What I'd like to see is the governments investing in things like high-voltage DC that ultimately make it more profitable to burn less CO2. Imagine if solar from the Sahara could reach the cloudy UK with only 30% energy loss, hydro dams could store more water when it's windy or sunny anywhere in their hemisphere and generate more when other renewables are not available, and energy users who are able to time-shift their demand (e.g. electric car charging at home) can do so in a free market that naturally helps blunt the peak hour demands.

I think a planet-scale free-market grid is not quite feasible with off-the-shelf technology quite yet. However it seems to me that if we had one, in most cases you'd have to subsidize fossil fuels to keep them competitive with renewables. My hope is that rich countries can try to pave the way to a planet-scale electricity grid, the consequence of which will be that fossil fuels die a natural death with no subsidy needed.

If this sounds like science fiction, check this video out: https://www.youtube.com/watch?.... Some early long-distance transmission projects have paid back their initial investment 100% in just one year. I would love to see governments at least getting out of the way of, or even better investing in, seriously high-power long distance electrical transmission.

Lucky you! My guess is that the widespread vaccination against the lung-targeting spike protein destroyed most of the food for the lung-targeting original strain, and that's why subsequent strains are less of a problem to the lungs. Even if you didn't get vaccinated, you may have benefited from folks who did.

The Covid vaccines create mostly IgG antibodies that mainly appear in the bloodstream, not IgA antibodies that appear in our mucous. That's the norm for injection (as opposed to live) vaccines, and you are right that we should not expect injected vaccines to do a perfect job of protecting against respiratory infections. That's the norm for all vaccines against respiratory infections - you expect vaccines to improve the second-layer defenses quite a bit, but not necessarily to eliminate infections.

Pre-vaccine, the original virus in circulation targeted the lungs with its spike protein. The case fatality rate was 1% - 2%. About 1 out of 600 people in my county died of Covid. My boss's younger brother in his 30s died of Covid; he didn't get the vaccine and got an early strain.

The vaccine targeted the spike protein that enabled serious lung infections. Post-vaccine, we're not seeing a ton of serious lung infections. The strains of Covid that can circulate in the population are not the ones that are using the lung-targeting spike proteins that the vaccination protected against.

Overall, I think the vaccines did a great job of preventing higher death rates, as well as chronic lung damage and associated complications. Expectations of the vaccine were too high; people hoped Covid would be like Smallpox, when in fact it's more like the Spanish Flu: not a mortal threat for many people anymore, but its descendants are still out there.

To summarize, the vaccines were awesome, life-saving tools that may have reduced mortality by better than 90% and have enabled us to get back to a much more normal lifestyle a lot sooner. Their PR has been bad, especially lately, but please, PLEASE consider the evidence before you try to persuade anybody else that they were basically worthless.

From Wikipedia (https://en.wikipedia.org/wiki/Lunar_resources#Helium-3):

Materials on the Moon's surface contain helium-3 at concentrations estimated between 1.4 and 15 parts per billion (ppb) in sunlit areas, and may contain concentrations as much as 50 ppb in permanently shadowed regions. For comparison, helium-3 in the Earth's atmosphere occurs at 7.2 parts per trillion (ppt).

OK, so He-3 is perhaps a factor of 1000 more abundant on the moon than on the Earth. It's still way less than one part per million. Is the juice really worth the squeeze, or is this venture more accurately described as a way of mining naive investors?

Yes, the Sun is incredibly important as a vast source of free energy. The consensus is that photosynthesis evolved a little bit after life arose (see https://www.ncbi.nlm.nih.gov/p... for example). There are some geochemical processes like alkaline vents and hot springs that provide a steady enough source of free energy to get life going, but as soon as photosynthesis evolved the Sun became our main meal ticket.

I completely agree!

I would say a star doesn't require an external energy gradient to keep being a star. It needed free energy to become a star, but it isn't reliant on continued external flows of free energy to keep on being a star.

Channeling free energy from an external source is a key property of living matter.

Hats off to understanding more about RNA world and how it might have worked. That's great. That said, a growing number of physicists like me think that RNA world could not have been the first life as the headline implies. There's a debate between genome-first and metabolism-first scientists, and I fall into the latter camp. We think metabolism of some kind probably predated RNA. Here's why.

RNA is cool because it can both catalyze reactions and act as a template for making more of itself. Each sugar has an extra hydrogen bond (compared to DNA) which makes RNA able to twist into functional shapes kind of like enzymes, but RNA can also serve as a template for a complementary strand to be made. RNA thus can do a half-assed job of both of DNA and of proteins, and RNA is an intermediate in the DNA -> RNA -> protein synthesis that happens in today's cells, so it's very likely to have been a precursor to both DNA and proteins, and avoids a lot of chicken-and-egg problems with all three having to appear at the same time.

That said, lots of physicists today are pretty confident that the first life had to include some form of metabolism: a channel through which free (i.e. low-entropy) energy is flowing. Any chemical reaction in thermodynamic equilibrium will by definition progress as fast forwards as backwards. "Life" without free energy would statistically be exactly as likely to shrink as to grow in size. Suppose there were a soup of elements at equilibrium and you added RNA to it. It would just sit there or decompose; without a source of free energy any movie of what it does would necessarily be equally likely played forwards or backwards. That's what equilibrium means.

