In the last several decades, telecoms have routinely been paid billions for projects that didn't produce, and then won money in the next round as well. Indeed the fight isn't usually about whether to continue this, but who should win this round (established telecoms, municipal broadband, etc).
It was a surprise when Starlink won in the first place. But it is no surprise that it is being held to a higher standard than, say, Charter.
It's totally fair to criticize subsidies that have been historically given major telecom companies when they failed to deliver (and they have). But, I don't think it's fair to say "historically the US government has given subsidies to telecoms that have failed to deliver, so you should keep doing that".
Given this history, I am inclined to believe that actually not giving money to Starlink for a fairly small performance miss is motivated by something other than just applying the rules.
If in a few years it is clear that policies have changed, and telecoms no longer get the gravy train either, then I'll change my opinion. Ending this corruption would be a welcome change.
The solution to “the old people weren’t being vetted enough” is not “the new players should also not be vetted enough.”
Bottom line: I don't think these $886 million now going elsewhere are going to benefit even a single person up here near the Canadian border in the next few years. Fiber is still not going to appear here. The border is over 3000 miles long. In contrast Starlink can do something now (if you look at the Starlink map [0], there are very few filled cells in northern New England or northern Midwest/Pacific West up near Canada), and has the real promise of doing better. Now that will be harder, and I do not believe anything better is going to come of it for any of us in that time frame. If fiber comes in 2028 wonderful, Starlink could be cancelled then. That's a long time though to still be on 56k/HEOsat.
To sibling comments talking crap like "do you work for SpaceX" if you wish to volunteer and come up to lay fiber for us right fucking now by all means.
----
0: https://www.starlink.com/map
1: I'm editing in response to tyrfing below because I suspect I'll hit the HN reply speed cap sometime shortly. This got tons of coverage in the 00s in particular but has fallen away over time as these things tend to. I had a bunch bookmarked at one point from old /. stories and the like and actual paper books but all the sites now seem to be dead. Here's a more recent one though that seems to cover the gist and could be a place to start for more: "The Book Of Broken Promises: $400 Billion Broadband Scandal" https://www.huffpost.com/entry/the-book-of-broken-promis_b_5...
Can you share a source for this? I'm very curious to see the breakdown of funding sources.
> there are others less well off within a hundred miles who were hoping to get Starlink if the monthly cost could be brought down by this program
Why would the company receiving free money make it cheaper, are there price caps that SpaceX wouldn't meet? You've clearly laid out that they have no competition.
> Funding very promising technology that is aggressively trying to ramp hard and has a clear 1-3 year horizon
They had 1.5 years. "1-3 year horizon" is thus clearly misleading, since halfway through that timeframe they are unable to meet requirements.
Sounds great. However, I've seen ↓31/↑4 and ↓20/↑10 out of it. Anecdata doesn't prove much.
The network is improving, the tech is capable, but the real life performance is not up to snuff. Starlink, much like any company Musk seems to get himself involved with, is promising more than it can deliver, and that hurts when it comes to subsidies. They can have their billions when their network does what it promised to do.
Heres ping info from starlink:
--- google.com ping statistics --- 38 packets transmitted, 38 packets received, 0.0% packet loss round-trip min/avg/max/stddev = 26.879/45.073/95.627/11.582 ms
Hughesnet was one of the best options prior to Starlink.
Anecdotes cannot prove population statistics, you need an unbiased random sample for that. But anecdotes absolute can prove limits though. Remember, this is what I was responding to, emphasis added:
>Starlink isn't CAPABLE of meeting its performance commitments
That's what I took issue with and think is important. Starlink is absolutely capable of those speeds; I've seen them. So the problem isn't the technology, the problem is balancing serving the most people (which is a public good too) with maintaining enough performance on average for each one, all while scaling and evolving the technology fast. I'm assuming SpaceX could have submitted private data showing the public results by ookla were off in some way that would have been sufficient and that the FCC didn't ignore that out of lobbying/politics, so I'm taking it as a given that they did fall behind there. But it's not like Starlink is "done", they're continuing to increase sat numbers, build more ground stations, they're prepping to bring online intersat optical links and have no given a firm date for that (it's required for their marine service to get away from coastal waters), they've got much better sats and a solid process on the rocket to launch them. If we didn't need someone to take on some more risk and help accelerate things, this entire program would be meaningless and there would be no need for government in this space at all! I guess it feels like they're being punished for trying to move faster when that's exactly what lots of Americans need. We needed something better then the crap we've had 10 years ago let alone now.
