Many people were expecting this to happen sooner or later.
Thanks to nearly a century of lobbying, the USA has a pedestrian-hostile culture and many states have car-centric laws (eg, jaywalking) which state that pedestrians must give way to cars except at designated crossing points and even then can only cross on a green light.
Growing up with these kinds of rules translate into assumptions that people generally don't walk ontot he road unexpectedly, which translate into robotic rules that people NEVER do that - meaning they're not setup to "expect the unexpected"
A human driver should have seen the pedestrian on the median some distance off and reacted (slowed down or changed lanes) even if she was stationary because someone standing there is likely to move. The robot just kept going thanks to cultural assumptions programmed into it that became rigid operating rules.
(Anticipating kids running onto the road is one of the classic tests for danger perception in many countries. Virtually all US drivers fail such tests badly, which is why trading your USA license in for any EU one generally involves at least a full license test.)
Sure, there shouldn't be a pedestrian on the freeway (or a deer either) but I'm not going to run him over. I sure as hell don't want my robot car to do so. Nor do I want it to ignore the 12 point stag, or cow on the road - one of those would come through the windscreen and kill the vehicle occupants. Ditto for the unexpected solo road cone. There might be a pothole behind it.
Arizona has to be one of the worst possible places to test self-driving vehicles. It's extremely car-centric, shitty and hostile for pedestrians and the number of edge/corner cases that happen is low, meaning the human supervisors get complacent. My bet is that the twit in the driver seat didn't even have eyes outside the cabin until after the poor woman in question was bouncing down the road.
"I don't think phytoplankton are affected by minor changes in ocean CO2 levels."
Except it's not minor.
The CO2 concentration in the oceans has changed so much that actual acidity levels have changed 30% in the last 200 years. (pH is a logarithmic scale)
Bear in mind also that in geological history, oceanic CO2 spikes go hand in hand with anoxic events (most of our oil comes from the last couple of these) and half our breathable oxygen comes from the seas.
Can you cope with a drop of oxygen from 19% to 15% at sea level? Hint: 90% of the human population cannot and will die of cardiac congestion caused by physiological changes as their bodies attempt to deal with such a long-term change.
That's quite apart from losing a large chunk of the food chain in an anoxic event. In general very few species averaging larger than about 40kg bodymass appear to have survived on land when this kind of thing has happened in the last and it's likely to stay the same rule of thumb in future.
Increased CO2 gives increased plant activity, but is also stresses the respiration part of their cycles and seems to result in less effective mineral fixation along with greater water uptake, do you might get more greenery, but it looks like it will be much lower quality.
The history of the world shows that cities get abandoned by their populations long before the leadership gives up. There are thousands of ghost towns around the planet which were once thriving cities and some of them only date back a few hundred years (gold, oil and mineral rushes, old coal towns, etc etc)
New Orleans is on track to become the first modern climatic ghost town but there will be plenty of others.
The article glosses other another point to bear in mind - given the nature of the subduction fault offshore, in a good sized earthquake there's a good chance that large portions of the land between the fault and the San Andreas fault will sink rapidly as compression stresses are released and this may be accompanied by sections further inland rising a bit (think of a squashed sponge uncrumpling.)
Saying "it's complex" doesn't even come close to explaining the various interactions occuring. Are there any ancient raised beaches along the californian coastline? if so, they're evidence of that area of land being systematically jacked up by sucessive quakes, etc.
When nuclear power goes bad it doesn't do that either. Perhaps you should look at actual stats for nuclear accidents.
Furthermore, if we'd actually moved to next-gen molten salt designs 50 years ago when they were proven workable and safe instead of killing them for political reasons, we wouldn't have giant steam bombs surrounding our nuclear piles, but a non-pressurised, non-reactive liquid that freezes solid when it goes below 300C and doesn't start boiling until well _ABOVE_ the inherent limiting temperatures of nuclear reactions.
