ITS has an unusually large gamble involved, even by the standards of Musk's companies. Just to pick issue one of many: it's cryogenic composite tanks. Composites and cryogenics don't play well together; there have been attempts in the past, and they were failures. Musk is wanting to take us from "zero launch vehicles of any size using composite cryogenic tanks" to "by far the largest launch vehicle ever built, fully reusable up to a thousand times (for the booster), out of composites". That's a huge jump....
They're also working on insanely high pressure, full flow staged combustion engines with a rarely used propellant mix, used up to a thousand times each with low maintenance...
Ordinarily I'd agree with you. If we were talking about the usual suspects (NASA/Boeing/LockMart), they'd have a pile of paper at this stage and not much else.
But SpaceX has (had) a giant carbon fiber tank which they successfully burst tested to 2/3rds the design pressure back in November, then blew up testing with liquid nitrogen on February 17th 2017. (Judging by the pictures, it failed at the equatorial seam.)
They've built and tested a 1/3rd scale Raptor engine (which I presume you already knew, but other readers might not). It's the first full flow methane fueled rocket engine ever to be test fired, and only the second full flow design in history. (The first was Russia's RD-270, tested back in 1967.)
Having done those things is impressive enough, but the absurdly fantastic part is how rapidly they've done it. They were in Mississippi at the Stennis Space Center in late 2013 to refurbish and modify the E2 test stand to handle methane. Slashdot covered that. They were done with that process April 21st, 2014. Slashdot didn't notice that part. They used that test stand to validate their design and conducted the scale model test firing on September 26th, 2016, just 2 years, 5 months, and 5 days later. And it worked. They were so sure it would work, they didn't even bother with the customary 'burp' test to be sure it would ignite properly. That's a ridiculously rapid development process for any rocket motor, let alone for a design that's been done only once before in history and never for the fuel they selected. For comparison, development of the F-1 used on the Saturn V started in 1955 for the Air Force and it wasn't until 1965 that it underwent a successful test firing without destroying itself, after three years of self-destructive test firings.
SpaceX have definitely set themselves some very hard tasks, but their demonstrated ability to actually get to the test article stage, and from there to the production stage, and to do so quickly, is unmatched in modern times.
It definitely puts things in perspective. If only you'd signed up a few hours earlier...
Hours? It was 1997. A new site that had a signup option was lucky to get a signup per day. I sure didn't. I'd been on the Internet four years by then. I wasn't about to sign up for some random site that would probably be gone in a few years. Besides, FreshMeat was much more pertinent to my interests.
They are large horses used for work, primary farm, hauling, etc. the most common example are clydesdales.
Clydesdales may be the most common example, but the most common breed is the Belgian Brabant. As it happens, that's the breed being used for the work in Wisconsin. Clydesdales are the second most common.
Right, my experience with log pullers and the like is they have such a strong instinct/training that as soon as they feel a load they tend to want to pull, often before you are ready.
I think it's fair to call it instinct at this point. Belgian Brabants have been selectively bred to pull for 1000 years. At this point the desire to pull is built into their brain pathways. A Brabant who isn't born wanting to pull is basically a sport.
Is DHS really this easy to manipulate? And are they really this stupid? If a laptop shell packed with explosives is enough to "bring down an airplane" (and with the right explosive, it certainly is)(but it would have to be packed with explosive, in which case what's all that swabbing and x-raying of passengers for if it couldn't detect that modification?), how does having it in the cargo hold help? It still makes a giant hole in the fuselage and down goes the plane.
I guess my real question is, are people stupid enough to be convinced by this security theater? And then I realize P.T. Barnum was right: you can't go broke overestimating people's stupidity.
Australia might have better healthcare now, compared to us (according to our president) but in just a few short years: Dieselpunk hellscape.
Well sure, but only because they want it that way. And who wouldn't? Dieselpunk hellscapes make for great movies. The US should do it too, at least in the southwest where there aren't so many trees in the way.
The real problem is we need a single payer system like the rest of the developed world, where we would pay about half as much to insure everyone while getting superior medical outcomes.
Correction. Half as much to care for everyone, not insure everyone. The distinction is important.