Of course, this may be 100% the correct legal decision depending on the exact language, administration and so on of the program. I could easily see the FCC not doing well on that but that doesn't mean they aren't required to stick to it by law. But that's not what I see most people criticizing it on, they're talking technical and societal angles instead. And on those grounds, I can be disappointed if the program didn't have some leeway. "Needs to be ready by 2028" isn't good enough!
Starlink is probably below three nines of reliability on even having a connection, and may only be at one to two nines or so on the speed demands. Many people on Starlink get bursts of decent speed and also stretches where the network operates at a crawl.
Thus, they can offer you good speed and still be incapable of meeting performance commitments.
They've actually been below their commitment for all of 2021 and 2022 so far: they started 2021Q1 with a 16.29 average and 2022Q1 had dropped to 9.33 average. That's a pretty worrying trend and I don't think its the FCC's job to evaluate literal rocket science to decide whether or not they believe SpaceX will meet their obligations.
SpaceX chose what speeds to pitch in their bid, how many satellites and ground stations to deploy, and how many customers to let onto their network. While I agree that Starlink has a bright future I can't help but see this as SpaceX clutching defeat from the jaws of victory.
We've been far too lenient with ISPs in the past with broadband funding, and this is what keeping them to their word looks like.
https://www.ookla.com/articles/starlink-hughesnet-viasat-per...
Citation needed. I don’t see why anyone would lobby against starlink or LTD here. Neither won. They money from the fund is not distributed. They can reapply.
What the hell are you talking about? If there is lobbying, it’s certainly not against Starlink or LTD getting the funds as subsidiaries. The goals were set, starlink didn’t meet them, they don’t get subsidized. Simple as that.
RDOF aims to hold providers to a much higher standard for rural broadband, and part of that commitment is whether you can deliver speeds across the whole network at scale, not just relying on theoretical top line speeds. The evidence so far per Ookla demonstrates they don't meet muster currently, and since the tech is so new and (mostly) unvetted, it makes sense for the FCC to play it safe here on behalf of the consumer.
That said I don't have a problem with the funding being held until the conditions are demonstrably met instead of as funding to meet them. The latter has failed too many times with providers just pocketing money to then stop executing once they have it.
And there are definitely reasonable scaling concerns here. They're using a finite set of frequencies and the laws of physics still apply. Any concurrent transmissions are going to be impacted by both their phy layer modulation strategy, in addition to the mac layer. But we know so little about those layers currently that its hard to comment without being entirely speculative.
Yes they must obey the laws of physics but this isn't a WISP, concurrent transmissions can't just stomp on each other and find out fast enough to hide it they've had to be broken out per client in the cell from day one. Other than mentioning the MAC layer exists is there a particular concern you have about it scaling? The satellites only have 2 paths to forward things and only a small patch of cells to do it for. Scaling has been brought by putting up orders of magnitude more satellites to provide coverage at more than the dense northern coverage from the orbits, it has not been by putting a ton more clients through the satellites themselves.
Is RF and framing something Starlink has had to design for? Absolutely, they do after all need these things. That doesn't mean it makes sense to jump to it as the source of any perceived issues, particularly on the basis it's something unknown.
I've seen the write up on the individual who was able to spin up an SDR to receive beacons from their satellites, but that doesn't really tell you anything about how the signal is modulated.
Maybe it "makes sense" to those who aren't actually one of those consumers. But if "playing it safe" means "you get to stay on 56k/4-1-ADSL/viasat/hughes for another 5 years and maybe you get fiber then or maybe another administration comes along and crushes this whole thing who knows! <3" then no that's not great right? The more positive angle is that this doesn't actually slow SpaceX down at all, they aren't a public company so this doesn't mean their stock plunges immediately and shareholders demand a course change or anything like that, they're obviously committed to improvements and obviously have a fair amount of private capital to tap. So it won't necessarily slow them down much if any, at least hopefully. But it does mean all the growing capacity and improvements will go exclusively to the better off who can afford the $110/mo privately in the immediate future and that's too bad. It's more too bad as well because even with someday tens of thousands of Starlink v2.0 and intersat optical mesh and greatly increased cell density, there will still be tight density limits relatively speaking. So less well off people may find themselves effectively shut out of Starlink due to having to have waited until others move/pass away/get fiber. It might not amount to much but I'd really like to see my tax dollars helping to get them something asap on a more even ground I guess.
