Hacker News Reader

Comment by I_Am_Nous | Hacker News Reader

I read that letter, and was unconvinced that it's anything more than the FCC not wanting to gamble with nearly 1/16th of the total RDOF grant money (for that round) and would rather give it to a company that can be reasonably expected to hit the obligatory throughput.

If Starlink bid for 25/3 they might have made it.

View on HN

You can arrive at your own conclusion. I think its pretty obvious whats happening here (the commissioners voted along party lines right down the middle). And theres no other company thats even close to Starlink now or in the medium term future. So I dont know who would practically fill this spot.

For below comment: This is for "rural" connection. You're not laying wire for that regardless of what Comcast wants you to believe. They can barely service what they have and the cost/benefit of laying 30 miles of wire to reach someone in the woods is never going to make sense.

It's a letter from one FCC commissioner, of which there are currently 5. He dissents from the decision the commission as a whole came to. There are a lot of companies on the ground that could benefit from that ~$900 million so a single company replacing Starlink is not necessary. The main concern is if the FCC give Starlink money to reach 100/20 and they don't do it (because there are legitimate technical issues to solve before it's possible for Starlink to supply over half a million people with 100/20), it's wasted money. The FCC didn't think it was doable on that time scale.

Doing some math, currently each satellite launch sends up 22 satellites at around 2.8 Gbps per satellite. For each launch, Starlink adds ~61.6 Gbps of capacity. If we cut that up into 100/20 slices, each launch supports 616 customers at 100/20. To support 650,000 subscribers at 100/20, it would take about 1055 perfect launches.

I don't think the FCC was wrong when they said Starlink could not reach 650,000 people at 100/20 by 2025. There aren't enough days to launch one rocket a day to even try to catch up.

Did you miss the other dissent which would mean 40% of the commission disagreed with the decision?

DISSENTING STATEMENT OF COMMISSIONER NATHAN SIMINGTON

>I wholeheartedly agree with the entirety of Commissioner Carr’s dissent. I write separately to further highlight some of the meretricious logic that underlies the Bureau’s, and now Commission’s, rescinding of SpaceX’s RDOF award. ... >I was disappointed by this wrongheaded decision when it was first announced, but the majority today lays bare just how thoroughly and lawlessly arbitrary it was. If this is what passes for due process and the rule of law at the FCC, then this agency ought not to be trusted with the adjudicatory powers Congress has granted it and the deference that the courts have given it

https://docs.fcc.gov/public/attachments/FCC-23-105A3.pdf

Well if you want to really dig into the numbers here and get down to the gnat's ass of uselessness, Simington was confirmed with a 49-46 vote which means that less than 50% of the Senate agreed with him being on the Commission and hence he shouldn't even serve because he couldn't garner a majority of Senate approval. So, while 40% of the Commission disagreed with the decision, we should recognize that 20% of that 40% comes from someone undemocratically serving on the Commission and hence should be ignored. Meaning that, in actuality, only 25% of the democratically appointed Commission (1 out of 4) disagreed with the decision, not 40%.

All of that to say: this whole point you're making about "40% disagreed" or "20% disagreed" because the decision wasn't unanimous is really fucking dumb. The decision by the Commission is the decision, it doesn't matter how many dissents there are.

Where does that dissent say 40% disagreed? It only uses the term majority.

There are 5 FCC commissioners (as @I_Am_Nous's comment points out). @I_Am_Nous references one dissent. @hnburnsy links to another. That's 2 dissents. 2 out of 5 dissenting is 40%.

you're ignoring over-provisioning which generally is ~10x

The terms of these subsidies only allow 4X oversubscription.

ok, so that still cuts down the amount of launches by 4x which takes them from 1055 launches to 260 launches. Over 2 years that would require doubling Starlink's launch cadence which is a lot, but does seem plausible.

So to make the 2025 deadline they would have had to perfectly launch more rockets than they ever have before...sounds like the FCC made the correct choice.

SpaceX has done that every year since 2020. In 2020 they had 26 successful Falcon 9/Heavy launches, 31 in 2021, 61 in 2022, and 91 to date in 2023.

