Month: December 2022

You've probably heard by now that the Department of Energy will announce tomorrow that one of their fusion machines has achieved the holy grail: net energy production. That is, the energy coming out will be greater than the energy going in.

Not to put a damper on the whole thing, but there's a little less here than meets the eye. Here's a very rough block diagram of the energy flow in the National Ignition Facility:

The very final stage, where lasers heat up the fusion fuel, does indeed produce a net energy gain. This is known as Q, and in this example equals 1.25, which is about what we're expecting DOE to announce tomorrow.

But the wall-to-wall energy consumption is quite different: Taken as a whole, the system is likely to produce about 1-2% of the energy initially used to charge the capacitors that drive the lasers. This is nowhere close to net energy gain.

There's no deception here. At least, I assume there won't be when DOE makes the official announcement. The name of the lab, after all, is the National Ignition Facility, and ignition refers specifically to creating a self-sustaining fusion reaction, not to producing a net energy gain. This was the original goal of the NIF, which is used for nuclear bomb research and is by no means state of the art in fusion energy production.

Achieving ignition is a landmark, and it certainly deserves a bit of hoopla. But net energy gain in a commercial reactor is still a very long way away.

Every once in a while I find myself browsing through old photos for one reason or another, and occasionally I find a good one that never got its day in the sun. Today's photo is the original of a Milky Way panorama that didn't get used because it was too distorted to correct properly (though you can see my best effort here). I happened to run across it while doing something else, and it immediately struck me that the distorted version was actually kind of nice. So here it is. I saturated the image more than usual to give you a good idea of all the colors that are there in the dust clouds surrounding the Milky Way core.

For what it's worth, I would like to remind everyone about the true nature of Twitter:

Of the most followed accounts, two out of 50 belong to political actors of any kind. That's about 7% of all followers for these accounts. Other ways of looking at Twitter readership put politics at around 10% or so. The other 90% is music, news, sports, and a smattering of other topics.¹

This is not to say that political Twitter is of no consequence. It's only to say that everyone has their own Twitter, which they think of as "Twitter" even though it's unique to them. If you follow extreme political accounts, Twitter seems like the Wild West. If you follow normal people who are interested in politics, it seems like an OK platform for keeping up with things. And if you're part of the 90% that follows the Steelers or Rihanna or NASA, it's a platform that has nothing to do with politics at all.

Keep this in mind. Elon Musk is obviously obsessed with politics but his customers aren't. The future of Twitter depends on celebrities and musicians, not the left or the right.

¹Obviously this can change from week to week and also differs depending on who's doing the counting. Don't take the precise numbers too seriously.

NOTE: Each of these charts actually has only 24 Twitter accounts. That's because I was interested only in American political players, so I deleted two accounts belonging to the prime minister of India.

Twitter sure is annoying these days. I mean, it's always been annoying thanks to trolls and character limits and mob pile-ons and so forth. But now Elon Musk is determined to make it more annoying still by hiring a few writers to trawl through internal documents in order to show that before he took over it was was a cesspool of woke liberalism and anti-conservative bias.

Fine. It's his company, after all. The problem—one of them, anyway—is that he insists on the results of these investigations being released on Twitter. This may seem natural for the owner of Twitter to do, but it's really not at all a natural environment for this kind of thing. I just finished reading Part 2 of the portentously named "Twitter Files" project, and when I was done I felt like there really wasn't much there. And there wasn't. It was 30 tweets long but that amounts to only about 1,000 words. In most places that's the length of a fluffy feature about a library cat.

To condense those thousand words even more, it's about shadow banning, which is a way of preventing tweets from being seen without the author knowing that anything is going on. For years Republicans have been complaining about this any time a tweet gets less attention than they think it should, and for years Twitter has been denying it. But if you read the fine print, what Twitter really said was that although they didn't shadow ban, they did "de-amplify."

