Fibre versus Starlink Discussion

[banned-66611]

The idiocy, it burns! SpaceX (Starlink) and Telesat were invited to give testimony in front of Canadian politicians on rural broadband access.

Starlink isn't a good solution for rural broadband. It's better than nothing but countries like Canada should be aiming much higher, otherwise people in rural areas will just be stuck with outdated broadband again.

Fibre is the only technology fit for the next decades, gigabit+ speeds and ultra low, consistent latency. Starlink is great and all but the speeds and latency are already on the low side and will only continue to fall behind as fibre becomes more common and the Starlink system gets more subscribers (and therefore less bandwith per user).

 

[ItsNotAboutTheMoney]

Fibre is the only technology fit for the next decades, gigabit+ speeds and ultra low, consistent latency. Starlink is great and all but the speeds and latency are already on the low side and will only continue to fall behind as fibre becomes more common and the Starlink system gets more subscribers (and therefore less bandwith per user).

That is total rubbish.

For a start, a large proportion of the world's population uses a cellphone for Internet access, so clearly fiber isn't necessary to have a large impact.

As long as the Internet service is fast enough to stream video and audio, providing access to government services, education and videoconferencing the essentials are covered. With Starlink, it is. Anything better is just gravy.

One of the best side-effects of services like Starlink is to shut people up about the need to subsidize rural broadband.

 

[Cosmacelf]

Starlink isn't a good solution for rural broadband. It's better than nothing but countries like Canada should be aiming much higher, otherwise people in rural areas will just be stuck with outdated broadband again.

Fibre is the only technology fit for the next decades, gigabit+ speeds and ultra low, consistent latency. Starlink is great and all but the speeds and latency are already on the low side and will only continue to fall behind as fibre becomes more common and the Starlink system gets more subscribers (and therefore less bandwith per user).

I think you are underestimating the cost/benefit of pulling fiber hundreds of miles to serve a small hamlet. There are a lot of those isolated places and, indeed, there always will be. You do realize a goodly portion of the earths population uses local diesel generators for electricity? It’s an expensive way of producing electricity, but they do it because running power lines from a more cost effective central power source just isn’t economically feasible.

You are also thinking Starlink is like geosats. When/if Starlink becomes saturated, they will simply loft more cheap satellites to fill in bandwidth. Also note that unlike geosats, Starlink serves the globe. The Galapagos island are nice to visit, but any kind of telecom there is a challenge (and Ecuador is not going to spend the $$ for an expensive undersea fiber cable any time soon). Starlink will immediately give a fairly large population of people access to high quality Internet.

Finally, Starlink has use cases that fiber will never be used for. Mobile, airplane, oceans, emergency response, “temporary” resource extraction (think far flung mining operations) to name a few.

Edit: As I was saying ... (Greece has a lot of Islands where pulling fiber cable isn’t economically feasible): Read this article with Text Fish : Musk’s SpaceX to bring Starlink internet to Greece | Kathimerini

 

[bxr140]

Starlink isn't a good solution for rural broadband. It's better than nothing but countries like Canada should be aiming much higher, otherwise people in rural areas will just be stuck with outdated broadband again.

Fibre is the only technology fit for the next decades, gigabit+ speeds and ultra low, consistent latency. Starlink is great and all but the speeds and latency are already on the low side and will only continue to fall behind as fibre becomes more common and the Starlink system gets more subscribers (and therefore less bandwith per user).

In addition to what's been said already, I think it would be a worthwhile exercise for you to actually play out this logic. Actually develop a schedule with some practical user demand over time. I think what you'll find is that the current incarnation of Starlink pretty well serves the mission of providing broadband to those under/not served by terrestrial infrastructure, and if your projection is in any way realistic it should be easy to contemplate a future incarnation that can pretty efficiently and, given SpaceX's MO, pretty quickly scale to demand. Future being things like: going higher in spectrum, using larger satellites, using more satellites, leveraging tech/silicon advancements, etc. Remember that each starlink is really only good for ~5 years--after that it gets replaced with whatever makes sense at the time.

