Kevin Drum

Here is Elon Musk's version of the story about how and why Starlink satellite coverage around Crimea was inactive when the Ukrainian government wanted to use it to guide a drone attack on the Russian navy:

There was an emergency request from government authorities to activate Starlink all the way to Sevastopol.

The obvious intent being to sink most of the Russian fleet at anchor.

If I had agreed to their request, then SpaceX would be explicitly complicit in a major act of war and…

— Elon Musk (@elonmusk) September 7, 2023

And here is Walter Isaacson's version:

Although he had readily supported Ukraine, he believed it was reckless for Ukraine to launch an attack on Crimea, which Russia had annexed in 2014....The [Russian] ambassador had explicitly told him that a Ukrainian attack on Crimea would lead to a nuclear response. Musk explained to me in great detail, as I stood behind the bleachers, the Russian laws and doctrines that decreed such a response.

Throughout the evening and into the night, he personally took charge of the situation. Allowing the use of Starlink for the attack, he concluded, could be a disaster for the world. So he secretly told his engineers to turn off coverage within 100 kilometers of the Crimean coast. As a result, when the Ukrainian drone subs got near the Russian fleet in Sevastopol, they lost connectivity and washed ashore harmlessly.

In Musk's version, he was asked to activate coverage around Crimea to support an "obvious" Ukrainian attack. In Isaacson's version, he deactivated coverage that already existed and didn't tell anyone. The attack went forward and failed.

There's a big difference between the two. Isaacson's version seems most likely to be the true one, which means Musk is lying to make himself look better. I'm surprised that more people haven't commented on this difference in their stories.

UPDATE: Walter Isaacson has changed his story:

To clarify on the Starlink issue: the Ukrainians THOUGHT coverage was enabled all the way to Crimea, but it was not. They asked Musk to enable it for their drone sub attack on the Russian fleet. Musk did not enable it, because he thought, probably correctly, that would cause a…

— Walter Isaacson (@WalterIsaacson) September 9, 2023

This puts a different spin on things. It's one thing to secretly shut down existing coverage, but quite another to do no more than deny a request for additional service. If this is really how things went down, there's little reason to condemn Musk's actions.

Bob Somerby is annoyed that liberals continue to say that Florida's Stop WOKE Act bans any classroom discussion that "could prompt students to feel discomfort" because of their race. In fairness, early drafts of the law did use that word, and it was only dropped in the final bill. Nonetheless, he's right that the bill as passed says nothing about discomfort.

But it does make several pronouncements about what can and can't be taught, and it words them in different ways in different places. Here's the most understandable version:

Instruction and supporting materials on the topics enumerated in this section¹ must be consistent with the following principles of individual freedom:

• No person is inherently racist, sexist, or oppressive, whether consciously or unconsciously, solely by virtue of his or her race or sex.
• No race is inherently superior to another race.
• No person should be discriminated against or receive adverse treatment solely or partly on the basis of race, color, national origin, religion, disability, or sex.
• Meritocracy or traits such as a hard work ethic are not racist but fundamental to the right to pursue happiness and be rewarded for industry.
• A person, by virtue of his or her race or sex, does not bear responsibility for actions committed in the past by other members of the same race or sex.
• A person should not be instructed that he or she must feel guilt, anguish, or other forms of psychological distress for actions, in which he or she played no part, committed in the past by other members of the same race or sex.

The Florida law bans any instruction suggesting that students "must feel guilt, anguish, or other forms of psychological distress" for actions in the historical past, which is not at all the same thing as "prompting discomfort." But it's also not entirely different. The upshot, I suppose, is that a little bit of classroom discomfort over our past might be OK as long as it doesn't edge into instruction that students should feel anguish or distress?

It's also worth noting that this isn't the only law ever passed by the Florida legislature. For example, they also passed a law that bans instruction in both Critical Race Theory and systemic racism. That's why they aren't mentioned here.

¹These include American history; the founding of the republic; the Holocaust; slavery, racism, and African American history; principles of agriculture; kindness to animals; state history; and a few other odds and ends.

Cactus photographed against the sky along Laguna Canyon Road.

Measurements of the COVID virus in wastewater have now quadrupled since their low point in June. But what does that mean?

Wastewater measurements are now about the same as they were at this time in both 2021 and 2022. In 2021 that led to a huge outbreak later in the year. In 2022 it led to nothing.

So there's no telling what will happen this year. I suppose the betting money says that 2023 is more similar to 2022 thanks to high vaccination and immunization rates. But you never know.

Here are the results of a new CNN/SSRN poll about Joe Biden's involvement with Hunter Biden's business dealings. I had to interpolate a few numbers because CNN inexplicably provided only some of them, but they're all pretty close.

It's almost impossible to overstate how completely nuts this is. There is literally zero evidence that Joe Biden was involved in Hunter's business affairs. There is, if anything, less than zero evidence that Joe Biden did anything illegal. And there is about negative infinity evidence that he even remotely influenced the investigation in any way.

Every allegation against Joe Biden has been nothing but a malignant stew of innuendo and insinuation. Mysterious hard drives! Offshore accounts! Diamonds the size of your fist! Phone calls between Joe and his son! Emails sent under pseudonyms! Arguments between DOJ investigators! Getting a Ukrainian prosecutor fired! VP meeting notes sent to Hunter!