That's a mighty thin distinction, so score one for the Twitter critics. Still, that's not very much all by itself: it's one thing for Twitter to de-amplify folks who break their rules, but what we really want to know is whether they routinely de-amplified conservative voices more than liberal ones. Bari Weiss, a writer whose primary motivation in life seems to be her deep resentment of liberals, is the author of Part 2, and she provides the names of three political actors who were de-amplified:¹

• Charlie Kirk
• Dan Bongino
• Chaya Raichik (Libs of TikTok)

And that's it. Weiss doesn't explain why she picked these three to highlight. She doesn't say why these three were de-amplified. Nor does she say if she also knows of any liberals who were de-amplified. And even though Musk has given her access to anything she wants to see, she provides no statistics about how many people were de-amplified; the reasons they were de-amplified; how long they were de-amplified; or anything else. She just provides three hand-picked names and that's the end of it.

So once again Elon Musk's peculiar act of revenge has spat forth a mouse. Sure, it's now more widely known that Twitter de-amplifies folks for breaking the rules, but since Musk himself has said he plans to do the same thing this isn't really much of a revelation.

So let's move on to Part 3. You can read it here, but honestly I wasn't able to do much more than skim it. It basically shows Twitter execs talking internally about stuff that seems mostly routine. And then there's Part 4, which is yet more internal discussion about how to handle assholes.

I confess to some ongoing confusion about this whole project. Why is Musk doing it? It's publicity, sure, but not the kind that will do him any good. Nor will it bolster his reputation as a free speech defender unless he makes everything public, not just a few items cherry picked by a trio of lib-hating journalists. Basically, all he's doing is pissing off liberals and getting little in return since conservatives were expecting way bigger smoking guns than this. It is a mystery.

¹There's a fourth, Dr. Jay Bhattacharya, but he was de-amplified over some kind of COVID beef.

A new study has concluded that "minorities are 24 to 33 percent more likely to be stopped for speeding" than white people who are driving the exact same speed. This is based on research among Lyft drivers in Florida

That sounded striking, but when I was reading through the study I noticed a brief mention of a statistical procedure the authors were unable to perform because "speeding citations are rare." Hmmm. Then I reached the main result:

Note the units. The y-axis represents citations per 10,000 hours of driving. A normal working year is about 2,000 hours, which means that white drivers got 0.022 citations per year and minority drivers got 0.029 citations per year. (This is derived from the fixed effects (FE) model on the left.)

Put another way, white drivers got 2.2 citations per century and minority drivers got 2.9 citations per century. That's a difference of 0.7 citations per century.

That's using the fixed effects model. The authors also used a machine learning model, and it concluded that the difference was 0.5 citations per century. In Florida. Among Lyft drivers.

The authors insist that their results are statistically significant. Maybe so. But who cares? When the absolute numbers are so minuscule it's hardly possible to draw any meaningful conclusions.

A few Bitcoin fans have dredged up a post I wrote four years ago today that celebrated its plummet over the preceding year. Sure enough, it turned out to be at its lowest point in a while and promptly took an upward turn. Notch a victory for the crypto groupies.

Since then I've written multiple posts suggesting that Bitcoin is basically a collectible, not a substitute for money, and we should all expect it to be as volatile as any other collectible that randomly finds itself in and out of favor. For example, since we're celebrating anniversaries, here is Bitcoin over the past two years:

If you were investing your money on December 9, 2020, you'd have been better off putting it in an S&P 500 index fund than putting it into Bitcoin.

Of course, this is all a game of cherry picking dates. Choose a different date and you can get a very different result. Nevertheless, if I were investing for the next ten or twenty years, I'd feel a whole lot more comfortable about the S&P keeping its value than I would about Bitcoin.

This is Charlie hiding behind the drapes in my study. He doesn't do this a lot anymore, but that's mainly because I don't close the drapes very much this time of year. As always, Charlie has to adapt to the weird habits of his humans.

Near the beginning of the movie Arthur, Arthur (played by Dudley Moore) is asked what he does for a living. He answers, "I race cars, I play tennis, I fondle women, but I have weekends off and I am my own boss."

Forty years later Apple executive Tony Blevins was asked the same question as he exited his Mercedes-Benz SLR McLaren at a car show. He answered, “I race cars, play golf and fondle big-breasted women. But I take weekends and major holidays off.”

