Fibre versus Starlink Discussion

[banned-66611]

They're supposed to start on my half next year. If my neighborhood didn't meet their ROI requirements, how in the world would a couple of homes out in the sticks???

2. My SIL lives in the sticks on several acres. Power and POTS but no cable. The local power company, in a low-density county west of SAT, decided to run fiber on their poles recently to serve customers. SIL got it because her husband works for the power company. It took six guys in three trucks to manage the install. Maybe it was teething pains but I just don't see how that method would be economical at large scale. However, I think that the power company setting up "wireless nodes" occasionally on their poles would be economical with the last mile being shared wireless (something like 4G or 5G but tuned for internet access). That way they could run fiber to their nodes and drop off a wireless transceiver at any house that opted in.

I wonder how they calculate ROI. Seems unlikely that it wouldn't be worth it over a reasonably long time period for fibre. 50+ years at least.

If it took 6 guys to upgrade some copper to fibre then either they were doing some kind of teaching exercise or they are getting massively ripped off. I've seen two guys come and do it, and one of them was only there to make sure nobody walked under the work area why they were hanging the new cables.

 

[ItsNotAboutTheMoney]

Yes, AT&T paid $50B for DirecTv and now want/need to dump it for $20B since they have no ability to raise money to invest in new infrastructure. Their bad decisions have come home to roost and startups with clean balance sheets can outraise AT&T to build fiber, but alas, not on a large scale. It’ll take hundreds of startups, which will take time, but it’ll eventually happen. Meanwhile we’ve got Starlink!

They're trying to dump DirecTV in particular because satellite TV is dying, disrupted by OTT services on broadband.

The irony is that DirecTV was a disruptive company, as satellite TV was able to beat cable.
But broadband tipped the scales back to cable services.

And now we have advances that can allow LEO broadband.

But, there are also many people with mobile-only Internet. With people prioritizing their cellphones, it could be a big challenge for the economics of fiber-based ISPs.

 

[Cosmacelf]

But, there are also many people with mobile-only Internet. With people prioritizing their cellphones, it could be a big challenge for the economics of fiber-based ISPs.

Interesting thought.

 

[FoverM]

I wonder how they calculate ROI. Seems unlikely that it wouldn't be worth it over a reasonably long time period for fibre. 50+ years at least.

If it took 6 guys to upgrade some copper to fibre then either they were doing some kind of teaching exercise or they are getting massively ripped off. I've seen two guys come and do it, and one of them was only there to make sure nobody walked under the work area why they were hanging the new cables.

Don't know.

IIRC, the fiber went from the poles along the road 100 yards or so to the short power pole 20 ft from the house, then underground to the side of the house for distribution. Seemed like a lot of customized work to go the last 150 yards when a wireless node at the street and receiver in the house would have been simpler and more robust long term.

 

[banned-66611]

They're trying to dump DirecTV in particular because satellite TV is dying, disrupted by OTT services on broadband.

On wired broadband. Starlink will get hammered when that shared bandwidth is saturated by streaming, same as cellular broadband networks do. That's why many cellular networks have data caps and downgrade video quality.

 

[banned-66611]

Don't know.

IIRC, the fiber went from the poles along the road 100 yards or so to the short power pole 20 ft from the house, then underground to the side of the house for distribution. Seemed like a lot of customized work to go the last 150 yards when a wireless node at the street and receiver in the house would have been simpler and more robust long term.

Seems like a very odd way of doing things. Why go underground right at the end? Did they have to dig a new trench?

Odd that it would be custom work too, telcos are normally good at standardizing everything.

 

[hmcgregoraz]

Seems like a very odd way of doing things. Why go underground right at the end? Did they have to dig a new trench?

Odd that it would be custom work too, telcos are normally good at standardizing everything.

You missed the part that it was an electric utility, not a telco.

Electric utilities are used to working with very dangerous high voltage, and have large teams to manage that risk.

They did not reduce the size of the team for the Fiber deployment, also no mention if the Fiber was deployed in the "protected" high voltage space on the poles, or if it was deployed in the "safe zone" for telecommunications cables.

If it was deployed in the high voltage space, additional training, insurance, and capabilities are needed on the part of the installer.

-Harry

 

[banned-66611]

They don't usually deploy HV to homes, it's all low voltage.

 

[ItsNotAboutTheMoney]

Seems like a very odd way of doing things. Why go underground right at the end? Did they have to dig a new trench?

