Hacker News Reader

Comment by _fat_santa | Hacker News Reader

My first thought is "ok well then give the cash to a company that is actually willing to do it". IMO Starlink would be the best option for these rural areas, simply based off the fact that they have a track record of providing broadband to very rural areas + they don't have to run any fiber to connect a new location. Starlink is very expensive though which is where I would expect a govt subsidy to step in and bring the cost down to around the cost of existing DSL for consumers.

EDIT: My comment failed to account for congestion on the Starlink network which other commenters have pointed out. Honestly after reading that I've concluded this is just a shit situation, fiber is expensive as hell to lay and Starlink poses congestion issues both for customers and up in space.

View on HN

Starlink has about 4 million customers right now, but considering that Starlink is struggling with congestion in the US, there will be a gap of millions of rural US households who would be best served by fiber. And regardless, fiber is the best long-term internet access medium, because whatever latency problems fiber has now can be improved by changing the devices sending data through the fibers without replacing the fibers themselves.

[1] https://www.techdirt.com/2024/10/01/capacity-crunch-causes-m...

Personally I think 5g is the way to go. Free more spectrum from the old TV channels (DVB-T is a lot more efficient in terms of spectrum use), this will cover rural use.

10 years ago I would never think I would say that, but I'm happy with LTE at 50MB down, 15up. The problem? No new lte equipment manufactured in western countries. You want cat 16+ modems? ZTE and Huawei are the only ones to sell you one. And their software not only spies for the Chinese Communist Party. It is also shit (I know, I bought a zte modem to marvel at my 100MB down speeds, o ly to rip it out 2 days later in discust after all my voice calls sounded as if I was on half a Meg if I didn't restart the thing daily (no, it only has automated restart as a weekly option in software). So my old tp-link was installed back in its box on the mast.

So we need more work done here to make 5g happen. Chinese will no do it for us (especially now that their economy is failing).

5G networks need wired infrastructure to send internet data to towers within a few miles of the customer; fiber is the best option for those wires. And for the last few miles, fiber still has throughput and latency advantages over 5G [1].

[1] https://www.eff.org/wp/case-fiber-home-today-why-fiber-super...

The most cost effective approach is hybrid.

Force ATT et al. to build out fiber trunks in rural areas, then allow competitive wireless offerings to connect last mile. (Excluding the trunk owner)

And if ATT doesn't want to do that, then local governments should be free to do so.

This seems like a terrible idea. Satellite internet doesn't just cost money; having all those satellites in orbit have real non-monetary costs. They have a huge impact on astronomy, and they increase the probability of accidents which cause space debris.

Except the satellites are already up and functioning, the service is available today. We will look crazy for burying thousands of miles of wires to residential in the future.

Except we won't, because the satellites are wildly over congested and wires are, frankly, far more reliable.

You can't add 10 million more subscribers and maintain broadband speeds without massively increasing the number of satellites.

Fair point though I would make a counterargument that with or without those subsidies, those satellites are still going up to space so wouldn't it be better to offer the service to as many folks as possible so that the most people get the benefit of satellites and we while we wouldn't offset the costs of putting sats up in space, at least we can offset the per-capita cost.

I am 100% in favor of stricter regulations on how we use limited space resources.

>They have a huge impact on astronomy

I'm of the opinion that commercial satellites should be required to be painted as dark as possible. If that requires extra engineering to dissipate heat, so be it.

It's reasonable that satellites should mitigate the impact on astronomy as much as possible. But it's not reasonable that we should seriously constrain the use of satellites because some astronomers using land-based optical telescopes are inconvenienced.

How would painting a satellite black 'seriously constrain' them? Why do we let corporations foist externalities on science and the general public? Either you can engineer a solution, or you can't, and if you can't, you don't belong in space.

I don't design satellites so I don't know. As I wrote, if there are reasonable measures that can be taken to reduce the effect of satellites on astronomers (or potentially creating space debris) they should be taken. However, there's a clear trend towards more (cheaper) satellites that serve useful purposes. And I very much doubt the public cares much about any inconveniences it presents to academics in this case.

Painting a satellite black wouldn't solve the problem that they would still block the stars behind them when passing overhead. Instead of light streaks, you now have dark streaks across a region of the sky.

