Hacker News Reader

Comment by _wldu | Hacker News Reader

I live in a rural college town on the East Coast USA. I paid Verizon 90 USD per month for 1.5Mbit DSL and a home phone number.

I had this phone service for about 12 years. It was horrible. When the phone rang, the DSL would stop working. When I ended calls, the DSL would stop working. There was so much static on the line I could barely talk/hear others. But it grew worse and started impacting my ability to work from home after COVID.

Verizon said they could "roll a truck" to my house in about two months if I agreed to pay several hundred dollars upfront. I told them no. I had had enough. I ported my phone number to Consumer Cellular and use 4G for Internet access now. It's faster, $30 cheaper each month and much more reliable although I do limit my use as it has a 35 GB per month cap and "may" slow down if I go over that. I wonder, would it slow down more than Verizon's 1.5Mbit?

I'm still Waiting on Elon and Starlink to save us.

View on HN

>I'm still Waiting on Elon and Starlink to save us.

What leads you to believe Elon's internet will, in any way, be pro consumer? Tesla certainly isn't, and their customer service is not good.

We went from .4Mb download, with a 7 second ping, to ~100Mb and ~20ms ping when we changed from Telus (Alberta, Canada) to Starlink.

Maybe Starlink won't be perfect, but the bar is lower than you can even imagine.

Competition is always good and will force other providers to up their game.

Starlink, unfortunately, will not present meaningful competition. It will really only be viable in sparsely populated areas--even once the full constellation is deployed, there simply won't be the bandwidth for large numbers of users in densely populated areas like major cities--and both Starlink spokespeople and Musk himself have not been particularly coy about that fact...though I suppose neither have they tried particularly hard to push back against the prevailing tech-media narrative that once Starlink is fully armed and operational it will have Comcast, Spectrum, Cox, et al. quaking in their boots.

I think Starlink has a chance to raise the floor in rural areas.

Telcos (usually) do an OK job when there's a competing cable company, and vice versa, but when it's just one offering, there's often low reliability, low speed or both. This is most often in rural areas, where Starlink should be viable.

Some urban areas do fall into the same single provider, low quality offering, and Starlink won't be viable there.

Starlink will be good for people in rural areas, but it won’t positively change the existing economics of terrestrial internet service.

Frontier doesn’t offer slow DSL to rural areas because they’re rolling in cash from having no competitors. And when they lose customers to Starlink, it’s going to make their subscriber revenue to infrastructure expenditure ratio even worse.

If I'm Frontier, why spend on infra to provide a better product to people in areas with no competitors? That's just money spent with no change in revenue.

Now if there's a viable competitor, it's a choice between spending the money to keep customers or not spending the money and losing the customers over time. Terrestrial networking can easily beat Starlink on bandwidth, availability, latency, jitter, etc, all the quality metrics (except portability), but only if the terrestrial network makes the effort.

Telcos have the ability to quickly roll out new networks when they're inspired. AT&T ran fiber to the home (GPON) in several markets within about 18 months when Google Fiber announced they were entering those markets (which they then decided not to). Telcos have an army of people (many consultants) who can run outdoor wiring and do home installation, they have easy access to poles, and (mostly) a decent database of addresses and line lengths to get multifiber lines factory manufactured to roll out with minimal field splicing.

And that's the hard way; things like better remote terminal equipment and/or backhaul don't require a lot of capital or labor individually (but may in aggregate), or just more adaptive DSL speed profiles just developer labor and configuration pushes. I could probably get 50% more downstream speed and 2-3x my upstream speed, but my ISP has inflexible speed profiles; that's probably not the issue for people on 1.5 M dsl, but it's got to be an issue for some.

If you're not familiar with Frontier -- they currently aren't profitable and recently filed bankruptcy. Their network consists mainly of the old rural systems that the big telecoms didn't want due to a lack of profitability. The big and profitable telecoms have concentrated their service to denser areas, and pulled out of lesser profitable rural areas entirely, even though they had a monopoly in those areas.

Their bad service is not the result of a lack of competition, it's a lack of a profitable business model. It is simply more expensive to connect 1 customer with a mile of wire than it is to connect 100 or 1000 customers with a mile of wire. I don't think there's any way to solve this problem without subsidization of rural internet.

> Frontier doesn’t offer slow DSL to rural areas because they’re rolling in cash from having no competitors.

They actually do. The reason they’re going bankrupt is because competitors have started showing up due to the Connect America Fund causing them to lose customers in droves.

Simultaneously they took funds themselves and didn’t bother using them to upgrade infrastructure.

Source: previous frontier customer who couldn’t leave fast enough. Also convinced every single neighbor to do the same, even the ones who thought it was “fine” and wanted to be “loyal”.

Rural broadband (non-satellite) has always required more infrastructure per subscriber. More revenue, like through the subsidies you mention, are the only way to fix the problem. The lack of competition in rural areas is more of a symptom of the inherent infrastructure problem, rather than the cause.