Odd that it would be custom work too, telcos are normally good at standardizing everything.

Could depend on bye-laws requiring underground cabling.

Or just for storm or other protection. Lowers the potential cost of repair if they don't need to run high cables to every property.

 

[banned-66611]

Could depend on bye-laws requiring underground cabling.

Or just for storm or other protection. Lowers the potential cost of repair if they don't need to run high cables to every property.

Sure, but then wouldn't they already have trunking there for the existing POTS and electricity stuff?

Anyway, all this can be sorted out with a bit of though. Once they have fibre in there they can not only sell internet access, they can sell video and phone service. They can sell businesses access to the trunk side as well, e.g. for cellular masts or wireless links for places that really are too remote for fibre.

Hard to see how it wouldn't be very profitable over a few decades.

 

[hmcgregoraz]

They don't usually deploy HV to homes, it's all low voltage.

That does not impact the placement on the pole along the primary poles.

Also depending on deployment, I have seen the 13KV go all the way to the last pole before hitting the house, with a pole mounted transformer at the house to step it down to more residential split 240V.

-Harry

 

[ItsNotAboutTheMoney]

Sure, but then wouldn't they already have trunking there for the existing POTS and electricity stuff?

Anyway, all this can be sorted out with a bit of though. Once they have fibre in there they can not only sell internet access, they can sell video and phone service. They can sell businesses access to the trunk side as well, e.g. for cellular masts or wireless links for places that really are too remote for fibre.

Hard to see how it wouldn't be very profitable over a few decades.

They aren't going to sell much phone service. By 2019 53% of the UK population had no landline, and 54% of the US population. People have cellphones.

They also aren't going to sell much video. People are using independent OTT services. In the USA, cable subscriptions fell by 6.6% in 2019.

There's also a large chunk of people who rely on mobile Internet, and I expect that to grow as the cost of data continues to fall. In the USA, Pew's 2019 survey returned 17% of people solely using mobile Internet.

Now, things could change if more people work at home via the Internet, so have a justification to pay for cable Internet in addition to their mobile, but the need would depend on the nature of their work, and also what happens with mobile data pricing. US Mobile data prices keeps falling and are forecast to drop to $2.75/GB by 2023.

In addition cell companies offer plans with streaming excluded from the data. T-Mobile has its Magenta plans with "unlimited" 4G data and unlimited video streaming from most services, that start at $70/mo (Netflix 1 SD) or $85/mo (Netflix 2 HD) for 1 line, and the price per line falls for households with more lines. For comparison, here cable Internet is $70 with taxes.

Given the value people place on their cellphone and given the falling cost of data, that's a significant competitive challenge for wired services.

 

[banned-66611]

That does not impact the placement on the pole along the primary poles.

What's the issue with using the communications space?

Here is a pole in Japan:

It's a bit messy but as you can see they can easily fit an awful lot of comms cables on there. The Japanese utility companies turned their utility poles into sources of profit. The little thing at the bottom is a street light, which they provide as a free service to the public as a way of asking thanks/sorry for the pole being there.

 

[banned-66611]

They also aren't going to sell much video. People are using independent OTT services. In the USA, cable subscriptions fell by 6.6% in 2019.

Depends if net neutrality goes away I guess.

In the UK companies like BT, which provides internet and phone access, also provide premium TV channels, especially sport. Premium TV is a big money maker for them, and you get discounts if you have their internet service as well.

OpenReach own the copper lines and ISPs pay them a monthly fee per customer for access.

There's also a large chunk of people who rely on mobile Internet, and I expect that to grow as the cost of data continues to fall. In the USA, Pew's 2019 survey returned 17% of people solely using mobile Internet.

High speed 5G is fairly short range, meaning you need to bring fibre backhaul to the neighbourhood anyway and the install a base station near the subscriber.

Mobile is basically the last choice if DSL is too poor in your area and there is no fibre, it's not something people see as good value or fast.

For comparison, here cable Internet is $70 with taxes.

But you are not comparing like for like, cable will be much more reliable and consistent than a shared OTA service.

 

[Cosmacelf]

They don't usually deploy HV to homes, it's all low voltage.

Wrong. The poles carry anywhere from 4 kV to 12 kV (at least in the US). For such locations, they would have pole mounted transformers to step down for each or a few houses.

 

[banned-66611]

Wrong. The poles carry anywhere from 4 kV to 12 kV (at least in the US). For such locations, they would have pole mounted transformers to step down for each or a few houses.

