AT&T Spent Hundreds Of Billions On Mergers And All It Got Was A Big Pile Of Cord Cutters
from the synergies,-yo dept
Over the last few years AT&T and Verizon have been desperately trying to pivot from stodgy, protectionist old telcos -- to sexy new Millennial media juggernauts. And while this pivot attempt has been notably expensive, the net result has been somewhat underwhelming. Verizon, for example, spent billions to gobble up AOL and Yahoo, but its lack of savvy in the space has so far culminated in a privacy scandal, a major hacking scandal, a quickly shuttered website where reporters couldn't write about controversial subjects, and a fairly shitty Millennial streaming service even Verizon's own media partners have called a "dud."
AT&T's efforts have been notably more expensive, but just as underwhelming. The company first decided to shell out $70 billion for a satellite TV provider (DirecTV) on the eve of the cord cutting revolution. And the company's putting the finishing touches on shelling out another $89 billion for Time Warner in a quest to gain broader media and advertising relevance. That was paired with the launch of a new streaming service, DirecTV Now, which the company hoped would help it beat back the tide of cord cutting.
But things aren't really working out quite like AT&T planned. The company's stock took a beating last week after it acknowledged it would be facing a 390,000 reduction in pay TV subscribers this quarter. AT&T, in an 8K filing with the SEC, tried to partially blame hurricanes for the mass exodus occurring at the company:
"The video net losses were driven by heightened competition in traditional pay TV markets and over-the-top services, hurricanes and our stricter credit standards. The decline of traditional video subscribers negatively impacts our Entertainment Group revenues and margins, resulting in an adjusted consolidated operating income margin that will be essentially flat versus the year-ago third quarter."
Unmentioned is that AT&T also lost 351,000 pay TV subscribers the quarter before, as the company gets hit harder by cord cutting than most pay TV providers. One of the real reasons for these departures? While AT&T was willing to spend hundreds of billions on megamergers, it has spent the better part of the last decade (especially in places where poor people live) neglecting necessary network upgrades. As a result, in countless markets Verizon & AT&T users on last-generation DSL lines are switching to cable providers for faster broadband, and bundling cable TV service that's priced cheaper than broadband alone.
In short AT&T neglected its core business in order to daydream about matching Google or Facebook's ad revenues, but (so far) lacks the core competency to jump the gap from telecom to Silicon Valley-esque Millennial marketing. Both AT&T and Verizon have spent so many years operating as government-pampered protected duopolies, they believed they could pivot on a dime, ignoring that years of regulatory capture left them with only a few key skills: charging too much for too little, lobbying to thwart competition and bullshit.
To its credit, AT&T was at least willing to take a risk and launch DirecTV Now, a streaming alternative. And while the company did manage to add 300,000 streaming customers on the quarter, those users pay a fraction of what traditional cable TV customers do - and AT&T still saw a net loss of 90,000 pay TV users. Still, most other incumbent pay TV providers have responded to the cord cutting threat by raising cable TV rates (ingenious!) or by pretending to keep pace via the launch of streaming alternatives that are intentionally designed to be underwhelming, lest they cannibalize more lucrative legacy customers.
One of the core problems here is that Wall Street isn't satisfied with ISPs simply doing a good job and making a reasonable profit. The relentless, myopic need for quarterly improvements has companies like AT&T and Verizon trying to use megamergers and vertical integration to magically elbow their way into markets it's unclear they lack the competency for. And only after mindlessly cheering these deals do some Wall Street analysts realize some of these arrangements don't even make coherent business sense given the current market climate:
"Though the company partly blamed recent hurricanes for these trends, MoffettNathanson analyst Craig Moffett notes that weather was only the third of four reasons that AT&T listed. “Heightened competition in traditional pay TV markets and over-the-top services” came first. In other words: cord cutting. “It is becoming increasingly clear that the wheels are falling off of satellite TV,” he writes, meaning that Dish Network might announce similar results."
In an ideal world, AT&T would realize its core competencies (building and maintaining wireless and fixed-line networks) should take priority. That $70 billion spent on buying a doomed satellite TV company could have gone a long way in shoring up broadband service that in many regions still doesn't even meet the FCC's base 25 Mbps definition. But in the world we live in that's simply not sexy enough for Wall Street, and the need to grow simply for growth's sake will likely result in AT&T making even more expensive deals of dubious net benefit down the line. Next up: Waffle House?