I'd be happy to be proven wrong though and for everyone to get fiber instead!
>and since the tech is so new and (mostly) unvetted
Shouldn't government should have more, not less, appetite for risk of this nature? If it's risk free, the market would already be doing it. Particularly after decades of throwing billions down the crapper for "safe" and "proven" tech by "reliable, established" companies which then used it to do mergers or stock buybacks or lobbying against municipal broadband and getting laws against it enacted in much of the country and perhaps a few token deployments.
I completely understand and applaud the pendulum swinging back against the decades of established monopolies trying to water down definitions of "broadband" to meaningless so they could market and suck subsidies with no capex required. But all pendulums can swing back too far in the opposite direction too, and now that rapid deployment transitional tech is possible (Starlink and WISPs) it'd be nice to see timelines factored in as well. There is value to society in someone getting 50-150 via Starlink or fixed wireless until 2028 even if fiber then makes it in and obsoletes it. And there's parts of the US I doubt see it even then honestly.
Also to your sibling:
>They're using a finite set of frequencies and the laws of physics still apply
This is nice of you to write but would you care to crunch us the numbers on Shannon limit for the Ku-, Ka-, and E- bands (ignoring V-band since nothing for that exists for Starlink yet)? Do you actually have reason to think they're maxing out potential bandwidth right now in 12-18 GHz or 26.5-40 GHz right now vs that being a tech limitation that will be improved with better phased arrays and bigger sats like v2.0? Because it's kind of a significant amount of spectrum.
> Do you actually have reason to think they're maxing out potential bandwidth right now in 12-18 GHz or 26.5-40 GHz right now vs that being a tech limitation that will be improved with better phased arrays and bigger sats like v2.0?
I do think its a tech limitation, yes. The capacity limitations in many new RF solutions are unrelated to the theoretical max of any of those bands (re: Shannon limit). As to whether the that can be improved, that depends on where the bottle necks are. I brought up the mac layer note for that reason, I don't know that they're enveloping ethernet frames over that link, but if they are - what is the strategy there? TDMA? CDMA? There's just not a ton of public information out there to make a deductively valid argument that the FCC should expect it to improve.
(EDIT: Realized that I didn't convey my point well. It is a tech limitation imposed by the physical limitations of the medium itself. No matter how much spectrum you have, you still need a modulation scheme to allow for multiplexing over that spectrum. There are limitations there where the implementation details really do matter.)
The fastest thing I can get otherwise is a 10M down, 768K up DSL link. I had been working remote on that since 2008.
AFAIK the minimum speed for "broadband" is 25 down.
I am a highly online person living in a large city with many choices of ISP at many different speeds, and I still only have a 110mbps connection. It seems strange to me to suggest that speed is universally inadequate, though I can understand why large households may need higher bandwidth.
I have starlink and mostly see between 150-300 down and uploads in the 8-15MB range, all for $99/month - verizon was charging me $156/month for 3MB down and 0.768MBP up.
Nice to meet you - now you know at least one person 'with starlink has gotten better than 100 down'
I’d be lucky to get 5/1.5 from frontier, and their DSL service in the area goes down every few weeks.
This seems like a case of lobbyist-induced double standards.
If they failed their qualifying metrics and did win the subsidies anyway, wouldn’t mirror universe hedora be here complaining about that double standard?
To me at least, this seems less obviously suspect than that would have.
This collection of speed test data shows the US average has been 66/9 for the last month. At 50ms they're also nowhere near their latency requirement. https://starlinktrack.com/speedtests/region/us
Their plans for further scaling are solid and in progress. The tech is obviously and objectively capable. The lobbying against it has been intense for completely befuddling reasons though, but even so this is a really disappointing decision.