The thing about high level radioactive products is that they're both easily detectable AND shortlived. All you have to do is wait them out and that's measured in human scales.
Three Mile Island's infamous reactor is in the process of being dismantled and cleaned up. Chernobyl will be able to be done in another 20-30 years and Fukushima in 30-40 - and bear in mind that nowhere in Fukushima is more radioactive than Denver Colorado or Downtown Helsinki (both due to natural sources) unless you're silly enough to want to climb inside the nuclear reactors.
Even that "heavily contaminated water" is _less_ radioactive than most BEER
As soon as you add mechanical complexity then you're adding maintenance costs.
If the worker being replaced is paid a low wage, the cost:benefit is low. The low-hanging fruit for automation are more-or-less intellectual but repetitive tasks with higher pay rates and that's been going on for 40 years (when was the last time you saw a room full of accounts ledger clerks scratching away?)
The more likely targets for unemployment are accountants, junior lawyers and suchlike. The investment to do so is lower and the rewards are higher.
Minimum wages won't fall, but the number of higher-paid jobs will decline, bringing more people DOWN to minimum wage.
"That suggests to me that a visible, imitation camera would have accomplished the same."
Until people simply start using a baseball cap pulled down over their face - which is what happens regularly around my (visible) UK surveillance cameras.
Too bad they don't see the low-mounted ones with far-IR LEDs...
Video doorbells are usually awful pieces of crap and they almost never integrate into a decent DVR system.
"I have a Ring doorbell. I give it 2 out of 5 stars."
Ring has been an established brand in Europe for a while. 2 out of 5 stars is being generous. Most of what they sell is "whatever was cheapest in Shenzhen that week"
"The concept is ok, but the implementation is terrible."
This, in spades.
Ring was ok when it was basic overpriced (and fragile) dumb shit. Occasionally there would be a sale of old stock and it would be worthwhile buying.
The newer "smart" stuff is so terrible that I wouldn't touch it with a bargepole and their 'security' products rate a "Not even with someone else's bargepole" status. Seriously.
The latest generation of UK frigates are noisier than an explosion in a spanner factory - not a good look for something supposed to be capable of antisubmarine warfare.
Any submarines would hear them coming from 100 miles away and make themselves scarce.
And that's quite apart from being as spectacularly unreliable as the german boats - the difference being that the Royal Navy accepted their ones instead of being sensible and sending them back.
Then again this is the same Royal Navy that's bought a pair of Nimitz-size aircraft carriers with non-nuclear fuel source, whilst not having enough ships in the rest of the navy to form a single support group OR any aircraft to fly off them.... (HMS Sitting Duck and HMS White Elephant)
It was - until the USAF got involved. It was their unreasonable demands for payload capacity and cross-glide capabilities (single orbit missions) which caused the massive orbiter growth necessitating a sidemounted configuration.
Having caused the horse to turn into a camel, the USAF looked at it and said "thanks but no thanks"
Demonstrating that capability gave the Soviets a major attack of heartburn to the point that diplomatic channels ran red hot for a while and reportedly led to the including of explosive on-orbit self-destruct systems being fitted to USSR military payloads.
The demonstration proved that the cost of recovering a commercial satellite was higher than simply launching the flight spare - so whilst it _could_ recover a satellite it simply wasn't economic for commercial operations (or most military ones) and touching anything belonging to someone else would probably result in a self-destruct mechanism being triggered resulting in the loss of shuttle+crew.
I'd be more inclined to believe the rumours that classified missions were investigating the practicality of sex in zero gravity than in them going within a few hundred miles of rendezvousing with uncooperative orbiting soviet hardware (The more likely reality is that they were probably NRO missions)
This stuff is right at the edge of the "is it possible?" envelope. Recall that the initial grid-fin landing failed due to running out of hydraulic fluid.
Whatever the cause, SpaceX has been open with its failures up until now - although I'm disturbed by the reaction to losing the centre core when it was clear from the webcast that after the smoke cleared that there was no rocket on the droneship deck (screen on the righthand lower third of the background as the presenters were talking).