The American tendency to conflate insurance with healthcare is the number one, number two, and number three reasons why the American healthcare system is so badly broken. Number one, when you buy insurance and not care, you are no longer the customer of the care. The insurance company is, since they're the one actually paying the care provider. This has knock-on effect number two that, as a buyer of insurance, you often have no idea what the price of the care is going to be until long after you've received it. Number three, insurance companies have a profit motive to prevent healthcare, since paying for healthcare is a cost for them, the very definition of a perverse incentive. They're incentivized to the tune of billions of dollars per year to see to it that as little healthcare happens as possible.
If Americans implement single-payer with insurance companies as an integral part of the process, the amount paid will go up, not down. And all of it will go to support a parasitic industry. The fact that "medical coding and billing" is an actual job for which you can get specialized training should be a hint that you're doing it wrong.
Yeah, you have to fill out the csv template they give.
It's pretty fucking clear they want as few people as possible to be able to comment, by making technologically prohibitive to do so. It's fucking scummy as shit.
The site has reverted to the previous (slightly less scummy) behavior. The gofccyourself.com URL redirection works again, and the form accepts data normally.
Blaming the poor economic situation in Cuba on their socialized healthcare system is a bit silly. The continued cause of their poor economy is more likely rooted in the trade embargo that the USA instituted more than half a century ago.
And the fact they're a tiny little island with diddly squat for natural resources that gets hit by hurricanes all the time...
Let's face it, tiny little islands in warm seas are good for tourism or as a tax haven or for Larry Ellison to buy, and not much else. And yeah, the embargo eliminated all of those options.
Cisco is a harder nut to crack - no one would dare go up against them, but yet, if someone did, we'd all be better off as a result.
You make Juniper very sad. Also Brocade Communications, who acquired Foundry Networks. Hell, even HP and Huawei sell enterprise routers. And the backdoor in Huawei routers for the Chinese government is no worse than the NSA backdoor in Cisco gear.
As others have said repeatedly in this thread, "enterprise" switch gear really is a commodity at this point. Not precisely a cheap commodity, since their customers are primarily businesses so they charge all the traffic will bear, but still. Cisco is far from alone. They just want you to think they are. For a company that does very little consumer advertising, they have a peculiarly large amount of mindshare.
I'm dying to know how consumers benefit when you sell their private information. Please elaborate on this. Are you counting on them all being sadomasochists?
You speak of the young-and-dumb welcoming such technological invasion of their privacy (privacy that they apparently don't value at all, thanks so much to social medias' indoctrination)
A common meme, but I'm not so sure it's true. Yeah, early Millenials shared friggin' everything. They came to regret it. Late Millenials and the generation after them that hasn't grown up yet (or all been born yet) are considerably less forthcoming. Every kid I know puts a piece of tape over their laptop's built-in camera. Many of them have multiple social media accounts on the same service, and segregate who is connected to which account. Mom and Dad get to see the Family account and don't even know the other account exists. Kids use Snapchat and other at least nominally ephemeral communications channels, rather than persistent channels like Facebook.
As near as I can tell, sharing habits shifted very very quickly. The youngest generation is already more concerned with data hygiene than any generation before them.
How many Chinese speak English compared with Français?
When I was spending time in Germany, one of the people I talked to a lot wanted to practice his English with me. Why? Because his (German) company outsources production to China, and the language he had in common with their suppliers was, you guessed it, English. Their English was as terrible as his, but they still managed to maintain the relationship.
I suppose his launch costs will be lower, but I have a feeling that this won't work out as cheaply as he thinks. It will be viable in places where there are no providers now, but usually those are the same places that don't have the 2 things this scheme needs... 1. People who need/want internet access and 2. People who have money to pay for it. You got to have all that to make this work out...
A surprisingly large number of people in North America have money and no decent ISP. The US is big but people still live in the wide open spaces, and Canada is even bigger. My parents are less than 30 minutes from a major metro area and can't get either cable or even crappy DSL for any price. Offer to pay for the cable run for a few miles and the local monopolies still aren't interested. So yeah, SpaceX has a market for this. A market in the millions if they can avoid Iridium's mistake in pricing.
Yer bot is getting closer but still not quite right.
