I'm happy to see that you don't hate systemd. That was the last shoe to drop. I'll complete the switch to BSD now!

Dragonfly BSD works quite well on the desktop, as does Funtoo Linux, which is systemd-free. Gentoo also works and still uses OpenRC by default, although there is growing concern some of the devs are quietly preparing to push a systemd agenda (kdbus patches in the kernel, one of the devs commenting he hopes systemd would become the Gentoo default, and a habit of the moderators in the Gentoo forums to shut down any discussions critical of systemd).

Linus may not be showing good leadership in this instance, but not everyone has drunk the urine just yet, and there are others stepping up to the plate to maintain or create alternatives.

I currently have a web radio transceiver front panel application that works on Linux, Windows, MacOS, Android, Amazon Kindle Fire, under Chrome, Firefox, or Opera. No porting, no software installation. See blog.algoram.com for details of what I'm writing.

The one unsupported popular platform? iOS, because Safari doesn't have the function used to acquire the microphone in the web audio API (and perhaps doesn't have other parts of that API), and Apple insists on handicapping other browsers by forcing them to use Apple's rendering engine.

I don't have any answer other than "don't buy iOS until they fix it".

Or we build storage.

You don't HAVE to generate it within a fraction of a second of when it's used - that's just been convenient so far. So it's not with renewables? Fine! Store that juice!

When Bill Gates says:

"There's no battery technology that's even close to allowing us to take all of our energy from renewables and be able to use battery storage in order to deal not only with the 24-hour cycle but also with long periods of time where it's cloudy and you don't have sun or you don't have wind."

he's totally wrong.

For starters, there's Vanadium Redox. A flow battery (pumped electrolyte): Power limited by the size of the reaction device's electrode and membrane assembly. Energy storage limited by the size of the tanks. It's mainly used for utility-level energy storage down under (Oz or Nz, I think), because the patents are still fresh and the little startup doesn't want to license it to others. Vanadium is some substantial percentage of the Earth's crust so there's no shortage. Using the same element (in different sets of oxidation states - vanadium has (at least) 6 of 'em) for BOTH electrodes means leakage of small amounts of the element through the dilectric membrane doesn't poison the battery.

Lithum cells are already good enough to run laptops, cars, and houses, and are improving at a Moore's Law like rate. The elements are also not rare and the use of several nanotech techniques on the electrodes have drastically increased the lifetime and other useful properties. (We just had reports of yet another breakthrough within the last day or so, doubling the capacity and extending the life.) The fast-charge/discharge cells are also extremely efficient. (They have to be, because every horsepower is 3/4 kW, so even a few percent of loss would translate to enormous heat in an automotive application.) The main problem is to get companies to "pull the trigger" on deploying them - and risk their new production line being rendered obsolete before the product hits the market by NEXT month's breakthroughs.

Lead-acids need to be replaced once or twice per decade. But they have been the workhorses for off-grid since Edison's and Nikola Tesla's days, and still are today (though not for long, if Elon Musk and the five billion dollars of investments in his lithium battery plant have anything to say about it).

Nickel-Iron wet cells are a technology developed by Edison. They have more loss than lead-acids. But they literally last for centuries. If you have a moderately steady renewable source (like some combination of enough wind and a big enough windmill, enough sun and a big enough solar array, or a stream and a big enough hydro system) you'll have enough more power than you need to keep them topped off. They're just fine for covering days, or even a couple weeks, of bad generation weather, or down-for-maintenance situations. That IS what they were in at least one hydro plant I know of. (The problem is finding them: They last so long you only need to buy them ONCE, so there aren't many plants.)

That's just four FAMILIES of entirely adequate solutions. There ARE more.

So Bill is either uninformed, talking through his hat, or starting on the "embrace" stage of yet another:
  - Embrace
  - Extend
  - Extinguish
  - PROFIT!

we have a way to turn electricity directly into heat. But there is no direct way to turn heat into electricity. It has to go thru a second step of mechanical energy to spin a magnet to create electricity.

You can go from electricity directly to heat because that increases entropy. You can't go from heat to anything useful because that decreases entropy, and entropy of a closed system only increases. The best you can do is a heat engine, working off a temperature DIFFERENCE. (Some of them also work backward as heat pumps, to go from electricity to heat more effectively, by also grabbing some heat from elsewhere to include in the hot end output.)