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: broadband, cable, competition, cord cutting, dsl, mergers, satellite, tv, video
Companies: at&t, directv
Reader Comments
Subscribe: RSS
View by: Time | Thread
[ link to this | view in thread ]
What about AT&T's merger commitments?
>>While AT&T was willing to spend hundreds of billions on
>>megamergers, it has spent the better part of the last
>>decade (especially in places where poor people live)
>>neglecting necessary network upgrades.
About ten years or so ago, an AT&T employee told me that, as part of an agreement with the federal government, they were going to start introducing a flat $10/month DSL to particularly add access to the underserved areas. As I remember it, this was one of the conditions attached to a merger approval. A few years later, I signed up for AT&T's DSL. It was considerably more than $10/month, and they didn't even have the competence to hook me up for 3 months, despite already having DSL customers in my building.
[ link to this | view in thread ]
You can offer all the coolest things in the world, but when your own lines don't offer enough speed or reliability... why do we need to point out where you failed over and over?
You will never be Facebook or Google.
You will never manage to make billions in advertising.
It might be to late for you to still be a provider, because you delayed all of the investment & as you fail out of areas it'll be hard to stop upstarts rolling into town.
Stop dreaming about dollars you are never going to make & decide being reasonable would be a much better deal.
Its better to get $90 a head for giving them the pipe to use all the other services, than to engage in a war of attrition where your content will never be good enough & pissed off consumers will find something else.
Imagine the shock if you offered speeds starting at 50 without 400 other bundled things & recaptured the cord cutters into the fold. We want fast pipes, we don't care about your dreams of how you can use all of this to get more money for ads that we're never going to see.
[ link to this | view in thread ]
If these satellite companies can provide what they say, competitive speeds and pricing similar to or better than cable, then all of the current broadband ISP's are in for a world of hurt.
Granted they have to deliver first but as I heard someone say, "I've learned not to bet against Musk. He may not deliver on time, but he does deliver.".
[ link to this | view in thread ]
Re:
Best line of the week.
[ link to this | view in thread ]
Who could have foreseen that AT&T's downfall would be...
[ link to this | view in thread ]
Re: What about AT&T's merger commitments?
[ link to this | view in thread ]
Re:
How are you going to deliver all that cool stuff?
[ link to this | view in thread ]
Re:
WiFi does pretty much the same thing (handing off to the next AP with the same SSID when you get out of range), but I wonder how they'd deal with the spike in latency during the hand-off. It's not so much a problem when you're moving and the AP is staying put (to avoid latency due to a hand-off, you just stop moving), but if you're being handed off every few minutes, with no good way to stay connected? I can see it being more of a problem.
[ link to this | view in thread ]
Re:
[ link to this | view in thread ]
slightly more than 8
>>that many up there.
How about around 500 times that number:
https://www.theverge.com/2017/5/4/15539934/spacex-satellite-internet-launch-2019
[ link to this | view in thread ]
70 + 89 = 159. WHERE the "hundreds of billions" in title?
"Losses" no longer matter with the Federal Reserve inflating the "money supply" by a trillion per year since at least 2008: interest rates are so low that it's essentially free, and it driving up stock "prices" may actually turn out FAR BETTER than free!
Besides that, this is your endless attack on ATT while totally neglecting Google / Facebook / Amazon / Microsoft / Apple, though far more important and they're on Drudge Report daily.
[ link to this | view in thread ]
Line Service vs Data Services
In the old days, there was only ONE data service on the line. It may have been cable TV service or phone service, but that was it, and so it made no sense to break out the the line service from the data service in your bills. That would have just been confusing to consumers.
But then came the internet.
Providers quickly realized they could make new money on the existing infrastructure by offering new data services on the same lines. Thus, DSL and Cable internet services arrived.
Here's the problem: the newer services are priced assuming the line infrastructure is already there. The older services -- landline phones and Cable TV -- are still priced as if they have to maintain and build out all this infrastructure. In some ways, telcos see all this new fiber going in to support new internet services as if it's a cost of running your land line or cable TV subscription.
I think there's an argument to be made that I should see two line items on my bill: First, the basic line / core-network charge, for getting the cable to my home and maintaining that service. Second, the internet service charge, for providing internet access. And if I want an additional data services, like a land line or TV programming, those services can be separate line items... but they don't need to include the basic line as part of the pricing.