Bigger picture, terrestrial infrastructure will also be pushing into those under/not served areas over time as well, effectively reducing the potential user base of satellite internet, which is actually a bit of an upside for users of satellite internet since generally fewer users means everyone gets a bigger piece of the pie. We'll likely see a blurring of mobile and fixed wireless along and beyond the fringes of hardwired service with the upcoming increase in spectrum (as mobile service jumps to/above current terrestrial speeds), and combined with concepts like airborne or mega-tower broadcast (like FB's Supercell) one could imagine choice actually being a reality in low density regions, with ultra low density regions pretty much always being served by satellite.

Like the future of energy there's no one solution to the problem of connectivity; what needs to be implemented is an aggregate of multiple solutions.

 

[DMC-Orangeville]

Starlink isn't a good solution for rural broadband. It's better than nothing but countries like Canada should be aiming much higher, otherwise people in rural areas will just be stuck with outdated broadband again.

Fibre is the only technology fit for the next decades, gigabit+ speeds and ultra low, consistent latency. Starlink is great and all but the speeds and latency are already on the low side and will only continue to fall behind as fibre becomes more common and the Starlink system gets more subscribers (and therefore less bandwith per user).

Really?

Perhaps you aren't familiar with Canadian geography - understandable. Canada is the second largest country in the world, by land mass. 85% of the population of 38 million people are within 100 miles of the US border. Outside of this area (Edmonton aside), it is a huge land mass, with very few people, separated by 100's of miles apart. There are thousands of small communities that cannot be reached by road, as they are too remote. Running fiber to them is an impossible job.

We can't even get many of the indigenous settlements clean water (bad on us). A lot of them run generators, and some have switched to solar (if not too far north - winter brings very short days). - It's too far and expensive to run power to them. Same with fiber.
Heck, I'm just outside of Toronto, a city of 7 million, and I can't get fiber.........so your statement is correct for those (like me) who are in that 100 mile "banana belt", but not for the really remote communities

Satellite internet Starlink and/or Telesat (right @bxr140 ) will be a godsend, and currently the best foreseeable option, to these communities.

 

[banned-66611]

That is total rubbish.

For a start, a large proportion of the world's population uses a cellphone for Internet access, so clearly fiber isn't necessary to have a large impact.

As long as the Internet service is fast enough to stream video and audio, providing access to government services, education and videoconferencing the essentials are covered. With Starlink, it is. Anything better is just gravy.

One of the best side-effects of services like Starlink is to shut people up about the need to subsidize rural broadband.

I was talking about in Canada, not the rest of the world.

Will it be fast enough to work from home? Maybe not too bad today but it will slowly get worse and worse as file sizes increase or people want higher quality RDP. That's the issue, it just pushes the divide back a bit but doesn't solve it.

Fibre is a forever solution.

 

[banned-66611]

I think you are underestimating the cost/benefit of pulling fiber hundreds of miles to serve a small hamlet. There are a lot of those isolated places and, indeed, there always will be. You do realize a goodly portion of the earths population uses local diesel generators for electricity? It’s an expensive way of producing electricity, but they do it because running power lines from a more cost effective central power source just isn’t economically feasible.

We are talking about Canada. They have phone service in these small hamlets. It was apparently worth running copper wire for phone and electricity to those places. In fact much of the infrastructure is already there, the same poles and conduits can carry fibre.

Japan has already done this. Even out in the sticks you get fibre. It's not that expensive and mostly re-uses existing hardware.

 

[ItsNotAboutTheMoney]

I was talking about in Canada, not the rest of the world.

Will it be fast enough to work from home? Maybe not too bad today but it will slowly get worse and worse as file sizes increase or people want higher quality RDP. That's the issue, it just pushes the divide back a bit but doesn't solve it.

Fibre is a forever solution.

"As file sizes increase".

Why would file sizes increase? It's a solution looking for a problem.

Without Gigabit we already have video streams fast enough and with low enough latency for remote desktops, streaming video and videoconferencing at reasonable resolution. Nothing else is needed. I know this because we're doing it over ADSL that isn't particularly good ADSL.

 

[ItsNotAboutTheMoney]

We are talking about Canada. They have phone service in these small hamlets. It was apparently worth running copper wire for phone and electricity to those places. In fact much of the infrastructure is already there, the same poles and conduits can carry fibre.

Japan has already done this. Even out in the sticks you get fibre. It's not that expensive and mostly re-uses existing hardware.