Being open about losing the core during the broadcast and not just in the followup press conference would have been more confidence inspiring. The mission was a success the moment it cleared the pad without exploding, a major success when they achieved orbit and an outstanding success when the roadster boosted out of earth orbit. Even if _one_ of the secondary goals wasn't quite achieved I'll bet they don't dunk another centre core in subsequent missions.
"Turbines are routinely designed to contain rotor failure (ever see the jet engine tests?)"
Yup.
Turbines are most emphatically NOT designed to contain rotor failure. This would make them so heavy they can't fly, so they design to make the rotors are reliable as possible and to try and protect critical components (uncontained compressor/turbine rotor failure won't usually down an aircraft but it makes for an exciting day as far as the pilot's concerned and the cause is usually oil starvation)
What they ARE designed to contain is _fan_ blade failure - these are more than capable of bringing down an aircraft if not contained and the fan takes the brunt of anything being ingested (birds, humans, baggage carts....)
In both rocketry and aviation, statistics is used to determine what may fail and what the consequences of failure might be. So far SpaceX has been getting it right and that has a lot to do with not being told how to make every part of the rocket that NASA or the USAF is ordering.
In the case of aviation, 4 engines is heavier and uses more fuel than 2. The first eats into payload and the second reduces range.
The reason airlines are moving to Big Twins is that they now have the range to go long haul and 50+ years of operation has shown the reliability of turbofans so high that they statistically don't need to worry about engine failure as long as the aircraft can climb and maintain altitude on one engine (6 hour ETOPS is mind boggling reliability)
Back in the days of piston engines it was more common than not for transcontinental/transoceanic flights to arrive with one engine dead and they needed a full teardown after virtually every long haul flight. These days inflight engine failure is such a vanishingly rare occurence that a pilot may encounter it only once in their career (the most common failure is down to birdstrike at low level causing a compressor surge which is a strict "turnaround and land again" matter) and engines go months between teardowns.
It's telling that SpaceX's original plans to go from 9 small engines to one big one have been quietly shelved.
"Perfect" is the mortal enemy of "good enough" and the Merlin engines have proven "good enough" for the job, in the same way that Microsoft Word was "good enough" to take market share from competing word processors which cost 10 times as much.
(I suspect that the plans for reusing second stages will also be quietly shelved in favour of simply using older/EOL reflown engines outfitted for vacuum operation. It will achieve the same economies.)
"Even in terms of computer science, there are algorithms that cannot be done in parallel."
In a lot of cases, they CAN be done in parallel, but the people tasked to do the job lack the competence to perform it.
This is a particular problem in astronomy and space physics. I know this because I have to try and support those people and the thing they want is "ten times faster cores", not "ten more cores". This leads to much wailing and gnashing of teeth when they find that dropping a GPU into a system doesn't make their singlethreaded IDL code run any quicker or that a 32 core system is slower than a 4 core one for the tasks they want to run.
For that matter there's the issue of using IDL as a computation tool. It's standard practice despite it being arguably one of the worst possible tools for the job.
Yes, but not so much for the reason you think - the soviet military refused to back the moon missions as they saw no military purpose in them (only propaganda which in this case they saw no point in), which left the N1 program decidedly underfunded.
It didn't help that the N1's technical lead died before the first launch. Without his vision the project foundered due to constant power rivalries within the Soviet structure that somehow the USA managed to avoid.
The Brazailians discovered something similar when they managed to blow up a rocket and most of their technical lead staff on the pad - rockets are replaceable, visionary drive is not and ignoring basic safety protocols can have horribly unforgiving results.
"Many planetary missions had duplicate sets of hardware built"
_ALL_ missions have (at least) duplicate sets built. They're flight spares before launch and used as test articles for software updates or engineering assessments after a successful one.
