Waste is very subjective. When you have 100x the supply of something, it becomes cheaper to do stuff, which means other more valuable resources can be saved instead.
We don't manage "information" as much as we do abstract objects. You can have a 5 mebibyte, gibibyte or tebibyte objects, and they will still consume the same amount of mental capacity.
The switch to a new wireless standard is not a cliff, it will happen gradually, there is no need to throw away everything at once. Energy per bit transferred is lower for 5G devices, which makes them more efficient in the future to compensate the need to transfer more bits.
I don't think there is a problem with the former statement at all:
Amplifying stupidity must mean it can solve problems without external intelligence. If it required more intelligence, you'd have a less useful machine.
As I've been saying, you don't know the point at which there is no more to learn. Even minor things can have major impact when combined with further research.
Things like "enormous" are subjective, and it's indeed not impossible to get such amounts cheaply in the future, even if you can't at the time reduce it's need.
It's easy to look in the past and point out the obvious when you already have all the answers, but trying to divine something that changes the way you think about current physics is not as simple.
I think the author is giving lots of value to computation itself, which is not how humans generally allocate it. It's results are seen just as another generic resource.
What do people do when clean water becomes nearly free? They do ridiculously inefficient things. It's easy, efficient use of truly limited resources, like _time_.
There is a reason we don't optimize things beyond their economic value; because it was designed to optimize total resource use.
There is certainly efficiency to consider when making energy. We can't tap into the full potential of any source, there are "hard limits" to many of our current methods, but that doesen't stop a paradigm shift from happening and making it so.
Many things studied before had no "tangible" use then, it doesen't come like that. It's iteration upon iteration, using previous work to achieve a little bit of something new. You cannot foresee the future beforehand.
It isn't "exactly impossible" though, it never is. You can find out a more efficient way to create it, or you might have more power available in the future.
All manmade elements have been extremely expensive to make the first ti-me, and that hasn't stopped them from being constantly used.
The word "auto"(oneself), often used with meaning to do by itself.
Automation is a very broad term, which mechanization is a part, but not a requirement of.
Anything that moves work from human to other forces is automation, having a thing do things by itself. An automobile is the automating of our movement.
The simpler the task seems, the harder it must be. It's exactly why the task exists at all, it could not have been improved on before either and stopped there
In fact automation often goes top-down, reeucing complexity of difficult jobs and allowing lower-skilled workers to deal with the rest that isn't yet profitable to change.
It should be noted that as efficiency increases, the company can compete and lower prices, meaning people don't _have to_ go to work as much to achieve the same level of living.
All things automation. Really a chainsaw is automation, and so is a compressor. Automation is a pen. They all remove work and increase efficiency.
There are lots of examples which don't really count though, such as self-checkout, which have exactly the same amount of work left, only who works it is a different person.
Any load balancer will let you do that. Scaling is a few lines of scripts to do on any platform, and well worth the couple minutes, you don't even need a tool for just that.
No tool "works the rest out". It's _always_ a compromise, because inherent complexity can never be removed, only moved. What you gain in one area, encumbers another.
You may freely use k8s, but it's not magical nor easier to use than any existing systems. In fact adopting it often takes non-trivial time and the web is full of failure stories with very benign warnings and catastropic results.
_any_ non-proprietary tool is "cloud agnostic". Kubernetes is bundled software, and achieves the things those pieces achieve. There is nothing holy about k8s specifically, the tools you train with are easy, and it's very easy to get skewed opinions on that.
For example a lot of people would find writing scripts cumbersome, but not a person who's written a lot of them. They're not any more fragile than other logic error capable software is
Observing complex reactions with chemicals requires much longer study than simply beaming things with microwaves. Results are nearly immediate, and any long term effect would have to be shared with everything else that warms things.
Increasing attention available should be good regardless of how it would be used, even the worst-case scenario would be where we are now, only we get more done
Increasing the responsiveness of everything should have the exact opposite effect, because attention is most affected by latency. If you're always immediately getting what you're looking for immediately you can't be distracted either.
Depending on how many people a single point serves, it can imply a pretty sizeable chunk of energy. Smaller things add up when you're talking of hundreds of millions of people
Attention is highly latency sensitive, and bringing reaction for your action a second sooner has major implications, and significantly increases your ability to interact with and understand content