The solution to that problem is to get municipalities to exit their monopoly agreements with broadband providers.

Spending more tax dollars on satellites isn't helpful. That just gives your representatives an easy way out in the near-term, and probably results in everyone getting upset over wasted money in the long term.

What "tax dollars" are we talking about here? It's a private system.

This right here is what they're banking on :)

And they have plenty of reasons to bank of that ;) .

Comcast being slow hasn't been a problem for the majority, Comcast being expensive may have been.

Comcast having terrible customer support is another big problem.

How often do you need to call the customer support? I have had Comcast for 6 years now and I call them once a year to renegotiate the deal or move the connection.

That's a terrible excuse for having bad customer support.

Everyone is super excited about this and I am sure its going to capture a significant market, but you have to realise what are alternatives and how much they cost.

I expect them to be utterly unaffordable at least for first half of decade.

Their biggest competitor is Iridium which offer similar service at $150 per 2kbps. I will let you extrapolate real life prices from here.

Iridium is not their competitor. Different market. Even the CEO of the company said so.

Sure, terminals are likely to be different, but service is the same. I don't see enough differences between two.

SpaceX should be able to undercut that some since they can launch their satellites at cost and using a booster that's already been up once or twice really brings that cost down.

Sure, but why would they undercut them by 6-7 orders of magnitude in price? Seems silly to just leave money on the table.

Bigger less profitable userbase vs larger more profitable user base. The question is if the bigger userbase option is big enough to make overall profits larger than they would have been with the smaller one. I have no idea.

Heck, I already have FIOS, but if it's reasonably priced I'll subscribe just to support it (and to have a backup). Musk is basically the closest thing we have to Tony Stark. If that's not worth supporting, then I don't know what is.

High latency satellite internet already exists (re: HughesNet)

Interesting, I see reports of latency of 60ms on average [1]. Starlink claims it will offer as low as 25ms [2]

Is 35ms a significant value proposition, considering 5G is on the horizon?

> As of March 2019, HughesNet's average download speed is 11.04 Mbps. Throughout their coverage area, the average latency on HughesNet Internet speed tests is 61.87ms. For context, wired terrestrial Internet connections usually output latency results in the 5–70ms range. [1]

[1] https://broadbandnow.com/HughesNet-speed-test

[2] https://www.zdnet.com/article/spacexs-starlink-takes-a-big-s...

That doesn't seem right. Napkin math:

GEO Satellite is 22,236 miles above sea level[1]. Speed of light is 186,282 miles per second[2].

Therefore the absolute fastest information could travel on a GEO network is 22236/186282 = 119 milliseconds (one way at the speed of light), which is 238 ms round trip. In reality there are going to be internal and network latencies so I would expect something more like ~300-500ms RTT.

[1] https://en.wikipedia.org/wiki/Geosynchronous_orbit

[2] https://en.wikipedia.org/wiki/Speed_of_light

That's right. Geo will typically be around 600ms. The speed metric is incorrect/misleading as well. Both Hughes and ViaSat offer multiple speed plans that are vastly different. You can't simply take an average, or your speed isn't representative of what the service could do. Of course, satellite internet will pack a lot of users in there to fill up the satellite. I don't be naive about that, SpaceX will do the exact same thing.

> You can't simply take an average, or your speed isn't representative of what the service could do.

Why not? We're just comparing current satellite broadband offerings' known latency to what Starlink could potentially offer.

If your cable service offers 10, 25, 50, and 100Mbps plans, how is an average speed at all meaningful? It's just representative of how many people sign up per plan.

An existing offering does 60ms average, meaning its best case is better. Starlink would offer 25ms best-case scenario.

My post is clearly about bandwidth, not latency. You don't choose your latency.

> That doesn't seem right.

broadbandnow.com is a legit site that tracks internet providers and speeds by aggregating data.

Yeah but... it's claiming information can travel faster than light...

It must be averaging in terrestrial network latencies or something, because GEO satellites are literally hundreds of light-milliseconds away from earth

I'm genuinely curious, can you have it without your landlord (if you rent) approval and also install the ground transciever without HughesNet technicians? Sometimes they refuse to install in particular locations on the grounds of perceived landlords disapproval (?)

Yes you can. More here: https://www.fcc.gov/media/over-air-reception-devices-rule

Not a lawyer but there is definitely a law in the US which prevents landlords from not allowing tenants to install satellites for television. I'd imagine that the same law might cover generally any consumer telecommunications equipment