They need to do 180 a year to put enough satellites up to even try to hit the 2025 deadline. That's not even counting any satellites which may fail between now and then and need replaced. This is a major reason why the FCC didn't think they could have met the 2025 obligation to reach ~650,000 subscribers with 100/20 and rejected their application.

They're upgrading Vandenburg to do 100/year and Kennedy/Canaveral to do a ~daily cadence.

That will be sweet when they can get it done and reliably launch Starship! Starlink isn't bad, it just wasn't capable of meeting the RDOF deadline according to the information available at the time.

The calculation above assumes all satellites are available to provide bandwidth to the customers. That means essentially the 260 satellites need to be above the US (let's ignore that the visible horizon is different across the US). Now starlink are LEO, so 260 essentially we need to divide the 260 by the fraction the globe area the US is.

The 260 is a significant underestimate. It's likely 4-10x more

Sure but the assumption made already say, that SpaceX uses _all_ capacity for this program (and nothing else) and it doesn't require any double hops (I would think you need to at least add a factor of two for the up/down thing). And that you can see all satellites all the time. So it was a _very_ conservative assumption. And it would still require ~all launch capacity of 2024 and 2025. SpaceX calculations is extremely optimistic to the point of being delusional.

At least without Starship, which I _personally_ think that they will manage to iron out their problems of the course of next year. But even then _this_ timeline they won't be able to keep

Oversubscription where?

ISPs are not buying anywhere near that much transit bandwidth.

I'd rather the federal government just roll out fiber and not put Starlink and Elon in a position of power. That fiber will always be in the ground and available. Elon has shown himself to be unworthy of any position where trust and good judgement is required. If it costs more, that is a premium worth paying. Fool me once.

https://www.internetforall.gov/

https://www.whitehouse.gov/briefing-room/statements-releases...

https://spacenews.com/senate-armed-services-committee-to-pro...

https://www.washingtonpost.com/opinions/2023/08/30/elon-musk...

https://www.cnas.org/press/in-the-news/elon-musks-control-of...

https://babel.ua/en/news/98461-elon-musk-partially-transferr...

(disclosure: starlink customer)

> Elon has shown himself to be unworthy of any position where trust and good judgement is required.

That's an insane statement given the unprecedented success of SpaceX.

The success of SpaceX is placing Musk in a position to decide where America's allies have access to the internet and choosing what region of the world can be cut off just through meeting politicians he likes.

Surely there is no risk the US will be cut off.

That doesn’t negate the fact that he wields power against others when it meets his needs. He’s effective, I don’t dispute that, but still needs a metaphorical cage built around him to protect others.

He "wields power against others"? What are you talking about?

https://news.ycombinator.com/item?id=38127745

For someone who is such a fan, do you not know who your hero is?

That's a bunch of cherry-picked gossip. You could make another list with a lot of people saying positive things about him.

I don't follow

> just roll out fiber

I worked provisioning internet for the Telco that serves basically all of Northern Canada. 33% of Canada's landmass and only 0.3% of its population.

We're not talking about cities or even towns here, we're talking about very rural customers. Have you been to rural Alaska, or Montana or Wyoming?

I have, and you drive for hours with no cell service, let alone wires in the ground.

You are seriously underestimating the expense to run fibre to each of these customers. Some of our communities it was well over $1mil per customer.

Indeed, satellite or long haul fixed wireless will be the only option for some locations. I have been to rural Montana and Wyoming, but not Alaska.

Customers will have to pay for their own StarLink where the FCC won’t. Perhaps we should not be subsidizing folks where it costs $1M to deliver terrestrial connectivity to you. Cheaper to pay them to move.

> Cheaper to pay them to move

They are not going to move. Period. I know this sounds snarky, but in all honesty if you had been to Alaska you would understand.

> Perhaps we should not be subsidizing folks where it costs $1M to deliver terrestrial connectivity to you

Or serve them with fast, reliable internet that is not terrestrial, and does not cost anything remotely close to $1M.

Doesnt sound snarky at all. "paying them to move" sure does though.

I've seen subsidy numbers of $200k. I'm pretty sure a million is possible.

