Bruce Schneier over at Schneier on Security writes:

The web has become so interwoven with everyday life that it is easy to forget what an extraordinary accomplishment and treasure it is. In just a few decades, much of human knowledge has been collectively written up and made available to anyone with an internet connection.

But all of this is coming to an end. The advent of AI threatens to destroy the complex online ecosystem that allows writers, artists, and other creators to reach human audiences.

The internet let us all become creators. Writers. Artists. Publishers. You Name It. Content was no longer the province of gate-keeping institutions like book and music publishers, movie studios, art galleries, or libraries. Anyone and everyone had a voice. Naturally, it got a little loud.

So the folks at Google came up with SEO or Search Engine Optimization. These were things you could do to amplify your voice. The idea was that SEO would help the user to better find what they wanted. Of course the scammers were all over that and had counter-measures for everything else (like PageRank) that came their way. The result is that there is no way to know if the hits you get on your search are legit. I think we’ve all seen the shrinking range of search results over the years. And who hasn’t been frustrated by the long list of links that reference the same source? Cory Doctorow calls this degradation of quality enshittification and I think it’s a perfect word.

On the scene these days are ginormous computer programs called LLMs or Large Language Models. ChatGPT is an example of such a thing. All this stuff gets lumped under the umbrella term AI.

LLMs are fed all the data they can handle. ChatGPT is so voracious there’s talk it will run out of material to consume in a few years. Everything produced by all of us will some day be absorbed by these machines and reduced to impersonal bits.

I love a classic BLT. Bacon. Lettuce. Tomato. Toasted bread, some mustard and mayo. Yumm. But put it in a blender and turn it into a BLT Smoothie? Yeeucch.

That’s what our friends over at AI, Inc. want. They think chatbots will do a better job of managing all that information out there. And that is certainly possible if all we care about is INFORMATION. You add SEO, that, is “optimization” to LLMs, and you’ll get LLMOs. These unholy constructs will hoover up all the information in the world and customize it into any number of flavored potions. As Schneier says:

If you want to know about climate change, or immigration policy or any other contested issue, there are people, corporations, and lobby groups with strong vested interests in shaping what you believe. They’ll hire LLMOs to ensure that LLM outputs present their preferred slant, their handpicked facts, their favored conclusions.

The internet was like a flea market in the early days. Then it evolved to more like a suburban mall and soon it will be more like a cable-TV subscription. Lots of channels but not much choice. Search engines, in the early days, made it possible for people to connect with other people. Now we have AI-synthesized answers that cut people out of the equation.

That makes the internet a sterile place. Perhaps people will discover what they are missing and begin to demand (and create) an alternative.

The ancients were able to predict eclipses. That’s because they aren’t random. Rather, they are periodic. They occur in cycles.

The current eclipse is part of Saros Series 139. A saros is 6585.3 days long which is 18 years, 11 days, and 8 hours. Eclipses separated by one saros have a similar geometry. A series will typically last 12 to 13 centuries and contain 70 to 80 eclipses. This particular series began on May 17th in 1501 and will end on July 3rd in 2753. Saros Series 139 is 1262.11 years long.

Saros was a Greek word chosen by Edmund Halley to represent 222 lunar months (another way to express 6585.3 days).

Here at home it was just a wee bite. You had to travel to see the total eclipse. I hope the eclipse-chasers got to experience totality, wherever they were.

Otherwise known as “five hundred billion” dollars or $500,000,000,000. Does calling it “half a trillion” make it seem like more?

Regardless, it is a lot of money. For comparison, California’s annual state budget is about three hundred billion dollars ($300,000,000,000).

The folks over at Planet Tracker have this piece about the consequences of seafloor mining. I’ve written about this stuff before. I stumbled here because I read the article Deep Sea Mining could cost $500 billion in lost value study says on mining.com (a good site for those interested in natural resources).

If you can’t grow it then you have to mine it. We’ve all seen what mining does to terrestrial landscapes. Can you imagine what it would do to the seafloor?

Environmental damage and ecological disruption are costly. In actual dollars. If capitalism is to be dissuaded from its lust for short-term profits—which results in the degradation of the quality of life for all—it has to be dissuaded on its own terms. Making a mess has to be bad for business. It has to hurt the bottom line. Only then will our corporate overlords take the time to do things properly.

We don’t have to mine the seafloor. We can get what we need up here. And we can reduce, reuse, and recycle. The innovations we develop to exploit an exotic new landscape like the seafloor could just as well be used topside, too. I say we need to tackle the e-waste problem. Imagine the heaping piles of old TVs, computers, and phones we’ve thrown away in just the last few years. It’s a treasure trove. An unexploited resource. Let’s climb that mountain instead of the ones on bottom of the ocean.