Done right, this fixes a lot of things for telcos with regards to getting pricing right. Of course, they're used to running as a monopoly, so they'll screw it up.
[ link to this | view in thread ]
Re: Re:
[ link to this | view in thread ]
Re: Hundreds of satellites will NOT happen.
Just make up some numbers. 500 * $10,000,000 each to make and launch (I deliberately choose a ridiculously LOW number for cost) = 5 billion, but doesn't include the ground-based equipment... Point is, where's the payoff over the vastly cheaper already installed ground-based? Picking up cattle ranchers in Wyoming isn't going to cut it.
Oh, and I'm just ignoring the technical difficulties with coverage, reliability, cost of ground based receivers, and so on.
That project will not happen, any more than Musk's notion of boring tunnels under LA: too costly, too risky.
For a laugh, remember that clown proposed heating Mars by exploding fusion bombs at the poles. He seems to get ideas from 1940s science-fiction with lurid covers.
Musk is a nutter who somehow got rich -- first from out of the blue had enough to buy into Paypal -- and then somehow has enough cred in DC to get trusted for rocket launches, besides 5 billion in subsidies every year, besides tax breaks for those buying the cars...
[ link to this | view in thread ]
Re: Re:
[ link to this | view in thread ]
[ link to this | view in thread ]
Re: 70 + 89 = 159. WHERE the "hundreds of billions" in title?
Wow, talk about head in the sand. Have stock in ATT I see.
[ link to this | view in thread ]
[ link to this | view in thread ]
Re: slightly more than 8
Satellite is great and all but the latency sucks for gaming :/
[ link to this | view in thread ]
Re: Re: Hundreds of satellites will NOT happen.
We need more "nutters" like Musk to get our government looking forward instead of inward. Who cares where he got his money?
[ link to this | view in thread ]
Re: 70 + 89 = 159. WHERE the "hundreds of billions" in title?
lolwut?
[ link to this | view in thread ]
Aren't they the ones running Sling, aka "the guys who got OTT right"? Seems that would offset it quite a bit.
[ link to this | view in thread ]
Re: Re: Re:
But they're not competing on purely technical merits. They're competing against Comcast, Verizon and the rest who have used regulatory capture to gain government enforced monopolies.
And they're competing in rural areas where the same effective monopolies offer only cellular coverage and will kick people off the service by the thousands for using 1GB in a month.
Not to mention many places where cellular coverage simply isn't available. A VERY large fraction of the world. And not just the oceans, wilderness and second and third world countries. I have property in cottage country an hour and a half outside of Ottawa - in the direction of Toronto - and there's no cellular coverage in the area.
[ link to this | view in thread ]
Re: Re: Re:
Doesn't surprise me; I was thinking of the absolute bare minimum, and not really considering latitude.
As for the math, I got that from Wikipedia (with all of the disclaimers that should accompany that) which states that low Earth orbit has an orbital period of about 84-127 minutes.
With a number that high, I can't see more than five minutes between hand-offs, which, yeah... I don't know how they're going to solve the issue of latency during hand-offs.
[ link to this | view in thread ]
Re: Line Service vs Data Services
[ link to this | view in thread ]
Re: Re: Hundreds of satellites will NOT happen.
[ link to this | view in thread ]
I think it's a bit unfair to say that they're competent at this. As noted elsewhere in the article, they're much better at regulatory capture than they are at any form of market service.
[ link to this | view in thread ]
Re: Re: Hundreds of satellites will NOT happen.
This also ignores his other successful business ventures that have been revolutionary, such as Tesla.
All that aside, Musk isn't the only company planning to do this. There are at least two other separate companies that have basically the exact same plan. LEO satellites to provide high speed low latency internet service that can directly compete with cable.
And actually, the cost of getting such a service is not as much as you propose. They can launch multiple satellites on one rocket and with reusable rockets the cost is decreased. Additionally, they don't need to lay all the cables in the ground (which is VERY expensive and not at all cheap, especially when you start talking about providing coverage for 100% of the US) and the signal will be beamed to some ground stations, yes, but it will also be beamed directly to the customer's endpoint which is no bigger than a laptop.
And getting ideas from 1940s science fiction? Please, go look up how many of those absurd ideas are now actually a reality. Touchscreen interfaces? Computers that fit in your pocket? Virtual reality interfaces? Must I go on?
