Hacker News Reader

> that providing cell service is 100x to 300x harder than providing electricity?

Yes, unfortunately. It's at least that much harder. For a similar comparison think of traditional radio - you buy the radio, and that's it. You might wonder how that's possible when it costs so much to beam data to your phone wirelessly? It's a complex question with a lot of nuance, but there are two important reasons that float to the top:

First, Internet requires reliable two way communication. Neither radio or electricity require that. These means economies of scales work in favor of electricity generation and radio transmission - they can put up a huge plant or radio tower, crank the power up, and serve a lot of customers far away. The customer's device doesn't have to talk back. Imagine if each home had to have a mini generator to send electricity back to the power plant - that would cost a lot!

Second, Internet service requires much, much more precise electronics than power or radio transmission. The Ethernet port on your laptop can understand signalling at (at least) 1gbps (1 billion bits per second, more or less). And that's not even that much - large networks today are doing 100 of times that on single physical connections, with electronics on both sides both modulating and demodulating those signals simultaneously. All that expensive (both to design and to manufacture) electronics adds up, and is accounted for in the cost of getting those bits to you.

I worked for a North American Telco in IT, and I like everything about your answer.

I will also add that Telcos are wickedly inefficient. Multi-million dollar projects just to update a public website, tens of millions in advertising, more PMs and Business Analysts and Middle Managers than people doing actual work, etc. etc.

Of course they want to make a tidy profit too.

This doesn't really make sense to me. The big difference I see is the cost to buy spectrum and the inherent geographic monopoly of owning it. Transformers and switchcraft are heavy-iron things which cost a lot because they're safety-rated and built with a large amount of actual matter. Transceivers and antennas are comparatively cheap to manufacture (but with higher margins) and generally don't incinerate things if they fail.

In my area the cost to deploy service seems to be primarily driven by the cost of labor to bury things. That impacts electricity more than cell data, as you actually have to physically reach every one of your customers.

Also your meter is two-way. It's just sending your usage data back, not power.

> Also your meter is two-way. It's just sending your usage data back, not power.

This is actually a pretty good illustration of what I'm talking about - sending a few bytes a month for power usage data is much different than maintaining a constant bit rate of >0mbps from each customer. It's the difference between broadcasting 40 channels to everyone and reserving 1 channel for a bit of upload (fairly easy with old technology) and giving everyone their own channel for upload (much more difficult).

You're not wrong that a huge portion of the costs is putting things in the ground, though. The advantage that power companies have is that a lot of that work has already been done, and they don't have to upgrade as often because power usage requirements aren't changing as rapidly as bandwidth requirements. They also have the (arguable) advantage of usually being a government granted monopoly, so they don't have to worry about building something out and then being undercut by a competitor.

There is a lot to dig in to here, but consider this from a different perspective - there is actually quite a bit of competition among mobile providers, and yet prices aren't dropping very quickly. You could argue that that's because of corruption and collusion (in some cases it probably is) but also consider that any one of those companies would absolutely love to take market share from the cable companies. If Verizon could undercut Comcast and give you Internet and TV at home over their wireless network they would do that in a heartbeat. The only reason they're not doing it is because it would be too expensive. Or, more precisely, they are trying to do it, that's what 5G is, and it's costing them a fortune.

EDIT: Also consider another perspective - the power companies themselves would love to provide Internet service! On paper they are in a great position to be able to do that. They already have infrastructure, technicians, billing platforms, captive customers, etc. The reason they're not doing it is because their infrastructure doesn't support it, and couldn't really be made to do it without spending so much money on it that they could not be cost-competitive with existing products.

Personally I think mobile is tolerably (but not ideally) competitive. I think the costs are driven by the high costs of spectrum and marketing/customer acquisition. And the product is reasonably high-value, so prices _can_ be high.

A useful comparison is the cost of fixed wireless internet service vs mobile service. The price/GB of the former is usually far lower than the latter despite running on exactly the same infrastructure.

I merely point out that I struggle to believe that the physical infrastructure costs are large components of the cost of service. Unlike power delivery, where the physical distribution infrastructure seems to be the majority share. Or at least it is if my bill breakdown is to be believed.

> A useful comparison is the cost of fixed wireless internet service vs mobile service. The price/GB of the former is usually far lower than the latter despite running on exactly the same infrastructure.

Glad you brought this up, fixed wireless is my career so this is something I've thought about a lot over the last 15 years or so. I would argue that the difference in pricing between fixed wireless and mobile is actually almost entirely driven by the 'fixed' qualifier - it's just much easier and cheaper to deliver Internet service to a fixed point where you can put a high gain antenna (like a rooftop) than it is to deliver to a cell phone in someone's pocket inside a building.