Japan: 0.146 million mi²
UK: 0.094 million mi²
Canada: 3.855 million mi²
USA: 3.797 million mi²

The UK has companies doing rural fiber and reasonable cost using volunteer labor and community-wide commitment. But there are different definitions of rural in different countries.

 

[bxr140]

Will it be fast enough to work from home? Maybe not too bad today but it will slowly get worse and worse as file sizes increase or people want higher quality RDP. That's the issue, it just pushes the divide back a bit but doesn't solve it.

Again, play it out. What do you think load will be in 2030, for instance?

 

[banned-66611]

Again, play it out. What do you think load will be in 2030, for instance?

Well for example work had to give me a laptop because RDP was unusable on a 150mbit internet connection. Too much latency, even editing source code was a real pain, and even if I dropped the resolution right down to 2k which reduced readability.

Now I'm back to being productive with 2x4k screens.

 

[ItsNotAboutTheMoney]

No we don't, the UK is an absolute backwater for broadband.

Have you not heard of B4RN?

Japan may be smaller but the terrain is some of the most challenging in the world.

Canada has much, much lower population density, especially in rural areas (consider Supercharger location challenge in Ontario wilderness), and has more severe weather for wired infrastructure to handle. Areas of Japan get a lot of snow, but they aren't as cold.

As an example of the challenge, here in Maine when there's a strong storm, such as Saturday's Nor'easter with heavy rain to 20" of wet snow and 20+mph winds gusting to 40 mph, people have outages because trees fall and take down power lines. If they take down power lines, they take down telephone lines.

One of my wife's colleagues has lost all wired services for 3 days. You can use a generator to replace power, but you can't use it to replace wired telecom.

 

[ItsNotAboutTheMoney]

Well for example work had to give me a laptop because RDP was unusable on a 150mbit internet connection. Too much latency, even editing source code was a real pain, and even if I dropped the resolution right down to 2k which reduced readability.

Now I'm back to being productive with 2x4k screens.

You didn't answer his question.

Your problem is your latency. Latency is the key reason why geostationary satellite Internet is inadequate.

In Starlink or other LEO services low latency is part of the design, because it's known how important it is for modern Internet use.

 

[Cosmacelf]

No we don't, the UK is an absolute backwater for broadband.

You need to do some research. B4RN (B4RN: The World's Fastest Rural Broadband) is a stellar example of rural coops installing their own fiber to the home networks. On an even smaller scale and even more rural, I documented a 200 person farming community that installed their own fiber network here: FTTH.Build

That sort of can do attitude is very refreshing to see and appears to be a uniquely British thing. I haven’t seen many other communities that literally built their own FTTH networks, even though it is entirely possible.

 

[bxr140]

Well for example work had to give me a laptop because RDP was unusable on a 150mbit internet connection. Too much latency, even editing source code was a real pain, and even if I dropped the resolution right down to 2k which reduced readability.

Now I'm back to being productive with 2x4k screens.

As noted your issue is latency...Unless you're saying 150mbit service is inadequate for editing code...

And, just as with the parallel RV conversation going on in this thread, you may want to contemplate the difference between an edge case and an average case. There are OF COURSE Starlink users that will require boatloads of service to do whatever it is they do...but when factored against all the other users, the vast majority of whom are fixin' to binge the latest season of The Crown or whatever, the total impact on the system from those edge/corner users should be quite manageable.

So, again, how do you envision system load/demand evolving over the next decade?

 

[hmcgregoraz]

Networking is not a one size fits all situation.

Fiber to the home makes a LOT of sense in most areas, even rural areas that are spread out, as long as the cost to physically install can be kept reasonable. Fiber to the home in most city and suburban areas makes a ton of sense, but the telcos are going VERY slow with this.

Personally I feel that in most cities the "water department" should be building fiber networks anytime they are doing main replacements, and the physical layer of the network should be run as an essential service like water service. The transport layer (layer 3) should be provided by commercial and co-op groups on top of the city owned fiber.

While I tend to dislike Home Owners Associations (HOAs), I feel for new developments, the HOA should put in a active or passive fiber network, and probably HOA managed WiFi that covers everyone everywhere in the community. The HOA could then go out for bit to bring in bandwidth to the community as a whole, and do that every couple of years. Heck if it's large enough, they could get their own IPv4 (not cheap) and IPv6 allocations, as well as AS number.