For your reading pleasure:
https://arstechnica.com/information-technology/2016/11/spacex-plans-worldwide-satellite-int ernet-with-low-latency-gigabit-speed/
https://arstechnica.com/information-technology/2017/05/spacexs- falcon-9-rocket-will-launch-thousands-of-broadband-satellites/
[ link to this | view in thread ]
Re: Re: slightly more than 8
Also, they are planning to put up an entire mesh that is fully redundant. They are really aiming high. Pardon the pun. :)
[ link to this | view in thread ]
Re: Re: Line Service vs Data Services
[ link to this | view in thread ]
Re: Re: Re: Re:
Also, by using LEO satellites instead of traditional geosynchronous satellites that traditional satellite ISP's use, and additional improvements in satellite tech, they claim to have solved the latency problem.
[ link to this | view in thread ]
[ link to this | view in thread ]
Re: 70 + 89 = 159. WHERE the "hundreds of billions" in title?
Please, tell me, what has ATT/Verizon done lately to completely revolutionize and improve consumer technology? Also, all of those companies get better customer satisfaction ratings than any of the traditional cable and telco companies.
[ link to this | view in thread ]
Re: Re: slightly more than 8
Traditional satellite internet uses Geostationary satellites. Since they're thousands of miles high, you get latency issues but, LEO is less than 5 milliseconds away at light speed so those issues start to disappear.
[ link to this | view in thread ]
There's far more than gaming
[ link to this | view in thread ]
There's far more than gaming
[ link to this | view in thread ]
Re: Re:
[ link to this | view in thread ]
Not just cable TV service that's cheaper
Sometimes even the combination of cable TV service and broadband is cheaper than broadband alone.
[ link to this | view in thread ]
Re: Re: What about AT&T's merger commitments?
[ link to this | view in thread ]
Re: 70 + 89 = 159. WHERE the "hundreds of billions" in title?
[ link to this | view in thread ]
Re: Re: Re:
Is that a limit of physics or just of our current technology?
[ link to this | view in thread ]
Re: Re: Re: Re:
[ link to this | view in thread ]
Re: Line Service vs Data Services
It only works well if those are run by separate companies, like in the UK. In places like Canada, the incumbents inflate this "basic" cost when they legally have to let third-party providers use it, but then undercut it when their own services use it. E.g., a third-party provider using Bell fiber pays more for the fiber than a direct Bell customer would pay for fiber+internet (with the current "interim" tariff).
[ link to this | view in thread ]
Re: What about AT&T's merger commitments?
About ten years or so ago, an AT&T employee told me that, as part of an agreement with the federal government, they were going to start introducing a flat $10/month DSL to particularly add access to the underserved areas.
Some years ago (2009 maybe), I discovered that ATT was offering naked DSL in my area. At the time I was paying something like $20.00 for a landline that I was required to have and $19.99 for DSL. Crappy DSL! I cancelled the landline, signed up for naked DSL thinking I'd save $20.00/mo but this is ATT and that would be crazy.
$5.00 for naked DSL. Yep, they charged me for, um, nothing. Followed by a $5.00 increase each year to my DSL.
In 3-years the fee was exactly what it had been before and I didn't even get a landline. And, I didn't even discuss how ATT F'd up my billing by giving me a new account # for naked DSL.
So, before those customers knew it, their $10/mo was probably $29.99/mo.
[ link to this | view in thread ]
The core problem affects all publicly traded companies
>One of the core problems here is that Wall Street isn't satisfied with ISPs simply doing a good job and making a reasonable profit. The relentless, myopic need for quarterly improvements has companies like AT&T and Verizon
It's not just AT&T and Verizon, all companies are suffering from stockholders demands for stock prices to rise every single quarter.
The perfect example of the tail wagging the dog has caused nearly immeasurable damage to US manufacturing and workers.
First you cheapen the product to raise profits and stock price, then you have to drop the price of the goods, then you have to stop providing health insurance, raises, retirement packages, Once you have squeezed all you can out of the workforce, you outsource to a country with cheaper labor rates, after THAT quarter, you start selling off assets until there is nothing left.
So many companies have gone down that drain. And yet privately held corporations like Koch industries are wildly profitable and extending their reach into new markets every day.
