rarely hard to do, considering them being available with coatings that are allowed to be in most low-flammability zones like stairways, quite thin (so it's easy to hide the complete trunk behind a baseboard or along doorframes (if you can slightly round their corners to conform to the bending radius you'll need to adhere to for the many bends you'll have when snaking up a stairwell)), and non-conducting, so there's no isolation/ground fault/lightning protection to consider.
If my street would have gotten dug up just a little further when they did so a few months ago for subsurface electricity and more direct water lines (some buildings, including ours, had their water house mains come from a neighboring building through a meter between their intake meter and the other (e.g. our) building's main cut-off valve/initial fan-out piping.
Note, I would have literally dug into my meager savings and bankrolled/installed fiber conduits in accordance with what this ISP would accept/recommend, going from the houses (talking to the communal utility for sharing the wall feed-through, because at reasonable rates that's literally cheaper than buying the part/renting the drill to DIY at all the neighbour's houses that would take my offer of installing it for free) to a connector/hub box or just something that keeps the insides of these pipes clean from dirt and significant moisture ingress (maybe with a slight distribution to blow dried air through from my house's connection, and flow-regulating valves at the other buildings to extract any moisture that will diffuse through the walls, or something like that).
Once the street between there and the next conduit where the ISP can easily get fiber to (without digging) would get torn up (happens eventually), I'd make sure that more of this is placed, likely with some larger pipes to stuff further innerducts through for all the other residents on this street (low hundreds in units, about half or so single-unit buildings, half multi-unit buildings, but quite a few build in rows of a couple, so, with appropriate fireproofing, they could share a single house-to-street trench digging).
It's not worth it to dig the street up for this alone, but at least the part I'd have placed fiber isn't even paved, it's literally just compressed gravel/dirt mix, and will hopefully get a nice surface sooner or later. It get's rather dangerous to walk when frozen, and because it's not mine I can't use mineral products (salt/sand) to make it safe.
Yes, I also offered to fund a coating with crushed basalt, which would restrict maximum torque (the structural integrity/load bearing depends on a certain ratio of downwards vs. horizontal force), but eliminate most issues with water (at times there is no dry footpath that does not involve a >150cm leap from dry bank to dry bank (if you miss-judge, you will jump into a puddle and might slip in the mud)) and thus also some of the ice problems.
Sorry for the rant.
[0]: https://www.tal.de/produkte/fibertoyou/internet-anschluesse/ https://www.tal.de/produkte/fibertoyou/internet-anschluesse/
> Basically every telecom network is built as fiber to the node not fiber to the premises.
My town is quite literally stringing up fiber (and laying it in some cases) throughout the town limits over the next 90 days.
So unless you mean that most residents/businesses won't technically have a physical fiber-optic type cable dangling inside their abode, I don't understand what you're talking about.
With the capillary road network, don't most cities make developers of new blocks pay for the roads too? I think the difference although is developers are many small independent firms, while telecoms are large national corporations, and don't give ownership up once they are done developing.
With your post, it made me realize that telecom & power cable runs should really be owned by the resident city, like roads are.
The brilliant part is how quickly Comcast makes back the money spent on cables laid to apartment complexes and projects. They can wire up 20-1000 apartments with copper coax and distribution boxes for something like $50,000 and have the infra and labor paid off in under a year. After that first year, the business is essentially a money printer - no incentive to innovate or rewire neighborhoods when you have no competition.
Since the cable internet standards haven't required new copper (DOCSIS 1->2->3.0->3.1) in over a decade, Comcast can simply upgrade one or two $100 boxes per complex, raise their prices and provide "faster" service while spending nothing on actual R&D.
They can outsource customer support to Southeast Asia, standards to CableLabs, modems to nearly any company in China, and even piracy detection to random firms like IP-Echelon. Don't forget - Comcast owns an enormous number of film studios and production companies. If there's anything approaching the digital media panopticon, it's Comcast.
I don't know if there's a better-positioned software company in the ISP business.