It's a little bit alarming, but bots now pass the Turing Test online, as long as we stipulate that they only have to be indistinguishable from a dumb human.
And the big thing for SpaceX, is that by ensuring a continuous market for large numbers of launches, they can ensure economies of scale for launches as well. Which gets launch costs even lower.
There's a hidden benefit as well. Tesla cars currently have cellular antennas and Tesla has to pay cell phone companies to enable the over-the-air updates so highly touted for their vehicles. Once the SpaceX constellation is in orbit, Tesla vehicles can start using it instead (with a service call for existing vehicles and a hardware revision for new ones), and Elon Musk's two biggest companies become very cozy. Either the Tesla company gets a really great price break on vehicle connectivity, or SpaceX starts enjoying a mass market revenue stream, depending entirely on which direction Elon Musk wants to funnel money.
And as was noted, latency will be low, as this is a LEO constellation, not GEO. LEO isn't actually very high up. In many cases latency will be lower than with ground networks - fewer hops, no getting routed through particularly out-of-the-way locations, etc.
Yes and no. While the first couple of hops are quite straightforward, most of the Internet, and especially all the big server farms, is still connected to the major fiber backbones, and therefore still subject to the vagaries of whatever whacky peering agreements the MBAs can dream up. No longer being subjected to the wankeries of routing Comcast indulges in will certainly help, but the other end isn't exactly pristine.
If they succeed, the revenue potential is almost unthinkably large - they could become the entire planet's ISP.
Also yes and no. Sure they potentially have a global reach, but securing spectrum licenses in 250 jurisdictions will take longer than actually launching the constellation (given the SpaceX launch tempo). More to the point, while phased array antennas are indeed a little bit magical, they're still subject to bandwidth limitations. There's only so many bytes per second that can be crammed into the spectrum they're using, and I suspect that number is considerably less than the number of bytes per second currently being shoved around the world at any given time.
Don't get me wrong. I'll gleefully subscribe to SpaceX Internet if the numbers work out, but I was a good deal more excited about Google Fiber (and oh boy was that misplaced) than I am about SpaceX satellite internet service. Fiber throughput is upgradeable simply by replacing the transceivers, and has a much much higher bandwidth limit than any chunk of spectrum SpaceX is likely to get their hands on.
With all the streaming services available, it's not much cheaper than cable.
Cheaper? It costs more than cable. I should know. Charter has been spamming the ever-living shit out of me with actual paper mail every week for literally years, trying to get me to pay them $30/month to add cable TV to my Internet service. They're a major fraction of the contents of my recycle bin. This service appears to have even fewer channels (Charter's list includes AMC, Comedy Central, and MTV), but adds the on-demand stuff already available from Hulu for $8/month. Which you can buy for, effectively... $10/month. Wat?
Basically the TV networks think they've found a way to shore up their sagging viewership numbers by offering the same damn commercial-riddled thing they've been offering for the past 50 years, but "on the Internet!" Which somehow makes it better? Mmmm, no. Maybe they think viewers are as easily bamboozled as advertising buyers, who are spending way more money per eyeball than they used to, but haven't noticed yet. If they ever do... Oh boy.
Remember when you were a kid? Remember when your parents packed you "something healthy" for the lunch break? What did you do? Eat it?
Yes. Yes. And yes.
You win that one when they're toddlers. As a parent, it's your job to win that one. When your two year old says, "No! I don't want it!" your answer is, "Eat it anyway. It's all you're going to get, and you're not leaving the table until you eat it." And that's how it must be. It's often the first test of wills between parent and child, and if you fail that test, you're a bad parent. If you lose, you're forever at a disadvantage when dealing with that child, because they learn that they can just be a little shit and you'll cave. That lesson goes deep, and can have disastrous consequences later in life.
I'm not saying you can't compromise later. Some textures, especially, kids just have a physical reaction to. Some kids can't handle lima beans. Others can't handle raw tomatoes. Most people grow out of it. But the fundamental dynamic must be you're the parent and they're the child and they'll eat what they're given and that's an end of it. That is the origin of the American childhood obesity epidemic. Single parents too rushed and harried with their new toddler to wait them out when that day comes. The result is a nation of McDonalds addicts and fat kids.