[ link to this | view in thread ]
Re: There's far more than gaming
However, it seemed much more convincing the first time.
[ link to this | view in thread ]
Re: Re: Re: Re:
The ground set up would require at least two, and preferably three or more steerable beams, so as to be able to keep contact with at least one satellite. To get a narrow beam, even at satellite frequencies needs large panels (electronic beam steering) or dishes, several meters in diameter. This would be a rather expensive, complex and non portable ground system, and likely highly visible installation with all the problems of acceptance that can bring.
[ link to this | view in thread ]
Re: Re: Re: Re: Re:
And yet that's not the case with phones on existing satellite constellations.
Sure, the original Iridium system had low bandwidth. But that was 20 years ago. Globalstar has done better, and SpaceX has been launching Iridium NEXT satellites ten at a time.
Satellite phones tend to look like 15-year-old cell phones. No large panels or dishes needed.
[ link to this | view in thread ]
Re: What about AT&T's merger commitments?
[ link to this | view in thread ]
Re:
There's also going to be a lower bandwidth CAP, and lower speeds then a Wired Connection. That's just a given.
I think you're going to be disappointed at this over hype.
[ link to this | view in thread ]
Re: Re: 70 + 89 = 159. WHERE the "hundreds of billions" in title?
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re:
The only way to improve that is to serve a smaller number of customers per satellite, which means the customers signals must be directed at a particular satellite, and not all those that are above their horizon. That is where tight beam steerable beam aerials come in.
That is as I keep pointing out, satellites are good for serving remote, low population density areas, but are not suited to serving high population density areas, at least with simple aerials which can see and work all the satellites above the horizon, but cannot deliberately select one of a dozen satellites above the horizon.
[ link to this | view in thread ]
Re: Re: Re: Line Service vs Data Services
No one in their right mind are going to pay all these different company's $6, $15, $35 a month for their service. I laugh at CBS All Access. Not worth it. HULU+? Not worth it. I have Netflix, Amazon Prime, though I rarely watch anything there, got it for free 2nd day shipping. Then I have my Antenna I put up. The only one I can see anywhere around my house from my roof.
I already have to much crap to watch that what I can't see, I don't miss anyway. If you don't have enough content with a Antenna and Netflix, (Limit yourself to 1 at most 2 streaming services)and that's it. If you have any free time after that, go for a walk. Visit friends. Play on your Game Console. Read a Book. Do a little Gardening. Clean your Garage. I'd go out Jetsking, but I need to fix my Engine, so I go out for a ride on my Motorcycle to anywhere, just use a Motorcycle navigation App, and tell it to do a round trip from my starting place for a 2 hour ride, or a 3 hour ride, lots of turns and let it do it's thing, and go to someplace new.
My Dad retired and lives with me, and watches to much TV. All the crap my TIVO is recording from the Antenna almost non-stop. At least he's gone on Vacation this week.
The point being, To much TV is a bad thing. Signing up for a bunch of services is not going to be cheaper then cutting the cord. I signed up for DirectTVNow for 3 months to get the Apple TV 4. I watched it only a couple times in those 3 months. There's no DVR. It's like going back in time before the TIVO. Except you can't channel skip. You have to watch on the show airs. It's worse then normal antenna TV.
[ link to this | view in thread ]
Re: Re: Re:
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re:
[ link to this | view in thread ]
Re: 70 + 89 = 159. WHERE the "hundreds of billions" in title?
[ link to this | view in thread ]
Re: Re:
Given that they are stating they can reach speeds averaging at Gbps instead of Mbps and latencies of 25-30 ms, I don't see how you can't do gaming quite well on that kind of a connection.
I do CS:GO and Overwatch on 60 Mbps and around 50 ms latency with cable. If they can provide the performance they claim, I will ditch cable in a heartbeat.
If you want to know more about what they propose to do and tech details on how they are going to do it, please read here:
https://arstechnica.com/information-technology/2016/11/spacex-plans-worldwide-satellite-int ernet-with-low-latency-gigabit-speed/
and here: https://cdn.arstechnica.net/wp-content/uploads/2016/11/spacex-Technical-Attachment.pdf
[ link to this | view in thread ]
Re: Re: Re:
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re:
The way to increase overall bandwidth is to increase the density of reuse of the available channels, which mobile phone systems do but using lower power lower towers with smaller cells each covering fewer customers.