Dmitri Mendeleev predicted the existence of Gallium when he created his first periodic table. He expected to find an element similar to aluminum (and in the same column, group 13). Mendeleev called the element eka-aluminum, meaning “beyond” aluminum, and his prediction was confirmed by a French scientist in 1875. Paul-Émile Lecoq de Boisbaudran named the new metal after his native country (Gallia is Latin for France).

Gallium is surprisingly abundant in the earth’s crust but it is too reactive to exist in its native state and it forms no significant minerals by itself. It is obtained as a by-product from the processing of aluminum and zinc ores. China and Russia have the largest reserves.

Only a few hundred tonnes of gallium are produced worldwide but it is nonetheless an essential material in the semiconductor industry. Gallium arsenide, gallium nitride, and indium gallium phosphide compounds are critical parts of integrated circuits, logic chips, diodes, pre-amplifiers, lasers, and solar cells.

If you had a chunk of gallium it would melt in your hand (much like chocolate!). Gallium has no known biological role but clearly we can’t live without the stuff.

When I watch a sporting event on TV I have to remind myself that I’m not watching a sporting event. I’m watching a television program about a sporting event. It’s not the same thing.

The Super Bowl is the obvious example. I was interested in yesterday’s game, and it was a compelling contest, but it was a small part of the broadcast. For some time now people have shown more interest in the commercials than the result of the game, for example. And the halftime shows often generate more excitement or controversy.

Modern TV is as immersive as cinema and sufficiently convincing that we forget it is manufactured. TV presents a particular, organized view of things, something scripted and controlled. The outcomes in the game are still subject to chance (I don’t believe NFL games are rigged!) but the TV show is prepared for that.

There has always been this layer between the event (the game on the field) and the TV show (the experience of the viewer). Now it is more sophisticated. The technology is better, certainly. And the show-makers now have decades of practice (and audience feedback) to more expertly craft their products. The modern filter provided by our media makes the event seem more real and gives us the illusion of greater involvement. But it requires a more elaborate system to come between us and the thing and thus makes it even more remote.

Watching sports on TV always makes me think of a phrase I learned in school (attributed to Alfred Korzybski): “the map is not the territory.” And that remark always makes me think of René Magritte’s The Treachery of Images:

The translation is “this is not a pipe.”

Now let’s get on with some TV shows about baseball games!

Magnesium is an essential nutrient for plants and animals. Plants need chlorophyll to photosynthesize, and chlorophyll molecules contain magnesium ions. ATP (adenosine triphosphate), the energy molecule, binds to magnesium in order to get its work done. Hundreds of enzymes that are necessary to metabolism need magnesium to function. In the human body most of the magnesium is stored in the bones. Magnesium deficiency shows itself in a myriad of ways, reflecting its crucial role in so many processes.

For that alone we should know something of magnesium. The highest dietary sources are leafy greens, beans, nuts, seeds, and whole grains. Magnesium is in fact found in almost all the food we eat and so deficiencies are rare in well-fed places.

And then there’s mag wheels:

That was a big thing in the 70s—outfitting your car with “mag wheels.” They were originally made from magnesium but they are mostly aluminum now. Magnesium is lighter than aluminum but it corrodes more readily and is more flammable. The image above is from an F1 site. True “mag” wheels are expensive.

Magnesium alloys are ubiquitous. There is hardly a place where they can’t be found. The world produced one million tonnes of magnesium metal in 2022, mostly from magnesite and dolomite ores. Magnesium compounds are also extracted from seawater. One million tonnes does not seem like much when compared to the annual production of copper (~20 million tonnes), chromium (~40 million), and aluminum (~70 million) but it’s still a heaping pile of stuff.

In the lab you could get strips of magnesium to ignite and they produced an intensely bright white flame. The old flash bulbs of yesteryear had magnesium in them.

A guy named Scott Dewing writes a technology column (Inside the Box) for the Jefferson Journal. In the latest issue he talked about Artificial Intelligence* (“AI Gets Personal”). He’s not optimistic, or rather, he’s not optimistic about current versions of the technology and sees this for our future:

. . . the universe will descend into a morass of machine-generated content that will self-optimize until genuine human creativity is obliterated and all art becomes a bland panoply of monotonous mediocrity.