It says "to homes". That means HV to the dwelling. A HV line into the house. High voltage from the pole to your house. The bit of wire that is attached to your building carrying >1000V.

Now admittedly I don't live in the States but everything I can find suggests that, as you seem to have realized, the have a transformer on the pole that sends low voltage to homes. Can you get HV service at domestic locations? What would you do with it?

 

[Cosmacelf]

It says "to homes". That means HV to the dwelling. A HV line into the house. High voltage from the pole to your house. The bit of wire that is attached to your building carrying >1000V.

Now admittedly I don't live in the States but everything I can find suggests that, as you seem to have realized, the have a transformer on the pole that sends low voltage to homes. Can you get HV service at domestic locations? What would you do with it?

No, the post immediately before yours (you didn't quote a post) said:

"They did not reduce the size of the team for the Fiber deployment, also no mention if the Fiber was deployed in the "protected" high voltage space on the poles, or if it was deployed in the "safe zone" for telecommunications cables.

If it was deployed in the high voltage space, additional training, insurance, and capabilities are needed on the part of the installer."

He was indeed talking about poles.

Anyways, this is pointless. The thing people are trying to tell you, and you are trying not to believe, is that deploying fiber, be it on poles, or underground, or via an electric utility, or whatever, isn't cheap, nor a slam dunk. It is doable, for sure, and it is being done. But deployment is limited by many, many factors.

 

[Cosmacelf]

Here's an interesting article from Ars: Jared Mauch didn’t have good broadband—so he built his own fiber ISP

A guy in rural Michigan built his own fiber network to supply himself and 30 other neighbors with a 1 Gbps fiber network, all running out of his house. He expects about a 4 year payback (pretty sure he isn't factoring in his own time spent on the project). I love these do-it-yourself networks.

Building the local fiber plant, while complicated in its own right, isn't usually the problem. The problem for these rural networks is that they are usually hundreds of miles away from the nearest fiber node. Getting fiber to the project is the cost prohibitive part of the equation for most rural users. In this case, he was only 2 miles away from the nearest fiber node. Ironically, any telco could have profitably served the area, but it just wasn't worth their time (too small a service area, and too spread out for their large overheads to get any decent ROE).

 

[hmcgregoraz]

Here's an interesting article from Ars: Jared Mauch didn’t have good broadband—so he built his own fiber ISP

A guy in rural Michigan built his own fiber network to supply himself and 30 other neighbors with a 1 Gbps fiber network, all running out of his house. He expects about a 4 year payback (pretty sure he isn't factoring in his own time spent on the project). I love these do-it-yourself networks.

Building the local fiber plant, while complicated in its own right, isn't usually the problem. The problem for these rural networks is that they are usually hundreds of miles away from the nearest fiber node. Getting fiber to the project is the cost prohibitive part of the equation for most rural users. In this case, he was only 2 miles away from the nearest fiber node. Ironically, any telco could have profitably served the area, but it just wasn't worth their time (too small a service area, and too spread out for their large overheads to get any decent ROE).

I am quite familiar with Jared Mauch, he is a regular on the NANOG email list. I have even looked at doing some projects modeled after his network and others.

This fits in with exactly what I was talking about, we will see small expansions of fiber, locally done, co-ops, small providers, and municipal participation. Some large telco and cable co expansion of fiber, but most of that will be in urban and suburban areas.

We will not see a huge large scale deployment of fiber across suburban, urban, and rural America or Canada anytime soon. Ie StarLink has at least two network generations of opportunity, if not far more.

-Harry

 

[Cosmacelf]

I am quite familiar with Jared Mauch, he is a regular on the NANOG email list. I have even looked at doing some projects modeled after his network and others.

This fits in with exactly what I was talking about, we will see small expansions of fiber, locally done, co-ops, small providers, and municipal participation. Some large telco and cable co expansion of fiber, but most of that will be in urban and suburban areas.

We will not see a huge large scale deployment of fiber across suburban, urban, and rural America or Canada anytime soon. Ie StarLink has at least two network generations of opportunity, if not far more.

-Harry

I agree. The telcos seemed to be focused on wireless right now and aren't doing much with fiber (as far as investment $ go). Having said that, there are small startup companies (Netly Fiber and Home - SiFi Networks - Corporate) who are building fiber networks in urban areas. But it all takes time. Rural - you are on your own. Co-ops and entrepreneurs are the only answer - or Starlink.