Even a small community in rural Canada could probably justify a Passive Optical Network deployment. What they CAN NOT justify is the long distance fiber to get to the community, let alone a redundant fiber that loops from the community through other communities and back to real civilization. For these communities Satellite based up-link will be the norm for decades to come.

Individual residences and businesses will be using LEO and GEO satellite as the only practical choice due to the costs of long distance fiber, even once communities and telcos and cable cos get their head out of their rear end and start deploying fiber networks where economically practical.

One friend of mine had (since sold) a house that was ~2500 feet from Comcast Cable (HFC, coax network), this is in an area with above ground utilities, pole mounted, not under ground, no trenching involved. Comcast wanted $28,000 to build the network out the additional 2500 feet, for him to be able to become a paying customer. So I don't expect Telco and Cable cos to really step up.

If you look a map of where Century Link offers passive fiber in Tucson, it is TINY. Take a look at the map for your self: https://www.google.com/maps/d/viewer?mid=1Etqi-VRSYrlvgkpwibGofoNtmKtyOByP&ll=32.17381955262175%2C-110.9551205&z=9

Latency is the most important factor for non-streaming applications, and StarLink will solve that properly. The ability to "cheaply" expand the amount of bandwidth by launching more satellites (and yes, when compared to long distance fiber to remote locations or even suburban passive network deployment, it is cheap per customer), StarLink will be an amazing addition to the bandwidth options around the world.

A LEO network gives one thing that most people miss, it flips the normal "density = cheap" bandwidth para-dime on it's head. The more in the middle of no where a location is, the more excess bandwidth SpaceX will have in that location, once they have more ground stations and ISLs, the incremental cost of adding these users will be far lower than the incremental cost of adding dense population users.

Basically it is not StarLink vs Fiber, it is StarLink + Fiber. Fiber where and when we can, StarLink with ground stations where we can not.

Add to that, for specific applications of very long distances with Inter Satellite Links (ISL), latency will actually be lower on StarLink vs Fiber, as the speed of laser light in space is actually far faster vs the speed of light in optical fiber. For most users this won't matter, but for certain financial users, they will pay greatly for this service.

-Harry

 

[banned-66611]

Have you not heard of B4RN?

Yes. Fibre is only available to a tiny fraction of the UK, most people are stuck on copper.

 

[banned-66611]

You didn't answer his question.

Your problem is your latency. Latency is the key reason why geostationary satellite Internet is inadequate.

In Starlink or other LEO services low latency is part of the design, because it's known how important it is for modern Internet use.

Then why is latency on Starlink so poor? On fibre I'm seeing sub 1ms pings to sites like Google, which is a small fraction of the time it takes the signal to even get to the satellite.

It should hopefully be obvious that if it has to go first up and then back down it's never going to be as fast as taking a more direct route to the nearest datacentre.

 

[banned-66611]

You need to do some research. B4RN (B4RN: The World's Fastest Rural Broadband) is a stellar example of rural coops installing their own fiber to the home networks. On an even smaller scale and even more rural, I documented a 200 person farming community that installed their own fiber network here: FTTH.Build

That sort of can do attitude is very refreshing to see and appears to be a uniquely British thing. I haven’t seen many other communities that literally built their own FTTH networks, even though it is entirely possible.

Do you have any idea what proportion of the UK they actually serve?

 

[banned-66611]

As noted your issue is latency...Unless you're saying 150mbit service is inadequate for editing code...

I am saying that. Sure with a static screen only editing code it's okay, but when you switch tabs or windows etc. a large amount of data needs to be sent. Clicking a link in the browser creates huge delays as the whole screen has to refresh.

In any case, as for load over the next few years, well the current generation of consoles are already coming in digital only versions with games already in the 60-70GB range for the initial install. We can expect that to keep growing.

YouTube 4k videos are clocking in around the 40-50Mb/sec range, which will continue to increase as 120 fps and 8k become more common. And then you have VR and streaming games. You may think these things are unimportant but they all go towards creating a gap, and as new technologies become available it will only continue to increase.

Remember that Starlink will continue to get slower as more users are added to the system. It's shared bandwidth over the air, right now the low number of users is an advantage but that will slowly fade.