For production line equipment like GPS birds, the flight spare is the next launch's flight model.
Of course there may be dozens or even hundreds of test models built before the flight models for interplanetary missions. That's why the overall cost is so high but the incremental one not much of an issue.
All things considered, the rocket is usually the least expensive part of the entire bundle but there's an entire virtuous circle in getting launch costs down in terms of the amount of money and time committed to prototyping. Outside of cubesats, science missions are generally incredibly risk-averse due to launch costs, leading to rapidly ballooning R&D costs. Memories of the first Ariane 5 launch (and the loss of the Cluster mission) are probably behind everyone's refusal to take the offered launch space on FH.
(Disclosure, I work in a space laboratory. There's a fair bit of grumbling about "Elon didn't ask _us_ if we'd like to launch something" but I suspect he didn't want to put a european bird on a US launch)
The point about ERTMS is that it's a unified target standard and as such the bugs can be ironed out. The proliferation of incompatible standards and wildly varying operation rules across Europe is a far worse problem than making ERTMS work for all use cases.
It's a bit like going from the various competing old analog mobile phone systems to GSM, with a requirement to keep being able to use the old networks until everyone's cut over.
It's the power to weight ratio of engine AND fuel that matters. For pistons/turbines the mass difference is in the engines, but for electric vs turbines the it's the mass of the fuel vs mass of batteries.
Petroleum-based products are around 10kWh usable per kg. Batteries are a few hundred Wh at very best. (ie, 250-700Wh for raw cells, minus whatever you put around them for protection - less than 1/10 the density)
Some numbers: Even a shorthaul 737 is carrying 10-20 tons of fuel for its day's work
- 737-800 = 26 tons max fuel capacity, 79 ton MTOW and 65tons max landing weight - with batteries the aircraft doesn't lose weight as it flies, so you're effectively instantly losing 14 tons of equivalent fuel/cargo capacity when you go electric by having MOTW=MLW
You start appreciating the scale of the problem. You can have range or carrying capacity but not both.
Puddlejumper airlines don't have the luxury of being able to wait 3-5 hours for recharging after each leg - the vehicle has to carry enough charge at the start of the day to last the day, minus whatever you can stuff into the batteries in 30-90 minutes
EVs have the same problem when used all day. An electric urban delivery vehicle is a great idea and an electric personal hire vehicle/bus can charge in its off-periods. An electric long-haul transporter is going to have its cargo capacity constrained by the battery weight or its usable range constrained by battery capacity - and at least with terrestrial solutions we may be able to develop some form of power feeding to moving cars/trucks that doesn't involve trolley wires
Unless/until we find something that can approach the energy density of petroleum products, the best solution may well be nuclear-sourced synfuel processes (hydrogen -> methane -> diesel/kerosene(*)) that may not be overly efficient, but allow specific vehicles to retain long range without increasing carbon emissions.
(*) There's a _lot_ more hydrogen in a litre of diesel/kerosene/gasoline than a litre of liquid hydrogen and you don't have the faff of having to handle cryogenic liquids and the unsolved problem of hydrogen embrittlement as you would need for fuel cells/electric motor combinations.
The installations may be stationary, but locations in roadside cabinets or hilltops aren't conducive to keeping felines - especially when it may be months to years between visits. Rodents also love to eat cables in ducts.
"Then, there's the autonomous bit, which might get fried by radioactivity leaking from the bomb fuel."
Bomb material isn't particularly radioactive (in fact being too radioactive is usually what makes them fizzle). Electronics-frying levels tend to come from things like caesium and that's not a component of weapons.
Slashdot Mirror
Many people were expecting this to happen sooner or later.
Thanks to nearly a century of lobbying, the USA has a pedestrian-hostile culture and many states have car-centric laws (eg, jaywalking) which state that pedestrians must give way to cars except at designated crossing points and even then can only cross on a green light.
