Month: March 2023

Here's the latest from the Wall Street Journal editorial page:

If I were Tim Trevan I'd be pissed. Nowhere in his op-ed does he say that scientists got the lab leak theory wrong. He says only that a lab leak is "a possibility" that's "self-evidently plausible."

Trevan does accuse some unspecified "others" of trying to suppress dissent about the origin of COVID, but that's it. He doesn't say that the scientific consensus about a natural origin of the virus is wrong. He never comes close.

But I guess that was unacceptable to the editors who run things at the Journal editorial page, so they just made up a headline more to their liking. Boo-yah.

NOTE: If you want to check up to make sure I summarized the article fairly, just click the free link here.

A year after Obamacare passed we had reduced the uninsured population from around 17% to 11%. Since then, aside from a few small wobbles during the pandemic, it's stayed almost completely flat.

There's no excuse for this. The correct number is 0%.

I've written many times about the Flint water crisis, and after all the data was in my conclusion was pretty simple:

• The screw-up with Flint's water was a terrible tragedy that never should have happened.
• However, in the end there was little damage done. Lead levels never got all that high and the problem was fixed fairly quickly. There were probably no more than a handful of children who were seriously affected.

To this day, conventional wisdom is just the opposite: namely that lead levels in children skyrocketed and produced a huge spike in special education. One of the scientists who was among the first to sound the alarm over Flint was transformed from hero to villain in a heartbeat when he declined to go along with this.

He's back now with some co-authors to take a retrospective look at what happened. Here's the key chart:

Even at the height of the crisis, testing in children showed blood lead levels that were essentially the same as the Michigan average and far lower than Detroit, which had a safe water supply the entire time. During the whole of the crisis (which encompassed 18 months in 2014-15), the number of Flint children with elevated lead levels was 3.9%. In Detroit it was 8.1%.

Why does this matter? It's simple: continual panic over a nonexistent crisis is bad for residents, who have lived for years with elevated outrage and stress, and bad for their children, who internalize the idea that they're going to grow up stupid.

But if the crisis was never that bad, what's with the spike in special education? That's what the new paper is about:

Did a nocebo effect contribute to the rise in special education enrollment following the Flint, Michigan water crisis?

A nocebo effect is when someone experiences a negative effect merely because they expect it, not because there's any concrete underlying cause. The authors conclude that this is indeed what happened in Flint:

Despite an equal number of overall special education outcomes worsening and improving, only those that superficially appeared to be worsening were publicized in the media. Our detailed analysis shows these outcomes are insignificant or inconsistent with the actual lead exposure that occurred. Specifically, the seven-fold jump in suspension/expulsion rates of special education students had occurred in 2013-14 before the onset of the crisis.

....A nocebo effect is consistent with the trend of rising special education enrollment after the [Flint water crisis] was exposed. As a top news story of 2016, the crisis engendered negative psychological effects described by residents as “Flint fatigue,” and the surrounding international media coverage has continued for over five years with negative headlines. The news reports and their popularity on social media and negative perceptions of Flint community leaders and parents could have heightened negative expectations about the effects on children, who readily accept and act on information from those they trust.

This is a tragedy on top of a tragedy. The damage done to children from lead was small, but the damage done from fear of lead may very well be many times higher. Partly this is because the progressive media persistently made heroes of the residents and refused to follow the science, and partly because Flint residents had every reason to distrust politicians and various hired guns who tried to belittle the crisis from the start.

With only a few exceptions, there's nothing wrong with Flint's kids. They should not have to grow up thinking there is.

In the Washington Post today, Matt Bai writes about the Sierra Club's new style guide. It has the usual woke prescriptions ("stand with us" is offensive to the disabled, for example), which I understand is a topic of considerable interest these days.

Therefore I'm going to ignore it and instead focus on a trivial pet peeve of mind. Here is Bai explaining "embodied cognition":

I’m not exactly Bill Nye the Science Guy (having failed high school physics), but the basic idea is . . .

You'll have to click the link if you want to know what embodied cognition is. I, however, want to urge writers (and others) to stop bragging about how bad you are at math or science. It's hard to imagine someone like Bai explaining, say, the objective correlative, but interjecting that he might get it wrong since he failed high school English.

Now, fine, I know you're thinking that when I said this was trivial I was understating things. But it's still annoying. My complaint is nothing new (google C.P. Snow for more), and I imagine this kind of humblebragging is less common than it used to be, but we should get rid of it entirely. Being innumerate is not something you should be able to sheepishly admit in order to build camaraderie. It's a shortcoming—and if you're a journalist it's one you should probably try to address.

