I see you ignored the part of my post where I showed up the foolishness of your argument from increduality already, seeing as how you've listed electricity at the top and I've dealt with it already.
You've also failed to show why any of those resources should be considered impossible to provide to a fixed population of approximately todays size, or would be completely impossible to substitute with alternatives. Instead, you've simply been pointing to the absolute size of the numbers without any reference to the size of the supplies we're dealing with.
OK, I don't think you get it yet. Take the sample population of Paruro, Peru, about 3500 people living in a stunningly beautiful valley. To dramatically over-simplify, assume that all 3000 poor people use X-amount of resources, and all 500 middle class people use 10X resources.
(3000*X) + (500*10X) = 8000X
Wave your magic wand and make everyone educated and middle class. The population is now breeding at the replacement level. Great, right? Frack, no. They're now using 3500*10X = 35000X resources.
There's the problem with feel-good solutions to the population issue, there really isn't enough to go around. The only way to have everyone living comfortably enough that this solution works is to dramatically reduce the world's population first. Which works out to a really feel-bad solution before you can implement the feel-good solution.
Sorry.
35,000x the resources is not an unsustainable increase. 35,000x of what total? Do you know the total? Are they all the same resource or just a broadening diversity of resources? People living opulently don't consume 1000x the food and the resources they consume when increasing their living standard are largely self-produced in the first place across the breadth of the society they create.
So the question is, how do we find those resources and what are they? Well, obviously they need to be sustainable in some way, and the nature of the problem answers most of these questions: in Peru for example, the average energy consumption is 27.2 kWh / month (http://www.esmap.org/sites/esmap.org/files/ESMAP_PeruNationalSurvey_Web_0.pdf). This is pretty staggeringly low. Per day my house consumes about 5x that amount, which should it in perspective.
Per month, this equates to about 20x the energy consumption of a Perusian household per person.
And wow that's a lot of energy - only it's not. 66% of that number is generated by solar panels on our roof. That number is 66% because our roof geometry is tilted and non-optimal for any more. So...despite consuming 20x the amount of energy - per person - then the average Peruvian household, it's simply not an issue. The volume of available energy of that form is enormous and can brook exponentially more consumption - and this is before accounting for, again, efficiency improvements - which allow everyone to have proportionally more realized resources per unit raw resources.
promote the development of middle-classes... track record of dropping birthrates
Fail. Complete and utter logical failure. I see this stupid argument every time population levels are talked about. My sister-in-law lives a nice middle class (by Peruvian standards) life with her two kids in Puno. Her cousin is a poor farmer in Paruro with six or seven kids. Which family uses more resources? The smaller family, with their running water, propane stove, brick house, refrigerator, and electric lights, uses easily ten times the resources (probably more) than their cousins. Smaller families != less resources used unless their standard of living is the same.
Do you not see then why the population problem solves itself? People's resource consumption on the whole rises, but they drop their average fertility rate down to the replacement level (two kids). There resource use thus ceases to rise because there is not exponentially more people in successive generations. Every efficiency measure or new resource source thus benefits them enormously and raises their standard of living - it's not eaten up by needing to be divided amongst a growing population.
And, by contrast, poor regions also solve the same problem - they're poor. There resource use is lower, and stays low so long as they're poor and having (on average) more then the replacement rate of children.
We will have to replace a large fraction of our current oil consumption within 20 to 30 years. Coal is not an option since CTL emits way more CO2 than burning oil directly. We're still going to do I'm afraid.
Oil is a very different question to electricity production, and hardly intractable. Aircraft will be powered by biofuels - there's a lot of work being done on producing suitable blends from biomass to run in current generation turbines without any need for conversions.
As for light vehicles - honestly who cares? People drive their cars too much already, they'll be phased out largely by price increases leading people to walk more, use public transport more, and drive less. It'll be the least worrisome free market adaptation ever. Tesla and the Japanese will move in and buy the US auto industry when they fail to adapt again.
That leaves long-haul trucking and container ships. Short term we'll see local goods get more popular and that trend will probably stay - container shipping will move to nuclear power (since companies are already trying to do it). Farming wise I don't know, but no one's managed to dissuade me of the notion that you couldn't power a combine harvester off of batteries, or just a really long cable, seeing as they run over fixed areas with a relatively short haul distance. There's just been no incentive to figure it out yet because gas has been so cheap.