Growing up with these kinds of rules translate into assumptions that people generally don't walk ontot he road unexpectedly, which translate into robotic rules that people NEVER do that - meaning they're not setup to "expect the unexpected"
A human driver should have seen the pedestrian on the median some distance off and reacted (slowed down or changed lanes) even if she was stationary because someone standing there is likely to move. The robot just kept going thanks to cultural assumptions programmed into it that became rigid operating rules.
(Anticipating kids running onto the road is one of the classic tests for danger perception in many countries. Virtually all US drivers fail such tests badly, which is why trading your USA license in for any EU one generally involves at least a full license test.)
Sure, there shouldn't be a pedestrian on the freeway (or a deer either) but I'm not going to run him over. I sure as hell don't want my robot car to do so. Nor do I want it to ignore the 12 point stag, or cow on the road - one of those would come through the windscreen and kill the vehicle occupants. Ditto for the unexpected solo road cone. There might be a pothole behind it.
Arizona has to be one of the worst possible places to test self-driving vehicles. It's extremely car-centric, shitty and hostile for pedestrians and the number of edge/corner cases that happen is low, meaning the human supervisors get complacent. My bet is that the twit in the driver seat didn't even have eyes outside the cabin until after the poor woman in question was bouncing down the road.
"I don't think phytoplankton are affected by minor changes in ocean CO2 levels."
Except it's not minor.
The CO2 concentration in the oceans has changed so much that actual acidity levels have changed 30% in the last 200 years. (pH is a logarithmic scale)
Bear in mind also that in geological history, oceanic CO2 spikes go hand in hand with anoxic events (most of our oil comes from the last couple of these) and half our breathable oxygen comes from the seas.
Can you cope with a drop of oxygen from 19% to 15% at sea level? Hint: 90% of the human population cannot and will die of cardiac congestion caused by physiological changes as their bodies attempt to deal with such a long-term change.
That's quite apart from losing a large chunk of the food chain in an anoxic event. In general very few species averaging larger than about 40kg bodymass appear to have survived on land when this kind of thing has happened in the last and it's likely to stay the same rule of thumb in future.
Increased CO2 gives increased plant activity, but is also stresses the respiration part of their cycles and seems to result in less effective mineral fixation along with greater water uptake, do you might get more greenery, but it looks like it will be much lower quality.
Will they?
The history of the world shows that cities get abandoned by their populations long before the leadership gives up. There are thousands of ghost towns around the planet which were once thriving cities and some of them only date back a few hundred years (gold, oil and mineral rushes, old coal towns, etc etc)
New Orleans is on track to become the first modern climatic ghost town but there will be plenty of others.
The article glosses other another point to bear in mind - given the nature of the subduction fault offshore, in a good sized earthquake there's a good chance that large portions of the land between the fault and the San Andreas fault will sink rapidly as compression stresses are released and this may be accompanied by sections further inland rising a bit (think of a squashed sponge uncrumpling.)
Saying "it's complex" doesn't even come close to explaining the various interactions occuring. Are there any ancient raised beaches along the californian coastline? if so, they're evidence of that area of land being systematically jacked up by sucessive quakes, etc.
Forget the batteries and look at the environmental disasters unfolding in the areas around PV panel manufacturing plants.
Just because it's not in your backyard doesn't mean it's not a problem.
When nuclear power goes bad it doesn't do that either. Perhaps you should look at actual stats for nuclear accidents.
Furthermore, if we'd actually moved to next-gen molten salt designs 50 years ago when they were proven workable and safe instead of killing them for political reasons, we wouldn't have giant steam bombs surrounding our nuclear piles, but a non-pressurised, non-reactive liquid that freezes solid when it goes below 300C and doesn't start boiling until well _ABOVE_ the inherent limiting temperatures of nuclear reactions.
The thing about high level radioactive products is that they're both easily detectable AND shortlived. All you have to do is wait them out and that's measured in human scales.