That ends today's super earnest griping. And my apologies to Matt, who just happened to catch my eye right after a bone marrow biopsy.

This isn't really a health update per se, but I did just return from a bone marrow biopsy in preparation for my CAR-T treatment next month. This procedure draws bone marrow from a hip bone and is short but painful.

This time, however, it was very, very painful. Apparently my bone marrow was too soft and the doctor had a hard time extracting it. Plus the outside of my bone was like "eggshell." After a bit of (literal) screaming and thrashing that I couldn't help, my doctor gave up and decided to try another location. This one produced slightly less screaming and thrashing and was successful.

Apparently this was because the bone in the second location was like "porcelain," which I guess is a step up from eggshell. I don't quite know why my bones are in such fragile shape, since I had three years of treatment with a bone medication and then chemotherapy that kept the level of multiple myeloma low. I gather that it's yet another thing related to long-term use of the Evil Dex™.

But how bad can my bones be? Nobody has told me to restrict my activities in any way, so they must basically be OK. Right?

Anyway, now that it's all done it's only slightly sore. Until the lidocaine wears off, that is.

Poppy season has begun. It's not widespread yet, but there are a few patches of California poppies here and there. In a couple of weeks we're likely to have a superbloom, but due to past transgressions The Man has closed off all access this year:

Riverside County Sheriff Chad Bianco said law enforcement will take a zero-tolerance policy for people who are caught trespassing in the area. “Your warning is right now,” Bianco said. “If you are going to come here and you are going to park your car, you are subject to citation and possibly the towing of your vehicle.”

That's not very friendly, is it?

The Wall Street Journal is skeptical about economic doomsaying:

I don't know about that. I'd put it at about five months right now. My read of the evidence is that the economy is in OK but fragile shape, and by summer a whole bunch of things are likely to coalesce. This will be triggered by the Fed's interest rate hikes, which will finally start to seriously bite, followed by declines in consumer spending thanks to exhausted stimulus savings and three years of flat wages. The housing slowdown will get worse and investment levels will crater. All of this will cause GDP to flatten and fall, followed by a significant rise in the unemployment rate.

But I could be wrong! I hope so.

I hauled out the telescope a couple of days ago for some testing after I added a new part, and as long as it was all set up I decided to take some pictures. As usual in my backyard, I had very few choices and I eventually picked M106, a Type 2 Seyfert galaxy with a massive black hole at its core. Its claim to fame is that it contains a water vapor megamaser that enabled the first direct measurement of the distance to a galaxy.

I didn't realize this while I was creating the image, because they were too faint, but the final stacked image clearly shows two other galaxies. The one just below M106 is NGC 4248; the larger one at bottom left is NGC 4217. The final image was constructed from a stack of 300 subs of 60 seconds each, taken between 9 pm and 3 am.

In case you're interested, a lovely professional photo of M106 taken with the 4-meter telescope at Kitt Peak is below my backyard picture.

Emmanuel Macron wants to raise the retirement age in France from 62 to 64, and the French people are not happy about it. Rokhaya Diallo explains why:

If it is adopted, Macron’s reform would disproportionately hurt more vulnerable populations: low-income workers, women and citizens from the French overseas territories, who are overwhelmingly people of color. This is what has outraged the French public.

Uh huh. Sure it is.

What. The. Fuck.

This is from a paper written in December by Yu Takagi and Shinji Nishimoto of the Graduate School of Frontier Biosciences in Osaka. Subjects were put in an MRI machine and shown a series of pictures. The fMRI data was collected, and then software using an algorithm called Stable Diffusion attempted to reconstruct the image. It did pretty well!

The algorithm uses data from both image and text signals in the fMRI data. Here's a sample of what the reconstructions look like using just the image or just the text data:

I gather from reading the paper that this is not a big deal. Researchers have been working on this stuff for years, and the particular innovation of the Stable Diffusion algorithm is that it works "without the need for any additional training and fine-tuning of complex deep-learning models."

But I've never seen anything quite like this. I've seen vague blobs from fMRI reconstructions, but that's about it. I guess I haven't been keeping up.

This has possibilities that it's hard to even list. But go ahead and give it your best shot in comments. As always, the question is not so much what this technology can do right now, but what something already this good can do in a few years.