Naturally the Blevins version showed up on TikTok:

@itsdanielmac Quite the occupation this man has ✍️ #mercedesbenz #supercarstiktok #slr #car ♬ original sound - DANIEL MAC

A few days later Apple fired Blevins. The Wall Street Journal tells the story:

Mr. Blevins said he was asked to resign and he declined. Days after the video surfaced Sept. 5, he was fired....The firing has left him in disbelief that something he intended as a joke, uttered to a non-Apple employee while he wasn’t working, could erase a lifetime of dedication for a company he loved and believed in, Mr. Blevins said. He said he is sorry to those he offended with the comment, but he sees his firing as a blunder by Apple and a capitulation to broader cultural pressure.

Some of his former employees support him and say that while he sometimes would use humor to deflate tense situations, they hadn’t witnessed any other sexist comments or unprofessional behavior.

....In his two decades at the company, Mr. Blevins had emerged as one of an essential cadre of executives who helped build Apple into a colossus through deals with suppliers of chips, displays and other parts that were mostly assembled in Asia. With brazen tactics—such as once putting two companies competing for the same deal in neighboring rooms in a hotel while he went back and forth between them—Mr. Blevins helped Apple keep costs down and the supply of its iPhones and other gadgets secure.

....After his firing, company executives were still planning a going-away party for him with invitations extended to more than 100 people, but the party was canceled when communication began breaking down between Mr. Blevins and Apple, said Mr. Blevins. He also said he didn’t like the idea of Apple throwing him a party after having just fired him, calling it a “hypocritical” action. He said he didn’t receive severance upon his departure.

I've never heard of firing someone and then throwing a going-away party. That's very peculiar behavior.

But the big question, of course, is whether Apple should have fired him in the first place. His joke was only risque in a 1950s Borscht Belt sense, it was off campus, and obviously part of a TikTok routine.

But then again, stuff that flew in the 1950s (or in 1981) doesn't fly today. Maybe Blevins should have known better no matter where he happened to be.

For myself, I'd say it was worth a slap on the wrist and an apology, not a dismissal after 22 years with the company. Opinions?

In the current Senate, with the parties tied at 50-50 and Kamala Harris providing the tiebreaking vote, all committees are split evenly. Since Harris is not available to break ties on committee votes, this slows things down a lot for Democrats. The upcoming 51-49 Senate, where Ds will have majorities on all committees, should smooth things considerably.

And there's something else this will do: it allows Democrats to issue subpoenas and conduct depositions all by themselves. They no longer need any Republican votes to exercise oversight.

Over in the House, Kevin McCarthy has already made it clear what they're going to do with their majority: investigate Hunter Biden; consider impeachment charges against Homeland Security Secretary Alejandro Mayorkas; figure out once and for all where COVID came from; and so much more. I could provide a whole list, but why bother? McCarthy has already done it.

In the LA Times today, Kurt Bardella says Democrats should retaliate in force. And he should know. Bardella is a Democrat these days, but he was a spokesman for the House Oversight Committee back when Republicans were using it to investigate Benghazi. He knows how the game is played:

There is no reason Senate Democrats should not turn the tables on Republicans by finally investigating the conflicts of interest created by Donald Trump, Ivanka Trump and Jared Kushner during their time in the White House.

Documents recently released by Congress reveal that foreign nations were spending hundreds of thousands of dollars at President Trump’s hotel at the same time they were trying to influence our foreign policy.

Records obtained by Congress exposed that when agents had to stay in Trump hotels, Trump’s companies charged the Secret Service as much as five times more than the government rate, costing taxpayers more than $1.4 million.

Six months after leaving the White House, Jared Kushner received a $2 billion “investment” from a fund controlled by the Saudi crown prince.

This sounds like a bit of weak tea to me. But maybe that doesn't matter. Maybe the point is just to annoy the other side; provide fodder for news and pundits; and hope that if you keep at it something unexpected will pop up. After all, it was by pulling a long string on Benghazi that Republicans got lucky and discovered that Hillary Clinton had some issues with her email.