adrianN

„Something similar” can completely change asymptotic bounds though. Compare for example Integer Linear Programming and normal Linear programming. They look “very similar”, but there is an exponential worst case gap (unless P=NP).

Do we actually know enough about neural networks to be able to say how many parameters you need for a given performance? I’m under the impression that we’re still in the phase of trying things out and we can’t rule out different architectures massively reducing computational needs.

It is my understanding that mainland China is the rebel province ;)

Critical dependencies on foreign countries can be problematic if those countries start doing things you don't like, for example invading Taiwan.

An important function of fiction is showing us things to aspire to that are not quite possible in real life.

Heat pumps lose a lot of efficiency when they need to provide higher temperatures. Fossil fuel often heats the water in the radiators up to 70C in old systems, so that reasonably sized radiators have enough power to heat the poorly insulated building. You can do that with heat pumps too, but you get rubbish COP. Adding insulation or bigger radiators can be necessary to make the heat pump economically viable.

Heating up a pile of dirt to a few hundred degrees sounds easier and more energy efficient.

There were no solar panels because we didn’t know how to build them until the 20th century. Today you can run the factories that make them on renewable energy without a problem.

Oil also loses quite a bit during refining and transport.

An electric dehumidifier in winter keeps moisture in check and you get the energy of evaporation back as heat.

Of course, but with a heat pump you get the combo of lower demand and much higher efficiency. In Germany for example due to high electricity prices you really need COP>3 or so to beat fossil fuel for heating economically.

Did you have to improve insulation or do the heat pumps provide enough power?

Gas is used as backup in a grid dominated by cheap renewable. High capex sources like nuclear are a poor replacement.

Nuclear power replaced coal in Germany, not gas. A big dependence exists because people heat their homes with gas and lots of industry needs gas, nuclear doesn’t help with either.

Even if the brain were a quantum computer (I think the jury is still out on that one) we could simulate it just fine on a classical computer, it just has exponential slowdown in the worst case.

It is my understanding that energy produces a gravitational field regardless of the form it takes.

Pinworms are mostly harmless and trivially treated.

Siemens is making a lot of money selling inertial stabilizers. There is ongoing research on how to stabilize grids in the face of decreasing inertial frequency stabilization.

Over the lifetime of the battery it’s not that bad and gets better as the carbon intensity of the grid that runs the factories improves.

You can get through a Computer Science degree with just a few thousand lines of code if you focus on theory courses.

I like scratches on my machines, it gives them character.

An error in the formalization could also be that you formally prove something different from what you think you proved.

It looks like we’re pretty much in perfect agreement, but I think it doesn’t hurt to already start planning what to do about the last couple of percent. These are all projects with long time horizons

Gas turbines can also run on synthetic methane or hydrogen, which is nice.

I’d keep gas plants around, but anyway, you need to cover that case and can’t just handwave it away.

You generally get more return on investment if you invest in building infrastructure in your own country rather than investing in blowing up infrastructure in a different country. So I’m implying that solar and batteries are a lot cheaper than the war in Iraq.

The war in Iraq cost a trillion dollars.

Rooftop solar has the benefit that you distribute the investment cost over a large number of people, even if the costs are a lot higher than grid-scale installations on the ground.

I'm surprised to hear that you need to replace roofs every fifteen years. What are they made of? The roofs I know last decades longer than that.

Thermal storage benefits from the square/cube scaling, so the amount of insulation goes down as you scale up the storage.

A week of low solar and low wind is a yearly or every-couple of years event that you definitely need to solve for a completely renewable grid.