By that logic, why do we teach kids things they don't want to learn?
Kids want to learn everything. They only learn to hate learning when you pack them into a room with 20 or 30 assholes (a.k.a. other kids) and subject them to Prussian factory learning. Kids who haven't had the love of learning beaten out of them are voracious sponges for knowledge. Hell, even the ones who have had the love of conventional learning beaten out of them still love to learn. Pick the right generation and you can find kids who can name every single frickin' Pokemon, their abilities, and their evolutions. That same kid will absolutely hate learning what she's supposed to learn, just because of the format it was presented in. (I say she because the lunatic Pokemon lover I knew personally was a girl.)
It would be worth getting Space-X's estimate for the goal though.
Elon Musk, speaking for SpaceX, has already published his estimates. He's been spending $20-$30 million per year on it since 2016 (he's cagey about the exact figure) and he anticipates ramping that up to $300 million per year beginning in early 2019, because he expects development of Falcon 9 and Falcon Heavy to be complete. At that spending rate, he anticipates landing an unmanned ITS spaceship on Mars within the next 10 years, a spaceship so large that the crewed version is expected to have a crew capacity in excess of 100 people. As the Anonymous Coward said, he thinks he will be investing $10 billion towards establishing a Mars colony. Not a footprints and flags mission for Donald Trump, but a self-sustaining colony.
$10 billion in one year does indeed require government money, and yes, at current spending rates globally, it would take a fair-sized chunk of the space budgets of multiple countries to do that without severe disruption to their existing programs. But $10 billion over 30 years is feasible for a single private individual named Elon Musk, regardless of what the governments of the world do. As he said in that announcement, he thinks that the world's governments will ride his coattails to Mars, rather than spending billions of their own to try to get there first (SLS notwithstanding).
However, dumping $10 billion per year onto SpaceX will not reduce the time to Mars colonization to one year from thirty. As with any complex and difficult endeavor, adding money doesn't necessarily make it possible to go much faster. Nine women can not deliver a baby in one month. If nothing else, development of a Mars mission is constrained by the fact that the launch window for a reasonable orbit is available only every 26 months. Even vast amounts of money can only widen that launch window so far before physical constraints start to get in the way. Given the need to test stuff, rather than to just chuck people into the void in untested equipment, it's just barely possible to get people to Mars while Trump is in a second term, if he gets one, and that's with only two test launches. It isn't possible in his first term. The next launch window is in early 2018. SpaceX will be sending an unmanned Dragon capsule to Mars using a Falcon Heavy launch at that time, to test stuff. Developing the entire ITS system between now and then isn't possible no matter how much money you have. The people who know how simply aren't available.
The problem would be manufacturing and testing. Which is where the money comes in. The next decent Mars launch window is in April of 2018, then there's another in July of 2020. So you make a metric fuckton of rockets for 2018 and mount your payloads and shoot 'em off, then you follow up with humans a couple of years later.
Money. LOTS of money. Ludicrous amounts of money. But it would make a difference, and it could be done.
As it happens, there is someone who has ludicrous amounts of money. He owns a major share of an electric car company that is valued more highly than GM. And he has publicly stated that he plans to hit the 2018 launch window, sending an uncrewed Dragon capsule using the Falcon Heavy launcher, entirely on his own dime. He's currently spending somewhere in the neighborhood of $20-$30 million per year on the ITS project, and that first capsule is part of the project. He's hoping to hit Mars on the first try, something many other organizations have failed to do. He's also hoping his engineers don't confuse Imperial and metric units...
But even Elon Musk at his most optimistic doesn't think the first ITS spacecraft will arrive on Mars in much less than a decade, and that's even after he starts spending $300 million per year on the project starting in early 2019. If Donald Trump pulls off the statistical fluke twice, it's just barely possible he'll still be in office when the first ITS lands on Mars with people on board. Just barely.
Slashdot Mirror
ITS has an unusually large gamble involved, even by the standards of Musk's companies. Just to pick issue one of many: it's cryogenic composite tanks. Composites and cryogenics don't play well together; there have been attempts in the past, and they were failures. Musk is wanting to take us from "zero launch vehicles of any size using composite cryogenic tanks" to "by far the largest launch vehicle ever built, fully reusable up to a thousand times (for the booster), out of composites". That's a huge jump. ...