The first life therefore almost certainly was linked to some inorganic source of free energy; probably geochemical in origin. Molecules that shape the chemical reactions in specific, contagious ways would tend to propagate to the limits of the source of free energy. At some point, RNA probably became the dominant molecule enabling metabolism with contagious specific properties, but without the flow of free energy, you'd get no propagation.

Living is a verb. I'd even say that "metabolism with contagious specific properties" might be an interesting definition for life. (NASA's current definition is "a self-sustaining chemical system capable of Darwinian evolution" but self-sustaining discounts the necessity of the flow of free energy and you might not always have the time to see Darwinism in action. Is my spayed cat alive by NASA's definition? She needs cat food to sustain herself and cannot participate in Darwinian evolution anymore! Yet, her metabolism's specifics can be contagious and infect her cat food, causing its molecules to make more cat in a way the cat specifies.) The idea that the right brew of RNA in the absence of free energy flow and metabolism could "be" life is misguided and doesn't do justice to the centrality of metabolism in understanding what life really is.

Not to be a pedant, but "personification" and "anthropomorphism" are kind of opposites. A person can personify an abstract concept like virtue, and anthropomorphism means you are attributing a human quality to a nonhuman.

Do we want generative AI to be local? I think it would make sense for each machine to have a local AI processor only if it's used a good fraction of the time. Otherwise, the cost of the hardware and keeping the model up-to-date is going to be larger than the cost of having a centralized service generate your content for you.

Don't get me wrong. I wouldn't bet against generative AI being big. I'm just skeptical that everyone is going to want to have their own dedicated generative AI hardware sitting idle most of the time and needing to download massive and bespoke upgrades to its neural network weights. I'm also concerned enough about backdoors embedded in neural network weights that make untrustworthy behavior, even if the folks running the generative AI are major corporations with the resources to police the training process. Imagine how treacherous a set of weights you found on some dark corner of the internet could be, and how outmatched Joe Consumer would be when they are selecting trustworthy sources of trained generative network weights.

The strongest motivation for having your own hardware is so you can run generative models that can operate without the oversight of a major corporation. I kind of hope that this mode of operation won't become commonplace. It doesn't sound psychologically healthy.

Good questions. Sean Carroll has written a fantastic blog post about exactly your point: https://www.preposterousuniver....

Hope that helps!

Those of us who follow many-worlds as envisioned by Hugh Everett think that the world doesn't play dice either: the single wavefunction describing the configuration of the entire universe always evolves just according to the Schrodinger equation, which is not probabilistic. It takes some work, but from this postulate you can derive that the apparent classical universe can split in the same mathematical way that a compass needle can point northeast, and not just north or east. Through evolution under the Schrodinger equation, the wavefunction of the universe doesn't always align with just one basis vector we use to describe classical happenings.

Observers in this picture might feel like they see God playing dice, while in fact when God rolls a die there are 6 outcomes, all equally real, but with each classical observer seeing just one outcome.

That's a very good question. The answer is long and complicated though, and you might already know some of the answer, so I'm going to summarize it here rather than going into all the details. Feel free to search each term if you want more depth.

Quantum computers only work inasmuch as their Qbits follow the Schrodinger equation. The Schrodinger equation is a linear, unitary, partial differential equation. Any equation that's unitary can be run forward or backward in time without losing information - you cannot have two different input states create the same output state through the action of the Schrodinger equation. As long as the quantum computer is working as intended, logical Qbits therefore need to be reversible, in that there has to be a way for every earlier state to be recoverable given a later state.

Lots of classical computing operations are not reversible. Take the XOR operation. If the result of P XOR Q is FALSE, it's not possible to say whether both P and Q were FALSE or if they were both TRUE. You cannot make a reversible computer that has a bare XOR gate in it.

There's a clever workaround though. Suppose you have a gate that outputs both the result of P XOR Q and also tells you what P was. Now the inputs are recoverable from the outputs, and your computation can be reversible, which means there's no prohibition against implementing the XOR with a quantum computer anymore. You're stuck with the extra P output and it must be carried to the end of the calculation for the whole calculation to become reversible. The extra copy of P whose only purpose is to satisfy the reversibility of computation is called an ancilla (https://en.wikipedia.org/wiki/Ancilla_bit).

Shor's algorithm generates a bunch of ancillary bits. Offhand I don't know if 16k is precisely right in this case, but for what it's worth it seems to be in the right ballpark to me.

I would be OK with a scheme where to break it, you have to defeat both RSA and Kyber. I'm not a crypto expert, but wouldn't that be a good strategy both to appease those suspicious of rushing a post-quantum algorithm (your view) and those who think that there's at least a chance that the secrets my lawyer communicates that currently rely on RSA might still be sensitive at a time far in the future, when a reliable quantum computer might be feasible (my view)?

Dear Javaman25,

With respect, I'm with you all the way until you think that a person in their 40s is probably going to die from climate change. I suppose it's likely that climate change will reduce our life expectancy, but do you really think it will work out to years, or will it be more like weeks?

Don't get me wrong: by acting as slowly as we are on average we're doing far more harm to other humans and other species than we could be, so we should definitely address climate change. However, the attitude that even over 40s like you are about to die from climate change can make people react less, both by making the left lose hope and also by losing credibility with the right.

If you have time, please consider listening to this podcast, which is far more eloquent and informed than this Slashdot post and lays a much better case: https://www.preposterousuniver...

Warm regards,
LeDopore