4 ms·

The metro/urban bias on HN is rarely more prevalent than when the subject is cars or Internet access. I live on a wooded lot in a reasonably developed area in

by jcims 4y ago

The metro/urban bias on HN is rarely more prevalent than when the subject is cars or Internet access. I live on a wooded lot in a reasonably developed area in Ohio. DSL was brought out to my address in ~2004. Today it is limited to 15Mbps down, 768kbps up. 4G has been available since Verizon rolled it out in the area. I have no idea how long ago that was, maybe ten years ago. Today it's good for 25Mbps/2Mbps on a good day, no carrier on bad. I've had Starlink since Feb of 2021. Around November of last year it had become reliable enough that I could use it as my primary Internet access method. The numbers vary, but I would say my average is 140Mbps down and 25Mbps up. So in my specific case, Starlink is the best Internet access, hands down. It has the lowest latency, it has the highest raw and average performance, and it is the most reliable. It has been this way for approximately a year. So in circumstances where the customer is anywhere outside of a urban/suburban area, Starlink is incredibly competitive and may in fact be the best option available.

giobox 4y ago
I totally agree with you! But thanks for accusing me of bias anyway. The only question I ask is, "is there a sustainable market for affordable satellite internet access?". There is no debate it is the best connection for some people right now, but if it is unsustainable that's good for no one in long term. Before Starlink, satellite internet was far more expensive. Yet to turn a profit, and some big headwinds in way of competitive pressures from other cheaper technologies. I am also a Starlink customer and live in the middle of nowhere, so... For the little my example is worth as well, I haven't seen anything like as good as those speeds where I am in Washington State, for some time. Over the same period, cellular has appeared in places it wasn't before. Cellular doesn't need to reach everyone to be big enough to end the dream of affordable satellite access, and the cellular providers fight for the same government money many countries provide to improve connections for the most poorly connected regions. Assuming you are free of trees (a big problem!), Starlink so far only works well at really low density deployments - a huge detractor in many assessments comparing it to cellular deployments, even in remote regions. Leaked slide deck from company declared optimum deployments is 100 customers per 300km squared, their "100 per 300 rule". Maybe this is better today and gets better with time, but cellular has a lot of radio advantages and a lot of existing infra to expand from. Also doesn't need spaceships. > https://drive.google.com/file/d/1UOI7b5flgAjJrPs2HDa64p_ZABckasa_/view https://drive.google.com/file/d/1UOI7b5flgAjJrPs2HDa64p_ZABc...
- Dylan16807 4y ago
  > I totally agree with you! But thanks for accusing me of bias anyway. Well your first line was "By Musk's own admission, it can never be competitive against the cellular providers" and then the rebuttal was based around a strong example of it beating cellular providers inside their coverage range. You can understand the confusion.
  - freejazz 4y ago
    Not really? How are you going to be competitive with the cellular companies by selling something to what reflects a very small percentage of their customer base?
    - Dylan16807 4y ago
      > How are you going to be competitive with the cellular companies by selling something to what reflects a very small percentage of their customer base? By only needing 5000 "towers".
      - freejazz 4y ago
        Your point isn't clear to me. As I see it, there is significantly less money to be made.
        Dylan16807 4y ago
        Significantly less money to be made, and even more significantly less cost per square mile of coverage. Especially when a lot of that coverage range for towers can't do tens of megabits per second. So, it can be competitive.
      - giobox 4y ago
        If you want to make facetious arguments, starlink only needs up to “42,000 satellites, thousands of launches to install and ultimately maintain, plus a fleet of space ships”… I’m being charitable and not mentioning that each of those potential 42,000 needs replacing on a 5 year cycle - they don’t just go up there and stay there forever. 5000 or whatever towers ain’t insurmountable - the USA already has a huge percentage of the population covered today - and has nice benefit of being an already profitable business to be in, and construction can be done by anyone with hands, tools and a vehicle. The tower probably even lasts longer than 5 years.

by jcims 4y ago

giobox 4y ago

I totally agree with you! But thanks for accusing me of bias anyway. The only question I ask is, "is there a sustainable market for affordable satellite internet access?". There is no debate it is the best connection for some people right now, but if it is unsustainable that's good for no one in long term. Before Starlink, satellite internet was far more expensive. Yet to turn a profit, and some big headwinds in way of competitive pressures from other cheaper technologies. I am also a Starlink customer and live in the middle of nowhere, so... For the little my example is worth as well, I haven't seen anything like as good as those speeds where I am in Washington State, for some time. Over the same period, cellular has appeared in places it wasn't before. Cellular doesn't need to reach everyone to be big enough to end the dream of affordable satellite access, and the cellular providers fight for the same government money many countries provide to improve connections for the most poorly connected regions. Assuming you are free of trees (a big problem!), Starlink so far only works well at really low density deployments - a huge detractor in many assessments comparing it to cellular deployments, even in remote regions. Leaked slide deck from company declared optimum deployments is 100 customers per 300km squared, their "100 per 300 rule". Maybe this is better today and gets better with time, but cellular has a lot of radio advantages and a lot of existing infra to expand from. Also doesn't need spaceships. > https://drive.google.com/file/d/1UOI7b5flgAjJrPs2HDa64p_ZABckasa_/view https://drive.google.com/file/d/1UOI7b5flgAjJrPs2HDa64p_ZABc...

Dylan16807 4y ago
> I totally agree with you! But thanks for accusing me of bias anyway. Well your first line was "By Musk's own admission, it can never be competitive against the cellular providers" and then the rebuttal was based around a strong example of it beating cellular providers inside their coverage range. You can understand the confusion.
- freejazz 4y ago
  Not really? How are you going to be competitive with the cellular companies by selling something to what reflects a very small percentage of their customer base?
  - Dylan16807 4y ago
    > How are you going to be competitive with the cellular companies by selling something to what reflects a very small percentage of their customer base? By only needing 5000 "towers".
    - freejazz 4y ago
      Your point isn't clear to me. As I see it, there is significantly less money to be made.
      - Dylan16807 4y ago
        Significantly less money to be made, and even more significantly less cost per square mile of coverage. Especially when a lot of that coverage range for towers can't do tens of megabits per second. So, it can be competitive.
    - giobox 4y ago
      If you want to make facetious arguments, starlink only needs up to “42,000 satellites, thousands of launches to install and ultimately maintain, plus a fleet of space ships”… I’m being charitable and not mentioning that each of those potential 42,000 needs replacing on a 5 year cycle - they don’t just go up there and stay there forever. 5000 or whatever towers ain’t insurmountable - the USA already has a huge percentage of the population covered today - and has nice benefit of being an already profitable business to be in, and construction can be done by anyone with hands, tools and a vehicle. The tower probably even lasts longer than 5 years.