Every one of these things is either flatly untrue or, at most, implicates only Hunter, not Joe. I just don't know how much more clearly you can say that the entire Republican jihad against Hunter has produced nothing—absolutely nothing—implicating Joe Biden in any way.

Nothing.

It's just endless smoke intended to convince everyone there must be a fire somewhere. But there isn't. There is nothing.

In case you haven't heard, Tucker Carlson's latest is that in 1999 Barack Obama had a night of coke-fueled sex with a guy named Larry Sinclair, a serial con artist with a very long rap sheet and a slew of aliases. Seriously.

I say "latest," but only because it's new for Tucker. For the rest of us it's just ancient conspiracy theory twaddle. Sinclair wrote a whole book about it back in 2008 and it was as laughable then as it is now. He even dramatically agreed to take a lie detector test—and failed it. In true Trumpian style, he then insisted that one of Obama's aides had bribed the polygraph folks to rig the test.

So why do I care about this? I don't. But it goes to show something unexpected: apparently Fox News really did keep a lid on Tucker while he worked there. It often seemed as if his ranting could hardly get worse, but it's now obvious that it could—and has. No wonder they fired him. He must have been even more of a loose cannon than we imagined.

Via Tyler Cowen, a new study estimates the incarceration rate of immigrant men vs. native born Americans from 1870 to the present day. In the present context of the immigration debate we care mostly about immigrants from south of the border, so here's the chart for that:

Up until 1960 male Latino immigrants had a higher incarceration rate than natives, but since then it's been consistently lower. It's currently about 20% below the overall native rate.

The Wall Street Journal has alarming news:

Health-insurance costs are climbing at the steepest rate in years, walloping businesses and their workers. Costs for employer coverage are expected to surge around 6.5% for 2024, according to major benefits consulting firms Mercer and Willis Towers Watson, which provided their survey results exclusively to The Wall Street Journal.

I have trained you all well enough to know what's coming next, right? Here's the chart from the Journal article, but this time adjusted for inflation:¹

Assuming the Journal's numbers to be correct, the cost of health insurance is projected to grow about 2% in 2024—higher than the past couple of years but not a record or anything close to it. Why do people keep doing this stuff?

Always adjust for inflation. Always adjust for inflation. Always adjust for inflation. Always adjust for inflation. Always adjust for inflation. Always adjust for inflation. Always adjust for inflation. Always adjust for inflation. Always adjust for inflation. Always adjust for inflation. Always adjust for inflation.

¹I used an estimate for year-over-year inflation of 4% from this year to 2024.

This is apropos of nothing aside from running across something that made me wonder if demand for paper was up or down. I expected to see that it was still going up, which would prompt a snarky remark about the paperless office and related things. But no:

Global demand for ordinary office paper (i.e., not packaging or paper towels or pulp) peaked around 2007 and has been trending steadily down ever since. In rich countries demand peaked in 2000 and has declined by half since then. Maybe there's more to this whole paperless office thing than I thought.

You may have heard of the replication crisis. It mostly refers to studies in the social sciences that get a lot of media attention when they're first published but then fail to produce the same results when other researchers try to replicate them. The reason it's become a crisis is that a lot of studies fail to replicate, calling into question the whole enterprise of the social sciences.

But one problem with this is that it's hard to replicate a study. Small things can make a big difference sometimes, and doing a precise replication isn't always easy. A recent paper shows this pretty dramatically. First off, here's what the researchers were trying to replicate:

The study was published in 2021 and investigated an association of glucagon-like peptide 1 receptor agonists (GLP-1RA) and chronic lower respiratory disease (CLRD) exacerbation in a population with type 2 diabetes mellitus (T2D) and CLRD.

This is no dashed-off social science experiment. This is a big and very serious medical study with very specific methods and goals. What's more, the researchers weren't trying to replicate the whole thing. They were only trying to replicate one little part: deciding which patients to include in the study. Here are the criteria:

New users of GLP1-RA add-on therapy aged more than 17 years with at least 1 outpatient or 2 inpatient encounters with T2D and CLRD in the year before the index date with no prior insulin or dipeptidyl peptidase 4 inhibitors exposure and no prior type 1 diabetes mellitus, cystic fibrosis, lung cancer, pulmonary embolism, pulmonary hypertension, conditions requiring chronic systemic corticosteroid therapy within a year or pregnancy at the index date.

Got that? The researchers gathered together nine teams of qualified experts whose only task was to slice and dice a database to come up with a cohort of patients who met the inclusion criteria. The results were dismal. Interpretations of what the criteria meant were all over the map and none of the nine teams came close to matching the cohort from the original study. They couldn't even come close to agreeing on how many people were in the cohort:

The green bars represent patients chosen by both the original study and the team of replicators. The largest overlap is only 35%. The number of people chosen for the study ranges chaotically from 2,000 to 64,000. Every team deviated from the inclusion criteria in the original study in at least four different ways.

So replication is no walk in the park, even in the hard sciences where the procedures are presumably more concrete. But if replication is this hard, what chance do we have of properly replicating anything?