You're absolutely correct. The National Institute of Health has studies that show the Placebo effect can actually be effective treatment for certain diseases in certain cases, such as certain forms of mild depression (in the head). There's a real fine line between what the brain can control and real, physical or biological issues that can't be controlled by the brain. Placebos are quite cheap compared to actual medicine,depending on what one considers a valid Placebo.
The laying of a Priest's hands is financially cheap, and might be just enough to help someone if they believe enough in the action to activate the Placebo Effect.
On the other hand, the long-term cost of placebos due to the erosion of trust in the medical establishment which they require generally speaks against their use, not to mention the general erosion of the placebo effect as a result - it only works if you think you've been given real medicine.
But you don't make plans based on an undemonstrated ability. It's possible that tomorrow I may phase through the floor due to an unrealized psionic power in the human mind suddenly manifesting.
Of course this has never been demonstrated in practice, so I don't structure my life around it as a possibility - in fact it's so unlikely that I don't bother considering it at all, and until such time as that changes there's no reason too because there isn't even compelling theoretical evidence that such a thing is possible.
Why 20 or 30 years? There's no grand clock on this, other then concern about whatever may be a critical level of climate change. It's a long game, and we're not going to run out of coal in the next 20 or 30 years (oil, maybe). There's no physics which means we can't run the world off of renewables - we get 7400 times the energy we currently use from the sun every day.
In the next 100 years we're going to be done with oil - probably sooner, and Western nations are hardly going to brook putting dirty coal power stations up all over the place and electricity is the one thing you can't just outsource either.
You might want to start throwing down some more solid numbers on how untenable you think directly powering farming infrastructure would be, because this conversation is frightfully vague.
A new nuclear plant is currently averaging about $10 billion to construct, which includes various lifetime costs such as eventual decommissioning. So 1,000 of those (and it's unlikely at that scale of construction they'd all cost $10 billion) would put us about $10 trillion. That's a lot - but we're talking about massive restructuring of society and over a considerable timeframe (say, 50 to 100 years). So $2 trillion a year investment globally. This is not an untenable or unreasonable number, and again, it is likely far lower if we were building at a high rate. Moreover, I strongly suspect we don't actually need 1,000 nuclear powerplants to keep farming going.
For comparison, the US government's current debt is about $13 trillion, so that level of borrowing is also clearly sustainable to fund such a venture.
With regards to failures, well, what reactor type are we talking about? Any newer type for example, the probability of meltdown is 0. Not small, not some fractional %, but 0 - it is physically impossible within the reactor geometry for the plant to enter a meltdown phase - and as a result it also can't do the radioactive steam and hydrogen trick that we saw at Fukushima (which was a very old plant anyway).
I get it, arguments of sacrifice sound hippy and antiquated, and I picked a couple of touchy or cliched examples. But supply side manipulation can't solve complex problems like this. If you can balance your equation by also tweaking the demand side, you suddenly have many more potential solutions that don't rely on science fiction technology for supply. Just because folk live in a world that increasingly utilizes electricity doesn't mean we actually need all of those sources around us to still maintain a high quality of living. The author points to the demand side solution herself at the end of her article. But if these arguments sound like unacceptable sacrifices to you, how would you propose altering the demand side?
The article is so bad I'm not sure it's possible to give a coherent response in it's context though, because my argument is that we don't need to alter the demand side in any of the ways that's proposed. We can expedite a natural economic reality but incentivizing energy efficient appliances, and targeting our city and community planning to try and cut down on commute times or encourage the development of new urban centres (as opposed to sprawl progressively further out from one central location).
Globally we can fund women's education, and promote women's rights (which, by extension, benefits both genders - gender discrimination is as bad for men as it is for women and when it comes to education and development doubly so in a lot of cases), and in turn promote the development of middle-classes all over the world - which has a long track record of dropping birthrates. Specific enterprises - like trying to get refugees settled into permanent communities, are important in this regard too.
The idea that 2.9% increase in energy usage is inevitable into the future is just farcical though. Hundreds of years from now we could very well be travelling to other stars and harvesting black holes or whatever - it's a timeframe so long as to irrelevant.
If you study the past evolution of energy consumption in the US, it has been increasing continously, and that is despite equipments having been made more energy efficient. But you probably know more than the agencies doing the statistics for the government I suppose.
It has not however been increasing at the same rate. In fact in NSW in Australia for the past few years energy consumption has actually dropped. This has put some power companies in a bad place since they were upgrading and building transmission lines which are just not needed at all since the demand hasn't risen as it was expected to.