>just roll out fiber

As if this were a trivial task

This is in comparison to launching satellites into space. I think most people would agree it's probably more along the lines of "trivial" when compared to that.

Neither are trivial, the two just scale very differently.

I do see the benefit in resilience of building out fiber even to moderately unprofitable (from a unit economics point of view) regions, just like we also build roads to communities that will never "pay the investment back" in taxes. But there are cases where it just can't be justified.

But it's also not a simple either-or: There are other technologies than fiber and satellite; there can be more than one high-throughput LEO provider; we can have a few GEO satellites for redundancy (although with significantly worse latency) etc.

Outside of truly rural areas the question with fiber is how long is the payback period, not "will it be profitable". Especially if deployment is integrated with routine highway re-pavement projects (roads need torn up and redone roughly every 30 years, after all), the majority of the cost becomes the fiber bundles themselves - perfect for even a smaller county or city government to handle with a modest bond issue.

> the question with fiber is how long is the payback period, not "will it be profitable".

The "payback period" might well be infinite (with non-zero interest rates), in which case we're talking about a subsidy, not an investment. (Which might still be a good idea! It won't "pay for itself", though.)

Verizon was able to lay fiber all over rural New York in a pretty short amount of time due to a New York law for similar rural funding. Places that couldn't even get cable have fiber now. Just laying fiber is an alternative to satellite.

Do want to point out buildout requirements that are actually enforced in NY would be strongly compelling. Spectrum was heavily fined and had their license suspended on cable for failing to meet these commitments a few years back. Other states just dole out the money without punishing the companies that cash out dividends and use it for mergers.

Farmer's Telecom Coop service map, Jackson County and nearby, AL.

https://connect.farmerstel.com/front_end/zones

Yes, it's fiber. Yes, to the home. Currently, 93Mbps down, 83 Mbps up (but I have the cheap service). And the service is a crap-ton better than that of Spectrum in NC.

That's what I read too: you're not democratic enough elon

So basically now 15/16th of the money goes into a void to never actually get service to anyone.

Anecdotally, my dad lives in a rural area with no cable/DSL broadband available.

Cellular broadband only got him 10-15 Mbps. He was excited when Starlink was available. I think he was pretty early on the preorder list. Once he finally got access to Starlink (Feb 2022) the speeds were close to the advertised ~100 Mbps.

Now the price has increased and on average he's back to getting like 15-20 Mbps down.

Luckily, the EMC that services the area received some rural broadband grant money to roll out FTTH and that build out has been pretty quick. They have already run fiber down his road and said that service should be available in a couple of months. The EMC is offering 2 Gbps down / 1-2 Gbps up (!!!) for $100/mo.

So this money is actually being spent effectively when it goes to the right place. Starlink made a bunch of promises that they couldn't fulfill and the money is being redirected, as it should be.

I feel like in 90% of Starlinks use cases it is only the best option because they are the most motivated to succeed. Running traditional wired service is the more practical and permanent solution but the telecoms have made far to much money by taking money then not delivering.

>Running traditional wired service is the more practical and permanent solution

It's permanent but it depends on what the word practical means. Often the cost of setting up infrastructure for such low density population means the infrastructure will never pay for itself, or that the same money spent elsewhere would service many more customers, so its not necessarily practical.

> Running traditional wired service is the more practical and permanent solution

Not when you're 50+ miles from the nearest anything.

Don't think of people that live kind of near a town and still get LTE. Think of people that drive for hours and still don't get LTE.

"and said that service should be available in a couple of months"

Thats telcom speak for "the check is in the mail".

no, it will be awarded to other applicants instead.

Not necessarily. This round of grants is closed. There is no guarantee that this money will be rolled into the next round. In fact, that seems quite unlikely to me.

>Phase 1: Will provide up to $16.4 billion >Phase 2: Will provide at least $4.4 billion

When it says "at least $4.4 billion" that leaves the door open for phase 1 fund rollover. We'll see eventually. Maybe Starlink can get some money in phase 2.

1. https://rdof.com/rdof

I don't understand why you're just rephrasing my comment.

Reply on news.ycombinator.com