The Connect America Fund is also known as the "universal service high-cost program"

Isn't the point to target sparsely populated areas? I haven't seen anyone saying it will replace fibre optic networks in high density areas. 1Gbps is becoming common in New Zealand major centres and I've seen 4Gbps offered recently as well (with some minor upgrades required)

My parents live an hour out of our city (2 with traffic) and their service is around 3Mbps and it has just slowly degraded over the time they've lived there and been left behind by ISPs. They do have a great view of the sky though for their connectivity issues. I love the night sky there and soon I hope to love it even more knowing that its no longer a frustrating trade off for them.

Rural broadband and satellite providers are the ones looking to be disrupted with their low speeds, low data caps and high costs.

> Tesla certainly isn't, and their customer service is not good.

How is Tesla not pro-consumer? Genuinely curious. My thoughts as a non-customer are that they have blazed the trail for normalizing EVs. Without Tesla, the Porsche Taycan likely wouldn't exist, Rivian wouldn't exist in the same way, and electric vehicles altogether would probably be in the same place they were 10 years ago.

Even if their product isn't the best, at least it's an option, which leads to more competition, which in my eyes seems to be pro-consumer.

Starlink is already faster and will provide some competition too.

> 90 USD per month for 1.5Mbit DSL

I can’t believe it. I paid an equivalent of couple dollars per month for that 15 years ago in Eastern Europe.

There’s both a legitimate challenge (the land masses are huge in countries like Canada and the US) and functional monopolies that are leveraged by the companies who own the infrastructure that stretches out over those land masses.

Canada is an easier example because it’s essentially serviced by 3 companies who then lease out to the smaller “competitors”: for rural areas some people have the choice of either DSL over oxidized above-ground lines or tower-to-dish internet where the tower is daisy-chained from another tower or from it’s own oxidized lines. There is no cell coverage at all, and poor phone service. If someone in an area like that wants internet the company says that ~”the cost of the service must cover all the infrastructure and maintenance” and then they base that cost on only a few percent of people actually using it. In that way it’s hard to argue against “well, it costs us $1500/mo to provide service to 10 users, so we’re losing money when we charge a ‘reasonable’ $120”. But what’s really happening is that people in those areas are bandwidth starved, signing up for anyone who can provide decent service, and signing up in bulk while the company oversubscribes. It’s easy for them to still show losses publicly, but on the back end they are collecting government subsidies and using free capital to extend the network as far as they can.

Rural?

OP said he was in a town

Yes

> When the phone rang, the DSL would stop working. When I ended calls, the DSL would stop working.

In Russia such issues were just solved by adding some passive "splitter" device that supposedly filtered some of voice call from DSL data transfer. Every DSL ISP did it.

Same. I still have a couple of DSL splitters Verizon gave me when I had DSL ~20 years ago. They should have been included with the installation. One can probably find the splitters online for a few bucks if they need them.

All DSL subscribers everywhere get those.

We got Starlink in Placerville California and it’s great.

Just make sure you place the dish in a location with zero obstructions.

50-180mbps down and 24-40ms latency. ~25mbps up.

We just didn't get Starlink in Placerville, because we are surrounded by 100-ft tall trees. So we are still stuck on microwave links. (ATT ran fiber to within yards of the house more than a decade ago, and never lit it because the FCC allowed them not to. They pocketed the $half-trillion the FCC let them tax landlines for to build it, though.)

But it is good news that Pville is in a served area now. Where did you find a prairie in Pville?

https://towerdirect.net/product/used-120ft-sabre-self-suppor...

We're lucky - it's at the top of a hill overlooking a giant valley up mosquito road I think on the west side of town? It took a few tries and there is still a minor obstruction which causes occasional 10s drops. It's my in-laws house (I just bought the starlink set up for them so there would be better connectivity when I'm there).

If I had control over placement and such I think I could get a view without obstruction.

They still have calnet as a fallback network, but it's slow and has similar latency.

Have you looked at T-Mobile's home ISP product? I'm in an urban area, but it's been working great here @ $50/mo. Last speed test was 220Mbps down / 60Mbps up.

The fcc never gave them half a trillion. The fcc never had half a trillion to give. Can you show me where in ATTs financial statements they gained $500 billion or anywhere near that?

It was a $half billion. And obviously the FCC never gave it to anybody; as I wrote right there in the very posting you replied to, landline subscribers were taxed for it.

A $half billion seemed like a lot of money back then.

Per rasz's link, it is $400B and rising.

Only the reduced number of landlines in use makes it rise more slowly.

https://newnetworks.com/broadbandscandals.htm

Are you familiar with the concept of hyperbole?

A hyperbole is an exaggerated statement or claim not meant to be taken literally.

You can also try https://startyourownisp.com/

Is Verizon the only option there for the last 12 years?

I feel like geo internet such as HughesNet would be both cheaper and better than what you describe. Geo is high latency but at least better bandwidth than 1.5Mbps

the latency would make real-time audio/video impossible. That's not a solution in the world of Zoom/discord.

Reply on news.ycombinator.com