The Fe -> FeCl half-cell would require about 3 cubic meters of electrolyte at mild molarity (2M) to hold enough power for several days continuous operation of the average household.
With the right chemistry, you can store a staggering amount of power, the trick is to do it efficiently.
You don't strictly have to get an energy payoff - you could be energy negative, but still positive compared to the cost of synthesizing hydrocarbons from other sources. Even if it's no good for fuel, we still need large quantities of hydrocarbons for plastics production and to produce the basic solvents and low-end precursors for chemical synthesis (the complex ones we usually get from plants though - they're just better at asymmetric synthesis).
Economics is not a zero-sum game, and the prevalence of technology reduces it's price. Every industrial revolution has taken on average about half as long as the previous one. That's because every time one happens, the technology and know how for it has been able to be bought and imported, rather then needing to be built and invented.
More importantly, those declining population figures are all about female education and empowerment - two things, which are also correlated strongly with economic development and overall increases in the standard of living (probably as a little bit of cause and little bit of effect too).
It wasn't realistic to build a few hundred thousand oil wells when we first started (or drill for oil in the arctic, or out in deep oceans, or to open up oil fields in the middle east...)
"Not realistic" is stupid. If we can build more then 1, then we can build a few thousand presuming there's no actual limit on the fundamental resources (and for nuclear there isn't).
I like the notion that we're going to expand our energy consumption at a constant rate such that apparently, despite the massive amount of energy covering the entire planet in solar panels would net, it wouldn't be enough. But we're totally going to, what, run it off of coal and oil?
Solar energy is a great alternative energy source but the current technology for photovoltaic cells include too many rare earth elements that make them expensive and limit the amount of them that can be made. If you can make solar cells out of cheaper, more abundant materials then solar becomes more likely. Or, you go back to old fashioned boiler plate tech using mirrors to concentrate the light to generate heat to create steam or melt salt. Steam can drive a turbine directly or molten salt can be used as an intermediate storage so that steam can be generated through times when the sun isn't available.
The rare earth argument never makes any sense. It's suddenly become the "in thing" to throw out "oh too many rare earth's" when anyone mentions electronics or high technology, but no one ever comments on the specifics - it's just been picked up since the China-Japan thing happened.
PV cells are predominantly thin-wafers of silicon. Doped silicon - but the dopants are not rare-earth's, they're boron/arsenic which are both very common.
Rare-earth metals are used in high-performance high-speed transistors and permanent magnet electric motors (principally neodymium in that case). They're not used to the produce the materially voluminous bulk of solar cells, although they would be used in grid-feed inverters (presuming they're the non-isolated type - which is not certain either).
Use x2go. In my experience it's much better then FreeNX - faster and easier to consider, and also without the weird "everyone shares the same SSH keys" thing (it just connects to regular system SSH accounts, and sets everything up from there).
X is completely useless over anything more than a local network. Between South America and Europe, the only viable approach I've found is to run an Xvnc session remotely and then VNC to that. This is more or less what is proposed for remote Wayland, so we lose absolutely nothing by switching from X to Wayland (once all the necessary parts are implemented). Not sure where this unrealistic pro-X anti-Wayland FUD comes from. Nostalgia for dying 80s era technologies?
Experience with VNC. Slinging pixels sounds like a horrible idea if you've ever used VNC. Of course the issue is VNC is just slow, since obviously if I can stream 1080p from YouTube my quad core i7 should be capable of doing something similar with it's desktop output.
That said, it does seem like a missed opportunity in Wayland not to let applications pass some type of sub-information about drawable window space up to the compositor for this purpose - Wayland deals with them as flat buffers, there's no protocol way to send any information about the internal layout (i.e. it'd be great to be able to declare some sub-regions, and inform whichever remote end to restore cached information about them).
SSH is ridiculously slow for things that aren't text sessions (and the amount of advice which is "do your backup over SSH" is staggering considering this).
Over any reasonably fast link you need to be running the HPN-SSH patchset to get performance above 2 mb/s in my experience. The MT-AES cipher also helps (though breaks if your application forks, though I've found very few cases where this is a problem currently in practice).
Slashdot Mirror
I see you ignored the part of my post where I showed up the foolishness of your argument from increduality already, seeing as how you've listed electricity at the top and I've dealt with it already.
You've also failed to show why any of those resources should be considered impossible to provide to a fixed population of approximately todays size, or would be completely impossible to substitute with alternatives. Instead, you've simply been pointing to the absolute size of the numbers without any reference to the size of the supplies we're dealing with.