They're also working on insanely high pressure, full flow staged combustion engines with a rarely used propellant mix, used up to a thousand times each with low maintenance...
Ordinarily I'd agree with you. If we were talking about the usual suspects (NASA/Boeing/LockMart), they'd have a pile of paper at this stage and not much else.
But SpaceX has (had) a giant carbon fiber tank which they successfully burst tested to 2/3rds the design pressure back in November, then blew up testing with liquid nitrogen on February 17th 2017. (Judging by the pictures, it failed at the equatorial seam.)
They've built and tested a 1/3rd scale Raptor engine (which I presume you already knew, but other readers might not). It's the first full flow methane fueled rocket engine ever to be test fired, and only the second full flow design in history. (The first was Russia's RD-270, tested back in 1967.)
Having done those things is impressive enough, but the absurdly fantastic part is how rapidly they've done it. They were in Mississippi at the Stennis Space Center in late 2013 to refurbish and modify the E2 test stand to handle methane. Slashdot covered that. They were done with that process April 21st, 2014. Slashdot didn't notice that part. They used that test stand to validate their design and conducted the scale model test firing on September 26th, 2016, just 2 years, 5 months, and 5 days later. And it worked. They were so sure it would work, they didn't even bother with the customary 'burp' test to be sure it would ignite properly. That's a ridiculously rapid development process for any rocket motor, let alone for a design that's been done only once before in history and never for the fuel they selected. For comparison, development of the F-1 used on the Saturn V started in 1955 for the Air Force and it wasn't until 1965 that it underwent a successful test firing without destroying itself, after three years of self-destructive test firings.
SpaceX have definitely set themselves some very hard tasks, but their demonstrated ability to actually get to the test article stage, and from there to the production stage, and to do so quickly, is unmatched in modern times.
What a waste of a good space station. Attach boosters to it and send it to congress.
No member of Congress will notice unless you tape a few $100,000 bills to it.
I think the point was to deliver it straight down at terminal velocity...
It definitely puts things in perspective. If only you'd signed up a few hours earlier...
Hours? It was 1997. A new site that had a signup option was lucky to get a signup per day. I sure didn't. I'd been on the Internet four years by then. I wasn't about to sign up for some random site that would probably be gone in a few years. Besides, FreshMeat was much more pertinent to my interests.
They are large horses used for work, primary farm, hauling, etc. the most common example are clydesdales.
Clydesdales may be the most common example, but the most common breed is the Belgian Brabant. As it happens, that's the breed being used for the work in Wisconsin. Clydesdales are the second most common.
Right, my experience with log pullers and the like is they have such a strong instinct/training that as soon as they feel a load they tend to want to pull, often before you are ready.
I think it's fair to call it instinct at this point. Belgian Brabants have been selectively bred to pull for 1000 years. At this point the desire to pull is built into their brain pathways. A Brabant who isn't born wanting to pull is basically a sport.
Is DHS really this easy to manipulate? And are they really this stupid? If a laptop shell packed with explosives is enough to "bring down an airplane" (and with the right explosive, it certainly is)(but it would have to be packed with explosive, in which case what's all that swabbing and x-raying of passengers for if it couldn't detect that modification?), how does having it in the cargo hold help? It still makes a giant hole in the fuselage and down goes the plane.
I guess my real question is, are people stupid enough to be convinced by this security theater? And then I realize P.T. Barnum was right: you can't go broke overestimating people's stupidity.
Australia might have better healthcare now, compared to us (according to our president) but in just a few short years: Dieselpunk hellscape.
Well sure, but only because they want it that way. And who wouldn't? Dieselpunk hellscapes make for great movies. The US should do it too, at least in the southwest where there aren't so many trees in the way.
The real problem is we need a single payer system like the rest of the developed world, where we would pay about half as much to insure everyone while getting superior medical outcomes.
Correction. Half as much to care for everyone, not insure everyone. The distinction is important.