Three Mile Island's infamous reactor is in the process of being dismantled and cleaned up. Chernobyl will be able to be done in another 20-30 years and Fukushima in 30-40 - and bear in mind that nowhere in Fukushima is more radioactive than Denver Colorado or Downtown Helsinki (both due to natural sources) unless you're silly enough to want to climb inside the nuclear reactors.
Even that "heavily contaminated water" is _less_ radioactive than most BEER
Seriously.
As soon as you add mechanical complexity then you're adding maintenance costs.
If the worker being replaced is paid a low wage, the cost:benefit is low. The low-hanging fruit for automation are more-or-less intellectual but repetitive tasks with higher pay rates and that's been going on for 40 years (when was the last time you saw a room full of accounts ledger clerks scratching away?)
The more likely targets for unemployment are accountants, junior lawyers and suchlike. The investment to do so is lower and the rewards are higher.
Minimum wages won't fall, but the number of higher-paid jobs will decline, bringing more people DOWN to minimum wage.
They never relied on Windows for income, just for mindshare.
The REAL money maker is the Office suite and their other specialist software.
"That suggests to me that a visible, imitation camera would have accomplished the same."
Until people simply start using a baseball cap pulled down over their face - which is what happens regularly around my (visible) UK surveillance cameras.
Too bad they don't see the low-mounted ones with far-IR LEDs...
Video doorbells are usually awful pieces of crap and they almost never integrate into a decent DVR system.
"I have a Ring doorbell. I give it 2 out of 5 stars."
Ring has been an established brand in Europe for a while. 2 out of 5 stars is being generous. Most of what they sell is "whatever was cheapest in Shenzhen that week"
"The concept is ok, but the implementation is terrible."
This, in spades.
Ring was ok when it was basic overpriced (and fragile) dumb shit. Occasionally there would be a sale of old stock and it would be worthwhile buying.
The newer "smart" stuff is so terrible that I wouldn't touch it with a bargepole and their 'security' products rate a "Not even with someone else's bargepole" status. Seriously.
It's not just the Germans.
The latest generation of UK frigates are noisier than an explosion in a spanner factory - not a good look for something supposed to be capable of antisubmarine warfare.
Any submarines would hear them coming from 100 miles away and make themselves scarce.
And that's quite apart from being as spectacularly unreliable as the german boats - the difference being that the Royal Navy accepted their ones instead of being sensible and sending them back.
Then again this is the same Royal Navy that's bought a pair of Nimitz-size aircraft carriers with non-nuclear fuel source, whilst not having enough ships in the rest of the navy to form a single support group OR any aircraft to fly off them.... (HMS Sitting Duck and HMS White Elephant)
It was - until the USAF got involved. It was their unreasonable demands for payload capacity and cross-glide capabilities (single orbit missions) which caused the massive orbiter growth necessitating a sidemounted configuration.
Having caused the horse to turn into a camel, the USAF looked at it and said "thanks but no thanks"
The one and only example.
Demonstrating that capability gave the Soviets a major attack of heartburn to the point that diplomatic channels ran red hot for a while and reportedly led to the including of explosive on-orbit self-destruct systems being fitted to USSR military payloads.
The demonstration proved that the cost of recovering a commercial satellite was higher than simply launching the flight spare - so whilst it _could_ recover a satellite it simply wasn't economic for commercial operations (or most military ones) and touching anything belonging to someone else would probably result in a self-destruct mechanism being triggered resulting in the loss of shuttle+crew.
I'd be more inclined to believe the rumours that classified missions were investigating the practicality of sex in zero gravity than in them going within a few hundred miles of rendezvousing with uncooperative orbiting soviet hardware (The more likely reality is that they were probably NRO missions)
"I wonder if that was just a screw up... "
This stuff is right at the edge of the "is it possible?" envelope. Recall that the initial grid-fin landing failed due to running out of hydraulic fluid.