OK, I don't think you get it yet. Take the sample population of Paruro, Peru, about 3500 people living in a stunningly beautiful valley. To dramatically over-simplify, assume that all 3000 poor people use X-amount of resources, and all 500 middle class people use 10X resources.
(3000*X) + (500*10X) = 8000X
Wave your magic wand and make everyone educated and middle class. The population is now breeding at the replacement level. Great, right? Frack, no. They're now using 3500*10X = 35000X resources.
There's the problem with feel-good solutions to the population issue, there really isn't enough to go around. The only way to have everyone living comfortably enough that this solution works is to dramatically reduce the world's population first. Which works out to a really feel-bad solution before you can implement the feel-good solution.
Sorry.
35,000x the resources is not an unsustainable increase. 35,000x of what total? Do you know the total? Are they all the same resource or just a broadening diversity of resources? People living opulently don't consume 1000x the food and the resources they consume when increasing their living standard are largely self-produced in the first place across the breadth of the society they create.
So the question is, how do we find those resources and what are they? Well, obviously they need to be sustainable in some way, and the nature of the problem answers most of these questions: in Peru for example, the average energy consumption is 27.2 kWh / month (http://www.esmap.org/sites/esmap.org/files/ESMAP_PeruNationalSurvey_Web_0.pdf). This is pretty staggeringly low. Per day my house consumes about 5x that amount, which should it in perspective.
Per month, this equates to about 20x the energy consumption of a Perusian household per person.
And wow that's a lot of energy - only it's not. 66% of that number is generated by solar panels on our roof. That number is 66% because our roof geometry is tilted and non-optimal for any more. So...despite consuming 20x the amount of energy - per person - then the average Peruvian household, it's simply not an issue. The volume of available energy of that form is enormous and can brook exponentially more consumption - and this is before accounting for, again, efficiency improvements - which allow everyone to have proportionally more realized resources per unit raw resources.
promote the development of middle-classes ... track record of dropping birthrates
Fail. Complete and utter logical failure. I see this stupid argument every time population levels are talked about. My sister-in-law lives a nice middle class (by Peruvian standards) life with her two kids in Puno. Her cousin is a poor farmer in Paruro with six or seven kids. Which family uses more resources? The smaller family, with their running water, propane stove, brick house, refrigerator, and electric lights, uses easily ten times the resources (probably more) than their cousins. Smaller families != less resources used unless their standard of living is the same.
Do you not see then why the population problem solves itself? People's resource consumption on the whole rises, but they drop their average fertility rate down to the replacement level (two kids). There resource use thus ceases to rise because there is not exponentially more people in successive generations. Every efficiency measure or new resource source thus benefits them enormously and raises their standard of living - it's not eaten up by needing to be divided amongst a growing population.
And, by contrast, poor regions also solve the same problem - they're poor. There resource use is lower, and stays low so long as they're poor and having (on average) more then the replacement rate of children.
We will have to replace a large fraction of our current oil consumption within 20 to 30 years. Coal is not an option since CTL emits way more CO2 than burning oil directly. We're still going to do I'm afraid.
Oil is a very different question to electricity production, and hardly intractable. Aircraft will be powered by biofuels - there's a lot of work being done on producing suitable blends from biomass to run in current generation turbines without any need for conversions.
As for light vehicles - honestly who cares? People drive their cars too much already, they'll be phased out largely by price increases leading people to walk more, use public transport more, and drive less. It'll be the least worrisome free market adaptation ever. Tesla and the Japanese will move in and buy the US auto industry when they fail to adapt again.
That leaves long-haul trucking and container ships. Short term we'll see local goods get more popular and that trend will probably stay - container shipping will move to nuclear power (since companies are already trying to do it). Farming wise I don't know, but no one's managed to dissuade me of the notion that you couldn't power a combine harvester off of batteries, or just a really long cable, seeing as they run over fixed areas with a relatively short haul distance. There's just been no incentive to figure it out yet because gas has been so cheap.
I suggest wikipedia for some light reading: http://en.wikipedia.org/wiki/Economics_of_new_nuclear_power_plants
You're absolutely correct. The National Institute of Health has studies that show the Placebo effect can actually be effective treatment for certain diseases in certain cases, such as certain forms of mild depression (in the head). There's a real fine line between what the brain can control and real, physical or biological issues that can't be controlled by the brain. Placebos are quite cheap compared to actual medicine,depending on what one considers a valid Placebo.