The American tendency to conflate insurance with healthcare is the number one, number two, and number three reasons why the American healthcare system is so badly broken. Number one, when you buy insurance and not care, you are no longer the customer of the care. The insurance company is, since they're the one actually paying the care provider. This has knock-on effect number two that, as a buyer of insurance, you often have no idea what the price of the care is going to be until long after you've received it. Number three, insurance companies have a profit motive to prevent healthcare, since paying for healthcare is a cost for them, the very definition of a perverse incentive. They're incentivized to the tune of billions of dollars per year to see to it that as little healthcare happens as possible.
If Americans implement single-payer with insurance companies as an integral part of the process, the amount paid will go up, not down. And all of it will go to support a parasitic industry. The fact that "medical coding and billing" is an actual job for which you can get specialized training should be a hint that you're doing it wrong.
A spokesman for the henhouse comfirmed that he's thinking of replacing the roosters with "chicken experts" (i.e. foxes).
Yes, this is what regulatory capture looks like. It's usually not quite so rapid though. RIP EPA.
Yeah, you have to fill out the csv template they give.
It's pretty fucking clear they want as few people as possible to be able to comment, by making technologically prohibitive to do so. It's fucking scummy as shit.
The site has reverted to the previous (slightly less scummy) behavior. The gofccyourself.com URL redirection works again, and the form accepts data normally.
Blaming the poor economic situation in Cuba on their socialized healthcare system is a bit silly. The continued cause of their poor economy is more likely rooted in the trade embargo that the USA instituted more than half a century ago.
And the fact they're a tiny little island with diddly squat for natural resources that gets hit by hurricanes all the time...
Let's face it, tiny little islands in warm seas are good for tourism or as a tax haven or for Larry Ellison to buy, and not much else. And yeah, the embargo eliminated all of those options.
Cisco is a harder nut to crack - no one would dare go up against them, but yet, if someone did, we'd all be better off as a result.
You make Juniper very sad. Also Brocade Communications, who acquired Foundry Networks. Hell, even HP and Huawei sell enterprise routers. And the backdoor in Huawei routers for the Chinese government is no worse than the NSA backdoor in Cisco gear.
As others have said repeatedly in this thread, "enterprise" switch gear really is a commodity at this point. Not precisely a cheap commodity, since their customers are primarily businesses so they charge all the traffic will bear, but still. Cisco is far from alone. They just want you to think they are. For a company that does very little consumer advertising, they have a peculiarly large amount of mindshare.
I'm dying to know how consumers benefit when you sell their private information. Please elaborate on this. Are you counting on them all being sadomasochists?
No. Exhibitionists.
You speak of the young-and-dumb welcoming such technological invasion of their privacy (privacy that they apparently don't value at all, thanks so much to social medias' indoctrination)
A common meme, but I'm not so sure it's true. Yeah, early Millenials shared friggin' everything. They came to regret it. Late Millenials and the generation after them that hasn't grown up yet (or all been born yet) are considerably less forthcoming. Every kid I know puts a piece of tape over their laptop's built-in camera. Many of them have multiple social media accounts on the same service, and segregate who is connected to which account. Mom and Dad get to see the Family account and don't even know the other account exists. Kids use Snapchat and other at least nominally ephemeral communications channels, rather than persistent channels like Facebook.
As near as I can tell, sharing habits shifted very very quickly. The youngest generation is already more concerned with data hygiene than any generation before them.
How many Chinese speak English compared with Français?
When I was spending time in Germany, one of the people I talked to a lot wanted to practice his English with me. Why? Because his (German) company outsources production to China, and the language he had in common with their suppliers was, you guessed it, English. Their English was as terrible as his, but they still managed to maintain the relationship.
I suppose his launch costs will be lower, but I have a feeling that this won't work out as cheaply as he thinks. It will be viable in places where there are no providers now, but usually those are the same places that don't have the 2 things this scheme needs... 1. People who need/want internet access and 2. People who have money to pay for it. You got to have all that to make this work out...
A surprisingly large number of people in North America have money and no decent ISP. The US is big but people still live in the wide open spaces, and Canada is even bigger. My parents are less than 30 minutes from a major metro area and can't get either cable or even crappy DSL for any price. Offer to pay for the cable run for a few miles and the local monopolies still aren't interested. So yeah, SpaceX has a market for this. A market in the millions if they can avoid Iridium's mistake in pricing.