I mean that's like saying that a star's brightness isn't a problem when searching for exo planets, and they only block you from seeing the objects behind them. Obviously that's not true. Light pollution from orbital satellites is a real thing.

These "dark streaks" would be way less of an issue. The reason you get light streaks is that exposure is really long and the satellites move quickly. The fraction of time a satellite is covering any given portion of the picture is really small. If you use, say, a 30 minute exposure, these "dark streaks" will be parts of the picture which only got, say, 29 minutes and 55 seconds. That's not a big deal.

The problem with painted satellites is instead that they're not black enough; they still produce light streaks since they're still somewhat reflective. I don't know if it's possible to make them non-reflective enough to make this a non-issue.

So for the space debris argument, is humanity just not supposed to have satellites? Like should we phase out the entire concept? Do we expect to solve the issue somehow and then open up space?

Are some satellites okay and it's just a value judgment of the clear worth of one more satellite against the unquantifiable risk of possible downside?

We should recognize space in orbit as precious and make intelligent decisions about how to best make use of it. That means we don't waste a ton of it just because the government can't get AT&T to build what they're paid to build.

The downside for astronomy is immediately quantifiable and not theoretical...

It's not an issue of what companies to trust. The issue is trusting them at all.

Ideally this sort of thing should be a public work. Operators can be private but shared infrastructure should be publicly owned seeing as we're paying for it.

But if markets have to be involved...If we're willing to pay $X per customer we should give the tax subsidy to them. The market can sort out who is actually willing to get the work done and service customers to get the subsidy.

Yes, I have no idea how, in 2024, internet service is not considered a basic utility like power and water, and regulated as such.

My parents paid ~ $65 / mo for 6mb dsl for ages. They had constant outages, and ATT always stonewalled the repairs. Talking to the repair guy, he admitted that ATT basically hated supporting that DSL, and wished all their rural customers would go elsewhere. He even told them that if they canceled, ATT would not offer internet service to them again, they'd just decommission the service down their street.

They're now on starlink, despite being only a 1/4 mile outside of city limits, it's the best option they have.

This rings verrrry true.

I recall in 2005 in a major west coast city, having 6Mbps DSL service but it was very prone to disruption. The tech they would send out would be just trying different pairs from a big fat cable to see if he could find one that was not too deteriorated from age. Even two decades ago AT&T's infrastructure was falling apart, and again, this was in a major city everyone in the world has heard of. Even then, they didn't care enough to even fix their 50-year old copper (which is now 70 years old) let alone lay fiber. Meanwhile also in 2005 on the other side of the continent, FiOS was launched.

>Starlink is very expensive though

Really? It's not always the cheapest option but seems pretty competitive with land-based broadband generally.

I agree with the broader point though. In my experience, Starlink seems plenty good enough for most consumers. Once you get above a certain level where you can do things like video calls and stream a couple of movies with reasonable reliability, most people are just fine.

I'm sure standards will increase over time but I'd probably be just fine if Starlink were my only option today. Even considered getting it as a backup but Xfinity has been reliable and good enough of late where I live that didn't seem worth it.

The rural broadband subsidies would work much better if they were not available to large ISPs. The scaling effects of last mile internet installations are basically negligible.

For instance, a neighborhood of a few hundred houses in the mountains outside of Silicon Valley started their own fiber to the home ISP. It cost something like $10K per house to wire it up and monthly payments are much lower than starlink.

We need more of that. If some small rural ISPs decide they are better off expanding than getting subsidies, that’s fine. Stop paying future subsidies.

>The rural broadband subsidies would work much better if they were not available to large ISPs.

This. When you go on to the FCC website and look for good high speed internet, outside of major cities it's almost always available through some coop. The government should be supporting this, but it's not 'by the people, for the people' anymore, it's 'by the corporations, for the corporations'

Does Musk need more government charity? He's made a lot off begging, we can see whether he puts more restrictions on companies that are non-viable without government handouts but considering most of his haven't been, I can't imagine it.

He’s already come out saying he wants to remove EV subsidies after his company was able to get off the ground with them.

Age old pull-the-ladder-up!

And the rest of it... All of his companies are basically wanting government handouts.

Reply on news.ycombinator.com