The laying of a Priest's hands is financially cheap, and might be just enough to help someone if they believe enough in the action to activate the Placebo Effect.
On the other hand, the long-term cost of placebos due to the erosion of trust in the medical establishment which they require generally speaks against their use, not to mention the general erosion of the placebo effect as a result - it only works if you think you've been given real medicine.
But you don't make plans based on an undemonstrated ability. It's possible that tomorrow I may phase through the floor due to an unrealized psionic power in the human mind suddenly manifesting.
Of course this has never been demonstrated in practice, so I don't structure my life around it as a possibility - in fact it's so unlikely that I don't bother considering it at all, and until such time as that changes there's no reason too because there isn't even compelling theoretical evidence that such a thing is possible.
Why 20 or 30 years? There's no grand clock on this, other then concern about whatever may be a critical level of climate change. It's a long game, and we're not going to run out of coal in the next 20 or 30 years (oil, maybe). There's no physics which means we can't run the world off of renewables - we get 7400 times the energy we currently use from the sun every day.
In the next 100 years we're going to be done with oil - probably sooner, and Western nations are hardly going to brook putting dirty coal power stations up all over the place and electricity is the one thing you can't just outsource either.
You might want to start throwing down some more solid numbers on how untenable you think directly powering farming infrastructure would be, because this conversation is frightfully vague.
A new nuclear plant is currently averaging about $10 billion to construct, which includes various lifetime costs such as eventual decommissioning. So 1,000 of those (and it's unlikely at that scale of construction they'd all cost $10 billion) would put us about $10 trillion. That's a lot - but we're talking about massive restructuring of society and over a considerable timeframe (say, 50 to 100 years). So $2 trillion a year investment globally. This is not an untenable or unreasonable number, and again, it is likely far lower if we were building at a high rate. Moreover, I strongly suspect we don't actually need 1,000 nuclear powerplants to keep farming going.
For comparison, the US government's current debt is about $13 trillion, so that level of borrowing is also clearly sustainable to fund such a venture.
With regards to failures, well, what reactor type are we talking about? Any newer type for example, the probability of meltdown is 0. Not small, not some fractional %, but 0 - it is physically impossible within the reactor geometry for the plant to enter a meltdown phase - and as a result it also can't do the radioactive steam and hydrogen trick that we saw at Fukushima (which was a very old plant anyway).
I get it, arguments of sacrifice sound hippy and antiquated, and I picked a couple of touchy or cliched examples. But supply side manipulation can't solve complex problems like this. If you can balance your equation by also tweaking the demand side, you suddenly have many more potential solutions that don't rely on science fiction technology for supply. Just because folk live in a world that increasingly utilizes electricity doesn't mean we actually need all of those sources around us to still maintain a high quality of living. The author points to the demand side solution herself at the end of her article. But if these arguments sound like unacceptable sacrifices to you, how would you propose altering the demand side?
The article is so bad I'm not sure it's possible to give a coherent response in it's context though, because my argument is that we don't need to alter the demand side in any of the ways that's proposed. We can expedite a natural economic reality but incentivizing energy efficient appliances, and targeting our city and community planning to try and cut down on commute times or encourage the development of new urban centres (as opposed to sprawl progressively further out from one central location).
Globally we can fund women's education, and promote women's rights (which, by extension, benefits both genders - gender discrimination is as bad for men as it is for women and when it comes to education and development doubly so in a lot of cases), and in turn promote the development of middle-classes all over the world - which has a long track record of dropping birthrates. Specific enterprises - like trying to get refugees settled into permanent communities, are important in this regard too.
The idea that 2.9% increase in energy usage is inevitable into the future is just farcical though. Hundreds of years from now we could very well be travelling to other stars and harvesting black holes or whatever - it's a timeframe so long as to irrelevant.
If you study the past evolution of energy consumption in the US, it has been increasing continously, and that is despite equipments having been made more energy efficient.
But you probably know more than the agencies doing the statistics for the government I suppose.
It has not however been increasing at the same rate. In fact in NSW in Australia for the past few years energy consumption has actually dropped. This has put some power companies in a bad place since they were upgrading and building transmission lines which are just not needed at all since the demand hasn't risen as it was expected to.
The Fe -> FeCl half-cell would require about 3 cubic meters of electrolyte at mild molarity (2M) to hold enough power for several days continuous operation of the average household.
With the right chemistry, you can store a staggering amount of power, the trick is to do it efficiently.
For transport just use nuclear powered cargo ships. It works for aircraft carriers, it can certainly work here.
