Author: Kevin Drum

Let's do another triptych today. This one is of the San Onofre nuclear power plant, which has been closed for the past decade and is now undergoing decommissioning.

And just how long will this take? According to the ultimate authority, "Decommissioning San Onofre will take numerous years until the process is complete." Okey doke.

Also: "In February 1996, 4 kittens were found at the plant. The kittens were adopted by plant workers."

And now for the pictures. The top photo shows San Onofre from Interstate 5 in late afternoon. The middle photo shows it at dusk from the side of the road. The third one shows it at night.

Copyright news for 2023:

On January 1, 2023, copyrighted works from 1927 will enter the US public domain, including books from Hemingway, Woolf, and Agatha Christie; movies like Metropolis, Wings, and The Jazz Singer; and music by Irving Berlin & Louis Armstrong. https://t.co/3BHllesUlL

— kottke.org (@kottke) December 21, 2022

Only one year until Mickey Mouse is in the public domain.

Consumer confidence in December is up!

The increase in the Present Situation line looks mostly like noise to me, part of a downward trend that started more than a year ago. Conversely, the Expectations line looks like it really is rebounding a bit. However, as the Conference Board notes, "Expectations are still lingering around 80—a level associated with recession."

Better up than down, but I wouldn't make too much of this. There's a recession coming next year, and I'm not sure I buy the near-universal consensus that it will be "mild." When the Wall Street pack starts to unanimously agree on something is when I get nervous.

Conservatives are out in full force this morning to mock an attempt by the Stanford University IT Department to use less offensive language:

The Elimination of Harmful Language Initiative (EHLI) is a multi-phase, multi-year project to address harmful language in IT at Stanford. EHLI is one of the actions prioritized in the Statement of Solidarity and Commitment to Action, which was published by the Stanford CIO Council (CIOC) and People of Color in Technology (POC-IT) affinity group in December 2020.

Look, I'll admit that this already sounds pretty iffy. And the list of bad words that follows is unquestionably debatable. But it might help to know some of the principles that drive language recommendations like these. Here are a few:

• Don't reduce people to a single attribute, especially if that attribute has negative connotations. The Associated Press tries to follow this rule, which means, for example, that you should refer to "a person who immigrated" rather than "an immigrant." I find this rule awkward and don't always follow it, but it's hardly a massively woke imposition.
• Call people what they want to be called. I consider this nothing more than common decency. Don't get upset just because the new word is unfamiliar to you.
• Don't borrow phrases from other cultures, such as "bury the hatchet." This is my bête noir. Aside from avoiding obviously offensive uses of loan phrases, I think this principle is misguided and I don't follow it. Mixing and matching phrases from around the world is one of the glories of language and has been for millennia. (More generally, I think charges of cultural appropriation are usually ridiculous for a bunch of different reasons. Cultural mockery should be the standard for offensiveness.)
• Don't use words that are also used to describe disabilities. I'm on the fence with this one. On the one hand, some words simply become offensive over time and ought to be dropped: spaz, retard, cripple, and so forth. Others, like the perennial favorite crazy, I find ridiculous. Crazy has a perfectly ordinary colloquial meaning that's plainly not a slur on mentally ill people. (As an aside, I note that loco isn't on the Stanford list. It's tough deciding which oppressed group to favor, isn't it?)
• Don't use phrases like hit the bullseye, which have violent roots. I appreciate the sentiment here, but only if it's tightly restricted. It's really easy to go overboard on this and invite nothing but derision. It's also worth noting that violent imagery falls into the category of de gustibus, not cultural offensiveness. It should probably be treated separately.
• All words that might be redolent of slavery. There's some common sense to this one, but it can also go too far. The Stanford list, for example, includes white paper, white space, and yellow team (as one of several teams identified by colors). It's one thing to eliminate words where black and white have obvious derivations from black = bad and white = good, but sometimes a color is just a color.
• American. This is a special case because everyone is up in arms about it today. The argument here is that America represents all of the Americas, not just the United States, so we should use US when referring to our specific country. I'm not too keen on this stricture since, as far as I know, not a single other country in Central or South America uses America in its name and no other country's residents call themselves Americans. Conversely, pretty much everyone in the world uses American as shorthand for residents of the USA. I can't think of any reason to be offended by US citizen, but then, I'm not sure there's a legitimate reason to be offended by American either.

The key to all this is good judgment. It's way too easy to go overboard and create endless lists of allegedly offensive words that, in real life, offend almost no one. The best way to avoid this is to follow two rules. The first is to include multiple people of color on your language committee so you don't force a single person into the role of representing their entire minority group. The second is to make sure that lots of suggestions are rejected. Keep only the seriously offensive stuff instead of pretending you can teach hundreds of new usages to people. This is especially true of awkward replacements for common phrases that have been rejected solely due to obscure origins no one even remembers anymore. Perhaps secret balloting to keep/reject words is a good idea?

I may have opposed the release of Donald Trump's tax returns, but that doesn't mean I'm so prissy that I won't write about it now that the deed has been done. Here is tonight's New York Times piece about Trump's taxes:

The data, which includes details of Mr. Trump’s federal tax returns from 2015 through his full term in the White House, shows that he began his presidency suffering the sort of large business losses that had defined much of his career and paid almost nothing in income tax.