Yer bot is getting closer but still not quite right.
It's a little bit alarming, but bots now pass the Turing Test online, as long as we stipulate that they only have to be indistinguishable from a dumb human.
And the big thing for SpaceX, is that by ensuring a continuous market for large numbers of launches, they can ensure economies of scale for launches as well. Which gets launch costs even lower.
There's a hidden benefit as well. Tesla cars currently have cellular antennas and Tesla has to pay cell phone companies to enable the over-the-air updates so highly touted for their vehicles. Once the SpaceX constellation is in orbit, Tesla vehicles can start using it instead (with a service call for existing vehicles and a hardware revision for new ones), and Elon Musk's two biggest companies become very cozy. Either the Tesla company gets a really great price break on vehicle connectivity, or SpaceX starts enjoying a mass market revenue stream, depending entirely on which direction Elon Musk wants to funnel money.
And as was noted, latency will be low, as this is a LEO constellation, not GEO. LEO isn't actually very high up. In many cases latency will be lower than with ground networks - fewer hops, no getting routed through particularly out-of-the-way locations, etc.
Yes and no. While the first couple of hops are quite straightforward, most of the Internet, and especially all the big server farms, is still connected to the major fiber backbones, and therefore still subject to the vagaries of whatever whacky peering agreements the MBAs can dream up. No longer being subjected to the wankeries of routing Comcast indulges in will certainly help, but the other end isn't exactly pristine.
If they succeed, the revenue potential is almost unthinkably large - they could become the entire planet's ISP.
Also yes and no. Sure they potentially have a global reach, but securing spectrum licenses in 250 jurisdictions will take longer than actually launching the constellation (given the SpaceX launch tempo). More to the point, while phased array antennas are indeed a little bit magical, they're still subject to bandwidth limitations. There's only so many bytes per second that can be crammed into the spectrum they're using, and I suspect that number is considerably less than the number of bytes per second currently being shoved around the world at any given time.
Don't get me wrong. I'll gleefully subscribe to SpaceX Internet if the numbers work out, but I was a good deal more excited about Google Fiber (and oh boy was that misplaced) than I am about SpaceX satellite internet service. Fiber throughput is upgradeable simply by replacing the transceivers, and has a much much higher bandwidth limit than any chunk of spectrum SpaceX is likely to get their hands on.
It's most certainly AI, but... your voice activated toaster doesn't learn and adapt. Not locally anyway.
I beg to differ. I use a machine with quad nVidia Titan Xp cards in it to make toast. I thought that's what those boards were created for.
With all the streaming services available, it's not much cheaper than cable.
Cheaper? It costs more than cable. I should know. Charter has been spamming the ever-living shit out of me with actual paper mail every week for literally years, trying to get me to pay them $30/month to add cable TV to my Internet service. They're a major fraction of the contents of my recycle bin. This service appears to have even fewer channels (Charter's list includes AMC, Comedy Central, and MTV), but adds the on-demand stuff already available from Hulu for $8/month. Which you can buy for, effectively... $10/month. Wat?
Basically the TV networks think they've found a way to shore up their sagging viewership numbers by offering the same damn commercial-riddled thing they've been offering for the past 50 years, but "on the Internet!" Which somehow makes it better? Mmmm, no. Maybe they think viewers are as easily bamboozled as advertising buyers, who are spending way more money per eyeball than they used to, but haven't noticed yet. If they ever do... Oh boy.
Remember when you were a kid? Remember when your parents packed you "something healthy" for the lunch break? What did you do? Eat it?
Yes. Yes. And yes.
You win that one when they're toddlers. As a parent, it's your job to win that one. When your two year old says, "No! I don't want it!" your answer is, "Eat it anyway. It's all you're going to get, and you're not leaving the table until you eat it." And that's how it must be. It's often the first test of wills between parent and child, and if you fail that test, you're a bad parent. If you lose, you're forever at a disadvantage when dealing with that child, because they learn that they can just be a little shit and you'll cave. That lesson goes deep, and can have disastrous consequences later in life.
