Live data from Hacker News

Oh, 2022

antipope.org

221–230 of 557 posts

Re: Oh, 2022

#221
post #107

Earlier quoted context omitted.

Space is hugely, vastly, mind-bogglingly big. Useful low orbital shells around the Earth are an awful lot smaller. They're still big, like the sky is big, but airplanes do manage to, at least somewhat regularly, collide with each other "in the middle of nowhere" (most midairs are around airports, since there are a lot of airplanes heading in and out of them). Aircraft debris, however, falls out of the sky in short or…

Yes, space is big. Lower earth orbit is still very big. However if you have dozens of operators, each with thousands of satiates making multiple round the world trips. The chances of a specific satellite colliding perhaps is tiny (given how big space is). However the chances of one of the thousands colliding with any of the other thousands over any of their orbital periods increases exponentially as the number of sat…

Tens of thousands.

> There are currently over 1,600 Starlink satellites in orbit, and that number will continue to grow; SpaceX has filed paperwork for up to 42,000 satellites for the constellation.

https://www.space.com/spacex-starlink-internet-satellites-la...

They're aware of the risk of collisions, but toss in a few more vendors trying for the same size, and you can have potentially hundreds of thousands of satellites in the shells. They're big, but they're not that big.

Re: Oh, 2022

#222

Regarding spacex in this, my starlink terminal recently hit a milestone of 0.00% ICMP loss to terrestrial ISP things in the pacific northwest. Averaged over 5, 6 hour periods. 20 icmp pings every 60s, basically a default smokeping install set to once a minute. Previously it was running somewhere around 0.15% to 0.60% which was still totally usable... if you put a similar test setup on a noisy/oversubscribed coaxial c…

I'll go as far as to say that Starlink is now so good i honestly believe it's going to change the other industries and not just its own. As an example I've been looking at properties outside of suburbia since testing a friends Starlink terminal. I know i can get proper broadband on par with cable anywhere. It's changed my personal long term future plans. I think it's going to change a lot of things.

Isn’t starlink internet connection expensive atm? $99 per month plus one time h/w costs?

Re: Oh, 2022

#223
post #174
post #155

Earlier quoted context omitted.

Cars don’t last as long as you’d think, on average. There will also be a knee in the curve where the ICE fleet drops below critical mass and gas stations start shutting down - and the harder it becomes to find a station, the faster people will switch.

They last a hell of a lot longer than they did when we were still putting TEPb in the fuel.

The lifespans of cars have been elongating rapidly. The median car on the road recently passed 12 years. Especially sensible cars with medium-displacement engines making little power through a CVT, these will go a million miles with ease. The long tail of model year 2030 Honda Civics is going to be almost impossible to eradicate without controlling the supply of fossil motor fuels.

Re: Oh, 2022

#224

Earlier quoted context omitted.

I'll go as far as to say that Starlink is now so good i honestly believe it's going to change the other industries and not just its own. As an example I've been looking at properties outside of suburbia since testing a friends Starlink terminal. I know i can get proper broadband on par with cable anywhere. It's changed my personal long term future plans. I think it's going to change a lot of things.

Isn’t starlink internet connection expensive atm? $99 per month plus one time h/w costs?

Not just expensive. IIRC, the receiving dish and associated gear burns 120W or something like that.

Re: Oh, 2022

#225
post #207

Earlier quoted context omitted.

Low earth orbit (where Starlink and the ISS are) actually has a fair amount of atmospheric drag. Random google searching suggests an inert Starlink satellite (or fragments of one) would come down in 5-10 years.

That's the claim made, yes. Now, how many other satellites are likely to get hit in "5-10 years" of flak flying around at orbital velocities? It's nice if your goal is to say "Well, it probably won't render space useless forever ..." - but that assumes one collision, and no ongoing collisions. If it's "5-10 years per collision," and each new one resets the 5-10 years timeframe, well, there's a lot of stuff up there t…

> If it's "5-10 years per collision," and each new one resets the 5-10 years timeframe

It's not.

The concern is that unwanted collisions will become common enough that satellites break down or are completely destroyed within months. There will never be enough debris that we can't fly through LEO.

Re: Oh, 2022

#226

Earlier quoted context omitted.

Since space debris in the satellite zone is already a problem, that's not a solution.

There isn't just one satellite zone. Starlink satellites are about 4 times closer to earth than than they are to the geosynchronous Hughesnet satellites. The starlink satellites are close enough that even after full Gravity -style Kessler syndrome, you could just wait several years and start over from scratch. Obviously that is something to avoid, but it's not like we're filling up a non-renewable resource in low ear…

Note: Starlink orbits at an altitude of 340-350 miles . Hughesnet’s geosynchronous orbit has an altitude of 22,000 miles. That is over 62 times higher, not 4.

FWIW.

Re: Oh, 2022

#227

Earlier quoted context omitted.

> If 50% of the population are COVID invalids, who's going to be doing the (building) work for a new economic boom? This is a good question to raise, and IMHO does not invalidate the original piece: Two plausible predictions, but how do they interact? Which one gets the short end of the stick? Are the able-bodied hard at work throwing up temporary homes for invalids?

Combine this with "but robots", and a somewhat small group of actual human beings will still be needed to do work. Combine that with the eugenics trend we're currently on, and, well, happy dreams.

> the eugenics trend we're currently on

The what

Re: Oh, 2022

#228

Earlier quoted context omitted.

I'll go as far as to say that Starlink is now so good i honestly believe it's going to change the other industries and not just its own. As an example I've been looking at properties outside of suburbia since testing a friends Starlink terminal. I know i can get proper broadband on par with cable anywhere. It's changed my personal long term future plans. I think it's going to change a lot of things.

Isn’t starlink internet connection expensive atm? $99 per month plus one time h/w costs?

It's not a replacement for suburban internet. It's a way to pay just 50% more (in Australian pricing) for something that offers way more options.

Re: Oh, 2022

#229

Earlier quoted context omitted.

Isn’t starlink internet connection expensive atm? $99 per month plus one time h/w costs?

Not just expensive. IIRC, the receiving dish and associated gear burns 120W or something like that.

Living with more space though, you can offset that with solar pretty easily.

Re: Oh, 2022

#230

Regarding spacex in this, my starlink terminal recently hit a milestone of 0.00% ICMP loss to terrestrial ISP things in the pacific northwest. Averaged over 5, 6 hour periods. 20 icmp pings every 60s, basically a default smokeping install set to once a minute. Previously it was running somewhere around 0.15% to 0.60% which was still totally usable... if you put a similar test setup on a noisy/oversubscribed coaxial c…

What's your next hop time? Mine is about 20ms at home (FTTC - Fibre to the Cabinet) Satellite links are quite phenomenal to be honest (when they work).

The metrics to worry about are latency and loss and throughput. Jitter is really a SIPnRTP thing. Don't use ICMP to test for "quality" - it isn't even the same protocol that you worry about (TCP/UDP), it is what it is - ICMP (use it wisely).

Post reply on HN