You don't strictly have to get an energy payoff - you could be energy negative, but still positive compared to the cost of synthesizing hydrocarbons from other sources. Even if it's no good for fuel, we still need large quantities of hydrocarbons for plastics production and to produce the basic solvents and low-end precursors for chemical synthesis (the complex ones we usually get from plants though - they're just better at asymmetric synthesis).
Economics is not a zero-sum game, and the prevalence of technology reduces it's price. Every industrial revolution has taken on average about half as long as the previous one. That's because every time one happens, the technology and know how for it has been able to be bought and imported, rather then needing to be built and invented.
More importantly, those declining population figures are all about female education and empowerment - two things, which are also correlated strongly with economic development and overall increases in the standard of living (probably as a little bit of cause and little bit of effect too).
It wasn't realistic to build a few hundred thousand oil wells when we first started (or drill for oil in the arctic, or out in deep oceans, or to open up oil fields in the middle east...)
"Not realistic" is stupid. If we can build more then 1, then we can build a few thousand presuming there's no actual limit on the fundamental resources (and for nuclear there isn't).
Also "middle class" is a shifting definition. Middle class in 2075 is likely to mean just the 2 skycar family.
Everytime everytime we discuss energy policy people jump to the self-sacrifice well, which then torpedos the entire affair.
The refrigerator you buy today is 3 times as big, and uses 1/3rd the energy, of one you could buy in the 1980s.
Paganistic morality plays have no relevance to a world of ever increasing means and understanding, not everything requires "sacrifice".
I like the notion that we're going to expand our energy consumption at a constant rate such that apparently, despite the massive amount of energy covering the entire planet in solar panels would net, it wouldn't be enough. But we're totally going to, what, run it off of coal and oil?
Solar energy is a great alternative energy source but the current technology for photovoltaic cells include too many rare earth elements that make them expensive and limit the amount of them that can be made. If you can make solar cells out of cheaper, more abundant materials then solar becomes more likely. Or, you go back to old fashioned boiler plate tech using mirrors to concentrate the light to generate heat to create steam or melt salt. Steam can drive a turbine directly or molten salt can be used as an intermediate storage so that steam can be generated through times when the sun isn't available.
The rare earth argument never makes any sense. It's suddenly become the "in thing" to throw out "oh too many rare earth's" when anyone mentions electronics or high technology, but no one ever comments on the specifics - it's just been picked up since the China-Japan thing happened.
PV cells are predominantly thin-wafers of silicon. Doped silicon - but the dopants are not rare-earth's, they're boron/arsenic which are both very common.
Rare-earth metals are used in high-performance high-speed transistors and permanent magnet electric motors (principally neodymium in that case). They're not used to the produce the materially voluminous bulk of solar cells, although they would be used in grid-feed inverters (presuming they're the non-isolated type - which is not certain either).
Use x2go. In my experience it's much better then FreeNX - faster and easier to consider, and also without the weird "everyone shares the same SSH keys" thing (it just connects to regular system SSH accounts, and sets everything up from there).
X is completely useless over anything more than a local network. Between South America and Europe, the only viable approach I've found is to run an Xvnc session remotely and then VNC to that. This is more or less what is proposed for remote Wayland, so we lose absolutely nothing by switching from X to Wayland (once all the necessary parts are implemented). Not sure where this unrealistic pro-X anti-Wayland FUD comes from. Nostalgia for dying 80s era technologies?
Experience with VNC. Slinging pixels sounds like a horrible idea if you've ever used VNC. Of course the issue is VNC is just slow, since obviously if I can stream 1080p from YouTube my quad core i7 should be capable of doing something similar with it's desktop output.
That said, it does seem like a missed opportunity in Wayland not to let applications pass some type of sub-information about drawable window space up to the compositor for this purpose - Wayland deals with them as flat buffers, there's no protocol way to send any information about the internal layout (i.e. it'd be great to be able to declare some sub-regions, and inform whichever remote end to restore cached information about them).
SSH is ridiculously slow for things that aren't text sessions (and the amount of advice which is "do your backup over SSH" is staggering considering this).
Over any reasonably fast link you need to be running the HPN-SSH patchset to get performance above 2 mb/s in my experience. The MT-AES cipher also helps (though breaks if your application forks, though I've found very few cases where this is a problem currently in practice).
Getting away from the central server topology would be nice. Both of the two you listed don't quite manage it.
All the silicon in the world was produced by releasing oxygen into the atmosphere when it was refined. No danger there.