Of course the next question is why don't the mobile operators do home service with a fixed antenna on the rooftop like fixed wireless operators do, but using their massive spectrum and tower footprint advantages? I don't really know the answer to that. There have been a lot of efforts to do this, actually - the earliest one I know of was Sprint putting up these diamond shape antennas on rooftops back in the late 90s / early 00s. I think at least part of it is that it's hard to have high-gain antennas on rooftops coexist with low-gain antennas in cell phone on the same medium, and another part is that it's hard to hire and train and manage folks to do the home installs. I do believe that part of the push to 5G is to enable this type of service, though.

I love this question.

I don't know the answer, but one part of it might be that electricity is a "commodity" that you can effectively buy from anywhere in the country, or perhaps anywhere in the world, via the power grid. Whereas for cell service you're effectively buying from one of three companies, I think. So way less competition.

I'm sorry, but you think you can buy electricity from anywhere in the world? in any american city, you have one choice where to buy it from. 3 companies can put up their antennas on a tower. 3 companies cannot dig trenches through chicago to every house to run wires.

there are some small exceptions where it was traded on the power grid with deregulation. it caused the enron scandal. do you remember enron?

the issue is we have limited frequencies to use for cell phones. any time you have another player, you are wasting that bandwidth by splitting it up more. we should have 1 cell phone provider for each area of the country, and it should be regulated like a utility. that is what will make cost lower, not more competition. there are things that are physically limited, so running it capitalist style is not appropriate because of high barrier to entry. capitalism only works well with low startup costs.

i like how most of europe has internet for dirt cheap. it's public land, it's public cables. all an isp does is lease cables and provide service with their equipment connected.

Providing cell service takes at least the cost of electricity to run them. They're literally both electricity going down a wire. So yes, coal has to be mined and water reservoirs have to be created. You also have to build the equipment, and I imagine paying off that equipment's cost is more than the cost of electricity. That has to be mined, smelted, processed, worked, and assembled.

Watt-Hours and Gigabytes are fundamentally different units, so comparing the marginal cost of electricity and data is non-sensical.

In terms of the fixed costs. For electricity, the price you pay powering your phone is subsidized by the price you pay powering your HVAC. The portion of the electricity infrastructure that powers phones is negligable.

In contrasts, the cellular networks exist almost entirely to run phones; so the cost phone users pay needs to be enough to maintain the entire network.

To be a fair comparison, you would need to imagine a world where phones were the primary consumers of the electric grid. In such a world, I would expect to see electricity prices rise significantly

> imagine a world where phones were the primary consumers of the electric grid.

This is actually an interesting thought experiment. Maybe the best solution would be shipping powerbanks around?

Mobile data in US is very expensive. In India you can get 1GB data for $0.10.

Electricity has economies of scale at play there. It wouldn’t be economical at the normal prices per kw if people were only charging phones.

Even ignoring that bit, electricity has different scaling considerations. Everyone’s computers don’t need access to special Youtube electricity and Twitter electricity. It all just comes from a localized grid and can come from any standard generation source.

Internet access would certainly be way cheaper if it didn’t provide global connectivity.

Another interesting point of comparison is that data limits on a VPS like DigitalOcean are $5/month/terabyte, and this includes all other aspects of renting the VPS as well. There are also MVNOs that cost less than $15/month with 1GB data caps. So keeping high bandwidth traffic to a VPS with a low bandwidth cell connection like a terminal emulator could potentially allow someone to access large amounts of data much more efficiently.

It is not like it is a P2P mesh network.

Land leases are expensive for tower placement, and they need electricity and a backup generator, to power the comm. equipment and the air conditioning.

And then we can talk about HVAC maintenance.

Don't the utility companies have the same issues with infrastructure placement and land though? Or is there some eminent domain thing for power utilities that doesn't apply to telecom?

Well, in the US, electric and telephone have more rights and regulations, compared to cable television.

So, as a telecom, it is better to go in under the guise of a CATV company and use a franchise agreement, then as a phone company that also offers TV, and internet.

Also, if the poles are already there (usually phone or power), the new entrant is able to lease a portion of the pole to string infrastructure.

Sometimes the municipality owns the power company.

Cellular providers lease tower space from private companies that negotiate the tower placement with private land owners.

Power was built out at a time when infrastructure projects were still feasible and NIMBYs had less influence.

This is kind of a nonsensical comparison. Most people spend more on electricity (especially indirectly) than they do on cell phone service, in total.