Whatever the cause, SpaceX has been open with its failures up until now - although I'm disturbed by the reaction to losing the centre core when it was clear from the webcast that after the smoke cleared that there was no rocket on the droneship deck (screen on the righthand lower third of the background as the presenters were talking).
Being open about losing the core during the broadcast and not just in the followup press conference would have been more confidence inspiring. The mission was a success the moment it cleared the pad without exploding, a major success when they achieved orbit and an outstanding success when the roadster boosted out of earth orbit.
Even if _one_ of the secondary goals wasn't quite achieved I'll bet they don't dunk another centre core in subsequent missions.
"Turbines are routinely designed to contain rotor failure (ever see the jet engine tests?)"
Yup.
Turbines are most emphatically NOT designed to contain rotor failure. This would make them so heavy they can't fly, so they design to make the rotors are reliable as possible and to try and protect critical components (uncontained compressor/turbine rotor failure won't usually down an aircraft but it makes for an exciting day as far as the pilot's concerned and the cause is usually oil starvation)
What they ARE designed to contain is _fan_ blade failure - these are more than capable of bringing down an aircraft if not contained and the fan takes the brunt of anything being ingested (birds, humans, baggage carts....)
In both rocketry and aviation, statistics is used to determine what may fail and what the consequences of failure might be. So far SpaceX has been getting it right and that has a lot to do with not being told how to make every part of the rocket that NASA or the USAF is ordering.
"If one blows up you've had it."
For recent example: See Antares (Which was using N1 engines. Coincidence?)
"Falcon doesn't get censored in interviews"
Which is why SpaceX are the best in the falcon business.... :)
In the case of aviation, 4 engines is heavier and uses more fuel than 2. The first eats into payload and the second reduces range.
The reason airlines are moving to Big Twins is that they now have the range to go long haul and 50+ years of operation has shown the reliability of turbofans so high that they statistically don't need to worry about engine failure as long as the aircraft can climb and maintain altitude on one engine (6 hour ETOPS is mind boggling reliability)
Back in the days of piston engines it was more common than not for transcontinental/transoceanic flights to arrive with one engine dead and they needed a full teardown after virtually every long haul flight.
These days inflight engine failure is such a vanishingly rare occurence that a pilot may encounter it only once in their career (the most common failure is down to birdstrike at low level causing a compressor surge which is a strict "turnaround and land again" matter) and engines go months between teardowns.
It's telling that SpaceX's original plans to go from 9 small engines to one big one have been quietly shelved.
"Perfect" is the mortal enemy of "good enough" and the Merlin engines have proven "good enough" for the job, in the same way that Microsoft Word was "good enough" to take market share from competing word processors which cost 10 times as much.
(I suspect that the plans for reusing second stages will also be quietly shelved in favour of simply using older/EOL reflown engines outfitted for vacuum operation. It will achieve the same economies.)
"Even in terms of computer science, there are algorithms that cannot be done in parallel."
In a lot of cases, they CAN be done in parallel, but the people tasked to do the job lack the competence to perform it.
This is a particular problem in astronomy and space physics. I know this because I have to try and support those people and the thing they want is "ten times faster cores", not "ten more cores". This leads to much wailing and gnashing of teeth when they find that dropping a GPU into a system doesn't make their singlethreaded IDL code run any quicker or that a 32 core system is slower than a 4 core one for the tasks they want to run.
For that matter there's the issue of using IDL as a computation tool. It's standard practice despite it being arguably one of the worst possible tools for the job.
"That, and communism"
Yes, but not so much for the reason you think - the soviet military refused to back the moon missions as they saw no military purpose in them (only propaganda which in this case they saw no point in), which left the N1 program decidedly underfunded.
It didn't help that the N1's technical lead died before the first launch. Without his vision the project foundered due to constant power rivalries within the Soviet structure that somehow the USA managed to avoid.
The Brazailians discovered something similar when they managed to blow up a rocket and most of their technical lead staff on the pad - rockets are replaceable, visionary drive is not and ignoring basic safety protocols can have horribly unforgiving results.
