Hacker News Reader

> So to support 100Mbit customers, you need to launch a Satellite once for every 40 customers.

No, you don't. Oversubscription is a thing, a rather obvious thing when you go looking for it (evening Netflix time is horrible bandwidth on any ISP lately, including my Starlink connection), and you could comfortably handle hundreds (or low thousands) of customers at once on a single 40Gbit bird.

No, not everyone will be able to pull their max speed at the same time, but that's just not a use case you design an ISP for. Assume 20Mbit or so in the evenings for Netflix or Prime or whatever on most of your customers, then let the rest of the bandwidth shuffle around as needed.

Oversubscription proved to be a problematic thing early in 2020, because of everyone working from home. This will ease up, but the old oversubscription numbers won’t work anymore as people have permanently changed some of their habits.

And the bandwidth required by any given household will only really go up in the future.

You're assuming customers will be using all of their bandwidth 24/7, no ISP / telephone / utility company works like that.

And even then, his calculation is off by a factor of 5 or 10

There can be a lot of arguments against starlink as envisioned but the reasons you gave don't seem to hold up.

> cityfolk won't really use Starlink

I don't think starlink was ever meant to replace current high speed internet.

> 5G will cover many rural areas

5G means many different things. The high speed bands don't have enough range to make them economical to build in rural areas from what I've read.

> directional WiFi (Fixed Wireless) works surprisingly well out in rural areas for 50-mile ranges

Yes, if it's directional, and the directionality is pretty narrow if you want to get any reasonable performance. I don't think it's feasible for each town to have a dedicated antenna for each house within a 50 mile radius.

> Yes, if it's directional, and the directionality is pretty narrow if you want to get any reasonable performance. I don't think it's feasible for each town to have a dedicated antenna for each house within a 50 mile radius.

You're asking the wrong questions.

So lets say Starlink is 4000-customers per satellite (10x oversubscription of a 40Gbps underlying, 100Mbit per customer, seem fair to you?). It costs $1000 to $3000 for the Starlink Satellite dish right now.

The total costs of Starlink, per customer, is therefore 1/400th of a satellite launch + $1500 or so (depending on how cheaply the dish can get).

Can you run an antenna from a town center to a house 50-miles away for less than $1500 + SatelliteLaunch/4000 ??

EDIT: 4000, not 400.

I admit that I don't know how much a Satellite Launch costs, but Elon Musk is asking $30 Billion for a reason. I assume the satellite launches are a good chunk of those costs.

A dedicated point-to-point link costs $300 or less which is even cheaper than Starlink.

> Only the most rural of areas would use Starlink.

I mean yes? Rural and underserved areas always been the target market of Starlink in the first place. Starlink isn't attempting to target the urban market.

> I mean yes? Rural and underserved areas always been the target market of Starlink in the first place. Starlink isn't attempting to target the urban market.

Are there millions of people in rural areas who will trust a US-based space ISP over local ISPs?

The thing about targeting rural areas, is that you're by definition targeting a small subset of people. People generally live in cities, on the average.

There are millions of people in current rural US with terrible terrible ISP options that would happily pay this. And that isn’t counting other countries. Australia alone is terribly underserved right now by all national ISP’s.

Even if you put in the TOS that the NSA is going to jerk off to all their stuff, most would probably be super happy to sign up if it meant sub-300ms pings and > 5mbps speeds.

https://www.census.gov/programs-surveys/geography/guidance/g...

Of the 300-million people in the USA, 250-million live in "Urban" areas.

https://www2.census.gov/geo/pdfs/reference/GARM/Ch12GARM.pdf

By definition: this is all towns of 50,000 and above. The next big group of people are in 5,000+ person towns, which are a good target for 5G and fixed wireless internet. I've got an uncle who lives in one of these areas ("rural" for sure, but with line-of-sight to the town center so wireless / line of sight still works and so do cell phones / internet)

I think you're grossly overestimating the number of people where Starlink's $500 upfront costs + $99 / month is competitive.

Furthermore: line-of-sight is not a problem for a lot of people. Huge swaths of the USA's rural areas are extremely flat (ex: Kansas, Iowa, etc. etc.). So Fixed Wireless is reliable for many miles.

So you’re agreeing there are millions unserved and in truly rural areas right? That’s all it takes, and all I was referencing.

Fixed wireless has been an option for a decade plus. When it works, it’s great. Almost everyone I know in rural areas has thought of it or tried it once. Trees, distance, sometimes a hill usually get in the way.

Hughes and others don’t have the subscriber numbers they do because they’re good services that people like using over alternatives.

Don’t forget there are a lot of people that would love to live in a rural area that haven’t because of lack of a decent connection. If starlink keeps on this trajectory, a lot of folks will be moving who never could before.

The purpose of my comment wasn't to discuss the efficacy of Starlink's strategy on targeting rural / underserved areas. Only to point out that was indeed their strategy. Which I think was important to highlight given your criticism of Starlink being uncompetitive in urban markets.

My point about Urban markets is that 80% of the US's population is in an urban area in the 2010 census.

Which means that Starlink immediately disqualifies itself from 80% of the US population. The remaining 20% seem to be largely covered by direct line-of-sight wireless technology.

Not the most rural, remember that also this is unlimited data while most carriers still shy away from "truly" unlimited data plans.

Yes 5g, more bandwidth etc so plans may become more liberal but still.

WISP for rural areas are hard to manage, not just the engineering but the marketing and customer retention, now I feel (really believe/know) is not the time to even begin to be a rural WISP where starlink and telecoms are on technological breakthroughs to literally deicentivize using you.

I think you're assuming the satellites aren't oversubscribed? Generally consumer ISPs are marketed well in excess of what all customers could actually use simultaneously. The economics change somewhat if you can put 400-4000 customers on the average satellite instead of 40.

Your math is off by an order of magnitude. 40Gbps/100Mbps=400.

It should also be able to handle at least 4000 customers (10x) through oversubscrpition.

The math is more complicated than that. Oversubscription reduces the number of satellites needed while oceans increase it.

While that's true, my Physics teacher always said that a crappy calculation that's within-an-order of magnitude is always better than gut-feeling.

If we're talking about feasibility and/or if something is a good idea, then once all the math is added up, then we need to see that the final costs are "really really good".

EDIT: As long as we're clear on each other's math / assumptions, we can make "napkin level" estimates to guess at the feasibility of this project.