Essentially satellites cannot reduce their footprint on the ground to anything approaching the size of a mobile cell in a city. Therefore the ground station have to reduce their footprint on the sky, so that only one satellite (per aerial) is inside the footprint. That way the available bandwidth is multiplied by the number of satellites above the horizon for any location.
A base station the size of a laptop will not be narrow beam, and so they will be dividing the available channels between all the satellites above the horizon, and so while this will work fine for a remote mining community in the rockies, or the scientists in Antarctica, this will not replace mobile in any city, never mind DSL or cable.
It is also worth noting that the bandwidth available to customers is limited by the number of satellites which can see a ground station with an Internet connection, and how many other satellites are being routed via that satellite.
Now, a reliable service into rugged remote areas runs into the problem of location only being able to see a smaller area of sky, and a higher density satellite swarm is needed to ensure that any location can see several satellites all of the time.
I am not saying satellites won't be useful, they will provide broadband Internet to areas where the population densities are low, and the telcos do want to provide service because there are not enough customers per square mile.
There are probably more than enough customers spread out in remote and difficult terrain, the Andes, the Rockies, norther Canada, Siberia etc. plus Ships and Aircraft to make the system viable. It would also be attractive to long distant lorry drivers, so that they have Internet wherever they park up.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re:
WiFi mesh networks use clusters of AP's on the same channel and that is essentially what they are planning to do here. Yes it cuts down on bandwidth but that's why you use multiple AP's so that you get fewer users per AP.
We aren't talking about them throwing up a few dozen, maybe even a few hundred satellites. We're talking about them throwing up THOUSANDS of satellites. Total satellites orbiting earth was estimated to be around 1,459. SpaceX wants to put up over 4,425 in addition to what's already up there.
I really recommend reading this article for an overview of how they plan to accomplish it: https://arstechnica.com/information-technology/2017/05/spacexs-falcon-9-rocket-will-launch-thousands -of-broadband-satellites/
And this article is SpaceX's technical details of how it will all work. I haven't read the whole thing but in skimming they seem to touch on some of your concerns: https://cdn.arstechnica.net/wp-content/uploads/2016/11/spacex-Technical-Attachment.pdf
[ link to this | view in thread ]
Re: Re:
[ link to this | view in thread ]
[ link to this | view in thread ]
Re: 70 + 89 = 159. WHERE the "hundreds of billions" in title?
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
Thanks for the reference, and guess what the are using steerable beam aerials from the satellites. Note however that the satellite ground footprint has a 1,060kM radius. This isn't going to replace broadband in the cities, but will serve all those place in the US that have no or poor service presently, or those who spend a lot of time away from home, like long distance lorry drivers and traveling sales people.
Also note that the intention is to have 200 ground stations and by playing with polarization up to 8 satellites can connect to a single ground station at any one time. That means the whole of the USA is being served by the bandwidth equivalent to a few hundred fibers, (certainly less than one thousand). That is less bandwidth than any major data center, and less than is needed for most cities.
This is a nice system to fill in the areas that are not currently served, or not well served by the existing broadband providers. It could also be a nice system for anybody who spend a lot of time away from home, like long distance lorry drivers etc. giving a single reliable Internet connection when they are parked up.
This system is not going to kill the incumbents where they provide a broadband service, it just doen't have anything like the capacity to serve a single major city.
[ link to this | view in thread ]
Re: Re: Re: Re:
Or at least in my dreams...
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
Yes, I see your points about the ground stations and the footprint radius but the receivers are also going to be using steerable beams so they could point to a different satellite if one is saturated. In fact I believe they even mention something to that effect in the paper.
The ground stations are in addition to beaming directly to people's home receiver. Streaming video is very high bandwidth, yet current satellite tv providers have no issues with high density population areas streaming tv from only a few satellites.
I could potentially see your point if you were talking about cities like New York City or LA where you have millions of people in a very small area, but those are only a small portion of the US. The entire rest of population is more spread out.
If you are correct then I don't see this being viable at all except for areas like Antarctica because essentially the 1,060 KM radius basically covers an entire state. If it can't support one state then it won't work in the US at all because most states have at least one high density area. You might be able to do Alaska but not the rest of the country.
My understanding is this will operate in a mesh with multiple satellite footprints overlapping each other. That way if you do saturate one satellite, all the remaining connections can fail over to 1, 2, 3, or more satellites and easily handle the bandwidth from there.