"Many planetary missions had duplicate sets of hardware built"
_ALL_ missions have (at least) duplicate sets built. They're flight spares before launch and used as test articles for software updates or engineering assessments after a successful one.
For production line equipment like GPS birds, the flight spare is the next launch's flight model.
Of course there may be dozens or even hundreds of test models built before the flight models for interplanetary missions. That's why the overall cost is so high but the incremental one not much of an issue.
All things considered, the rocket is usually the least expensive part of the entire bundle but there's an entire virtuous circle in getting launch costs down in terms of the amount of money and time committed to prototyping. Outside of cubesats, science missions are generally incredibly risk-averse due to launch costs, leading to rapidly ballooning R&D costs. Memories of the first Ariane 5 launch (and the loss of the Cluster mission) are probably behind everyone's refusal to take the offered launch space on FH.
(Disclosure, I work in a space laboratory. There's a fair bit of grumbling about "Elon didn't ask _us_ if we'd like to launch something" but I suspect he didn't want to put a european bird on a US launch)
Only this time with reliable technology and safety systems.
The point about ERTMS is that it's a unified target standard and as such the bugs can be ironed out. The proliferation of incompatible standards and wildly varying operation rules across Europe is a far worse problem than making ERTMS work for all use cases.
It's a bit like going from the various competing old analog mobile phone systems to GSM, with a requirement to keep being able to use the old networks until everyone's cut over.
https://web.archive.org/web/20...
It's the power to weight ratio of engine AND fuel that matters. For pistons/turbines the mass difference is in the engines, but for electric vs turbines the it's the mass of the fuel vs mass of batteries.
Petroleum-based products are around 10kWh usable per kg. Batteries are a few hundred Wh at very best. (ie, 250-700Wh for raw cells, minus whatever you put around them for protection - less than 1/10 the density)
Some numbers:
Even a shorthaul 737 is carrying 10-20 tons of fuel for its day's work
- 737-800 = 26 tons max fuel capacity, 79 ton MTOW and 65tons max landing weight
- with batteries the aircraft doesn't lose weight as it flies, so you're effectively instantly losing 14 tons of equivalent fuel/cargo capacity when you go electric by having MOTW=MLW
You start appreciating the scale of the problem. You can have range or carrying capacity but not both.
Puddlejumper airlines don't have the luxury of being able to wait 3-5 hours for recharging after each leg - the vehicle has to carry enough charge at the start of the day to last the day, minus whatever you can stuff into the batteries in 30-90 minutes
EVs have the same problem when used all day. An electric urban delivery vehicle is a great idea and an electric personal hire vehicle/bus can charge in its off-periods. An electric long-haul transporter is going to have its cargo capacity constrained by the battery weight or its usable range constrained by battery capacity - and at least with terrestrial solutions we may be able to develop some form of power feeding to moving cars/trucks that doesn't involve trolley wires
Unless/until we find something that can approach the energy density of petroleum products, the best solution may well be nuclear-sourced synfuel processes (hydrogen -> methane -> diesel/kerosene(*)) that may not be overly efficient, but allow specific vehicles to retain long range without increasing carbon emissions.
(*) There's a _lot_ more hydrogen in a litre of diesel/kerosene/gasoline than a litre of liquid hydrogen and you don't have the faff of having to handle cryogenic liquids and the unsolved problem of hydrogen embrittlement as you would need for fuel cells/electric motor combinations.
The installations may be stationary, but locations in roadside cabinets or hilltops aren't conducive to keeping felines - especially when it may be months to years between visits. Rodents also love to eat cables in ducts.
"Then, there's the autonomous bit, which might get fried by radioactivity leaking from the bomb fuel."
Bomb material isn't particularly radioactive (in fact being too radioactive is usually what makes them fizzle). Electronics-frying levels tend to come from things like caesium and that's not a component of weapons.