But his fortunes changed in 2018, as he reported $24.3 million in adjusted gross income....His sudden burst of income occurred largely because he had sold properties or investments at a gain of $22 million.

....By 2020, however, Mr. Trump had returned to reporting losses. In fact, despite the capital gains that boosted his bottom line in 2018, the entirety of his core businesses — mostly real estate, golf courses and hotels — continued to report losses every year, totaling $60 million during his presidency.

There's not really anything here that we didn't already know, but I appreciate the bluntness of the Times report. The "entirety of his core business" reported losses every year, just as it always has. The truth is that Trump has never really made money from anything except The Apprentice and his licensing business.

Then, at the very end, the Times reporters throw in a hilarious little FU about the interest income Trump receives:

Nearly all of his interest income came from his share of profits earned by a partnership that is controlled by Vornado Realty Trust.

The partnership owns two valuable office towers: 1290 Sixth Avenue in Manhattan; and 555 California Street in San Francisco. Mr. Trump, who has a 30 percent share in the partnership, has no authority over its management, and it has consistently been his strongest-performing asset.

Twist that knife.

This month my M-protein level is up a bit:

These small ups and downs are pretty normal, but naturally I'd just as soon see more downs than ups. Still, as long as I'm below 1.0 things are OK.

That said, I'd sure like the Carvykti people to find a slot for me. There are not a lot of pleasant options left once we decide that the Empliciti is no longer working and needs to be replaced with something else.

Selfie, Times Square.

Construction starts for single-family homes were down 4.1% in November. That's hardly a surprise. But starts for new apartment units were up 4.8%:

In case you think this is unusual, it's actually part of a very long-term trend:

The single-family housing market is a boom-and-bust affair, but with a marked secular trend downward over the past three decades. During the same period, however, apartment unit starts have grown about 5x. Make of this what you will.

Yesterday I pondered a question about fentanyl: it first became available on the street in the late 1970s, but it wasn't until 2013 that its use suddenly skyrocketed in the US. Why did it take so long for that to happen, given that it's practically a perfect drug for narcosyndicates?

Via Twitter, Beau Kilmer at RAND referred me to a report he co-authored a few years ago. The single biggest factor, he thinks, was the growing number of easy ways to synthesize fentanyl outside a commercial lab:

The Siegfried Method came out of Russia and was specifically designed for illicit manufacture of fentanyl. In the aughts, several more simplified methods were discovered that could be made by anyone with even modest training:

The invention and dissemination of easier synthesis methods can render a much broader set of individuals qualified and capable of producing fentanyl....Largely gone are the days of a single highly-trained chemist synthesizing fentanyl in a professional laboratory. The potential scale of chemical and pharmaceutical manufacturing, especially in China, combined with more-accessible synthesis techniques allows for the untrained to manufacture fentanyl virtually anywhere, making supply disruption more challenging.

A second possible explanation lies in the postal service:

Rising e-commerce and the growth of inbound packages from China overlap with fentanyl’s arrival. In 2011, postal services of the United States and China entered into an agreement to streamline mail delivery and reduce shipping costs for merchandise originating in China. This “ePacket” service is designed for shipping consumer goods (under 2 kg) from China directly and rapidly to customers ordering items online.

....Today, shipping costs from China are negligible. A 1-kg parcel can be shipped from China to the United States for as little as $10 through the international postal system or for $100 by private consignment operator. The volume of mail and cargo from China gives adequate cover for smuggling minute quantities of fentanyl or other novel synthetic opioids. Online vendors realize this and often prefer sending packages through USPS, sometimes targeting processing centers that handle large volumes of mail.

Easier methods of synthesis broadened the base of manufacture while fentanyl's concentrated nature made it easy to ship via postal service—which is considerably safer than smuggling hundreds of bricks of heroin or cocaine in boats and planes.

Put together, this offers a plausible explanation for ~2013 as the year that fentanyl finally took off. Before then it was difficult to synthesize; required trained chemists and commercial-quality labs; was tricky to smuggle; and because of all this was probably about as profitable as more conventional drugs. After 2013 it was easy to make; easy to smuggle; required only cooks and makeshift labs; and was highly profitable. What more could an aspiring drug lord ask for?

Will House Democrats release Donald Trump's tax returns to the public?

A House committee is expected to vote Tuesday afternoon on whether to make public tax returns belonging to former President Donald J. Trump....A rarely used federal law allows the committee to obtain any U.S. taxpayer’s returns. While the statute generally requires lawmakers to keep such information confidential, it also empowers them to make it public by voting to report the material to the full House.

I hope they don't do this. The law allows Congress to demand tax returns for legislative purposes, so that's what they did (citing oversight of the presidential audit program). I think we all understand this was a thin pretext for getting Trump's returns and examining them for fraud, but nonetheless that's the pretext in play. Releasing everything to the public isn't, and there's no legitimate reason to do it.

I'm not naive. If Jim Jordan got hold of something similar he'd leak it in a minute. Hell, he'd make sure to leak just a few out-of-context excerpts that would make his victim look even worse than they are.

But Jim Jordan is an asshole. We don't have to be. Trump has the legal right to keep his tax returns private if he wants to, and we should make it clear that we believe in upholding legitimate legal rights. Vote no, Democrats.