I think you are correct in your understanding of how satellite technology works but are missing the scale and mesh capabilities that they are deploying it with.
[ link to this | view in thread ]
Re: Re: Re: 70 + 89 = 159. WHERE the "hundreds of billions" in title?
Bell rushed their "Fibe" infrastructure out so quick in my city that my head swam, even on our crappy apartment. Shame Bell's so expensive, I need my arms and hands to type. :P
[ link to this | view in thread ]
Re: The core problem affects all publicly traded companies
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
Those ground stations are a hard limit on capacity, and I suspect that more would not increase the systems capacity. As a satellite link has less bandwidth than a single fiber, the USA will be served by the equivalent of less than 1,000 fibers, and any area within it by far less than that.
What this system will do is fill in all the remote communities that the current ISPs have no intent of serving, along with those who spend a lot of time away from their home, like long distant lorry drivers. That will be a big enough customer base to make the system viable as a commercial venture.
[ link to this | view in thread ]
Re:
[ link to this | view in thread ]
Re: The core problem affects all publicly traded companies
Now, you can find a few companies using about 10 % of their budget consistently on research (which far too often negates the blow to workers conditions and the constant savings, since the new products will provide a higher value and thus automatically make it easier to raise margins). But these companies are unfortunately few and far between, just as they are too often living a quiet life with static stockprices and low volumes.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
I agree that the paper does state the frequencies will not overlap but even with that, it's still technically a mesh network. The satellite will be in low earth orbit, which means that unlike traditional geosynchronous satellite internet, these satellites are going to be moving all over the earth. The satellite you connect to today will likely not be the same satellite you connect to tomorrow or even 30 minutes from now. They have to operate in a mesh network otherwise you would drop connection every 15 minutes to reacquire the signal.
I also don't understand why you think adding more ground stations wouldn't increase capacity. The more ground stations you have, the more you can spread your links out meaning you don't have to saturate one satellite. You can shunt some of the traffic to an additional satellite that transmits to a separate ground station, thereby increasing overall bandwidth.
Additionally, why would they state that (loudly and proudly) that users will be able to get gigabit speeds at latencies of 25 - 30 ms if they knew that wasn't anywhere near realistic? What does it gain them? If all this system can do is service remote communities that current ISP's don't intend to serve, then how is it any different from current satellite ISP offerings? Given that, why go to all the trouble of launching almost 4 times the number of total operating satellites at altitudes thousands of miles less than current satellites?
Your arguments make sense if the company was simply doing what's already been done in satellite internet but they are not, they are completely changing the game. Not having to travel 26,000 miles, instead only travelling 1,000 miles is going to VASTLY decrease latency. And having 4,000+ satellites zipping all over the world with 200 ground stations should have no problem supplying plenty of bandwidth. If it doesn't, all they have to do is launch a few more satellites and build a few more ground stations.
[ link to this | view in thread ]
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
That is 200 ground stations for the USA alone, but also for the whole of the USA, with the limitation that each ground station can only support a maximum of 8 satellites at a time. Each satellite has a footprint with 1,060km radius, and the system is designed such that their is no frequency, (channel) re-use within that footprint. The initial system gives maximum capacity in the areas that it overs, and more satellites are planned to increase coverage to areas like Alaska which are not covered by the initial system. They will not however increase capacity in the areas covered by the initial system. Indeed the system design includes management of which satellites are active at any time, to avoid them interfering with each other.
While 200 *8 might seem like a lot of capacity, it is not all that impressive when that is to cover the whole of the USA, and with the modern Internet, fairly puny when compared to the fibers connecting any large city.
Also note, of those 4,000 satellites zipping over the globe, at any one time, a large number of them will be covering oceans, and still more covering other continents, and only a thousand or so will be over the USA.
Compared to existing satellite Internet, this system would be a huge increase in capacity, around a thousand satellites serving the USA, as opposed to single digit numbers. This is an impressive boost to the capacity to serve remote areas where the current ISP's don't go, but no threat to the current ISPs where cable and fiber are available.
[ link to this | view in thread ]
You need to regulate businesses.
In this specific example, the first and best step would be to separate carriers from content. ABSOLUTELY. Oh, you maintain the wires? You can't offer media. At all. And not holding companies either, we find that out and you're fined $50k a day for each violation.