I'm not saying you can't compromise later. Some textures, especially, kids just have a physical reaction to. Some kids can't handle lima beans. Others can't handle raw tomatoes. Most people grow out of it. But the fundamental dynamic must be you're the parent and they're the child and they'll eat what they're given and that's an end of it. That is the origin of the American childhood obesity epidemic. Single parents too rushed and harried with their new toddler to wait them out when that day comes. The result is a nation of McDonalds addicts and fat kids.
By that logic, why do we teach kids things they don't want to learn?
Kids want to learn everything. They only learn to hate learning when you pack them into a room with 20 or 30 assholes (a.k.a. other kids) and subject them to Prussian factory learning. Kids who haven't had the love of learning beaten out of them are voracious sponges for knowledge. Hell, even the ones who have had the love of conventional learning beaten out of them still love to learn. Pick the right generation and you can find kids who can name every single frickin' Pokemon, their abilities, and their evolutions. That same kid will absolutely hate learning what she's supposed to learn, just because of the format it was presented in. (I say she because the lunatic Pokemon lover I knew personally was a girl.)
It would be worth getting Space-X's estimate for the goal though.
Elon Musk, speaking for SpaceX, has already published his estimates. He's been spending $20-$30 million per year on it since 2016 (he's cagey about the exact figure) and he anticipates ramping that up to $300 million per year beginning in early 2019, because he expects development of Falcon 9 and Falcon Heavy to be complete. At that spending rate, he anticipates landing an unmanned ITS spaceship on Mars within the next 10 years, a spaceship so large that the crewed version is expected to have a crew capacity in excess of 100 people. As the Anonymous Coward said, he thinks he will be investing $10 billion towards establishing a Mars colony. Not a footprints and flags mission for Donald Trump, but a self-sustaining colony.
$10 billion in one year does indeed require government money, and yes, at current spending rates globally, it would take a fair-sized chunk of the space budgets of multiple countries to do that without severe disruption to their existing programs. But $10 billion over 30 years is feasible for a single private individual named Elon Musk, regardless of what the governments of the world do. As he said in that announcement, he thinks that the world's governments will ride his coattails to Mars, rather than spending billions of their own to try to get there first (SLS notwithstanding).
However, dumping $10 billion per year onto SpaceX will not reduce the time to Mars colonization to one year from thirty. As with any complex and difficult endeavor, adding money doesn't necessarily make it possible to go much faster. Nine women can not deliver a baby in one month. If nothing else, development of a Mars mission is constrained by the fact that the launch window for a reasonable orbit is available only every 26 months. Even vast amounts of money can only widen that launch window so far before physical constraints start to get in the way. Given the need to test stuff, rather than to just chuck people into the void in untested equipment, it's just barely possible to get people to Mars while Trump is in a second term, if he gets one, and that's with only two test launches. It isn't possible in his first term. The next launch window is in early 2018. SpaceX will be sending an unmanned Dragon capsule to Mars using a Falcon Heavy launch at that time, to test stuff. Developing the entire ITS system between now and then isn't possible no matter how much money you have. The people who know how simply aren't available.
The problem would be manufacturing and testing. Which is where the money comes in. The next decent Mars launch window is in April of 2018, then there's another in July of 2020. So you make a metric fuckton of rockets for 2018 and mount your payloads and shoot 'em off, then you follow up with humans a couple of years later.
Money. LOTS of money. Ludicrous amounts of money. But it would make a difference, and it could be done.
As it happens, there is someone who has ludicrous amounts of money. He owns a major share of an electric car company that is valued more highly than GM. And he has publicly stated that he plans to hit the 2018 launch window, sending an uncrewed Dragon capsule using the Falcon Heavy launcher, entirely on his own dime. He's currently spending somewhere in the neighborhood of $20-$30 million per year on the ITS project, and that first capsule is part of the project. He's hoping to hit Mars on the first try, something many other organizations have failed to do. He's also hoping his engineers don't confuse Imperial and metric units...
But even Elon Musk at his most optimistic doesn't think the first ITS spacecraft will arrive on Mars in much less than a decade, and that's even after he starts spending $300 million per year on the project starting in early 2019. If Donald Trump pulls off the statistical fluke twice, it's just barely possible he'll still be in office when the first ITS lands on Mars with people on board. Just barely.