It’s hard to state it more plainly and it’s hard to argue with! A capitalist system will always seek to optimize profit, and any new technology will ultimately survive only if it makes money. AI is very expensive. It’s going to have to pay for itself at some point. Absorbing all of human knowledge into a gigantic silicon Cuisinart and then spitting it out as personal “knowledge smoothies” for everyone on the planet ain’t gonna be cheap. And it may not be what is best for us. See above quote.

In the article Mr. Dewing expounds a notion of a more “personal” AI. That is, the current “generative” AI should exist to complement us, to enhance our work, and not to replace us. In fact, it ought to help us be better people. As Dewing says “generative AI learns about the world while personal AI learns about you.”

Indeed. All our inventions ought to help us. And if our gadgets don’t help us be better people, maybe we ought to stop making them.

*Artificial Intelligence is neither artificial nor intelligent. But the term “AI” is what we’ve got so we are stuck with it.

I’m starting to feel like Noah. The National Weather Service tells me this:

A half-inch of rain is a lot! The NWS has a collection of videos that give you a feeling for different rates of rainfall. Take a look.

Our little burg has its own funny little climate. But if I had to say what the climate is closest to I would say “high desert.” We get rain and snow, but only intermittently. The rain lets up in May and doesn’t return until November. Except for summer thundershowers, of course. The snow comes in December and quits in April. We average about foot of snow and about two feet of rain per year. It’s hot and very dry in the summer and it’s cold and dry in the winter.

But at this moment I’m in the midst of a deluge. It’s a biblical amount of rain, man. The snow is being washed away here in town but it is still clinging stubbornly to the hillsides.

We need the water, naturally. Well, we need the snowpack, to be precise. So let’s hope it is cold enough in the high places for this stuff to pile up. All I see out my window is runoff. And puddles. And my rain gutters clogged with melting snow!

Stay dry, my friends.

The lanthanoids, or rare-earth elements, are essential to the modern world. We interact with these substances but we don’t know them. We know about copper pipes and iron railings and aluminum cans and stuff like that, but the little bits and pieces of our high-tech world are mostly invisible to us. Nonetheless there is a growing global demand for elements 57 through 71.

Neodymium (Nd, #60) is a little different as most of us have heard of neodymium magnets. These were discovered in the 1980s at a couple of corporate labs and are currently the strongest commercially-available magnets. The alloy is a combination of neodymium, iron, and boron and is called NIB or Neo or NdFeB (the formula is Nd₂Fe₁₄B).

They are used in computer hard drives and electric motors and countless other places. You’ve seen them around because they can be purchased for home use:

You have to be careful with those things. Two NIB magnets will snap together suddenly and with a lot of force and good luck getting them apart! If you stick something to the fridge you can expect it to stay in place.

Next up: Magnesium (Mg), #12

So I was checking out the Northern California and Southern Oregon Offshore Wind Transmission Study (vol. 1) and I found some interesting bits. Here’s one:

The northern Coast of California and the southern coast of Oregon have some of the best wind resources in the United States, and development of these resources can contribute to meeting the clean energy goals of California, Oregon, and other western states.

Yay for us! But there’s more:

Because the existing transmission grid infrastructure that serves these coastal regions is limited in capacity, major investments in new transmission grid infrastructure will be needed.

There’s always a catch. You have to get the electricity that the turbines generate off-shore to facilities on-shore. And once there the power has to be distributed to the grid. That’s going to take new transmission lines of the 500-kv size. They look like this:

We all know what it is like to travel to and from the Southern Oregon or Northern California coasts—mountains, rivers, canyons, and winding roads. New transmission lines there won’t have the wide-open spaces of the Western deserts like the photo above. It will be a more challenging undertaking. Speaking of that, here’s another bit:

We note that some of the necessary technologies for large-scale development of floating OSW power are neither fully developed nor commercially available at this time.

Well, then. Infrastructure is do-able. It will take a while (more on that later) but the primary technology is not ready. Yet. We all know that wind turbines work. And they work on land and on sea. Offshore wind power is big stuff in the UK, for example. They have developments in the North Sea that can deliver at a gigawatt scale. Unfortunately in the Northern California and Southern Oregon offshore regions slated for wind development the continental shelf plunges rapidly away from the shore. The water is too deep and thus you need floating platforms. That tech is still experimental.

To sum up:

The development of tens of gigawatts of floating OSW power generation on the West Coast will not occur quickly; a successful effort would take decades, and the associated transmission upgrades would also take place over decades.

All complex problems will require complex solutions. I found this material and related stuff at the Schatz Energy Research Center at Cal Poly Humboldt.