You'd get your net neutrality for sure. There's plenty of good money -- not great, but good money -- in being in infrastructure. It requires capital to invest, but you have a captive audience of sorts. But if your only interest is maintaining the wires, you can't get involved in the fuckery of limiting TV channels, Internet sites, and so forth.
Oh yeah.
-C
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
You also can't keep shifting the goal posts. First you said that the satellites wouldn't have enough capacity, when I pointed out that there should be no problem with capacity on the satellites, you said "The Limit on the satellite system id its connection to the Internet backbone, which for the USA, is 200 ground stations....Those ground stations are a hard limit on capacity". So then I pointed out that the capacity of the ground stations is also very high and explained why having more ground station would increase capacity. Now you are trying to say that the limit is the satellites again. You can't have it both ways. Current satellite internet supports the entirety of the US, no questions asked, no limits on how many people can sign up. You're telling me that this new system, with vastly better coverage and capacity, can't do what the current satellite system has been doing for the better part of the last decade?
Just doing some quick math, each satellite is stated to have a maximum downlink capacity of 23 Gbps multiply that by 8 satellites per ground station multiplied by 200 ground stations and you get a total capacity of 36,800 Gbps or 36.8 Tbps. You're telling me that's not enough to support the entire network needs of the US given that traditional ISP's generally oversubscribe their networks by about 100:1? Taking oversubscription into account, that means that the SpaceX satellites could oversubscribe up to 3,680,000 Gbps or 3,680 Tbps or 3.68 Pbps. Really? That's not enough capacity for the entire US?
And we're talking if everyone in the US decided to jump ship and sign up. There's no realistic probability that would happen. However, just having that option and having even a quarter of current cable internet subscribers jump ship would be a huge threat to incumbent ISP's. That's a quarter of their current and future profits down the tube.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
What I am saying, it that the system will not have anywhere near the capacity to serve large towns and cities. It capacity is approximatively 7 million video streams for the whole of the USA, or about 4 thousand stream per satellite. (using 5Mbps per 1080p video, where they use 3.5 to 8 Mbps per stream).
Lets not confuse coverage, the whole of the united state, and capacity and therefore supportable customer base. What I keep on saying is that this system will not serve the population of towns and cities, not that it will not cover them. 7 million video streams distributed across the USA is a nice coverage filler, 4 to 12 thousand stream capacity, for a footprint sized area, depending on details of frequency allocations is not going to worry the ISPs who serve the larger towns and cities, and if a quarter of their subscribers jump ship, they will soon be back because they will have overloaded the capacity of the satellites.
By the way an 100 time over subscription, that is 100 time more subscribers than ability to make connections worked for phones, where use is occasional for any subscriber, but does not work for broadband where for part of the day many of your subscribers are online and streaming content of one sort or another. 3 or 4 time over subscription maybe, 1o times, and people will see buffering at peaks demand times.
[ link to this | view in thread ]
Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: Re:
However, you're still contradicting yourself. Again, current satellite internet provides service to the entire US, irregardless of how many people sign up and they don't seem to be worried about not having enough capacity.
Also, I don't think you understand how internet service works. Google Fiber offers gigabit speeds to several major cities all over the US. By your logic, their capacity would have to be several million Gbps per city. That just doesn't make any sense. No ISP has that much capacity. It was estimated in 2014 that 255 Tbps would be more than enough to handle the entirety of global internet traffic. Your understanding of how network service works is not realistic.
And 100:1 oversubscription may be on the higher side, but 50:1 is not uncommon and certainly 12-20:1 is extremely common. ISP's can't realistically serve everyone on a 1:1 or even 4:1 basis. It's called network management. They figure out what the likelihood of users hitting their max rate is and oversubscribe based on that because the vast majority of users will never hit that upper limit for long enough times to create a problem.
I am a heavy gamer and Netflix user on a 60/5 cable connection. I can support multiple HD Netflix streams and multiple gamers playing FPS games with no issues at all. The only time the network reaches full capacity and starts bogging down is if I am doing multiple sustained downloads of large files. This is a localized example but the principles are the same. You have a pipe of X size that has so much capacity but that doesn't mean you can't get a lot more people on it than just what the math works out to because of proper network management.
[ link to this | view in thread ]
Re: Re: Re: Line Service vs Data Services
[ link to this | view in thread ]