There ARE at least two major forms of electronic heat engines - direct from temperature differences to electricity, with only charge carriers as the moving parts: Thermoelectrics (thermocouples, peltier junctions, and thermopiles of them) and thermionics (both heat-driven vacuum diode generators and a FET-like semiconductor analog of them). Both are discussed in other responses to the parent post.

TEs are ridiculously inefficient and aren't looking to be much better anytime soon

Because thermoelectric effect devices leak heat big time.

However there's also thermionics. The vacuum-tube version is currently inefficient - about as inefficient as slightly behind-the-curve solar cells - due to space charge accumulation discouraging current, but I've seen reports of a semiconductor close analog of it (as an FET is a semiconductor close analog of a vacuum triode) that IS efficient, encouraging the space charge to propagate through the drift region by doping tricks (that I don't recall offhand). The semiconductor version beats the problems that plague thermoelectrics because the only charge carriers crossing the temperature gradient are the ones doing so in an efficient manner, so the bulk of the thermal leakage is mechanical rather than electrical, and the drift region can be long enough to keep that fraction down.

Super Dracos are for escape in flight too, including in and past MaxQ. But they are on Crew Dragon, not Cargo Dragon. Cargo Dragon did not carry a crew and wasn't programmed to save itself.

I got the impression from the (sketchy) article that repeater AMPLIFIERS were still needed but repeater REGENERATORS were not.

Then again - another part of the article makes it look like an additional result was that they could boost this less-subject-to-degradation-by-nonlinear-distortions signal at the start until the fibre itself was acting non-linearly, in order to get a signal strong enough to survive a much longer hop.

So it's not clear to me whether the distance was achieved by:
  - long hops enabled by strong signals, and NO amplifiers
  - longer propagation without regenaration using JUST amplifers
  - a combination of the two: Both getting long total length without regeneration AND being able to use stronger signals and thus use larger space between the amplifier-type repeaters.

Since the diameter of the earth is 7 926.3352 miles, this could conceivably remove any need for repeaters.

I got the impression from the (sketchy) article that repeater AMPLIFIERS were still needed but repeater REGENERATORS were not.

I.e. you still needed to boost the strength of the signal to make up for the losses. But the progressive degradation of the quality of the signal - with data from different frequency bands bleeding into other bands (especially in the amplifiers themselves) due to nonlinear "mixing" processes - had been headed off, by synchronizing the frequencies of all the carriers to exact multiples of a common basic difference-between-the-carriers frequency.

This apparently sets up a situation where the distortion products of each carrier's interaction with nonlinear processes cancel out with respect to trying to recover the signals on another carrier - much the way the modulation products do in OFDM modulation schemes. In OFDM it allows you to make essentially total use of the bandwidth. In this system it lets you use simple, cheap, amplifiers to get your signal boost, rather than ending the fibre before things get too intertwingled, demodulating all the signals back to data streams and recovered clocking, then generating a fresh set of modulated light streams for the next hop - MUCH more expensive and power hungry.

Most of us do have a need to transmit messages privately. Do you not make any online purchases?

Yes, but those have to use public-key encryption. I am sure of my one-time-pad encryption because it's just exclusive-OR with the data, and I am sure that my diode noise is really random and there is no way for anyone else to predict or duplicate it. I can not extend the same degree of surety to public-key encryption. The software is complex, the math is hard to understand, and it all depends on the assumption that some algorithms are difficult to reverse - which might not be true.

Most algorithms to do this use the time between keypresses, measured to very high precision so that the lower bits are chaotic. So it doesn't really matter what keys you hit, and it doesn't matter how rythmic your typing is.

The problem with FM static is that you could start receiving a station, and if you don't happen to realize you are now getting low-entropy data, that's a problem.

There are many well-characterized forms of electronic noise: thermal noise, shot noise, avalanche noise, flicker noise, all of these are easy to produce with parts that cost a few dollars.

True randomness comes from quantum mechanical phenomena. Linux /dev/random is chaotic, yes, enough to seed a software "R"NG. But we can do better and devices to do so are cheap these days.

I wouldn't trust anything but diode noise for randomness. If I had a need to transmit messages privately, I'd only trust a one-time pad.

... government regulators couldn't possibly find financial irregularities by grabbing you documents from the cloud service provider, ...

The courts said you have no expectation of privacy one you put your data in the hands of a third party. Great! Let's convince all those "evil corporations" to store all their data in the cloud. Then the government can go after them any time they want. B-b